summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
-rw-r--r--Documentation/kernel-parameters.txt11
-rw-r--r--Documentation/power/swsusp.txt51
-rw-r--r--Documentation/power/userland-swsusp.txt149
-rw-r--r--Documentation/power/video.txt74
-rw-r--r--arch/cris/kernel/irq.c10
-rw-r--r--arch/frv/kernel/irq.c10
-rw-r--r--arch/i386/Kconfig24
-rw-r--r--arch/i386/Kconfig.debug9
-rw-r--r--arch/i386/kernel/Makefile2
-rw-r--r--arch/i386/kernel/alternative.c321
-rw-r--r--arch/i386/kernel/apic.c1
-rw-r--r--arch/i386/kernel/cpu/centaur.c1
-rw-r--r--arch/i386/kernel/cpu/common.c47
-rw-r--r--arch/i386/kernel/cpu/cpufreq/powernow-k8.c4
-rw-r--r--arch/i386/kernel/cpu/intel.c12
-rw-r--r--arch/i386/kernel/cpu/intel_cacheinfo.c4
-rw-r--r--arch/i386/kernel/cpu/proc.c2
-rw-r--r--arch/i386/kernel/crash.c2
-rw-r--r--arch/i386/kernel/entry.S4
-rw-r--r--arch/i386/kernel/head.S5
-rw-r--r--arch/i386/kernel/io_apic.c25
-rw-r--r--arch/i386/kernel/kprobes.c4
-rw-r--r--arch/i386/kernel/module.c32
-rw-r--r--arch/i386/kernel/mpparse.c7
-rw-r--r--arch/i386/kernel/nmi.c6
-rw-r--r--arch/i386/kernel/process.c2
-rw-r--r--arch/i386/kernel/ptrace.c4
-rw-r--r--arch/i386/kernel/semaphore.c8
-rw-r--r--arch/i386/kernel/setup.c118
-rw-r--r--arch/i386/kernel/signal.c7
-rw-r--r--arch/i386/kernel/smpboot.c3
-rw-r--r--arch/i386/kernel/topology.c9
-rw-r--r--arch/i386/kernel/traps.c57
-rw-r--r--arch/i386/kernel/vmlinux.lds.S20
-rw-r--r--arch/i386/kernel/vsyscall-sysenter.S3
-rw-r--r--arch/i386/mach-es7000/es7000.h5
-rw-r--r--arch/i386/mach-es7000/es7000plat.c6
-rw-r--r--arch/i386/mm/fault.c210
-rw-r--r--arch/i386/mm/init.c45
-rw-r--r--arch/i386/oprofile/nmi_int.c7
-rw-r--r--arch/ia64/hp/sim/simserial.c7
-rw-r--r--arch/m32r/kernel/irq.c10
-rw-r--r--arch/m68k/bvme6000/rtc.c4
-rw-r--r--arch/mips/kernel/irq.c10
-rw-r--r--arch/mips/kernel/smp.c4
-rw-r--r--arch/mips/sgi-ip27/ip27-irq.c5
-rw-r--r--arch/parisc/kernel/smp.c25
-rw-r--r--arch/powerpc/kernel/irq.c5
-rw-r--r--arch/powerpc/kernel/kprobes.c4
-rw-r--r--arch/powerpc/kernel/setup-common.c5
-rw-r--r--arch/powerpc/kernel/setup_32.c5
-rw-r--r--arch/powerpc/platforms/powermac/smp.c4
-rw-r--r--arch/ppc/kernel/setup.c10
-rw-r--r--arch/s390/kernel/smp.c4
-rw-r--r--arch/sh/kernel/irq.c5
-rw-r--r--arch/sh/kernel/setup.c5
-rw-r--r--arch/sh64/kernel/irq.c5
-rw-r--r--arch/sparc/kernel/irq.c5
-rw-r--r--arch/sparc/kernel/smp.c24
-rw-r--r--arch/sparc/kernel/sun4d_irq.c8
-rw-r--r--arch/sparc/kernel/sun4d_smp.c8
-rw-r--r--arch/sparc/kernel/sun4m_smp.c6
-rw-r--r--arch/sparc64/kernel/irq.c4
-rw-r--r--arch/sparc64/kernel/smp.c30
-rw-r--r--arch/sparc64/mm/init.c4
-rw-r--r--arch/um/kernel/um_arch.c12
-rw-r--r--arch/x86_64/kernel/early_printk.c26
-rw-r--r--arch/x86_64/kernel/irq.c21
-rw-r--r--arch/x86_64/kernel/kprobes.c4
-rw-r--r--arch/x86_64/kernel/nmi.c4
-rw-r--r--arch/x86_64/kernel/signal.c4
-rw-r--r--arch/xtensa/kernel/irq.c15
-rw-r--r--arch/xtensa/platform-iss/console.c4
-rw-r--r--block/ioctl.c22
-rw-r--r--drivers/base/power/suspend.c5
-rw-r--r--drivers/block/cciss.c2
-rw-r--r--drivers/block/floppy.c17
-rw-r--r--drivers/block/loop.c18
-rw-r--r--drivers/block/nbd.c16
-rw-r--r--drivers/block/pktcdvd.c27
-rw-r--r--drivers/block/rd.c4
-rw-r--r--drivers/cdrom/cdrom.c874
-rw-r--r--drivers/cdrom/cdu31a.c8
-rw-r--r--drivers/cdrom/cm206.c44
-rw-r--r--drivers/cdrom/sbpcd.c710
-rw-r--r--drivers/cdrom/viocd.c2
-rw-r--r--drivers/char/amiserial.c18
-rw-r--r--drivers/char/generic_serial.c14
-rw-r--r--drivers/char/istallion.c1
-rw-r--r--drivers/char/n_tty.c10
-rw-r--r--drivers/char/nwflash.c11
-rw-r--r--drivers/char/raw.c23
-rw-r--r--drivers/char/ser_a2232.c4
-rw-r--r--drivers/char/snsc.c8
-rw-r--r--drivers/char/snsc_event.c5
-rw-r--r--drivers/char/stallion.c1
-rw-r--r--drivers/char/sx.c2
-rw-r--r--drivers/char/tty_io.c50
-rw-r--r--drivers/char/vme_scc.c2
-rw-r--r--drivers/char/vt.c22
-rw-r--r--drivers/char/watchdog/pcwd_usb.c7
-rw-r--r--drivers/connector/connector.c15
-rw-r--r--drivers/firmware/dcdbas.c23
-rw-r--r--drivers/ide/ide-cd.c110
-rw-r--r--drivers/ide/ide-disk.c11
-rw-r--r--drivers/ide/ide-floppy.c11
-rw-r--r--drivers/ide/ide-tape.c19
-rw-r--r--drivers/isdn/capi/kcapi.c17
-rw-r--r--drivers/isdn/hisax/config.c1
-rw-r--r--drivers/isdn/hisax/elsa.c1
-rw-r--r--drivers/net/loopback.c4
-rw-r--r--drivers/net/ppp_generic.c25
-rw-r--r--drivers/oprofile/cpu_buffer.c3
-rw-r--r--drivers/pnp/pnpbios/rsparser.c6
-rw-r--r--drivers/s390/block/dasd_ioctl.c8
-rw-r--r--drivers/scsi/ide-scsi.c11
-rw-r--r--drivers/scsi/sr.c37
-rw-r--r--drivers/scsi/sr.h1
-rw-r--r--drivers/scsi/sr_ioctl.c19
-rw-r--r--drivers/serial/68328serial.c9
-rw-r--r--drivers/serial/au1x00_uart.c11
-rw-r--r--drivers/serial/crisv10.c68
-rw-r--r--drivers/serial/m32r_sio.c15
-rw-r--r--drivers/serial/sunsu.c13
-rw-r--r--drivers/tc/zs.c9
-rw-r--r--fs/9p/mux.c11
-rw-r--r--fs/adfs/file.c4
-rw-r--r--fs/autofs4/autofs_i.h3
-rw-r--r--fs/autofs4/inode.c2
-rw-r--r--fs/autofs4/waitq.c16
-rw-r--r--fs/bio.c8
-rw-r--r--fs/block_dev.c28
-rw-r--r--fs/buffer.c6
-rw-r--r--fs/cifs/dir.c8
-rw-r--r--fs/cifs/fcntl.c4
-rw-r--r--fs/cifs/file.c4
-rw-r--r--fs/cifs/inode.c16
-rw-r--r--fs/cifs/link.c16
-rw-r--r--fs/cifs/readdir.c4
-rw-r--r--fs/cifs/xattr.c16
-rw-r--r--fs/devpts/inode.c76
-rw-r--r--fs/dquot.c167
-rw-r--r--fs/eventpoll.c32
-rw-r--r--fs/ext2/namei.c54
-rw-r--r--fs/ext3/dir.c52
-rw-r--r--fs/ext3/file.c4
-rw-r--r--fs/ext3/inode.c16
-rw-r--r--fs/ext3/ioctl.c4
-rw-r--r--fs/ext3/super.c6
-rw-r--r--fs/fat/fatent.c6
-rw-r--r--fs/fcntl.c9
-rw-r--r--fs/file.c34
-rw-r--r--fs/file_table.c10
-rw-r--r--fs/hpfs/hpfs_fn.h5
-rw-r--r--fs/hpfs/inode.c10
-rw-r--r--fs/hpfs/namei.c60
-rw-r--r--fs/hpfs/super.c4
-rw-r--r--fs/inode.c18
-rw-r--r--fs/inotify.c116
-rw-r--r--fs/jbd/checkpoint.c4
-rw-r--r--fs/jbd/journal.c4
-rw-r--r--fs/jbd/transaction.c4
-rw-r--r--fs/jffs/inode-v23.c86
-rw-r--r--fs/jffs/intrep.c6
-rw-r--r--fs/jffs/jffs_fm.c2
-rw-r--r--fs/jffs/jffs_fm.h5
-rw-r--r--fs/libfs.c14
-rw-r--r--fs/minix/namei.c48
-rw-r--r--fs/namei.c12
-rw-r--r--fs/ncpfs/file.c4
-rw-r--r--fs/ncpfs/inode.c6
-rw-r--r--fs/ncpfs/ncplib_kernel.c4
-rw-r--r--fs/ncpfs/sock.c34
-rw-r--r--fs/open.c8
-rw-r--r--fs/proc/proc_misc.c2
-rw-r--r--fs/qnx4/file.c3
-rw-r--r--fs/quota.c6
-rw-r--r--fs/quota_v2.c2
-rw-r--r--fs/ramfs/file-mmu.c11
-rw-r--r--fs/seq_file.c10
-rw-r--r--fs/super.c10
-rw-r--r--fs/sysv/namei.c48
-rw-r--r--fs/udf/balloc.c36
-rw-r--r--fs/udf/ialloc.c8
-rw-r--r--fs/udf/super.c2
-rw-r--r--fs/ufs/file.c10
-rw-r--r--fs/ufs/namei.c48
-rw-r--r--fs/xfs/Makefile-linux-2.640
-rw-r--r--fs/xfs/linux-2.6/kmem.h91
-rw-r--r--fs/xfs/linux-2.6/xfs_aops.c484
-rw-r--r--fs/xfs/linux-2.6/xfs_aops.h4
-rw-r--r--fs/xfs/linux-2.6/xfs_buf.c7
-rw-r--r--fs/xfs/linux-2.6/xfs_export.c37
-rw-r--r--fs/xfs/linux-2.6/xfs_file.c187
-rw-r--r--fs/xfs/linux-2.6/xfs_fs_subr.c6
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl.c138
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl32.c12
-rw-r--r--fs/xfs/linux-2.6/xfs_ioctl32.h4
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.c317
-rw-r--r--fs/xfs/linux-2.6/xfs_iops.h12
-rw-r--r--fs/xfs/linux-2.6/xfs_linux.h10
-rw-r--r--fs/xfs/linux-2.6/xfs_lrw.c51
-rw-r--r--fs/xfs/linux-2.6/xfs_stats.c7
-rw-r--r--fs/xfs/linux-2.6/xfs_super.c213
-rw-r--r--fs/xfs/linux-2.6/xfs_super.h7
-rw-r--r--fs/xfs/linux-2.6/xfs_sysctl.c3
-rw-r--r--fs/xfs/linux-2.6/xfs_vfs.c19
-rw-r--r--fs/xfs/linux-2.6/xfs_vfs.h3
-rw-r--r--fs/xfs/linux-2.6/xfs_vnode.c35
-rw-r--r--fs/xfs/linux-2.6/xfs_vnode.h33
-rw-r--r--fs/xfs/quota/xfs_dquot_item.c2
-rw-r--r--fs/xfs/quota/xfs_qm.c13
-rw-r--r--fs/xfs/quota/xfs_qm_bhv.c4
-rw-r--r--fs/xfs/support/ktrace.c4
-rw-r--r--fs/xfs/support/uuid.c15
-rw-r--r--fs/xfs/xfs_acl.h4
-rw-r--r--fs/xfs/xfs_attr.c59
-rw-r--r--fs/xfs/xfs_attr_leaf.c729
-rw-r--r--fs/xfs/xfs_attr_leaf.h47
-rw-r--r--fs/xfs/xfs_attr_sf.h8
-rw-r--r--fs/xfs/xfs_bmap.c1305
-rw-r--r--fs/xfs/xfs_bmap.h22
-rw-r--r--fs/xfs/xfs_bmap_btree.c10
-rw-r--r--fs/xfs/xfs_bmap_btree.h8
-rw-r--r--fs/xfs/xfs_clnt.h2
-rw-r--r--fs/xfs/xfs_da_btree.c409
-rw-r--r--fs/xfs/xfs_da_btree.h16
-rw-r--r--fs/xfs/xfs_dfrag.c4
-rw-r--r--fs/xfs/xfs_dir.c32
-rw-r--r--fs/xfs/xfs_dir2.h27
-rw-r--r--fs/xfs/xfs_dir2_block.c193
-rw-r--r--fs/xfs/xfs_dir2_block.h7
-rw-r--r--fs/xfs/xfs_dir2_data.c240
-rw-r--r--fs/xfs/xfs_dir2_data.h26
-rw-r--r--fs/xfs/xfs_dir2_leaf.c285
-rw-r--r--fs/xfs/xfs_dir2_leaf.h15
-rw-r--r--fs/xfs/xfs_dir2_node.c303
-rw-r--r--fs/xfs/xfs_dir2_node.h10
-rw-r--r--fs/xfs/xfs_dir2_sf.c8
-rw-r--r--fs/xfs/xfs_dir_leaf.c82
-rw-r--r--fs/xfs/xfs_dir_sf.h24
-rw-r--r--fs/xfs/xfs_dmapi.h10
-rw-r--r--fs/xfs/xfs_fsops.c1
-rw-r--r--fs/xfs/xfs_ialloc.c13
-rw-r--r--fs/xfs/xfs_iget.c6
-rw-r--r--fs/xfs/xfs_inode.c1303
-rw-r--r--fs/xfs/xfs_inode.h78
-rw-r--r--fs/xfs/xfs_iomap.c2
-rw-r--r--fs/xfs/xfs_itable.c5
-rw-r--r--fs/xfs/xfs_log_recover.c2
-rw-r--r--fs/xfs/xfs_mount.c646
-rw-r--r--fs/xfs/xfs_mount.h44
-rw-r--r--fs/xfs/xfs_rw.h1
-rw-r--r--fs/xfs/xfs_trans.c187
-rw-r--r--fs/xfs/xfs_trans.h2
-rw-r--r--fs/xfs/xfs_vfsops.c82
-rw-r--r--fs/xfs/xfs_vnodeops.c15
-rw-r--r--include/asm-alpha/mmu_context.h5
-rw-r--r--include/asm-alpha/topology.h4
-rw-r--r--include/asm-generic/bug.h4
-rw-r--r--include/asm-generic/percpu.h7
-rw-r--r--include/asm-i386/alternative.h129
-rw-r--r--include/asm-i386/arch_hooks.h3
-rw-r--r--include/asm-i386/atomic.h36
-rw-r--r--include/asm-i386/bitops.h7
-rw-r--r--include/asm-i386/cache.h2
-rw-r--r--include/asm-i386/cpufeature.h1
-rw-r--r--include/asm-i386/mach-default/do_timer.h2
-rw-r--r--include/asm-i386/mach-es7000/mach_mpparse.h10
-rw-r--r--include/asm-i386/mach-visws/do_timer.h2
-rw-r--r--include/asm-i386/mach-voyager/do_timer.h2
-rw-r--r--include/asm-i386/mpspec.h1
-rw-r--r--include/asm-i386/mtrr.h1
-rw-r--r--include/asm-i386/mutex.h6
-rw-r--r--include/asm-i386/pgtable-2level.h2
-rw-r--r--include/asm-i386/pgtable-3level.h2
-rw-r--r--include/asm-i386/rwlock.h56
-rw-r--r--include/asm-i386/semaphore.h8
-rw-r--r--include/asm-i386/spinlock.h34
-rw-r--r--include/asm-i386/system.h62
-rw-r--r--include/asm-i386/uaccess.h12
-rw-r--r--include/asm-i386/unistd.h36
-rw-r--r--include/asm-ia64/atomic.h8
-rw-r--r--include/asm-ia64/cache.h2
-rw-r--r--include/asm-m68k/atomic.h8
-rw-r--r--include/asm-parisc/cache.h2
-rw-r--r--include/asm-powerpc/percpu.h7
-rw-r--r--include/asm-s390/atomic.h18
-rw-r--r--include/asm-s390/percpu.h7
-rw-r--r--include/asm-sparc64/atomic.h10
-rw-r--r--include/asm-sparc64/cache.h2
-rw-r--r--include/asm-sparc64/percpu.h7
-rw-r--r--include/asm-um/alternative.h6
-rw-r--r--include/asm-x86_64/atomic.h8
-rw-r--r--include/asm-x86_64/cache.h2
-rw-r--r--include/asm-x86_64/percpu.h7
-rw-r--r--include/linux/cache.h4
-rw-r--r--include/linux/cdrom.h5
-rw-r--r--include/linux/eventpoll.h8
-rw-r--r--include/linux/ext3_fs.h9
-rw-r--r--include/linux/ext3_fs_i.h7
-rw-r--r--include/linux/file.h28
-rw-r--r--include/linux/fs.h22
-rw-r--r--include/linux/generic_serial.h4
-rw-r--r--include/linux/genhd.h14
-rw-r--r--include/linux/init_task.h10
-rw-r--r--include/linux/jbd.h7
-rw-r--r--include/linux/kernel.h3
-rw-r--r--include/linux/kprobes.h3
-rw-r--r--include/linux/loop.h3
-rw-r--r--include/linux/msdos_fs.h3
-rw-r--r--include/linux/nbd.h3
-rw-r--r--include/linux/ncp_fs_i.h2
-rw-r--r--include/linux/ncp_fs_sb.h5
-rw-r--r--include/linux/pm.h3
-rw-r--r--include/linux/profile.h2
-rw-r--r--include/linux/quota.h7
-rw-r--r--include/linux/raid/raid1.h2
-rw-r--r--include/linux/rcupdate.h2
-rw-r--r--include/linux/seq_file.h4
-rw-r--r--include/linux/swap.h5
-rw-r--r--include/linux/tty.h8
-rw-r--r--include/linux/tty_flip.h12
-rw-r--r--include/linux/udf_fs_sb.h4
-rw-r--r--include/linux/vt_kern.h5
-rw-r--r--init/do_mounts_initrd.c1
-rw-r--r--init/main.c26
-rw-r--r--kernel/cpuset.c212
-rw-r--r--kernel/exit.c4
-rw-r--r--kernel/fork.c8
-rw-r--r--kernel/kprobes.c14
-rw-r--r--kernel/kthread.c7
-rw-r--r--kernel/module.c53
-rw-r--r--kernel/panic.c97
-rw-r--r--kernel/posix-timers.c1
-rw-r--r--kernel/power/Makefile2
-rw-r--r--kernel/power/disk.c20
-rw-r--r--kernel/power/main.c2
-rw-r--r--kernel/power/pm.c21
-rw-r--r--kernel/power/power.h75
-rw-r--r--kernel/power/process.c61
-rw-r--r--kernel/power/snapshot.c335
-rw-r--r--kernel/power/swap.c544
-rw-r--r--kernel/power/swsusp.c887
-rw-r--r--kernel/power/user.c333
-rw-r--r--kernel/profile.c11
-rw-r--r--kernel/rcupdate.c14
-rw-r--r--kernel/sched.c13
-rw-r--r--kernel/signal.c11
-rw-r--r--kernel/spinlock.c9
-rw-r--r--kernel/sys.c46
-rw-r--r--lib/reed_solomon/reed_solomon.c11
-rw-r--r--mm/readahead.c1
-rw-r--r--mm/swapfile.c57
-rw-r--r--security/seclvl.c210
-rw-r--r--sound/oss/ac97_codec.c24
-rw-r--r--sound/oss/aci.c11
-rw-r--r--sound/oss/ad1889.c7
-rw-r--r--sound/oss/ad1889.h2
-rw-r--r--sound/oss/ali5455.c8
-rw-r--r--sound/oss/au1000.c44
-rw-r--r--sound/oss/au1550_ac97.c44
-rw-r--r--sound/oss/btaudio.c36
-rw-r--r--sound/oss/cmpci.c20
-rw-r--r--sound/oss/cs4281/cs4281m.c54
-rw-r--r--sound/oss/cs46xx.c75
-rw-r--r--sound/oss/dmasound/dmasound_awacs.c10
-rw-r--r--sound/oss/emu10k1/hwaccess.h2
-rw-r--r--sound/oss/emu10k1/main.c2
-rw-r--r--sound/oss/emu10k1/midi.c14
-rw-r--r--sound/oss/es1370.c71
-rw-r--r--sound/oss/es1371.c71
-rw-r--r--sound/oss/esssolo1.c50
-rw-r--r--sound/oss/forte.c11
-rw-r--r--sound/oss/hal2.c22
-rw-r--r--sound/oss/i810_audio.c8
-rw-r--r--sound/oss/ite8172.c20
-rw-r--r--sound/oss/maestro.c26
-rw-r--r--sound/oss/maestro3.c20
-rw-r--r--sound/oss/nec_vrc5477.c20
-rw-r--r--sound/oss/rme96xx.c17
-rw-r--r--sound/oss/sonicvibes.c48
-rw-r--r--sound/oss/swarm_cs4297a.c39
-rw-r--r--sound/oss/trident.c62
-rw-r--r--sound/oss/via82cxxx_audio.c49
-rw-r--r--sound/oss/vwsnd.c61
-rw-r--r--sound/oss/ymfpci.c14
-rw-r--r--sound/oss/ymfpci.h3
388 files changed, 10686 insertions, 7374 deletions
diff --git a/Documentation/kernel-parameters.txt b/Documentation/kernel-parameters.txt
index fc99075e0af4..7b7382d0f758 100644
--- a/Documentation/kernel-parameters.txt
+++ b/Documentation/kernel-parameters.txt
@@ -1008,7 +1008,9 @@ running once the system is up.
noexec=on: enable non-executable mappings (default)
noexec=off: disable nn-executable mappings
- nofxsr [BUGS=IA-32]
+ nofxsr [BUGS=IA-32] Disables x86 floating point extended
+ register save and restore. The kernel will only save
+ legacy floating-point registers on task switch.
nohlt [BUGS=ARM]
@@ -1053,6 +1055,8 @@ running once the system is up.
nosbagart [IA-64]
+ nosep [BUGS=IA-32] Disables x86 SYSENTER/SYSEXIT support.
+
nosmp [SMP] Tells an SMP kernel to act as a UP kernel.
nosync [HW,M68K] Disables sync negotiation for all devices.
@@ -1122,6 +1126,11 @@ running once the system is up.
pas16= [HW,SCSI]
See header of drivers/scsi/pas16.c.
+ pause_on_oops=
+ Halt all CPUs after the first oops has been printed for
+ the specified number of seconds. This is to be used if
+ your oopses keep scrolling off the screen.
+
pcbit= [HW,ISDN]
pcd. [PARIDE]
diff --git a/Documentation/power/swsusp.txt b/Documentation/power/swsusp.txt
index b28b7f04abb8..d7814a113ee1 100644
--- a/Documentation/power/swsusp.txt
+++ b/Documentation/power/swsusp.txt
@@ -17,6 +17,11 @@ Some warnings, first.
* but it will probably only crash.
*
* (*) suspend/resume support is needed to make it safe.
+ *
+ * If you have any filesystems on USB devices mounted before suspend,
+ * they won't be accessible after resume and you may lose data, as though
+ * you have unplugged the USB devices with mounted filesystems on them
+ * (see the FAQ below for details).
You need to append resume=/dev/your_swap_partition to kernel command
line. Then you suspend by
@@ -27,19 +32,18 @@ echo shutdown > /sys/power/disk; echo disk > /sys/power/state
echo platform > /sys/power/disk; echo disk > /sys/power/state
+. If you have SATA disks, you'll need recent kernels with SATA suspend
+support. For suspend and resume to work, make sure your disk drivers
+are built into kernel -- not modules. [There's way to make
+suspend/resume with modular disk drivers, see FAQ, but you probably
+should not do that.]
+
If you want to limit the suspend image size to N bytes, do
echo N > /sys/power/image_size
before suspend (it is limited to 500 MB by default).
-Encrypted suspend image:
-------------------------
-If you want to store your suspend image encrypted with a temporary
-key to prevent data gathering after resume you must compile
-crypto and the aes algorithm into the kernel - modules won't work
-as they cannot be loaded at resume time.
-
Article about goals and implementation of Software Suspend for Linux
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
@@ -333,4 +337,37 @@ init=/bin/bash, then swapon and starting suspend sequence manually
usually does the trick. Then it is good idea to try with latest
vanilla kernel.
+Q: How can distributions ship a swsusp-supporting kernel with modular
+disk drivers (especially SATA)?
+
+A: Well, it can be done, load the drivers, then do echo into
+/sys/power/disk/resume file from initrd. Be sure not to mount
+anything, not even read-only mount, or you are going to lose your
+data.
+
+Q: How do I make suspend more verbose?
+
+A: If you want to see any non-error kernel messages on the virtual
+terminal the kernel switches to during suspend, you have to set the
+kernel console loglevel to at least 5, for example by doing
+
+ echo 5 > /proc/sys/kernel/printk
+
+Q: Is this true that if I have a mounted filesystem on a USB device and
+I suspend to disk, I can lose data unless the filesystem has been mounted
+with "sync"?
+
+A: That's right. It depends on your hardware, and it could be true even for
+suspend-to-RAM. In fact, even with "-o sync" you can lose data if your
+programs have information in buffers they haven't written out to disk.
+
+If you're lucky, your hardware will support low-power modes for USB
+controllers while the system is asleep. Lots of hardware doesn't,
+however. Shutting off the power to a USB controller is equivalent to
+unplugging all the attached devices.
+
+Remember that it's always a bad idea to unplug a disk drive containing a
+mounted filesystem. With USB that's true even when your system is asleep!
+The safest thing is to unmount all USB-based filesystems before suspending
+and remount them after resuming.
diff --git a/Documentation/power/userland-swsusp.txt b/Documentation/power/userland-swsusp.txt
new file mode 100644
index 000000000000..94058220aaf0
--- /dev/null
+++ b/Documentation/power/userland-swsusp.txt
@@ -0,0 +1,149 @@
+Documentation for userland software suspend interface
+ (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
+
+First, the warnings at the beginning of swsusp.txt still apply.
+
+Second, you should read the FAQ in swsusp.txt _now_ if you have not
+done it already.
+
+Now, to use the userland interface for software suspend you need special
+utilities that will read/write the system memory snapshot from/to the
+kernel. Such utilities are available, for example, from
+<http://www.sisk.pl/kernel/utilities/suspend>. You may want to have
+a look at them if you are going to develop your own suspend/resume
+utilities.
+
+The interface consists of a character device providing the open(),
+release(), read(), and write() operations as well as several ioctl()
+commands defined in kernel/power/power.h. The major and minor
+numbers of the device are, respectively, 10 and 231, and they can
+be read from /sys/class/misc/snapshot/dev.
+
+The device can be open either for reading or for writing. If open for
+reading, it is considered to be in the suspend mode. Otherwise it is
+assumed to be in the resume mode. The device cannot be open for reading
+and writing. It is also impossible to have the device open more than once
+at a time.
+
+The ioctl() commands recognized by the device are:
+
+SNAPSHOT_FREEZE - freeze user space processes (the current process is
+ not frozen); this is required for SNAPSHOT_ATOMIC_SNAPSHOT
+ and SNAPSHOT_ATOMIC_RESTORE to succeed
+
+SNAPSHOT_UNFREEZE - thaw user space processes frozen by SNAPSHOT_FREEZE
+
+SNAPSHOT_ATOMIC_SNAPSHOT - create a snapshot of the system memory; the
+ last argument of ioctl() should be a pointer to an int variable,
+ the value of which will indicate whether the call returned after
+ creating the snapshot (1) or after restoring the system memory state
+ from it (0) (after resume the system finds itself finishing the
+ SNAPSHOT_ATOMIC_SNAPSHOT ioctl() again); after the snapshot
+ has been created the read() operation can be used to transfer
+ it out of the kernel
+
+SNAPSHOT_ATOMIC_RESTORE - restore the system memory state from the
+ uploaded snapshot image; before calling it you should transfer
+ the system memory snapshot back to the kernel using the write()
+ operation; this call will not succeed if the snapshot
+ image is not available to the kernel
+
+SNAPSHOT_FREE - free memory allocated for the snapshot image
+
+SNAPSHOT_SET_IMAGE_SIZE - set the preferred maximum size of the image
+ (the kernel will do its best to ensure the image size will not exceed
+ this number, but if it turns out to be impossible, the kernel will
+ create the smallest image possible)
+
+SNAPSHOT_AVAIL_SWAP - return the amount of available swap in bytes (the last
+ argument should be a pointer to an unsigned int variable that will
+ contain the result if the call is successful).
+
+SNAPSHOT_GET_SWAP_PAGE - allocate a swap page from the resume partition
+ (the last argument should be a pointer to a loff_t variable that
+ will contain the swap page offset if the call is successful)
+
+SNAPSHOT_FREE_SWAP_PAGES - free all swap pages allocated with
+ SNAPSHOT_GET_SWAP_PAGE
+
+SNAPSHOT_SET_SWAP_FILE - set the resume partition (the last ioctl() argument
+ should specify the device's major and minor numbers in the old
+ two-byte format, as returned by the stat() function in the .st_rdev
+ member of the stat structure); it is recommended to always use this
+ call, because the code to set the resume partition could be removed from
+ future kernels
+
+The device's read() operation can be used to transfer the snapshot image from
+the kernel. It has the following limitations:
+- you cannot read() more than one virtual memory page at a time
+- read()s accross page boundaries are impossible (ie. if ypu read() 1/2 of
+ a page in the previous call, you will only be able to read()
+ _at_ _most_ 1/2 of the page in the next call)
+
+The device's write() operation is used for uploading the system memory snapshot
+into the kernel. It has the same limitations as the read() operation.
+
+The release() operation frees all memory allocated for the snapshot image
+and all swap pages allocated with SNAPSHOT_GET_SWAP_PAGE (if any).
+Thus it is not necessary to use either SNAPSHOT_FREE or
+SNAPSHOT_FREE_SWAP_PAGES before closing the device (in fact it will also
+unfreeze user space processes frozen by SNAPSHOT_UNFREEZE if they are
+still frozen when the device is being closed).
+
+Currently it is assumed that the userland utilities reading/writing the
+snapshot image from/to the kernel will use a swap parition, called the resume
+partition, as storage space. However, this is not really required, as they
+can use, for example, a special (blank) suspend partition or a file on a partition
+that is unmounted before SNAPSHOT_ATOMIC_SNAPSHOT and mounted afterwards.
+
+These utilities SHOULD NOT make any assumptions regarding the ordering of
+data within the snapshot image, except for the image header that MAY be
+assumed to start with an swsusp_info structure, as specified in
+kernel/power/power.h. This structure MAY be used by the userland utilities
+to obtain some information about the snapshot image, such as the size
+of the snapshot image, including the metadata and the header itself,
+contained in the .size member of swsusp_info.
+
+The snapshot image MUST be written to the kernel unaltered (ie. all of the image
+data, metadata and header MUST be written in _exactly_ the same amount, form
+and order in which they have been read). Otherwise, the behavior of the
+resumed system may be totally unpredictable.
+
+While executing SNAPSHOT_ATOMIC_RESTORE the kernel checks if the
+structure of the snapshot image is consistent with the information stored
+in the image header. If any inconsistencies are detected,
+SNAPSHOT_ATOMIC_RESTORE will not succeed. Still, this is not a fool-proof
+mechanism and the userland utilities using the interface SHOULD use additional
+means, such as checksums, to ensure the integrity of the snapshot image.
+
+The suspending and resuming utilities MUST lock themselves in memory,
+preferrably using mlockall(), before calling SNAPSHOT_FREEZE.
+
+The suspending utility MUST check the value stored by SNAPSHOT_ATOMIC_SNAPSHOT
+in the memory location pointed to by the last argument of ioctl() and proceed
+in accordance with it:
+1. If the value is 1 (ie. the system memory snapshot has just been
+ created and the system is ready for saving it):
+ (a) The suspending utility MUST NOT close the snapshot device
+ _unless_ the whole suspend procedure is to be cancelled, in
+ which case, if the snapshot image has already been saved, the
+ suspending utility SHOULD destroy it, preferrably by zapping
+ its header. If the suspend is not to be cancelled, the
+ system MUST be powered off or rebooted after the snapshot
+ image has been saved.
+ (b) The suspending utility SHOULD NOT attempt to perform any
+ file system operations (including reads) on the file systems
+ that were mounted before SNAPSHOT_ATOMIC_SNAPSHOT has been
+ called. However, it MAY mount a file system that was not
+ mounted at that time and perform some operations on it (eg.
+ use it for saving the image).
+2. If the value is 0 (ie. the system state has just been restored from
+ the snapshot image), the suspending utility MUST close the snapshot
+ device. Afterwards it will be treated as a regular userland process,
+ so it need not exit.
+
+The resuming utility SHOULD NOT attempt to mount any file systems that could
+be mounted before suspend and SHOULD NOT attempt to perform any operations
+involving such file systems.
+
+For details, please refer to the source code.
diff --git a/Documentation/power/video.txt b/Documentation/power/video.txt
index 912bed87c758..d18a57d1a531 100644
--- a/Documentation/power/video.txt
+++ b/Documentation/power/video.txt
@@ -1,7 +1,7 @@
Video issues with S3 resume
~~~~~~~~~~~~~~~~~~~~~~~~~~~
- 2003-2005, Pavel Machek
+ 2003-2006, Pavel Machek
During S3 resume, hardware needs to be reinitialized. For most
devices, this is easy, and kernel driver knows how to do
@@ -15,6 +15,27 @@ run normally so video card is normally initialized. It should not be
problem for S1 standby, because hardware should retain its state over
that.
+We either have to run video BIOS during early resume, or interpret it
+using vbetool later, or maybe nothing is neccessary on particular
+system because video state is preserved. Unfortunately different
+methods work on different systems, and no known method suits all of
+them.
+
+Userland application called s2ram has been developed; it contains long
+whitelist of systems, and automatically selects working method for a
+given system. It can be downloaded from CVS at
+www.sf.net/projects/suspend . If you get a system that is not in the
+whitelist, please try to find a working solution, and submit whitelist
+entry so that work does not need to be repeated.
+
+Currently, VBE_SAVE method (6 below) works on most
+systems. Unfortunately, vbetool only runs after userland is resumed,
+so it makes debugging of early resume problems
+hard/impossible. Methods that do not rely on userland are preferable.
+
+Details
+~~~~~~~
+
There are a few types of systems where video works after S3 resume:
(1) systems where video state is preserved over S3.
@@ -104,6 +125,7 @@ HP NX7000 ??? (*)
HP Pavilion ZD7000 vbetool post needed, need open-source nv driver for X
HP Omnibook XE3 athlon version none (1)
HP Omnibook XE3GC none (1), video is S3 Savage/IX-MV
+HP Omnibook 5150 none (1), (S1 also works OK)
IBM TP T20, model 2647-44G none (1), video is S3 Inc. 86C270-294 Savage/IX-MV, vesafb gets "interesting" but X work.
IBM TP A31 / Type 2652-M5G s3_mode (3) [works ok with BIOS 1.04 2002-08-23, but not at all with BIOS 1.11 2004-11-05 :-(]
IBM TP R32 / Type 2658-MMG none (1)
@@ -120,18 +142,24 @@ IBM ThinkPad T42p (2373-GTG) s3_bios (2)
IBM TP X20 ??? (*)
IBM TP X30 s3_bios (2)
IBM TP X31 / Type 2672-XXH none (1), use radeontool (http://fdd.com/software/radeon/) to turn off backlight.
-IBM TP X32 none (1), but backlight is on and video is trashed after long suspend
+IBM TP X32 none (1), but backlight is on and video is trashed after long suspend. s3_bios,s3_mode (4) works too. Perhaps that gets better results?
IBM Thinkpad X40 Type 2371-7JG s3_bios,s3_mode (4)
+IBM TP 600e none(1), but a switch to console and back to X is needed
Medion MD4220 ??? (*)
Samsung P35 vbetool needed (6)
-Sharp PC-AR10 (ATI rage) none (1)
+Sharp PC-AR10 (ATI rage) none (1), backlight does not switch off
Sony Vaio PCG-C1VRX/K s3_bios (2)
Sony Vaio PCG-F403 ??? (*)
+Sony Vaio PCG-GRT995MP none (1), works with 'nv' X driver
+Sony Vaio PCG-GR7/K none (1), but needs radeonfb, use radeontool (http://fdd.com/software/radeon/) to turn off backlight.
Sony Vaio PCG-N505SN ??? (*)
Sony Vaio vgn-s260 X or boot-radeon can init it (5)
+Sony Vaio vgn-S580BH vga=normal, but suspend from X. Console will be blank unless you return to X.
+Sony Vaio vgn-FS115B s3_bios (2),s3_mode (4)
Toshiba Libretto L5 none (1)
-Toshiba Satellite 4030CDT s3_mode (3)
-Toshiba Satellite 4080XCDT s3_mode (3)
+Toshiba Portege 3020CT s3_mode (3)
+Toshiba Satellite 4030CDT s3_mode (3) (S1 also works OK)
+Toshiba Satellite 4080XCDT s3_mode (3) (S1 also works OK)
Toshiba Satellite 4090XCDT ??? (*)
Toshiba Satellite P10-554 s3_bios,s3_mode (4)(****)
Toshiba M30 (2) xor X with nvidia driver using internal AGP
@@ -151,39 +179,3 @@ Asus A7V8X nVidia RIVA TNT2 model 64 s3_bios,s3_mode (4)
(***) To be tested with a newer kernel.
(****) Not with SMP kernel, UP only.
-
-VBEtool details
-~~~~~~~~~~~~~~~
-(with thanks to Carl-Daniel Hailfinger)
-
-First, boot into X and run the following script ONCE:
-#!/bin/bash
-statedir=/root/s3/state
-mkdir -p $statedir
-chvt 2
-sleep 1
-vbetool vbestate save >$statedir/vbe
-
-
-To suspend and resume properly, call the following script as root:
-#!/bin/bash
-statedir=/root/s3/state
-curcons=`fgconsole`
-fuser /dev/tty$curcons 2>/dev/null|xargs ps -o comm= -p|grep -q X && chvt 2
-cat /dev/vcsa >$statedir/vcsa
-sync
-echo 3 >/proc/acpi/sleep
-sync
-vbetool post
-vbetool vbestate restore <$statedir/vbe
-cat $statedir/vcsa >/dev/vcsa
-rckbd restart
-chvt $[curcons%6+1]
-chvt $curcons
-
-
-Unless you change your graphics card or other hardware configuration,
-the state once saved will be OK for every resume afterwards.
-NOTE: The "rckbd restart" command may be different for your
-distribution. Simply replace it with the command you would use to
-set the fonts on screen.
diff --git a/arch/cris/kernel/irq.c b/arch/cris/kernel/irq.c
index 30deaf1b728a..b504def3e346 100644
--- a/arch/cris/kernel/irq.c
+++ b/arch/cris/kernel/irq.c
@@ -52,9 +52,8 @@ int show_interrupts(struct seq_file *p, void *v)
if (i == 0) {
seq_printf(p, " ");
- for (j=0; j<NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "CPU%d ",j);
+ for_each_online_cpu(j)
+ seq_printf(p, "CPU%d ",j);
seq_putc(p, '\n');
}
@@ -67,9 +66,8 @@ int show_interrupts(struct seq_file *p, void *v)
#ifndef CONFIG_SMP
seq_printf(p, "%10u ", kstat_irqs(i));
#else
- for (j = 0; j < NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+ for_each_online_cpu(j)
+ seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
#endif
seq_printf(p, " %14s", irq_desc[i].handler->typename);
seq_printf(p, " %s", action->name);
diff --git a/arch/frv/kernel/irq.c b/arch/frv/kernel/irq.c
index 27ab4c30aac6..11fa326a8f62 100644
--- a/arch/frv/kernel/irq.c
+++ b/arch/frv/kernel/irq.c
@@ -75,9 +75,8 @@ int show_interrupts(struct seq_file *p, void *v)
switch (i) {
case 0:
seq_printf(p, " ");
- for (j = 0; j < NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "CPU%d ",j);
+ for_each_online_cpu(j)
+ seq_printf(p, "CPU%d ",j);
seq_putc(p, '\n');
break;
@@ -100,9 +99,8 @@ int show_interrupts(struct seq_file *p, void *v)
#ifndef CONFIG_SMP
seq_printf(p, "%10u ", kstat_irqs(i));
#else
- for (j = 0; j < NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "%10u ", kstat_cpu(j).irqs[i - 1]);
+ for_each_online_cpu(j)
+ seq_printf(p, "%10u ", kstat_cpu(j).irqs[i - 1]);
#endif
level = group->sources[ix]->level - frv_irq_levels;
diff --git a/arch/i386/Kconfig b/arch/i386/Kconfig
index 5b1a7d46d1d9..bfea1bedcbf2 100644
--- a/arch/i386/Kconfig
+++ b/arch/i386/Kconfig
@@ -80,6 +80,7 @@ config X86_VOYAGER
config X86_NUMAQ
bool "NUMAQ (IBM/Sequent)"
+ select SMP
select NUMA
help
This option is used for getting Linux to run on a (IBM/Sequent) NUMA
@@ -400,6 +401,7 @@ choice
config NOHIGHMEM
bool "off"
+ depends on !X86_NUMAQ
---help---
Linux can use up to 64 Gigabytes of physical memory on x86 systems.
However, the address space of 32-bit x86 processors is only 4
@@ -436,6 +438,7 @@ config NOHIGHMEM
config HIGHMEM4G
bool "4GB"
+ depends on !X86_NUMAQ
help
Select this if you have a 32-bit processor and between 1 and 4
gigabytes of physical RAM.
@@ -503,10 +506,6 @@ config NUMA
default n if X86_PC
default y if (X86_NUMAQ || X86_SUMMIT)
-# Need comments to help the hapless user trying to turn on NUMA support
-comment "NUMA (NUMA-Q) requires SMP, 64GB highmem support"
- depends on X86_NUMAQ && (!HIGHMEM64G || !SMP)
-
comment "NUMA (Summit) requires SMP, 64GB highmem support, ACPI"
depends on X86_SUMMIT && (!HIGHMEM64G || !ACPI)
@@ -660,13 +659,18 @@ config BOOT_IOREMAP
default y
config REGPARM
- bool "Use register arguments (EXPERIMENTAL)"
- depends on EXPERIMENTAL
- default n
+ bool "Use register arguments"
+ default y
help
- Compile the kernel with -mregparm=3. This uses a different ABI
- and passes the first three arguments of a function call in registers.
- This will probably break binary only modules.
+ Compile the kernel with -mregparm=3. This instructs gcc to use
+ a more efficient function call ABI which passes the first three
+ arguments of a function call via registers, which results in denser
+ and faster code.
+
+ If this option is disabled, then the default ABI of passing
+ arguments via the stack is used.
+
+ If unsure, say Y.
config SECCOMP
bool "Enable seccomp to safely compute untrusted bytecode"
diff --git a/arch/i386/Kconfig.debug b/arch/i386/Kconfig.debug
index bf32ecc9ad04..00108ba9a78d 100644
--- a/arch/i386/Kconfig.debug
+++ b/arch/i386/Kconfig.debug
@@ -31,6 +31,15 @@ config DEBUG_STACK_USAGE
This option will slow down process creation somewhat.
+config STACK_BACKTRACE_COLS
+ int "Stack backtraces per line" if DEBUG_KERNEL
+ range 1 3
+ default 2
+ help
+ Selects how many stack backtrace entries per line to display.
+
+ This can save screen space when displaying traces.
+
comment "Page alloc debug is incompatible with Software Suspend on i386"
depends on DEBUG_KERNEL && SOFTWARE_SUSPEND
diff --git a/arch/i386/kernel/Makefile b/arch/i386/kernel/Makefile
index 65656c033d70..5b9ed21216cf 100644
--- a/arch/i386/kernel/Makefile
+++ b/arch/i386/kernel/Makefile
@@ -7,7 +7,7 @@ extra-y := head.o init_task.o vmlinux.lds
obj-y := process.o semaphore.o signal.o entry.o traps.o irq.o \
ptrace.o time.o ioport.o ldt.o setup.o i8259.o sys_i386.o \
pci-dma.o i386_ksyms.o i387.o dmi_scan.o bootflag.o \
- quirks.o i8237.o topology.o
+ quirks.o i8237.o topology.o alternative.o
obj-y += cpu/
obj-y += timers/
diff --git a/arch/i386/kernel/alternative.c b/arch/i386/kernel/alternative.c
new file mode 100644
index 000000000000..5cbd6f99fb2a
--- /dev/null
+++ b/arch/i386/kernel/alternative.c
@@ -0,0 +1,321 @@
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/list.h>
+#include <asm/alternative.h>
+#include <asm/sections.h>
+
+#define DEBUG 0
+#if DEBUG
+# define DPRINTK(fmt, args...) printk(fmt, args)
+#else
+# define DPRINTK(fmt, args...)
+#endif
+
+/* Use inline assembly to define this because the nops are defined
+ as inline assembly strings in the include files and we cannot
+ get them easily into strings. */
+asm("\t.data\nintelnops: "
+ GENERIC_NOP1 GENERIC_NOP2 GENERIC_NOP3 GENERIC_NOP4 GENERIC_NOP5 GENERIC_NOP6
+ GENERIC_NOP7 GENERIC_NOP8);
+asm("\t.data\nk8nops: "
+ K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6
+ K8_NOP7 K8_NOP8);
+asm("\t.data\nk7nops: "
+ K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6
+ K7_NOP7 K7_NOP8);
+
+extern unsigned char intelnops[], k8nops[], k7nops[];
+static unsigned char *intel_nops[ASM_NOP_MAX+1] = {
+ NULL,
+ intelnops,
+ intelnops + 1,
+ intelnops + 1 + 2,
+ intelnops + 1 + 2 + 3,
+ intelnops + 1 + 2 + 3 + 4,
+ intelnops + 1 + 2 + 3 + 4 + 5,
+ intelnops + 1 + 2 + 3 + 4 + 5 + 6,
+ intelnops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
+};
+static unsigned char *k8_nops[ASM_NOP_MAX+1] = {
+ NULL,
+ k8nops,
+ k8nops + 1,
+ k8nops + 1 + 2,
+ k8nops + 1 + 2 + 3,
+ k8nops + 1 + 2 + 3 + 4,
+ k8nops + 1 + 2 + 3 + 4 + 5,
+ k8nops + 1 + 2 + 3 + 4 + 5 + 6,
+ k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
+};
+static unsigned char *k7_nops[ASM_NOP_MAX+1] = {
+ NULL,
+ k7nops,
+ k7nops + 1,
+ k7nops + 1 + 2,
+ k7nops + 1 + 2 + 3,
+ k7nops + 1 + 2 + 3 + 4,
+ k7nops + 1 + 2 + 3 + 4 + 5,
+ k7nops + 1 + 2 + 3 + 4 + 5 + 6,
+ k7nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
+};
+static struct nop {
+ int cpuid;
+ unsigned char **noptable;
+} noptypes[] = {
+ { X86_FEATURE_K8, k8_nops },
+ { X86_FEATURE_K7, k7_nops },
+ { -1, NULL }
+};
+
+
+extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
+extern struct alt_instr __smp_alt_instructions[], __smp_alt_instructions_end[];
+extern u8 *__smp_locks[], *__smp_locks_end[];
+
+extern u8 __smp_alt_begin[], __smp_alt_end[];
+
+
+static unsigned char** find_nop_table(void)
+{
+ unsigned char **noptable = intel_nops;
+ int i;
+
+ for (i = 0; noptypes[i].cpuid >= 0; i++) {
+ if (boot_cpu_has(noptypes[i].cpuid)) {
+ noptable = noptypes[i].noptable;
+ break;
+ }
+ }
+ return noptable;
+}
+
+/* Replace instructions with better alternatives for this CPU type.
+ This runs before SMP is initialized to avoid SMP problems with
+ self modifying code. This implies that assymetric systems where
+ APs have less capabilities than the boot processor are not handled.
+ Tough. Make sure you disable such features by hand. */
+
+void apply_alternatives(struct alt_instr *start, struct alt_instr *end)
+{
+ unsigned char **noptable = find_nop_table();
+ struct alt_instr *a;
+ int diff, i, k;
+
+ DPRINTK("%s: alt table %p -> %p\n", __FUNCTION__, start, end);
+ for (a = start; a < end; a++) {
+ BUG_ON(a->replacementlen > a->instrlen);
+ if (!boot_cpu_has(a->cpuid))
+ continue;
+ memcpy(a->instr, a->replacement, a->replacementlen);
+ diff = a->instrlen - a->replacementlen;
+ /* Pad the rest with nops */
+ for (i = a->replacementlen; diff > 0; diff -= k, i += k) {
+ k = diff;
+ if (k > ASM_NOP_MAX)
+ k = ASM_NOP_MAX;
+ memcpy(a->instr + i, noptable[k], k);
+ }
+ }
+}
+
+static void alternatives_smp_save(struct alt_instr *start, struct alt_instr *end)
+{
+ struct alt_instr *a;
+
+ DPRINTK("%s: alt table %p-%p\n", __FUNCTION__, start, end);
+ for (a = start; a < end; a++) {
+ memcpy(a->replacement + a->replacementlen,
+ a->instr,
+ a->instrlen);
+ }
+}
+
+static void alternatives_smp_apply(struct alt_instr *start, struct alt_instr *end)
+{
+ struct alt_instr *a;
+
+ for (a = start; a < end; a++) {
+ memcpy(a->instr,
+ a->replacement + a->replacementlen,
+ a->instrlen);
+ }
+}
+
+static void alternatives_smp_lock(u8 **start, u8 **end, u8 *text, u8 *text_end)
+{
+ u8 **ptr;
+
+ for (ptr = start; ptr < end; ptr++) {
+ if (*ptr < text)
+ continue;
+ if (*ptr > text_end)
+ continue;
+ **ptr = 0xf0; /* lock prefix */
+ };
+}
+
+static void alternatives_smp_unlock(u8 **start, u8 **end, u8 *text, u8 *text_end)
+{
+ unsigned char **noptable = find_nop_table();
+ u8 **ptr;
+
+ for (ptr = start; ptr < end; ptr++) {
+ if (*ptr < text)
+ continue;
+ if (*ptr > text_end)
+ continue;
+ **ptr = noptable[1][0];
+ };
+}
+
+struct smp_alt_module {
+ /* what is this ??? */
+ struct module *mod;
+ char *name;
+
+ /* ptrs to lock prefixes */
+ u8 **locks;
+ u8 **locks_end;
+
+ /* .text segment, needed to avoid patching init code ;) */
+ u8 *text;
+ u8 *text_end;
+
+ struct list_head next;
+};
+static LIST_HEAD(smp_alt_modules);
+static DEFINE_SPINLOCK(smp_alt);
+
+static int smp_alt_once = 0;
+static int __init bootonly(char *str)
+{
+ smp_alt_once = 1;
+ return 1;
+}
+__setup("smp-alt-boot", bootonly);
+
+void alternatives_smp_module_add(struct module *mod, char *name,
+ void *locks, void *locks_end,
+ void *text, void *text_end)
+{
+ struct smp_alt_module *smp;
+ unsigned long flags;
+
+ if (smp_alt_once) {
+ if (boot_cpu_has(X86_FEATURE_UP))
+ alternatives_smp_unlock(locks, locks_end,
+ text, text_end);
+ return;
+ }
+
+ smp = kzalloc(sizeof(*smp), GFP_KERNEL);
+ if (NULL == smp)
+ return; /* we'll run the (safe but slow) SMP code then ... */
+
+ smp->mod = mod;
+ smp->name = name;
+ smp->locks = locks;
+ smp->locks_end = locks_end;
+ smp->text = text;
+ smp->text_end = text_end;
+ DPRINTK("%s: locks %p -> %p, text %p -> %p, name %s\n",
+ __FUNCTION__, smp->locks, smp->locks_end,
+ smp->text, smp->text_end, smp->name);
+
+ spin_lock_irqsave(&smp_alt, flags);
+ list_add_tail(&smp->next, &smp_alt_modules);
+ if (boot_cpu_has(X86_FEATURE_UP))
+ alternatives_smp_unlock(smp->locks, smp->locks_end,
+ smp->text, smp->text_end);
+ spin_unlock_irqrestore(&smp_alt, flags);
+}
+
+void alternatives_smp_module_del(struct module *mod)
+{
+ struct smp_alt_module *item;
+ unsigned long flags;
+
+ if (smp_alt_once)
+ return;
+
+ spin_lock_irqsave(&smp_alt, flags);
+ list_for_each_entry(item, &smp_alt_modules, next) {
+ if (mod != item->mod)
+ continue;
+ list_del(&item->next);
+ spin_unlock_irqrestore(&smp_alt, flags);
+ DPRINTK("%s: %s\n", __FUNCTION__, item->name);
+ kfree(item);
+ return;
+ }
+ spin_unlock_irqrestore(&smp_alt, flags);
+}
+
+void alternatives_smp_switch(int smp)
+{
+ struct smp_alt_module *mod;
+ unsigned long flags;
+
+ if (smp_alt_once)
+ return;
+ BUG_ON(!smp && (num_online_cpus() > 1));
+
+ spin_lock_irqsave(&smp_alt, flags);
+ if (smp) {
+ printk(KERN_INFO "SMP alternatives: switching to SMP code\n");
+ clear_bit(X86_FEATURE_UP, boot_cpu_data.x86_capability);
+ clear_bit(X86_FEATURE_UP, cpu_data[0].x86_capability);
+ alternatives_smp_apply(__smp_alt_instructions,
+ __smp_alt_instructions_end);
+ list_for_each_entry(mod, &smp_alt_modules, next)
+ alternatives_smp_lock(mod->locks, mod->locks_end,
+ mod->text, mod->text_end);
+ } else {
+ printk(KERN_INFO "SMP alternatives: switching to UP code\n");
+ set_bit(X86_FEATURE_UP, boot_cpu_data.x86_capability);
+ set_bit(X86_FEATURE_UP, cpu_data[0].x86_capability);
+ apply_alternatives(__smp_alt_instructions,
+ __smp_alt_instructions_end);
+ list_for_each_entry(mod, &smp_alt_modules, next)
+ alternatives_smp_unlock(mod->locks, mod->locks_end,
+ mod->text, mod->text_end);
+ }
+ spin_unlock_irqrestore(&smp_alt, flags);
+}
+
+void __init alternative_instructions(void)
+{
+ apply_alternatives(__alt_instructions, __alt_instructions_end);
+
+ /* switch to patch-once-at-boottime-only mode and free the
+ * tables in case we know the number of CPUs will never ever
+ * change */
+#ifdef CONFIG_HOTPLUG_CPU
+ if (num_possible_cpus() < 2)
+ smp_alt_once = 1;
+#else
+ smp_alt_once = 1;
+#endif
+
+ if (smp_alt_once) {
+ if (1 == num_possible_cpus()) {
+ printk(KERN_INFO "SMP alternatives: switching to UP code\n");
+ set_bit(X86_FEATURE_UP, boot_cpu_data.x86_capability);
+ set_bit(X86_FEATURE_UP, cpu_data[0].x86_capability);
+ apply_alternatives(__smp_alt_instructions,
+ __smp_alt_instructions_end);
+ alternatives_smp_unlock(__smp_locks, __smp_locks_end,
+ _text, _etext);
+ }
+ free_init_pages("SMP alternatives",
+ (unsigned long)__smp_alt_begin,
+ (unsigned long)__smp_alt_end);
+ } else {
+ alternatives_smp_save(__smp_alt_instructions,
+ __smp_alt_instructions_end);
+ alternatives_smp_module_add(NULL, "core kernel",
+ __smp_locks, __smp_locks_end,
+ _text, _etext);
+ alternatives_smp_switch(0);
+ }
+}
diff --git a/arch/i386/kernel/apic.c b/arch/i386/kernel/apic.c
index 776c90989e06..eb5279d23b7f 100644
--- a/arch/i386/kernel/apic.c
+++ b/arch/i386/kernel/apic.c
@@ -38,6 +38,7 @@
#include <asm/i8253.h>
#include <mach_apic.h>
+#include <mach_apicdef.h>
#include <mach_ipi.h>
#include "io_ports.h"
diff --git a/arch/i386/kernel/cpu/centaur.c b/arch/i386/kernel/cpu/centaur.c
index f52669ecb93f..bd75629dd262 100644
--- a/arch/i386/kernel/cpu/centaur.c
+++ b/arch/i386/kernel/cpu/centaur.c
@@ -4,6 +4,7 @@
#include <asm/processor.h>
#include <asm/msr.h>
#include <asm/e820.h>
+#include <asm/mtrr.h>
#include "cpu.h"
#ifdef CONFIG_X86_OOSTORE
diff --git a/arch/i386/kernel/cpu/common.c b/arch/i386/kernel/cpu/common.c
index e6bd095ae108..7e3d6b6a4e96 100644
--- a/arch/i386/kernel/cpu/common.c
+++ b/arch/i386/kernel/cpu/common.c
@@ -25,9 +25,10 @@ EXPORT_PER_CPU_SYMBOL(cpu_gdt_descr);
DEFINE_PER_CPU(unsigned char, cpu_16bit_stack[CPU_16BIT_STACK_SIZE]);
EXPORT_PER_CPU_SYMBOL(cpu_16bit_stack);
-static int cachesize_override __devinitdata = -1;
-static int disable_x86_fxsr __devinitdata = 0;
-static int disable_x86_serial_nr __devinitdata = 1;
+static int cachesize_override __cpuinitdata = -1;
+static int disable_x86_fxsr __cpuinitdata;
+static int disable_x86_serial_nr __cpuinitdata = 1;
+static int disable_x86_sep __cpuinitdata;
struct cpu_dev * cpu_devs[X86_VENDOR_NUM] = {};
@@ -59,7 +60,7 @@ static int __init cachesize_setup(char *str)
}
__setup("cachesize=", cachesize_setup);
-int __devinit get_model_name(struct cpuinfo_x86 *c)
+int __cpuinit get_model_name(struct cpuinfo_x86 *c)
{
unsigned int *v;
char *p, *q;
@@ -89,7 +90,7 @@ int __devinit get_model_name(struct cpuinfo_x86 *c)
}
-void __devinit display_cacheinfo(struct cpuinfo_x86 *c)
+void __cpuinit display_cacheinfo(struct cpuinfo_x86 *c)
{
unsigned int n, dummy, ecx, edx, l2size;
@@ -130,7 +131,7 @@ void __devinit display_cacheinfo(struct cpuinfo_x86 *c)
/* in particular, if CPUID levels 0x80000002..4 are supported, this isn't used */
/* Look up CPU names by table lookup. */
-static char __devinit *table_lookup_model(struct cpuinfo_x86 *c)
+static char __cpuinit *table_lookup_model(struct cpuinfo_x86 *c)
{
struct cpu_model_info *info;
@@ -151,7 +152,7 @@ static char __devinit *table_lookup_model(struct cpuinfo_x86 *c)
}
-static void __devinit get_cpu_vendor(struct cpuinfo_x86 *c, int early)
+static void __cpuinit get_cpu_vendor(struct cpuinfo_x86 *c, int early)
{
char *v = c->x86_vendor_id;
int i;
@@ -187,6 +188,14 @@ static int __init x86_fxsr_setup(char * s)
__setup("nofxsr", x86_fxsr_setup);
+static int __init x86_sep_setup(char * s)
+{
+ disable_x86_sep = 1;
+ return 1;
+}
+__setup("nosep", x86_sep_setup);
+
+
/* Standard macro to see if a specific flag is changeable */
static inline int flag_is_changeable_p(u32 flag)
{
@@ -210,7 +219,7 @@ static inline int flag_is_changeable_p(u32 flag)
/* Probe for the CPUID instruction */
-static int __devinit have_cpuid_p(void)
+static int __cpuinit have_cpuid_p(void)
{
return flag_is_changeable_p(X86_EFLAGS_ID);
}
@@ -254,7 +263,7 @@ static void __init early_cpu_detect(void)
}
}
-void __devinit generic_identify(struct cpuinfo_x86 * c)
+void __cpuinit generic_identify(struct cpuinfo_x86 * c)
{
u32 tfms, xlvl;
int junk;
@@ -307,7 +316,7 @@ void __devinit generic_identify(struct cpuinfo_x86 * c)
#endif
}
-static void __devinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
+static void __cpuinit squash_the_stupid_serial_number(struct cpuinfo_x86 *c)
{
if (cpu_has(c, X86_FEATURE_PN) && disable_x86_serial_nr ) {
/* Disable processor serial number */
@@ -335,7 +344,7 @@ __setup("serialnumber", x86_serial_nr_setup);
/*
* This does the hard work of actually picking apart the CPU stuff...
*/
-void __devinit identify_cpu(struct cpuinfo_x86 *c)
+void __cpuinit identify_cpu(struct cpuinfo_x86 *c)
{
int i;
@@ -405,6 +414,10 @@ void __devinit identify_cpu(struct cpuinfo_x86 *c)
clear_bit(X86_FEATURE_XMM, c->x86_capability);
}
+ /* SEP disabled? */
+ if (disable_x86_sep)
+ clear_bit(X86_FEATURE_SEP, c->x86_capability);
+
if (disable_pse)
clear_bit(X86_FEATURE_PSE, c->x86_capability);
@@ -417,7 +430,7 @@ void __devinit identify_cpu(struct cpuinfo_x86 *c)
else
/* Last resort... */
sprintf(c->x86_model_id, "%02x/%02x",
- c->x86_vendor, c->x86_model);
+ c->x86, c->x86_model);
}
/* Now the feature flags better reflect actual CPU features! */
@@ -453,7 +466,7 @@ void __devinit identify_cpu(struct cpuinfo_x86 *c)
}
#ifdef CONFIG_X86_HT
-void __devinit detect_ht(struct cpuinfo_x86 *c)
+void __cpuinit detect_ht(struct cpuinfo_x86 *c)
{
u32 eax, ebx, ecx, edx;
int index_msb, core_bits;
@@ -500,7 +513,7 @@ void __devinit detect_ht(struct cpuinfo_x86 *c)
}
#endif
-void __devinit print_cpu_info(struct cpuinfo_x86 *c)
+void __cpuinit print_cpu_info(struct cpuinfo_x86 *c)
{
char *vendor = NULL;
@@ -523,7 +536,7 @@ void __devinit print_cpu_info(struct cpuinfo_x86 *c)
printk("\n");
}
-cpumask_t cpu_initialized __devinitdata = CPU_MASK_NONE;
+cpumask_t cpu_initialized __cpuinitdata = CPU_MASK_NONE;
/* This is hacky. :)
* We're emulating future behavior.
@@ -570,7 +583,7 @@ void __init early_cpu_init(void)
* and IDT. We reload them nevertheless, this function acts as a
* 'CPU state barrier', nothing should get across.
*/
-void __devinit cpu_init(void)
+void __cpuinit cpu_init(void)
{
int cpu = smp_processor_id();
struct tss_struct * t = &per_cpu(init_tss, cpu);
@@ -670,7 +683,7 @@ void __devinit cpu_init(void)
}
#ifdef CONFIG_HOTPLUG_CPU
-void __devinit cpu_uninit(void)
+void __cpuinit cpu_uninit(void)
{
int cpu = raw_smp_processor_id();
cpu_clear(cpu, cpu_initialized);
diff --git a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
index e11a09207ec8..3d5110b65cc3 100644
--- a/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
+++ b/arch/i386/kernel/cpu/cpufreq/powernow-k8.c
@@ -1145,9 +1145,7 @@ static int __cpuinit powernowk8_init(void)
{
unsigned int i, supported_cpus = 0;
- for (i=0; i<NR_CPUS; i++) {
- if (!cpu_online(i))
- continue;
+ for_each_cpu(i) {
if (check_supported_cpu(i))
supported_cpus++;
}
diff --git a/arch/i386/kernel/cpu/intel.c b/arch/i386/kernel/cpu/intel.c
index 8c0120186b9f..5386b29bb5a5 100644
--- a/arch/i386/kernel/cpu/intel.c
+++ b/arch/i386/kernel/cpu/intel.c
@@ -29,7 +29,7 @@ extern int trap_init_f00f_bug(void);
struct movsl_mask movsl_mask __read_mostly;
#endif
-void __devinit early_intel_workaround(struct cpuinfo_x86 *c)
+void __cpuinit early_intel_workaround(struct cpuinfo_x86 *c)
{
if (c->x86_vendor != X86_VENDOR_INTEL)
return;
@@ -44,7 +44,7 @@ void __devinit early_intel_workaround(struct cpuinfo_x86 *c)
* This is called before we do cpu ident work
*/
-int __devinit ppro_with_ram_bug(void)
+int __cpuinit ppro_with_ram_bug(void)
{
/* Uses data from early_cpu_detect now */
if (boot_cpu_data.x86_vendor == X86_VENDOR_INTEL &&
@@ -62,7 +62,7 @@ int __devinit ppro_with_ram_bug(void)
* P4 Xeon errata 037 workaround.
* Hardware prefetcher may cause stale data to be loaded into the cache.
*/
-static void __devinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
+static void __cpuinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
{
unsigned long lo, hi;
@@ -81,7 +81,7 @@ static void __devinit Intel_errata_workarounds(struct cpuinfo_x86 *c)
/*
* find out the number of processor cores on the die
*/
-static int __devinit num_cpu_cores(struct cpuinfo_x86 *c)
+static int __cpuinit num_cpu_cores(struct cpuinfo_x86 *c)
{
unsigned int eax, ebx, ecx, edx;
@@ -96,7 +96,7 @@ static int __devinit num_cpu_cores(struct cpuinfo_x86 *c)
return 1;
}
-static void __devinit init_intel(struct cpuinfo_x86 *c)
+static void __cpuinit init_intel(struct cpuinfo_x86 *c)
{
unsigned int l2 = 0;
char *p = NULL;
@@ -205,7 +205,7 @@ static unsigned int intel_size_cache(struct cpuinfo_x86 * c, unsigned int size)
return size;
}
-static struct cpu_dev intel_cpu_dev __devinitdata = {
+static struct cpu_dev intel_cpu_dev __cpuinitdata = {
.c_vendor = "Intel",
.c_ident = { "GenuineIntel" },
.c_models = {
diff --git a/arch/i386/kernel/cpu/intel_cacheinfo.c b/arch/i386/kernel/cpu/intel_cacheinfo.c
index ffe58cee0c48..ce61921369e5 100644
--- a/arch/i386/kernel/cpu/intel_cacheinfo.c
+++ b/arch/i386/kernel/cpu/intel_cacheinfo.c
@@ -174,7 +174,7 @@ unsigned int __cpuinit init_intel_cacheinfo(struct cpuinfo_x86 *c)
unsigned int new_l1d = 0, new_l1i = 0; /* Cache sizes from cpuid(4) */
unsigned int new_l2 = 0, new_l3 = 0, i; /* Cache sizes from cpuid(4) */
- if (c->cpuid_level > 4) {
+ if (c->cpuid_level > 3) {
static int is_initialized;
if (is_initialized == 0) {
@@ -330,7 +330,7 @@ static void __cpuinit cache_shared_cpu_map_setup(unsigned int cpu, int index)
}
}
}
-static void __devinit cache_remove_shared_cpu_map(unsigned int cpu, int index)
+static void __cpuinit cache_remove_shared_cpu_map(unsigned int cpu, int index)
{
struct _cpuid4_info *this_leaf, *sibling_leaf;
int sibling;
diff --git a/arch/i386/kernel/cpu/proc.c b/arch/i386/kernel/cpu/proc.c
index 89a85af33d28..5cfbd8011698 100644
--- a/arch/i386/kernel/cpu/proc.c
+++ b/arch/i386/kernel/cpu/proc.c
@@ -40,7 +40,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
/* Other (Linux-defined) */
"cxmmx", "k6_mtrr", "cyrix_arr", "centaur_mcr",
NULL, NULL, NULL, NULL,
- "constant_tsc", NULL, NULL, NULL, NULL, NULL, NULL, NULL,
+ "constant_tsc", "up", NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,
diff --git a/arch/i386/kernel/crash.c b/arch/i386/kernel/crash.c
index d49dbe8dc96b..e3c5fca0aa8a 100644
--- a/arch/i386/kernel/crash.c
+++ b/arch/i386/kernel/crash.c
@@ -105,7 +105,7 @@ static int crash_nmi_callback(struct pt_regs *regs, int cpu)
return 1;
local_irq_disable();
- if (!user_mode(regs)) {
+ if (!user_mode_vm(regs)) {
crash_fixup_ss_esp(&fixed_regs, regs);
regs = &fixed_regs;
}
diff --git a/arch/i386/kernel/entry.S b/arch/i386/kernel/entry.S
index 4d704724b2f5..cfc683f153b9 100644
--- a/arch/i386/kernel/entry.S
+++ b/arch/i386/kernel/entry.S
@@ -226,6 +226,10 @@ ENTRY(system_call)
pushl %eax # save orig_eax
SAVE_ALL
GET_THREAD_INFO(%ebp)
+ testl $TF_MASK,EFLAGS(%esp)
+ jz no_singlestep
+ orl $_TIF_SINGLESTEP,TI_flags(%ebp)
+no_singlestep:
# system call tracing in operation / emulation
/* Note, _TIF_SECCOMP is bit number 8, and so it needs testw and not testb */
testw $(_TIF_SYSCALL_EMU|_TIF_SYSCALL_TRACE|_TIF_SECCOMP|_TIF_SYSCALL_AUDIT),TI_flags(%ebp)
diff --git a/arch/i386/kernel/head.S b/arch/i386/kernel/head.S
index e0b7c632efbc..3debc2e26542 100644
--- a/arch/i386/kernel/head.S
+++ b/arch/i386/kernel/head.S
@@ -450,7 +450,6 @@ int_msg:
.globl boot_gdt_descr
.globl idt_descr
-.globl cpu_gdt_descr
ALIGN
# early boot GDT descriptor (must use 1:1 address mapping)
@@ -470,8 +469,6 @@ cpu_gdt_descr:
.word GDT_ENTRIES*8-1
.long cpu_gdt_table
- .fill NR_CPUS-1,8,0 # space for the other GDT descriptors
-
/*
* The boot_gdt_table must mirror the equivalent in setup.S and is
* used only for booting.
@@ -485,7 +482,7 @@ ENTRY(boot_gdt_table)
/*
* The Global Descriptor Table contains 28 quadwords, per-CPU.
*/
- .align PAGE_SIZE_asm
+ .align L1_CACHE_BYTES
ENTRY(cpu_gdt_table)
.quad 0x0000000000000000 /* NULL descriptor */
.quad 0x0000000000000000 /* 0x0b reserved */
diff --git a/arch/i386/kernel/io_apic.c b/arch/i386/kernel/io_apic.c
index 39d9a5fa907e..311b4e7266f1 100644
--- a/arch/i386/kernel/io_apic.c
+++ b/arch/i386/kernel/io_apic.c
@@ -351,8 +351,8 @@ static inline void rotate_irqs_among_cpus(unsigned long useful_load_threshold)
{
int i, j;
Dprintk("Rotating IRQs among CPUs.\n");
- for (i = 0; i < NR_CPUS; i++) {
- for (j = 0; cpu_online(i) && (j < NR_IRQS); j++) {
+ for_each_online_cpu(i) {
+ for (j = 0; j < NR_IRQS; j++) {
if (!irq_desc[j].action)
continue;
/* Is it a significant load ? */
@@ -381,7 +381,7 @@ static void do_irq_balance(void)
unsigned long imbalance = 0;
cpumask_t allowed_mask, target_cpu_mask, tmp;
- for (i = 0; i < NR_CPUS; i++) {
+ for_each_cpu(i) {
int package_index;
CPU_IRQ(i) = 0;
if (!cpu_online(i))
@@ -422,9 +422,7 @@ static void do_irq_balance(void)
}
}
/* Find the least loaded processor package */
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_online(i))
- continue;
+ for_each_online_cpu(i) {
if (i != CPU_TO_PACKAGEINDEX(i))
continue;
if (min_cpu_irq > CPU_IRQ(i)) {
@@ -441,9 +439,7 @@ tryanothercpu:
*/
tmp_cpu_irq = 0;
tmp_loaded = -1;
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_online(i))
- continue;
+ for_each_online_cpu(i) {
if (i != CPU_TO_PACKAGEINDEX(i))
continue;
if (max_cpu_irq <= CPU_IRQ(i))
@@ -619,9 +615,7 @@ static int __init balanced_irq_init(void)
if (smp_num_siblings > 1 && !cpus_empty(tmp))
physical_balance = 1;
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_online(i))
- continue;
+ for_each_online_cpu(i) {
irq_cpu_data[i].irq_delta = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
irq_cpu_data[i].last_irq = kmalloc(sizeof(unsigned long) * NR_IRQS, GFP_KERNEL);
if (irq_cpu_data[i].irq_delta == NULL || irq_cpu_data[i].last_irq == NULL) {
@@ -638,9 +632,11 @@ static int __init balanced_irq_init(void)
else
printk(KERN_ERR "balanced_irq_init: failed to spawn balanced_irq");
failed:
- for (i = 0; i < NR_CPUS; i++) {
+ for_each_cpu(i) {
kfree(irq_cpu_data[i].irq_delta);
+ irq_cpu_data[i].irq_delta = NULL;
kfree(irq_cpu_data[i].last_irq);
+ irq_cpu_data[i].last_irq = NULL;
}
return 0;
}
@@ -1761,7 +1757,8 @@ static void __init setup_ioapic_ids_from_mpc(void)
* Don't check I/O APIC IDs for xAPIC systems. They have
* no meaning without the serial APIC bus.
*/
- if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL && boot_cpu_data.x86 < 15))
+ if (!(boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
+ || APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
return;
/*
* This is broken; anything with a real cpu count has to
diff --git a/arch/i386/kernel/kprobes.c b/arch/i386/kernel/kprobes.c
index 694a13997637..7a59050242a7 100644
--- a/arch/i386/kernel/kprobes.c
+++ b/arch/i386/kernel/kprobes.c
@@ -84,9 +84,9 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
void __kprobes arch_remove_kprobe(struct kprobe *p)
{
- down(&kprobe_mutex);
+ mutex_lock(&kprobe_mutex);
free_insn_slot(p->ainsn.insn);
- up(&kprobe_mutex);
+ mutex_unlock(&kprobe_mutex);
}
static inline void save_previous_kprobe(struct kprobe_ctlblk *kcb)
diff --git a/arch/i386/kernel/module.c b/arch/i386/kernel/module.c
index 5149c8a621f0..470cf97e7cd3 100644
--- a/arch/i386/kernel/module.c
+++ b/arch/i386/kernel/module.c
@@ -104,26 +104,38 @@ int apply_relocate_add(Elf32_Shdr *sechdrs,
return -ENOEXEC;
}
-extern void apply_alternatives(void *start, void *end);
-
int module_finalize(const Elf_Ehdr *hdr,
const Elf_Shdr *sechdrs,
struct module *me)
{
- const Elf_Shdr *s;
+ const Elf_Shdr *s, *text = NULL, *alt = NULL, *locks = NULL;
char *secstrings = (void *)hdr + sechdrs[hdr->e_shstrndx].sh_offset;
- /* look for .altinstructions to patch */
for (s = sechdrs; s < sechdrs + hdr->e_shnum; s++) {
- void *seg;
- if (strcmp(".altinstructions", secstrings + s->sh_name))
- continue;
- seg = (void *)s->sh_addr;
- apply_alternatives(seg, seg + s->sh_size);
- }
+ if (!strcmp(".text", secstrings + s->sh_name))
+ text = s;
+ if (!strcmp(".altinstructions", secstrings + s->sh_name))
+ alt = s;
+ if (!strcmp(".smp_locks", secstrings + s->sh_name))
+ locks= s;
+ }
+
+ if (alt) {
+ /* patch .altinstructions */
+ void *aseg = (void *)alt->sh_addr;
+ apply_alternatives(aseg, aseg + alt->sh_size);
+ }
+ if (locks && text) {
+ void *lseg = (void *)locks->sh_addr;
+ void *tseg = (void *)text->sh_addr;
+ alternatives_smp_module_add(me, me->name,
+ lseg, lseg + locks->sh_size,
+ tseg, tseg + text->sh_size);
+ }
return 0;
}
void module_arch_cleanup(struct module *mod)
{
+ alternatives_smp_module_del(mod);
}
diff --git a/arch/i386/kernel/mpparse.c b/arch/i386/kernel/mpparse.c
index e6e2f43db85e..8d8aa9d1796d 100644
--- a/arch/i386/kernel/mpparse.c
+++ b/arch/i386/kernel/mpparse.c
@@ -828,6 +828,8 @@ void __init find_smp_config (void)
smp_scan_config(address, 0x400);
}
+int es7000_plat;
+
/* --------------------------------------------------------------------------
ACPI-based MP Configuration
-------------------------------------------------------------------------- */
@@ -935,7 +937,8 @@ void __init mp_register_ioapic (
mp_ioapics[idx].mpc_apicaddr = address;
set_fixmap_nocache(FIX_IO_APIC_BASE_0 + idx, address);
- if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL) && (boot_cpu_data.x86 < 15))
+ if ((boot_cpu_data.x86_vendor == X86_VENDOR_INTEL)
+ && !APIC_XAPIC(apic_version[boot_cpu_physical_apicid]))
tmpid = io_apic_get_unique_id(idx, id);
else
tmpid = id;
@@ -1011,8 +1014,6 @@ void __init mp_override_legacy_irq (
return;
}
-int es7000_plat;
-
void __init mp_config_acpi_legacy_irqs (void)
{
struct mpc_config_intsrc intsrc;
diff --git a/arch/i386/kernel/nmi.c b/arch/i386/kernel/nmi.c
index be87c5e2ee95..9074818b9473 100644
--- a/arch/i386/kernel/nmi.c
+++ b/arch/i386/kernel/nmi.c
@@ -143,7 +143,7 @@ static int __init check_nmi_watchdog(void)
local_irq_enable();
mdelay((10*1000)/nmi_hz); // wait 10 ticks
- for (cpu = 0; cpu < NR_CPUS; cpu++) {
+ for_each_cpu(cpu) {
#ifdef CONFIG_SMP
/* Check cpu_callin_map here because that is set
after the timer is started. */
@@ -510,7 +510,7 @@ void touch_nmi_watchdog (void)
* Just reset the alert counters, (other CPUs might be
* spinning on locks we hold):
*/
- for (i = 0; i < NR_CPUS; i++)
+ for_each_cpu(i)
alert_counter[i] = 0;
/*
@@ -543,7 +543,7 @@ void nmi_watchdog_tick (struct pt_regs * regs)
/*
* die_nmi will return ONLY if NOTIFY_STOP happens..
*/
- die_nmi(regs, "NMI Watchdog detected LOCKUP");
+ die_nmi(regs, "BUG: NMI Watchdog detected LOCKUP");
} else {
last_irq_sums[cpu] = sum;
alert_counter[cpu] = 0;
diff --git a/arch/i386/kernel/process.c b/arch/i386/kernel/process.c
index 0480454ebffa..299e61674084 100644
--- a/arch/i386/kernel/process.c
+++ b/arch/i386/kernel/process.c
@@ -295,7 +295,7 @@ void show_regs(struct pt_regs * regs)
printk("EIP: %04x:[<%08lx>] CPU: %d\n",0xffff & regs->xcs,regs->eip, smp_processor_id());
print_symbol("EIP is at %s\n", regs->eip);
- if (user_mode(regs))
+ if (user_mode_vm(regs))
printk(" ESP: %04x:%08lx",0xffff & regs->xss,regs->esp);
printk(" EFLAGS: %08lx %s (%s %.*s)\n",
regs->eflags, print_tainted(), system_utsname.release,
diff --git a/arch/i386/kernel/ptrace.c b/arch/i386/kernel/ptrace.c
index 5c1fb6aada5b..506462ef36a0 100644
--- a/arch/i386/kernel/ptrace.c
+++ b/arch/i386/kernel/ptrace.c
@@ -34,10 +34,10 @@
/*
* Determines which flags the user has access to [1 = access, 0 = no access].
- * Prohibits changing ID(21), VIP(20), VIF(19), VM(17), IOPL(12-13), IF(9).
+ * Prohibits changing ID(21), VIP(20), VIF(19), VM(17), NT(14), IOPL(12-13), IF(9).
* Also masks reserved bits (31-22, 15, 5, 3, 1).
*/
-#define FLAG_MASK 0x00054dd5
+#define FLAG_MASK 0x00050dd5
/* set's the trap flag. */
#define TRAP_FLAG 0x100
diff --git a/arch/i386/kernel/semaphore.c b/arch/i386/kernel/semaphore.c
index 7455ab643943..967dc74df9ee 100644
--- a/arch/i386/kernel/semaphore.c
+++ b/arch/i386/kernel/semaphore.c
@@ -110,11 +110,11 @@ asm(
".align 4\n"
".globl __write_lock_failed\n"
"__write_lock_failed:\n\t"
- LOCK "addl $" RW_LOCK_BIAS_STR ",(%eax)\n"
+ LOCK_PREFIX "addl $" RW_LOCK_BIAS_STR ",(%eax)\n"
"1: rep; nop\n\t"
"cmpl $" RW_LOCK_BIAS_STR ",(%eax)\n\t"
"jne 1b\n\t"
- LOCK "subl $" RW_LOCK_BIAS_STR ",(%eax)\n\t"
+ LOCK_PREFIX "subl $" RW_LOCK_BIAS_STR ",(%eax)\n\t"
"jnz __write_lock_failed\n\t"
"ret"
);
@@ -124,11 +124,11 @@ asm(
".align 4\n"
".globl __read_lock_failed\n"
"__read_lock_failed:\n\t"
- LOCK "incl (%eax)\n"
+ LOCK_PREFIX "incl (%eax)\n"
"1: rep; nop\n\t"
"cmpl $1,(%eax)\n\t"
"js 1b\n\t"
- LOCK "decl (%eax)\n\t"
+ LOCK_PREFIX "decl (%eax)\n\t"
"js __read_lock_failed\n\t"
"ret"
);
diff --git a/arch/i386/kernel/setup.c b/arch/i386/kernel/setup.c
index ab62a9f4701e..2d8782960f41 100644
--- a/arch/i386/kernel/setup.c
+++ b/arch/i386/kernel/setup.c
@@ -1377,101 +1377,6 @@ static void __init register_memory(void)
pci_mem_start, gapstart, gapsize);
}
-/* Use inline assembly to define this because the nops are defined
- as inline assembly strings in the include files and we cannot
- get them easily into strings. */
-asm("\t.data\nintelnops: "
- GENERIC_NOP1 GENERIC_NOP2 GENERIC_NOP3 GENERIC_NOP4 GENERIC_NOP5 GENERIC_NOP6
- GENERIC_NOP7 GENERIC_NOP8);
-asm("\t.data\nk8nops: "
- K8_NOP1 K8_NOP2 K8_NOP3 K8_NOP4 K8_NOP5 K8_NOP6
- K8_NOP7 K8_NOP8);
-asm("\t.data\nk7nops: "
- K7_NOP1 K7_NOP2 K7_NOP3 K7_NOP4 K7_NOP5 K7_NOP6
- K7_NOP7 K7_NOP8);
-
-extern unsigned char intelnops[], k8nops[], k7nops[];
-static unsigned char *intel_nops[ASM_NOP_MAX+1] = {
- NULL,
- intelnops,
- intelnops + 1,
- intelnops + 1 + 2,
- intelnops + 1 + 2 + 3,
- intelnops + 1 + 2 + 3 + 4,
- intelnops + 1 + 2 + 3 + 4 + 5,
- intelnops + 1 + 2 + 3 + 4 + 5 + 6,
- intelnops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
-};
-static unsigned char *k8_nops[ASM_NOP_MAX+1] = {
- NULL,
- k8nops,
- k8nops + 1,
- k8nops + 1 + 2,
- k8nops + 1 + 2 + 3,
- k8nops + 1 + 2 + 3 + 4,
- k8nops + 1 + 2 + 3 + 4 + 5,
- k8nops + 1 + 2 + 3 + 4 + 5 + 6,
- k8nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
-};
-static unsigned char *k7_nops[ASM_NOP_MAX+1] = {
- NULL,
- k7nops,
- k7nops + 1,
- k7nops + 1 + 2,
- k7nops + 1 + 2 + 3,
- k7nops + 1 + 2 + 3 + 4,
- k7nops + 1 + 2 + 3 + 4 + 5,
- k7nops + 1 + 2 + 3 + 4 + 5 + 6,
- k7nops + 1 + 2 + 3 + 4 + 5 + 6 + 7,
-};
-static struct nop {
- int cpuid;
- unsigned char **noptable;
-} noptypes[] = {
- { X86_FEATURE_K8, k8_nops },
- { X86_FEATURE_K7, k7_nops },
- { -1, NULL }
-};
-
-/* Replace instructions with better alternatives for this CPU type.
-
- This runs before SMP is initialized to avoid SMP problems with
- self modifying code. This implies that assymetric systems where
- APs have less capabilities than the boot processor are not handled.
- Tough. Make sure you disable such features by hand. */
-void apply_alternatives(void *start, void *end)
-{
- struct alt_instr *a;
- int diff, i, k;
- unsigned char **noptable = intel_nops;
- for (i = 0; noptypes[i].cpuid >= 0; i++) {
- if (boot_cpu_has(noptypes[i].cpuid)) {
- noptable = noptypes[i].noptable;
- break;
- }
- }
- for (a = start; (void *)a < end; a++) {
- if (!boot_cpu_has(a->cpuid))
- continue;
- BUG_ON(a->replacementlen > a->instrlen);
- memcpy(a->instr, a->replacement, a->replacementlen);
- diff = a->instrlen - a->replacementlen;
- /* Pad the rest with nops */
- for (i = a->replacementlen; diff > 0; diff -= k, i += k) {
- k = diff;
- if (k > ASM_NOP_MAX)
- k = ASM_NOP_MAX;
- memcpy(a->instr + i, noptable[k], k);
- }
- }
-}
-
-void __init alternative_instructions(void)
-{
- extern struct alt_instr __alt_instructions[], __alt_instructions_end[];
- apply_alternatives(__alt_instructions, __alt_instructions_end);
-}
-
static char * __init machine_specific_memory_setup(void);
#ifdef CONFIG_MCA
@@ -1554,6 +1459,16 @@ void __init setup_arch(char **cmdline_p)
parse_cmdline_early(cmdline_p);
+#ifdef CONFIG_EARLY_PRINTK
+ {
+ char *s = strstr(*cmdline_p, "earlyprintk=");
+ if (s) {
+ setup_early_printk(strchr(s, '=') + 1);
+ printk("early console enabled\n");
+ }
+ }
+#endif
+
max_low_pfn = setup_memory();
/*
@@ -1578,19 +1493,6 @@ void __init setup_arch(char **cmdline_p)
* NOTE: at this point the bootmem allocator is fully available.
*/
-#ifdef CONFIG_EARLY_PRINTK
- {
- char *s = strstr(*cmdline_p, "earlyprintk=");
- if (s) {
- extern void setup_early_printk(char *);
-
- setup_early_printk(strchr(s, '=') + 1);
- printk("early console enabled\n");
- }
- }
-#endif
-
-
dmi_scan_machine();
#ifdef CONFIG_X86_GENERICARCH
diff --git a/arch/i386/kernel/signal.c b/arch/i386/kernel/signal.c
index 963616d364ec..5c352c3a9e7f 100644
--- a/arch/i386/kernel/signal.c
+++ b/arch/i386/kernel/signal.c
@@ -123,7 +123,8 @@ restore_sigcontext(struct pt_regs *regs, struct sigcontext __user *sc, int *peax
err |= __get_user(tmp, &sc->seg); \
loadsegment(seg,tmp); }
-#define FIX_EFLAGS (X86_EFLAGS_AC | X86_EFLAGS_OF | X86_EFLAGS_DF | \
+#define FIX_EFLAGS (X86_EFLAGS_AC | X86_EFLAGS_RF | \
+ X86_EFLAGS_OF | X86_EFLAGS_DF | \
X86_EFLAGS_TF | X86_EFLAGS_SF | X86_EFLAGS_ZF | \
X86_EFLAGS_AF | X86_EFLAGS_PF | X86_EFLAGS_CF)
@@ -582,9 +583,6 @@ static void fastcall do_signal(struct pt_regs *regs)
if (!user_mode(regs))
return;
- if (try_to_freeze())
- goto no_signal;
-
if (test_thread_flag(TIF_RESTORE_SIGMASK))
oldset = &current->saved_sigmask;
else
@@ -613,7 +611,6 @@ static void fastcall do_signal(struct pt_regs *regs)
return;
}
-no_signal:
/* Did we come from a system call? */
if (regs->orig_eax >= 0) {
/* Restart the system call - no handlers present */
diff --git a/arch/i386/kernel/smpboot.c b/arch/i386/kernel/smpboot.c
index 7007e1783797..4c470e99a742 100644
--- a/arch/i386/kernel/smpboot.c
+++ b/arch/i386/kernel/smpboot.c
@@ -899,6 +899,7 @@ static int __devinit do_boot_cpu(int apicid, int cpu)
unsigned short nmi_high = 0, nmi_low = 0;
++cpucount;
+ alternatives_smp_switch(1);
/*
* We can't use kernel_thread since we must avoid to
@@ -1368,6 +1369,8 @@ void __cpu_die(unsigned int cpu)
/* They ack this in play_dead by setting CPU_DEAD */
if (per_cpu(cpu_state, cpu) == CPU_DEAD) {
printk ("CPU %d is now offline\n", cpu);
+ if (1 == num_online_cpus())
+ alternatives_smp_switch(0);
return;
}
msleep(100);
diff --git a/arch/i386/kernel/topology.c b/arch/i386/kernel/topology.c
index 67a0e1baa28b..296355292c7c 100644
--- a/arch/i386/kernel/topology.c
+++ b/arch/i386/kernel/topology.c
@@ -41,6 +41,15 @@ int arch_register_cpu(int num){
parent = &node_devices[node].node;
#endif /* CONFIG_NUMA */
+ /*
+ * CPU0 cannot be offlined due to several
+ * restrictions and assumptions in kernel. This basically
+ * doesnt add a control file, one cannot attempt to offline
+ * BSP.
+ */
+ if (!num)
+ cpu_devices[num].cpu.no_control = 1;
+
return register_cpu(&cpu_devices[num].cpu, num, parent);
}
diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c
index b814dbdcc91e..de5386b01d38 100644
--- a/arch/i386/kernel/traps.c
+++ b/arch/i386/kernel/traps.c
@@ -99,6 +99,8 @@ int register_die_notifier(struct notifier_block *nb)
{
int err = 0;
unsigned long flags;
+
+ vmalloc_sync_all();
spin_lock_irqsave(&die_notifier_lock, flags);
err = notifier_chain_register(&i386die_chain, nb);
spin_unlock_irqrestore(&die_notifier_lock, flags);
@@ -112,12 +114,30 @@ static inline int valid_stack_ptr(struct thread_info *tinfo, void *p)
p < (void *)tinfo + THREAD_SIZE - 3;
}
-static void print_addr_and_symbol(unsigned long addr, char *log_lvl)
+/*
+ * Print CONFIG_STACK_BACKTRACE_COLS address/symbol entries per line.
+ */
+static inline int print_addr_and_symbol(unsigned long addr, char *log_lvl,
+ int printed)
{
- printk(log_lvl);
+ if (!printed)
+ printk(log_lvl);
+
+#if CONFIG_STACK_BACKTRACE_COLS == 1
printk(" [<%08lx>] ", addr);
+#else
+ printk(" <%08lx> ", addr);
+#endif
print_symbol("%s", addr);
- printk("\n");
+
+ printed = (printed + 1) % CONFIG_STACK_BACKTRACE_COLS;
+
+ if (printed)
+ printk(" ");
+ else
+ printk("\n");
+
+ return printed;
}
static inline unsigned long print_context_stack(struct thread_info *tinfo,
@@ -125,20 +145,24 @@ static inline unsigned long print_context_stack(struct thread_info *tinfo,
char *log_lvl)
{
unsigned long addr;
+ int printed = 0; /* nr of entries already printed on current line */
#ifdef CONFIG_FRAME_POINTER
while (valid_stack_ptr(tinfo, (void *)ebp)) {
addr = *(unsigned long *)(ebp + 4);
- print_addr_and_symbol(addr, log_lvl);
+ printed = print_addr_and_symbol(addr, log_lvl, printed);
ebp = *(unsigned long *)ebp;
}
#else
while (valid_stack_ptr(tinfo, stack)) {
addr = *stack++;
if (__kernel_text_address(addr))
- print_addr_and_symbol(addr, log_lvl);
+ printed = print_addr_and_symbol(addr, log_lvl, printed);
}
#endif
+ if (printed)
+ printk("\n");
+
return ebp;
}
@@ -166,8 +190,7 @@ static void show_trace_log_lvl(struct task_struct *task,
stack = (unsigned long*)context->previous_esp;
if (!stack)
break;
- printk(log_lvl);
- printk(" =======================\n");
+ printk("%s =======================\n", log_lvl);
}
}
@@ -194,21 +217,17 @@ static void show_stack_log_lvl(struct task_struct *task, unsigned long *esp,
for(i = 0; i < kstack_depth_to_print; i++) {
if (kstack_end(stack))
break;
- if (i && ((i % 8) == 0)) {
- printk("\n");
- printk(log_lvl);
- printk(" ");
- }
+ if (i && ((i % 8) == 0))
+ printk("\n%s ", log_lvl);
printk("%08lx ", *stack++);
}
- printk("\n");
- printk(log_lvl);
- printk("Call Trace:\n");
+ printk("\n%sCall Trace:\n", log_lvl);
show_trace_log_lvl(task, esp, log_lvl);
}
void show_stack(struct task_struct *task, unsigned long *esp)
{
+ printk(" ");
show_stack_log_lvl(task, esp, "");
}
@@ -233,7 +252,7 @@ void show_registers(struct pt_regs *regs)
esp = (unsigned long) (&regs->esp);
savesegment(ss, ss);
- if (user_mode(regs)) {
+ if (user_mode_vm(regs)) {
in_kernel = 0;
esp = regs->esp;
ss = regs->xss & 0xffff;
@@ -333,6 +352,8 @@ void die(const char * str, struct pt_regs * regs, long err)
static int die_counter;
unsigned long flags;
+ oops_enter();
+
if (die.lock_owner != raw_smp_processor_id()) {
console_verbose();
spin_lock_irqsave(&die.lock, flags);
@@ -385,6 +406,7 @@ void die(const char * str, struct pt_regs * regs, long err)
ssleep(5);
panic("Fatal exception");
}
+ oops_exit();
do_exit(SIGSEGV);
}
@@ -623,7 +645,7 @@ void die_nmi (struct pt_regs *regs, const char *msg)
/* If we are in kernel we are probably nested up pretty bad
* and might aswell get out now while we still can.
*/
- if (!user_mode(regs)) {
+ if (!user_mode_vm(regs)) {
current->thread.trap_no = 2;
crash_kexec(regs);
}
@@ -694,6 +716,7 @@ fastcall void do_nmi(struct pt_regs * regs, long error_code)
void set_nmi_callback(nmi_callback_t callback)
{
+ vmalloc_sync_all();
rcu_assign_pointer(nmi_callback, callback);
}
EXPORT_SYMBOL_GPL(set_nmi_callback);
diff --git a/arch/i386/kernel/vmlinux.lds.S b/arch/i386/kernel/vmlinux.lds.S
index 4710195b6b74..3f21c6f6466d 100644
--- a/arch/i386/kernel/vmlinux.lds.S
+++ b/arch/i386/kernel/vmlinux.lds.S
@@ -68,6 +68,26 @@ SECTIONS
*(.data.init_task)
}
+ /* might get freed after init */
+ . = ALIGN(4096);
+ __smp_alt_begin = .;
+ __smp_alt_instructions = .;
+ .smp_altinstructions : AT(ADDR(.smp_altinstructions) - LOAD_OFFSET) {
+ *(.smp_altinstructions)
+ }
+ __smp_alt_instructions_end = .;
+ . = ALIGN(4);
+ __smp_locks = .;
+ .smp_locks : AT(ADDR(.smp_locks) - LOAD_OFFSET) {
+ *(.smp_locks)
+ }
+ __smp_locks_end = .;
+ .smp_altinstr_replacement : AT(ADDR(.smp_altinstr_replacement) - LOAD_OFFSET) {
+ *(.smp_altinstr_replacement)
+ }
+ . = ALIGN(4096);
+ __smp_alt_end = .;
+
/* will be freed after init */
. = ALIGN(4096); /* Init code and data */
__init_begin = .;
diff --git a/arch/i386/kernel/vsyscall-sysenter.S b/arch/i386/kernel/vsyscall-sysenter.S
index 76b728159403..3b62baa6a371 100644
--- a/arch/i386/kernel/vsyscall-sysenter.S
+++ b/arch/i386/kernel/vsyscall-sysenter.S
@@ -21,6 +21,9 @@
* instruction clobbers %esp, the user's %esp won't even survive entry
* into the kernel. We store %esp in %ebp. Code in entry.S must fetch
* arg6 from the stack.
+ *
+ * You can not use this vsyscall for the clone() syscall because the
+ * three dwords on the parent stack do not get copied to the child.
*/
.text
.globl __kernel_vsyscall
diff --git a/arch/i386/mach-es7000/es7000.h b/arch/i386/mach-es7000/es7000.h
index f1e3204f5dec..80566ca4a80a 100644
--- a/arch/i386/mach-es7000/es7000.h
+++ b/arch/i386/mach-es7000/es7000.h
@@ -83,6 +83,7 @@ struct es7000_oem_table {
struct psai psai;
};
+#ifdef CONFIG_ACPI
struct acpi_table_sdt {
unsigned long pa;
unsigned long count;
@@ -99,6 +100,9 @@ struct oem_table {
u32 OEMTableSize;
};
+extern int find_unisys_acpi_oem_table(unsigned long *oem_addr);
+#endif
+
struct mip_reg {
unsigned long long off_0;
unsigned long long off_8;
@@ -114,7 +118,6 @@ struct mip_reg {
#define MIP_FUNC(VALUE) (VALUE & 0xff)
extern int parse_unisys_oem (char *oemptr);
-extern int find_unisys_acpi_oem_table(unsigned long *oem_addr);
extern void setup_unisys(void);
extern int es7000_start_cpu(int cpu, unsigned long eip);
extern void es7000_sw_apic(void);
diff --git a/arch/i386/mach-es7000/es7000plat.c b/arch/i386/mach-es7000/es7000plat.c
index a9ab0644f403..3d0fc853516d 100644
--- a/arch/i386/mach-es7000/es7000plat.c
+++ b/arch/i386/mach-es7000/es7000plat.c
@@ -51,8 +51,6 @@ struct mip_reg *host_reg;
int mip_port;
unsigned long mip_addr, host_addr;
-#if defined(CONFIG_X86_IO_APIC) && defined(CONFIG_ACPI)
-
/*
* GSI override for ES7000 platforms.
*/
@@ -76,8 +74,6 @@ es7000_rename_gsi(int ioapic, int gsi)
return gsi;
}
-#endif /* (CONFIG_X86_IO_APIC) && (CONFIG_ACPI) */
-
void __init
setup_unisys(void)
{
@@ -160,6 +156,7 @@ parse_unisys_oem (char *oemptr)
return es7000_plat;
}
+#ifdef CONFIG_ACPI
int __init
find_unisys_acpi_oem_table(unsigned long *oem_addr)
{
@@ -212,6 +209,7 @@ find_unisys_acpi_oem_table(unsigned long *oem_addr)
}
return -1;
}
+#endif
static void
es7000_spin(int n)
diff --git a/arch/i386/mm/fault.c b/arch/i386/mm/fault.c
index cf572d9a3b6e..7f0fcf219a26 100644
--- a/arch/i386/mm/fault.c
+++ b/arch/i386/mm/fault.c
@@ -214,6 +214,68 @@ static noinline void force_sig_info_fault(int si_signo, int si_code,
fastcall void do_invalid_op(struct pt_regs *, unsigned long);
+static inline pmd_t *vmalloc_sync_one(pgd_t *pgd, unsigned long address)
+{
+ unsigned index = pgd_index(address);
+ pgd_t *pgd_k;
+ pud_t *pud, *pud_k;
+ pmd_t *pmd, *pmd_k;
+
+ pgd += index;
+ pgd_k = init_mm.pgd + index;
+
+ if (!pgd_present(*pgd_k))
+ return NULL;
+
+ /*
+ * set_pgd(pgd, *pgd_k); here would be useless on PAE
+ * and redundant with the set_pmd() on non-PAE. As would
+ * set_pud.
+ */
+
+ pud = pud_offset(pgd, address);
+ pud_k = pud_offset(pgd_k, address);
+ if (!pud_present(*pud_k))
+ return NULL;
+
+ pmd = pmd_offset(pud, address);
+ pmd_k = pmd_offset(pud_k, address);
+ if (!pmd_present(*pmd_k))
+ return NULL;
+ if (!pmd_present(*pmd))
+ set_pmd(pmd, *pmd_k);
+ else
+ BUG_ON(pmd_page(*pmd) != pmd_page(*pmd_k));
+ return pmd_k;
+}
+
+/*
+ * Handle a fault on the vmalloc or module mapping area
+ *
+ * This assumes no large pages in there.
+ */
+static inline int vmalloc_fault(unsigned long address)
+{
+ unsigned long pgd_paddr;
+ pmd_t *pmd_k;
+ pte_t *pte_k;
+ /*
+ * Synchronize this task's top level page-table
+ * with the 'reference' page table.
+ *
+ * Do _not_ use "current" here. We might be inside
+ * an interrupt in the middle of a task switch..
+ */
+ pgd_paddr = read_cr3();
+ pmd_k = vmalloc_sync_one(__va(pgd_paddr), address);
+ if (!pmd_k)
+ return -1;
+ pte_k = pte_offset_kernel(pmd_k, address);
+ if (!pte_present(*pte_k))
+ return -1;
+ return 0;
+}
+
/*
* This routine handles page faults. It determines the address,
* and the problem, and then passes it off to one of the appropriate
@@ -223,6 +285,8 @@ fastcall void do_invalid_op(struct pt_regs *, unsigned long);
* bit 0 == 0 means no page found, 1 means protection fault
* bit 1 == 0 means read, 1 means write
* bit 2 == 0 means kernel, 1 means user-mode
+ * bit 3 == 1 means use of reserved bit detected
+ * bit 4 == 1 means fault was an instruction fetch
*/
fastcall void __kprobes do_page_fault(struct pt_regs *regs,
unsigned long error_code)
@@ -237,13 +301,6 @@ fastcall void __kprobes do_page_fault(struct pt_regs *regs,
/* get the address */
address = read_cr2();
- if (notify_die(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
- SIGSEGV) == NOTIFY_STOP)
- return;
- /* It's safe to allow irq's after cr2 has been saved */
- if (regs->eflags & (X86_EFLAGS_IF|VM_MASK))
- local_irq_enable();
-
tsk = current;
si_code = SEGV_MAPERR;
@@ -259,17 +316,29 @@ fastcall void __kprobes do_page_fault(struct pt_regs *regs,
*
* This verifies that the fault happens in kernel space
* (error_code & 4) == 0, and that the fault was not a
- * protection error (error_code & 1) == 0.
+ * protection error (error_code & 9) == 0.
*/
- if (unlikely(address >= TASK_SIZE)) {
- if (!(error_code & 5))
- goto vmalloc_fault;
- /*
+ if (unlikely(address >= TASK_SIZE)) {
+ if (!(error_code & 0x0000000d) && vmalloc_fault(address) >= 0)
+ return;
+ if (notify_die(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
+ SIGSEGV) == NOTIFY_STOP)
+ return;
+ /*
* Don't take the mm semaphore here. If we fixup a prefetch
* fault we could otherwise deadlock.
*/
goto bad_area_nosemaphore;
- }
+ }
+
+ if (notify_die(DIE_PAGE_FAULT, "page fault", regs, error_code, 14,
+ SIGSEGV) == NOTIFY_STOP)
+ return;
+
+ /* It's safe to allow irq's after cr2 has been saved and the vmalloc
+ fault has been handled. */
+ if (regs->eflags & (X86_EFLAGS_IF|VM_MASK))
+ local_irq_enable();
mm = tsk->mm;
@@ -440,24 +509,31 @@ no_context:
bust_spinlocks(1);
-#ifdef CONFIG_X86_PAE
- if (error_code & 16) {
- pte_t *pte = lookup_address(address);
+ if (oops_may_print()) {
+ #ifdef CONFIG_X86_PAE
+ if (error_code & 16) {
+ pte_t *pte = lookup_address(address);
- if (pte && pte_present(*pte) && !pte_exec_kernel(*pte))
- printk(KERN_CRIT "kernel tried to execute NX-protected page - exploit attempt? (uid: %d)\n", current->uid);
+ if (pte && pte_present(*pte) && !pte_exec_kernel(*pte))
+ printk(KERN_CRIT "kernel tried to execute "
+ "NX-protected page - exploit attempt? "
+ "(uid: %d)\n", current->uid);
+ }
+ #endif
+ if (address < PAGE_SIZE)
+ printk(KERN_ALERT "BUG: unable to handle kernel NULL "
+ "pointer dereference");
+ else
+ printk(KERN_ALERT "BUG: unable to handle kernel paging"
+ " request");
+ printk(" at virtual address %08lx\n",address);
+ printk(KERN_ALERT " printing eip:\n");
+ printk("%08lx\n", regs->eip);
}
-#endif
- if (address < PAGE_SIZE)
- printk(KERN_ALERT "Unable to handle kernel NULL pointer dereference");
- else
- printk(KERN_ALERT "Unable to handle kernel paging request");
- printk(" at virtual address %08lx\n",address);
- printk(KERN_ALERT " printing eip:\n");
- printk("%08lx\n", regs->eip);
page = read_cr3();
page = ((unsigned long *) __va(page))[address >> 22];
- printk(KERN_ALERT "*pde = %08lx\n", page);
+ if (oops_may_print())
+ printk(KERN_ALERT "*pde = %08lx\n", page);
/*
* We must not directly access the pte in the highpte
* case, the page table might be allocated in highmem.
@@ -465,7 +541,7 @@ no_context:
* it's allocated already.
*/
#ifndef CONFIG_HIGHPTE
- if (page & 1) {
+ if ((page & 1) && oops_may_print()) {
page &= PAGE_MASK;
address &= 0x003ff000;
page = ((unsigned long *) __va(page))[address >> PAGE_SHIFT];
@@ -510,51 +586,41 @@ do_sigbus:
tsk->thread.error_code = error_code;
tsk->thread.trap_no = 14;
force_sig_info_fault(SIGBUS, BUS_ADRERR, address, tsk);
- return;
-
-vmalloc_fault:
- {
- /*
- * Synchronize this task's top level page-table
- * with the 'reference' page table.
- *
- * Do _not_ use "tsk" here. We might be inside
- * an interrupt in the middle of a task switch..
- */
- int index = pgd_index(address);
- unsigned long pgd_paddr;
- pgd_t *pgd, *pgd_k;
- pud_t *pud, *pud_k;
- pmd_t *pmd, *pmd_k;
- pte_t *pte_k;
-
- pgd_paddr = read_cr3();
- pgd = index + (pgd_t *)__va(pgd_paddr);
- pgd_k = init_mm.pgd + index;
-
- if (!pgd_present(*pgd_k))
- goto no_context;
-
- /*
- * set_pgd(pgd, *pgd_k); here would be useless on PAE
- * and redundant with the set_pmd() on non-PAE. As would
- * set_pud.
- */
+}
- pud = pud_offset(pgd, address);
- pud_k = pud_offset(pgd_k, address);
- if (!pud_present(*pud_k))
- goto no_context;
-
- pmd = pmd_offset(pud, address);
- pmd_k = pmd_offset(pud_k, address);
- if (!pmd_present(*pmd_k))
- goto no_context;
- set_pmd(pmd, *pmd_k);
+#ifndef CONFIG_X86_PAE
+void vmalloc_sync_all(void)
+{
+ /*
+ * Note that races in the updates of insync and start aren't
+ * problematic: insync can only get set bits added, and updates to
+ * start are only improving performance (without affecting correctness
+ * if undone).
+ */
+ static DECLARE_BITMAP(insync, PTRS_PER_PGD);
+ static unsigned long start = TASK_SIZE;
+ unsigned long address;
- pte_k = pte_offset_kernel(pmd_k, address);
- if (!pte_present(*pte_k))
- goto no_context;
- return;
+ BUILD_BUG_ON(TASK_SIZE & ~PGDIR_MASK);
+ for (address = start; address >= TASK_SIZE; address += PGDIR_SIZE) {
+ if (!test_bit(pgd_index(address), insync)) {
+ unsigned long flags;
+ struct page *page;
+
+ spin_lock_irqsave(&pgd_lock, flags);
+ for (page = pgd_list; page; page =
+ (struct page *)page->index)
+ if (!vmalloc_sync_one(page_address(page),
+ address)) {
+ BUG_ON(page != pgd_list);
+ break;
+ }
+ spin_unlock_irqrestore(&pgd_lock, flags);
+ if (!page)
+ set_bit(pgd_index(address), insync);
+ }
+ if (address == start && test_bit(pgd_index(address), insync))
+ start = address + PGDIR_SIZE;
}
}
+#endif
diff --git a/arch/i386/mm/init.c b/arch/i386/mm/init.c
index 7ba55a6e2dbc..9f66ac582a8b 100644
--- a/arch/i386/mm/init.c
+++ b/arch/i386/mm/init.c
@@ -720,21 +720,6 @@ static int noinline do_test_wp_bit(void)
return flag;
}
-void free_initmem(void)
-{
- unsigned long addr;
-
- addr = (unsigned long)(&__init_begin);
- for (; addr < (unsigned long)(&__init_end); addr += PAGE_SIZE) {
- ClearPageReserved(virt_to_page(addr));
- init_page_count(virt_to_page(addr));
- memset((void *)addr, 0xcc, PAGE_SIZE);
- free_page(addr);
- totalram_pages++;
- }
- printk (KERN_INFO "Freeing unused kernel memory: %dk freed\n", (__init_end - __init_begin) >> 10);
-}
-
#ifdef CONFIG_DEBUG_RODATA
extern char __start_rodata, __end_rodata;
@@ -758,17 +743,31 @@ void mark_rodata_ro(void)
}
#endif
+void free_init_pages(char *what, unsigned long begin, unsigned long end)
+{
+ unsigned long addr;
+
+ for (addr = begin; addr < end; addr += PAGE_SIZE) {
+ ClearPageReserved(virt_to_page(addr));
+ init_page_count(virt_to_page(addr));
+ memset((void *)addr, 0xcc, PAGE_SIZE);
+ free_page(addr);
+ totalram_pages++;
+ }
+ printk(KERN_INFO "Freeing %s: %ldk freed\n", what, (end - begin) >> 10);
+}
+
+void free_initmem(void)
+{
+ free_init_pages("unused kernel memory",
+ (unsigned long)(&__init_begin),
+ (unsigned long)(&__init_end));
+}
#ifdef CONFIG_BLK_DEV_INITRD
void free_initrd_mem(unsigned long start, unsigned long end)
{
- if (start < end)
- printk (KERN_INFO "Freeing initrd memory: %ldk freed\n", (end - start) >> 10);
- for (; start < end; start += PAGE_SIZE) {
- ClearPageReserved(virt_to_page(start));
- init_page_count(virt_to_page(start));
- free_page(start);
- totalram_pages++;
- }
+ free_init_pages("initrd memory", start, end);
}
#endif
+
diff --git a/arch/i386/oprofile/nmi_int.c b/arch/i386/oprofile/nmi_int.c
index 0493e8b8ec49..1accce50c2c7 100644
--- a/arch/i386/oprofile/nmi_int.c
+++ b/arch/i386/oprofile/nmi_int.c
@@ -122,7 +122,7 @@ static void nmi_save_registers(void * dummy)
static void free_msrs(void)
{
int i;
- for (i = 0; i < NR_CPUS; ++i) {
+ for_each_cpu(i) {
kfree(cpu_msrs[i].counters);
cpu_msrs[i].counters = NULL;
kfree(cpu_msrs[i].controls);
@@ -138,10 +138,7 @@ static int allocate_msrs(void)
size_t counters_size = sizeof(struct op_msr) * model->num_counters;
int i;
- for (i = 0; i < NR_CPUS; ++i) {
- if (!cpu_online(i))
- continue;
-
+ for_each_online_cpu(i) {
cpu_msrs[i].counters = kmalloc(counters_size, GFP_KERNEL);
if (!cpu_msrs[i].counters) {
success = 0;
diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c
index 626cdc83668b..0e5c6ae50228 100644
--- a/arch/ia64/hp/sim/simserial.c
+++ b/arch/ia64/hp/sim/simserial.c
@@ -46,11 +46,6 @@
#define KEYBOARD_INTR 3 /* must match with simulator! */
#define NR_PORTS 1 /* only one port for now */
-#define SERIAL_INLINE 1
-
-#ifdef SERIAL_INLINE
-#define _INLINE_ inline
-#endif
#define IRQ_T(info) ((info->flags & ASYNC_SHARE_IRQ) ? SA_SHIRQ : SA_INTERRUPT)
@@ -237,7 +232,7 @@ static void rs_put_char(struct tty_struct *tty, unsigned char ch)
local_irq_restore(flags);
}
-static _INLINE_ void transmit_chars(struct async_struct *info, int *intr_done)
+static void transmit_chars(struct async_struct *info, int *intr_done)
{
int count;
unsigned long flags;
diff --git a/arch/m32r/kernel/irq.c b/arch/m32r/kernel/irq.c
index 1ce63926a3c0..a4634b06f675 100644
--- a/arch/m32r/kernel/irq.c
+++ b/arch/m32r/kernel/irq.c
@@ -37,9 +37,8 @@ int show_interrupts(struct seq_file *p, void *v)
if (i == 0) {
seq_printf(p, " ");
- for (j=0; j<NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "CPU%d ",j);
+ for_each_online_cpu(j)
+ seq_printf(p, "CPU%d ",j);
seq_putc(p, '\n');
}
@@ -52,9 +51,8 @@ int show_interrupts(struct seq_file *p, void *v)
#ifndef CONFIG_SMP
seq_printf(p, "%10u ", kstat_irqs(i));
#else
- for (j = 0; j < NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+ for_each_online_cpu(j)
+ seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
#endif
seq_printf(p, " %14s", irq_desc[i].handler->typename);
seq_printf(p, " %s", action->name);
diff --git a/arch/m68k/bvme6000/rtc.c b/arch/m68k/bvme6000/rtc.c
index 703cbc6dc9cc..15c16b62dff5 100644
--- a/arch/m68k/bvme6000/rtc.c
+++ b/arch/m68k/bvme6000/rtc.c
@@ -18,6 +18,7 @@
#include <linux/module.h>
#include <linux/mc146818rtc.h> /* For struct rtc_time and ioctls, etc */
#include <linux/smp_lock.h>
+#include <linux/bcd.h>
#include <asm/bvme6000hw.h>
#include <asm/io.h>
@@ -32,9 +33,6 @@
* ioctls.
*/
-#define BCD2BIN(val) (((val)&15) + ((val)>>4)*10)
-#define BIN2BCD(val) ((((val)/10)<<4) + (val)%10)
-
static unsigned char days_in_mo[] =
{0, 31, 28, 31, 30, 31, 30, 31, 31, 30, 31, 30, 31};
diff --git a/arch/mips/kernel/irq.c b/arch/mips/kernel/irq.c
index 7d93992e462c..3dd76b3d2967 100644
--- a/arch/mips/kernel/irq.c
+++ b/arch/mips/kernel/irq.c
@@ -68,9 +68,8 @@ int show_interrupts(struct seq_file *p, void *v)
if (i == 0) {
seq_printf(p, " ");
- for (j=0; j<NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "CPU%d ",j);
+ for_each_online_cpu(j)
+ seq_printf(p, "CPU%d ",j);
seq_putc(p, '\n');
}
@@ -83,9 +82,8 @@ int show_interrupts(struct seq_file *p, void *v)
#ifndef CONFIG_SMP
seq_printf(p, "%10u ", kstat_irqs(i));
#else
- for (j = 0; j < NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+ for_each_online_cpu(j)
+ seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
#endif
seq_printf(p, " %14s", irq_desc[i].handler->typename);
seq_printf(p, " %s", action->name);
diff --git a/arch/mips/kernel/smp.c b/arch/mips/kernel/smp.c
index 06ed90752424..78d171bfa331 100644
--- a/arch/mips/kernel/smp.c
+++ b/arch/mips/kernel/smp.c
@@ -167,8 +167,8 @@ int smp_call_function (void (*func) (void *info), void *info, int retry,
mb();
/* Send a message to all other CPUs and wait for them to respond */
- for (i = 0; i < NR_CPUS; i++)
- if (cpu_online(i) && i != cpu)
+ for_each_online_cpu(i)
+ if (i != cpu)
core_send_ipi(i, SMP_CALL_FUNCTION);
/* Wait for response */
diff --git a/arch/mips/sgi-ip27/ip27-irq.c b/arch/mips/sgi-ip27/ip27-irq.c
index 73e5e52781d8..2854ac4c9be1 100644
--- a/arch/mips/sgi-ip27/ip27-irq.c
+++ b/arch/mips/sgi-ip27/ip27-irq.c
@@ -88,12 +88,9 @@ static inline int find_level(cpuid_t *cpunum, int irq)
{
int cpu, i;
- for (cpu = 0; cpu <= NR_CPUS; cpu++) {
+ for_each_online_cpu(cpu) {
struct slice_data *si = cpu_data[cpu].data;
- if (!cpu_online(cpu))
- continue;
-
for (i = BASE_PCI_IRQ; i < LEVELS_PER_SLICE; i++)
if (si->level_to_irq[i] == irq) {
*cpunum = cpu;
diff --git a/arch/parisc/kernel/smp.c b/arch/parisc/kernel/smp.c
index 25564b7ca6bb..d6ac1c60a471 100644
--- a/arch/parisc/kernel/smp.c
+++ b/arch/parisc/kernel/smp.c
@@ -298,8 +298,8 @@ send_IPI_allbutself(enum ipi_message_type op)
{
int i;
- for (i = 0; i < NR_CPUS; i++) {
- if (cpu_online(i) && i != smp_processor_id())
+ for_each_online_cpu(i) {
+ if (i != smp_processor_id())
send_IPI_single(i, op);
}
}
@@ -643,14 +643,13 @@ int sys_cpus(int argc, char **argv)
if ( argc == 1 ){
#ifdef DUMP_MORE_STATE
- for(i=0; i<NR_CPUS; i++) {
+ for_each_online_cpu(i) {
int cpus_per_line = 4;
- if(cpu_online(i)) {
- if (j++ % cpus_per_line)
- printk(" %3d",i);
- else
- printk("\n %3d",i);
- }
+
+ if (j++ % cpus_per_line)
+ printk(" %3d",i);
+ else
+ printk("\n %3d",i);
}
printk("\n");
#else
@@ -659,9 +658,7 @@ int sys_cpus(int argc, char **argv)
} else if((argc==2) && !(strcmp(argv[1],"-l"))) {
printk("\nCPUSTATE TASK CPUNUM CPUID HARDCPU(HPA)\n");
#ifdef DUMP_MORE_STATE
- for(i=0;i<NR_CPUS;i++) {
- if (!cpu_online(i))
- continue;
+ for_each_online_cpu(i) {
if (cpu_data[i].cpuid != NO_PROC_ID) {
switch(cpu_data[i].state) {
case STATE_RENDEZVOUS:
@@ -695,9 +692,7 @@ int sys_cpus(int argc, char **argv)
} else if ((argc==2) && !(strcmp(argv[1],"-s"))) {
#ifdef DUMP_MORE_STATE
printk("\nCPUSTATE CPUID\n");
- for (i=0;i<NR_CPUS;i++) {
- if (!cpu_online(i))
- continue;
+ for_each_online_cpu(i) {
if (cpu_data[i].cpuid != NO_PROC_ID) {
switch(cpu_data[i].state) {
case STATE_RENDEZVOUS:
diff --git a/arch/powerpc/kernel/irq.c b/arch/powerpc/kernel/irq.c
index 24dc8117b822..771a59cbd213 100644
--- a/arch/powerpc/kernel/irq.c
+++ b/arch/powerpc/kernel/irq.c
@@ -135,9 +135,8 @@ skip:
#ifdef CONFIG_TAU_INT
if (tau_initialized){
seq_puts(p, "TAU: ");
- for (j = 0; j < NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "%10u ", tau_interrupts(j));
+ for_each_online_cpu(j)
+ seq_printf(p, "%10u ", tau_interrupts(j));
seq_puts(p, " PowerPC Thermal Assist (cpu temp)\n");
}
#endif
diff --git a/arch/powerpc/kernel/kprobes.c b/arch/powerpc/kernel/kprobes.c
index 258039fb3016..cb1fe5878e8b 100644
--- a/arch/powerpc/kernel/kprobes.c
+++ b/arch/powerpc/kernel/kprobes.c
@@ -81,9 +81,9 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
void __kprobes arch_remove_kprobe(struct kprobe *p)
{
- down(&kprobe_mutex);
+ mutex_lock(&kprobe_mutex);
free_insn_slot(p->ainsn.insn);
- up(&kprobe_mutex);
+ mutex_unlock(&kprobe_mutex);
}
static inline void prepare_singlestep(struct kprobe *p, struct pt_regs *regs)
diff --git a/arch/powerpc/kernel/setup-common.c b/arch/powerpc/kernel/setup-common.c
index be12041c0fc5..c1d62bf11f29 100644
--- a/arch/powerpc/kernel/setup-common.c
+++ b/arch/powerpc/kernel/setup-common.c
@@ -162,9 +162,8 @@ static int show_cpuinfo(struct seq_file *m, void *v)
#if defined(CONFIG_SMP) && defined(CONFIG_PPC32)
unsigned long bogosum = 0;
int i;
- for (i = 0; i < NR_CPUS; ++i)
- if (cpu_online(i))
- bogosum += loops_per_jiffy;
+ for_each_online_cpu(i)
+ bogosum += loops_per_jiffy;
seq_printf(m, "total bogomips\t: %lu.%02lu\n",
bogosum/(500000/HZ), bogosum/(5000/HZ) % 100);
#endif /* CONFIG_SMP && CONFIG_PPC32 */
diff --git a/arch/powerpc/kernel/setup_32.c b/arch/powerpc/kernel/setup_32.c
index db72a92943bf..dc2770df25b3 100644
--- a/arch/powerpc/kernel/setup_32.c
+++ b/arch/powerpc/kernel/setup_32.c
@@ -272,9 +272,8 @@ int __init ppc_init(void)
if ( ppc_md.progress ) ppc_md.progress(" ", 0xffff);
/* register CPU devices */
- for (i = 0; i < NR_CPUS; i++)
- if (cpu_possible(i))
- register_cpu(&cpu_devices[i], i, NULL);
+ for_each_cpu(i)
+ register_cpu(&cpu_devices[i], i, NULL);
/* call platform init */
if (ppc_md.init != NULL) {
diff --git a/arch/powerpc/platforms/powermac/smp.c b/arch/powerpc/platforms/powermac/smp.c
index 6d64a9bf3474..1065d87fc279 100644
--- a/arch/powerpc/platforms/powermac/smp.c
+++ b/arch/powerpc/platforms/powermac/smp.c
@@ -191,9 +191,7 @@ static void smp_psurge_message_pass(int target, int msg)
if (num_online_cpus() < 2)
return;
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_online(i))
- continue;
+ for_each_online_cpu(i) {
if (target == MSG_ALL
|| (target == MSG_ALL_BUT_SELF && i != smp_processor_id())
|| target == i) {
diff --git a/arch/ppc/kernel/setup.c b/arch/ppc/kernel/setup.c
index c08ab432e958..53e9deacee82 100644
--- a/arch/ppc/kernel/setup.c
+++ b/arch/ppc/kernel/setup.c
@@ -168,9 +168,8 @@ int show_cpuinfo(struct seq_file *m, void *v)
/* Show summary information */
#ifdef CONFIG_SMP
unsigned long bogosum = 0;
- for (i = 0; i < NR_CPUS; ++i)
- if (cpu_online(i))
- bogosum += cpu_data[i].loops_per_jiffy;
+ for_each_online_cpu(i)
+ bogosum += cpu_data[i].loops_per_jiffy;
seq_printf(m, "total bogomips\t: %lu.%02lu\n",
bogosum/(500000/HZ), bogosum/(5000/HZ) % 100);
#endif /* CONFIG_SMP */
@@ -712,9 +711,8 @@ int __init ppc_init(void)
if ( ppc_md.progress ) ppc_md.progress(" ", 0xffff);
/* register CPU devices */
- for (i = 0; i < NR_CPUS; i++)
- if (cpu_possible(i))
- register_cpu(&cpu_devices[i], i, NULL);
+ for_each_cpu(i)
+ register_cpu(&cpu_devices[i], i, NULL);
/* call platform init */
if (ppc_md.init != NULL) {
diff --git a/arch/s390/kernel/smp.c b/arch/s390/kernel/smp.c
index 7dbe00c76c6b..d52d6d211d9f 100644
--- a/arch/s390/kernel/smp.c
+++ b/arch/s390/kernel/smp.c
@@ -799,9 +799,7 @@ void __init smp_prepare_cpus(unsigned int max_cpus)
*/
print_cpu_info(&S390_lowcore.cpu_data);
- for(i = 0; i < NR_CPUS; i++) {
- if (!cpu_possible(i))
- continue;
+ for_each_cpu(i) {
lowcore_ptr[i] = (struct _lowcore *)
__get_free_pages(GFP_KERNEL|GFP_DMA,
sizeof(void*) == 8 ? 1 : 0);
diff --git a/arch/sh/kernel/irq.c b/arch/sh/kernel/irq.c
index 6883c00728cb..b56e79632f24 100644
--- a/arch/sh/kernel/irq.c
+++ b/arch/sh/kernel/irq.c
@@ -35,9 +35,8 @@ int show_interrupts(struct seq_file *p, void *v)
if (i == 0) {
seq_puts(p, " ");
- for (j=0; j<NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "CPU%d ",j);
+ for_each_online_cpu(j)
+ seq_printf(p, "CPU%d ",j);
seq_putc(p, '\n');
}
diff --git a/arch/sh/kernel/setup.c b/arch/sh/kernel/setup.c
index a067a34e0b64..c0e79843f580 100644
--- a/arch/sh/kernel/setup.c
+++ b/arch/sh/kernel/setup.c
@@ -404,9 +404,8 @@ static int __init topology_init(void)
{
int cpu_id;
- for (cpu_id = 0; cpu_id < NR_CPUS; cpu_id++)
- if (cpu_possible(cpu_id))
- register_cpu(&cpu[cpu_id], cpu_id, NULL);
+ for_each_cpu(cpu_id)
+ register_cpu(&cpu[cpu_id], cpu_id, NULL);
return 0;
}
diff --git a/arch/sh64/kernel/irq.c b/arch/sh64/kernel/irq.c
index 9fc2b71dbd84..d69879c0e063 100644
--- a/arch/sh64/kernel/irq.c
+++ b/arch/sh64/kernel/irq.c
@@ -53,9 +53,8 @@ int show_interrupts(struct seq_file *p, void *v)
if (i == 0) {
seq_puts(p, " ");
- for (j=0; j<NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "CPU%d ",j);
+ for_each_online_cpu(j)
+ seq_printf(p, "CPU%d ",j);
seq_putc(p, '\n');
}
diff --git a/arch/sparc/kernel/irq.c b/arch/sparc/kernel/irq.c
index 410b9a72aba9..4c60a6ef54a9 100644
--- a/arch/sparc/kernel/irq.c
+++ b/arch/sparc/kernel/irq.c
@@ -184,9 +184,8 @@ int show_interrupts(struct seq_file *p, void *v)
#ifndef CONFIG_SMP
seq_printf(p, "%10u ", kstat_irqs(i));
#else
- for (j = 0; j < NR_CPUS; j++) {
- if (cpu_online(j))
- seq_printf(p, "%10u ",
+ for_each_online_cpu(j) {
+ seq_printf(p, "%10u ",
kstat_cpu(cpu_logical_map(j)).irqs[i]);
}
#endif
diff --git a/arch/sparc/kernel/smp.c b/arch/sparc/kernel/smp.c
index c6e721d8f477..ea5682ce7031 100644
--- a/arch/sparc/kernel/smp.c
+++ b/arch/sparc/kernel/smp.c
@@ -243,9 +243,8 @@ int setup_profiling_timer(unsigned int multiplier)
return -EINVAL;
spin_lock_irqsave(&prof_setup_lock, flags);
- for(i = 0; i < NR_CPUS; i++) {
- if (cpu_possible(i))
- load_profile_irq(i, lvl14_resolution / multiplier);
+ for_each_cpu(i) {
+ load_profile_irq(i, lvl14_resolution / multiplier);
prof_multiplier(i) = multiplier;
}
spin_unlock_irqrestore(&prof_setup_lock, flags);
@@ -273,13 +272,12 @@ void smp_bogo(struct seq_file *m)
{
int i;
- for (i = 0; i < NR_CPUS; i++) {
- if (cpu_online(i))
- seq_printf(m,
- "Cpu%dBogo\t: %lu.%02lu\n",
- i,
- cpu_data(i).udelay_val/(500000/HZ),
- (cpu_data(i).udelay_val/(5000/HZ))%100);
+ for_each_online_cpu(i) {
+ seq_printf(m,
+ "Cpu%dBogo\t: %lu.%02lu\n",
+ i,
+ cpu_data(i).udelay_val/(500000/HZ),
+ (cpu_data(i).udelay_val/(5000/HZ))%100);
}
}
@@ -288,8 +286,6 @@ void smp_info(struct seq_file *m)
int i;
seq_printf(m, "State:\n");
- for (i = 0; i < NR_CPUS; i++) {
- if (cpu_online(i))
- seq_printf(m, "CPU%d\t\t: online\n", i);
- }
+ for_each_online_cpu(i)
+ seq_printf(m, "CPU%d\t\t: online\n", i);
}
diff --git a/arch/sparc/kernel/sun4d_irq.c b/arch/sparc/kernel/sun4d_irq.c
index 52621348a56c..cea7fc6fc6e5 100644
--- a/arch/sparc/kernel/sun4d_irq.c
+++ b/arch/sparc/kernel/sun4d_irq.c
@@ -103,11 +103,9 @@ found_it: seq_printf(p, "%3d: ", i);
#ifndef CONFIG_SMP
seq_printf(p, "%10u ", kstat_irqs(i));
#else
- for (x = 0; x < NR_CPUS; x++) {
- if (cpu_online(x))
- seq_printf(p, "%10u ",
- kstat_cpu(cpu_logical_map(x)).irqs[i]);
- }
+ for_each_online_cpu(x)
+ seq_printf(p, "%10u ",
+ kstat_cpu(cpu_logical_map(x)).irqs[i]);
#endif
seq_printf(p, "%c %s",
(action->flags & SA_INTERRUPT) ? '+' : ' ',
diff --git a/arch/sparc/kernel/sun4d_smp.c b/arch/sparc/kernel/sun4d_smp.c
index 4219dd2ce3a2..41bb9596be48 100644
--- a/arch/sparc/kernel/sun4d_smp.c
+++ b/arch/sparc/kernel/sun4d_smp.c
@@ -249,11 +249,9 @@ void __init smp4d_boot_cpus(void)
} else {
unsigned long bogosum = 0;
- for(i = 0; i < NR_CPUS; i++) {
- if (cpu_isset(i, cpu_present_map)) {
- bogosum += cpu_data(i).udelay_val;
- smp_highest_cpu = i;
- }
+ for_each_present_cpu(i) {
+ bogosum += cpu_data(i).udelay_val;
+ smp_highest_cpu = i;
}
SMP_PRINTK(("Total of %d Processors activated (%lu.%02lu BogoMIPS).\n", cpucount + 1, bogosum/(500000/HZ), (bogosum/(5000/HZ))%100));
printk("Total of %d Processors activated (%lu.%02lu BogoMIPS).\n",
diff --git a/arch/sparc/kernel/sun4m_smp.c b/arch/sparc/kernel/sun4m_smp.c
index fbbd8a474c4c..1dde312eebda 100644
--- a/arch/sparc/kernel/sun4m_smp.c
+++ b/arch/sparc/kernel/sun4m_smp.c
@@ -218,10 +218,8 @@ void __init smp4m_boot_cpus(void)
cpu_present_map = cpumask_of_cpu(smp_processor_id());
} else {
unsigned long bogosum = 0;
- for(i = 0; i < NR_CPUS; i++) {
- if (cpu_isset(i, cpu_present_map))
- bogosum += cpu_data(i).udelay_val;
- }
+ for_each_present_cpu(i)
+ bogosum += cpu_data(i).udelay_val;
printk("Total of %d Processors activated (%lu.%02lu BogoMIPS).\n",
cpucount + 1,
bogosum/(500000/HZ),
diff --git a/arch/sparc64/kernel/irq.c b/arch/sparc64/kernel/irq.c
index 8c93ba655b33..e505a4125e35 100644
--- a/arch/sparc64/kernel/irq.c
+++ b/arch/sparc64/kernel/irq.c
@@ -117,9 +117,7 @@ int show_interrupts(struct seq_file *p, void *v)
#ifndef CONFIG_SMP
seq_printf(p, "%10u ", kstat_irqs(i));
#else
- for (j = 0; j < NR_CPUS; j++) {
- if (!cpu_online(j))
- continue;
+ for_each_online_cpu(j) {
seq_printf(p, "%10u ",
kstat_cpu(j).irqs[i]);
}
diff --git a/arch/sparc64/kernel/smp.c b/arch/sparc64/kernel/smp.c
index 373a701c90a5..1b6e2ade1008 100644
--- a/arch/sparc64/kernel/smp.c
+++ b/arch/sparc64/kernel/smp.c
@@ -57,25 +57,21 @@ void smp_info(struct seq_file *m)
int i;
seq_printf(m, "State:\n");
- for (i = 0; i < NR_CPUS; i++) {
- if (cpu_online(i))
- seq_printf(m,
- "CPU%d:\t\tonline\n", i);
- }
+ for_each_online_cpu(i)
+ seq_printf(m, "CPU%d:\t\tonline\n", i);
}
void smp_bogo(struct seq_file *m)
{
int i;
- for (i = 0; i < NR_CPUS; i++)
- if (cpu_online(i))
- seq_printf(m,
- "Cpu%dBogo\t: %lu.%02lu\n"
- "Cpu%dClkTck\t: %016lx\n",
- i, cpu_data(i).udelay_val / (500000/HZ),
- (cpu_data(i).udelay_val / (5000/HZ)) % 100,
- i, cpu_data(i).clock_tick);
+ for_each_online_cpu(i)
+ seq_printf(m,
+ "Cpu%dBogo\t: %lu.%02lu\n"
+ "Cpu%dClkTck\t: %016lx\n",
+ i, cpu_data(i).udelay_val / (500000/HZ),
+ (cpu_data(i).udelay_val / (5000/HZ)) % 100,
+ i, cpu_data(i).clock_tick);
}
void __init smp_store_cpu_info(int id)
@@ -1282,7 +1278,7 @@ int setup_profiling_timer(unsigned int multiplier)
return -EINVAL;
spin_lock_irqsave(&prof_setup_lock, flags);
- for (i = 0; i < NR_CPUS; i++)
+ for_each_cpu(i)
prof_multiplier(i) = multiplier;
current_tick_offset = (timer_tick_offset / multiplier);
spin_unlock_irqrestore(&prof_setup_lock, flags);
@@ -1384,10 +1380,8 @@ void __init smp_cpus_done(unsigned int max_cpus)
unsigned long bogosum = 0;
int i;
- for (i = 0; i < NR_CPUS; i++) {
- if (cpu_online(i))
- bogosum += cpu_data(i).udelay_val;
- }
+ for_each_online_cpu(i)
+ bogosum += cpu_data(i).udelay_val;
printk("Total of %ld processors activated "
"(%lu.%02lu BogoMIPS).\n",
(long) num_online_cpus(),
diff --git a/arch/sparc64/mm/init.c b/arch/sparc64/mm/init.c
index ded63ee9c4fd..1539a8362b6f 100644
--- a/arch/sparc64/mm/init.c
+++ b/arch/sparc64/mm/init.c
@@ -1828,8 +1828,8 @@ void __flush_tlb_all(void)
void online_page(struct page *page)
{
ClearPageReserved(page);
- set_page_count(page, 0);
- free_cold_page(page);
+ init_page_count(page);
+ __free_page(page);
totalram_pages++;
num_physpages++;
}
diff --git a/arch/um/kernel/um_arch.c b/arch/um/kernel/um_arch.c
index 27cdf9164422..80c9c18aae94 100644
--- a/arch/um/kernel/um_arch.c
+++ b/arch/um/kernel/um_arch.c
@@ -491,6 +491,16 @@ void __init check_bugs(void)
check_devanon();
}
-void apply_alternatives(void *start, void *end)
+void apply_alternatives(struct alt_instr *start, struct alt_instr *end)
+{
+}
+
+void alternatives_smp_module_add(struct module *mod, char *name,
+ void *locks, void *locks_end,
+ void *text, void *text_end)
+{
+}
+
+void alternatives_smp_module_del(struct module *mod)
{
}
diff --git a/arch/x86_64/kernel/early_printk.c b/arch/x86_64/kernel/early_printk.c
index 6dffb498ccd7..a8a6aa70d695 100644
--- a/arch/x86_64/kernel/early_printk.c
+++ b/arch/x86_64/kernel/early_printk.c
@@ -17,11 +17,8 @@
#define VGABASE ((void __iomem *)0xffffffff800b8000UL)
#endif
-#define MAX_YPOS max_ypos
-#define MAX_XPOS max_xpos
-
static int max_ypos = 25, max_xpos = 80;
-static int current_ypos = 1, current_xpos = 0;
+static int current_ypos = 25, current_xpos = 0;
static void early_vga_write(struct console *con, const char *str, unsigned n)
{
@@ -29,26 +26,26 @@ static void early_vga_write(struct console *con, const char *str, unsigned n)
int i, k, j;
while ((c = *str++) != '\0' && n-- > 0) {
- if (current_ypos >= MAX_YPOS) {
+ if (current_ypos >= max_ypos) {
/* scroll 1 line up */
- for (k = 1, j = 0; k < MAX_YPOS; k++, j++) {
- for (i = 0; i < MAX_XPOS; i++) {
- writew(readw(VGABASE + 2*(MAX_XPOS*k + i)),
- VGABASE + 2*(MAX_XPOS*j + i));
+ for (k = 1, j = 0; k < max_ypos; k++, j++) {
+ for (i = 0; i < max_xpos; i++) {
+ writew(readw(VGABASE+2*(max_xpos*k+i)),
+ VGABASE + 2*(max_xpos*j + i));
}
}
- for (i = 0; i < MAX_XPOS; i++)
- writew(0x720, VGABASE + 2*(MAX_XPOS*j + i));
- current_ypos = MAX_YPOS-1;
+ for (i = 0; i < max_xpos; i++)
+ writew(0x720, VGABASE + 2*(max_xpos*j + i));
+ current_ypos = max_ypos-1;
}
if (c == '\n') {
current_xpos = 0;
current_ypos++;
} else if (c != '\r') {
writew(((0x7 << 8) | (unsigned short) c),
- VGABASE + 2*(MAX_XPOS*current_ypos +
+ VGABASE + 2*(max_xpos*current_ypos +
current_xpos++));
- if (current_xpos >= MAX_XPOS) {
+ if (current_xpos >= max_xpos) {
current_xpos = 0;
current_ypos++;
}
@@ -244,6 +241,7 @@ int __init setup_early_printk(char *opt)
&& SCREEN_INFO.orig_video_isVGA == 1) {
max_xpos = SCREEN_INFO.orig_video_cols;
max_ypos = SCREEN_INFO.orig_video_lines;
+ current_ypos = SCREEN_INFO.orig_y;
early_console = &early_vga_console;
} else if (!strncmp(buf, "simnow", 6)) {
simnow_init(buf + 6);
diff --git a/arch/x86_64/kernel/irq.c b/arch/x86_64/kernel/irq.c
index 30d2a1e545fe..d8bd0b345b1e 100644
--- a/arch/x86_64/kernel/irq.c
+++ b/arch/x86_64/kernel/irq.c
@@ -38,9 +38,8 @@ int show_interrupts(struct seq_file *p, void *v)
if (i == 0) {
seq_printf(p, " ");
- for (j=0; j<NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "CPU%d ",j);
+ for_each_online_cpu(j)
+ seq_printf(p, "CPU%d ",j);
seq_putc(p, '\n');
}
@@ -53,10 +52,8 @@ int show_interrupts(struct seq_file *p, void *v)
#ifndef CONFIG_SMP
seq_printf(p, "%10u ", kstat_irqs(i));
#else
- for (j=0; j<NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "%10u ",
- kstat_cpu(j).irqs[i]);
+ for_each_online_cpu(j)
+ seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
#endif
seq_printf(p, " %14s", irq_desc[i].handler->typename);
@@ -68,15 +65,13 @@ skip:
spin_unlock_irqrestore(&irq_desc[i].lock, flags);
} else if (i == NR_IRQS) {
seq_printf(p, "NMI: ");
- for (j = 0; j < NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "%10u ", cpu_pda(j)->__nmi_count);
+ for_each_online_cpu(j)
+ seq_printf(p, "%10u ", cpu_pda(j)->__nmi_count);
seq_putc(p, '\n');
#ifdef CONFIG_X86_LOCAL_APIC
seq_printf(p, "LOC: ");
- for (j = 0; j < NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs);
+ for_each_online_cpu(j)
+ seq_printf(p, "%10u ", cpu_pda(j)->apic_timer_irqs);
seq_putc(p, '\n');
#endif
seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
diff --git a/arch/x86_64/kernel/kprobes.c b/arch/x86_64/kernel/kprobes.c
index 8b866a8572cf..14f0ced613b6 100644
--- a/arch/x86_64/kernel/kprobes.c
+++ b/arch/x86_64/kernel/kprobes.c
@@ -222,9 +222,9 @@ void __kprobes arch_disarm_kprobe(struct kprobe *p)
void __kprobes arch_remove_kprobe(struct kprobe *p)
{
- down(&kprobe_mutex);
+ mutex_lock(&kprobe_mutex);
free_insn_slot(p->ainsn.insn);
- up(&kprobe_mutex);
+ mutex_unlock(&kprobe_mutex);
}
static inline void save_previous_kprobe(struct kprobe_ctlblk *kcb)
diff --git a/arch/x86_64/kernel/nmi.c b/arch/x86_64/kernel/nmi.c
index 5bf17e41cd2d..66c009e10bac 100644
--- a/arch/x86_64/kernel/nmi.c
+++ b/arch/x86_64/kernel/nmi.c
@@ -162,9 +162,7 @@ int __init check_nmi_watchdog (void)
local_irq_enable();
mdelay((10*1000)/nmi_hz); // wait 10 ticks
- for (cpu = 0; cpu < NR_CPUS; cpu++) {
- if (!cpu_online(cpu))
- continue;
+ for_each_online_cpu(cpu) {
if (cpu_pda(cpu)->__nmi_count - counts[cpu] <= 5) {
endflag = 1;
printk("CPU#%d: NMI appears to be stuck (%d->%d)!\n",
diff --git a/arch/x86_64/kernel/signal.c b/arch/x86_64/kernel/signal.c
index 5876df116c92..e5f5ce7909a3 100644
--- a/arch/x86_64/kernel/signal.c
+++ b/arch/x86_64/kernel/signal.c
@@ -443,9 +443,6 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
if (!user_mode(regs))
return 1;
- if (try_to_freeze())
- goto no_signal;
-
if (!oldset)
oldset = &current->blocked;
@@ -463,7 +460,6 @@ int do_signal(struct pt_regs *regs, sigset_t *oldset)
return handle_signal(signr, &info, &ka, oldset, regs);
}
- no_signal:
/* Did we come from a system call? */
if ((long)regs->orig_rax >= 0) {
/* Restart the system call - no handlers present */
diff --git a/arch/xtensa/kernel/irq.c b/arch/xtensa/kernel/irq.c
index 4cbf6d91571f..51f9bed455fa 100644
--- a/arch/xtensa/kernel/irq.c
+++ b/arch/xtensa/kernel/irq.c
@@ -83,9 +83,8 @@ int show_interrupts(struct seq_file *p, void *v)
if (i == 0) {
seq_printf(p, " ");
- for (j=0; j<NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "CPU%d ",j);
+ for_each_online_cpu(j)
+ seq_printf(p, "CPU%d ",j);
seq_putc(p, '\n');
}
@@ -98,9 +97,8 @@ int show_interrupts(struct seq_file *p, void *v)
#ifndef CONFIG_SMP
seq_printf(p, "%10u ", kstat_irqs(i));
#else
- for (j = 0; j < NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
+ for_each_online_cpu(j)
+ seq_printf(p, "%10u ", kstat_cpu(j).irqs[i]);
#endif
seq_printf(p, " %14s", irq_desc[i].handler->typename);
seq_printf(p, " %s", action->name);
@@ -113,9 +111,8 @@ skip:
spin_unlock_irqrestore(&irq_desc[i].lock, flags);
} else if (i == NR_IRQS) {
seq_printf(p, "NMI: ");
- for (j = 0; j < NR_CPUS; j++)
- if (cpu_online(j))
- seq_printf(p, "%10u ", nmi_count(j));
+ for_each_online_cpu(j)
+ seq_printf(p, "%10u ", nmi_count(j));
seq_putc(p, '\n');
seq_printf(p, "ERR: %10u\n", atomic_read(&irq_err_count));
}
diff --git a/arch/xtensa/platform-iss/console.c b/arch/xtensa/platform-iss/console.c
index 94fdfe474ac1..2a580efb58ec 100644
--- a/arch/xtensa/platform-iss/console.c
+++ b/arch/xtensa/platform-iss/console.c
@@ -31,10 +31,6 @@
#include <linux/tty.h>
#include <linux/tty_flip.h>
-#ifdef SERIAL_INLINE
-#define _INLINE_ inline
-#endif
-
#define SERIAL_MAX_NUM_LINES 1
#define SERIAL_TIMER_VALUE (20 * HZ)
diff --git a/block/ioctl.c b/block/ioctl.c
index e1109491c234..35fdb7dc6512 100644
--- a/block/ioctl.c
+++ b/block/ioctl.c
@@ -42,9 +42,9 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
return -EINVAL;
}
/* partition number in use? */
- down(&bdev->bd_sem);
+ mutex_lock(&bdev->bd_mutex);
if (disk->part[part - 1]) {
- up(&bdev->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
return -EBUSY;
}
/* overlap? */
@@ -55,13 +55,13 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
continue;
if (!(start+length <= s->start_sect ||
start >= s->start_sect + s->nr_sects)) {
- up(&bdev->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
return -EBUSY;
}
}
/* all seems OK */
add_partition(disk, part, start, length);
- up(&bdev->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
return 0;
case BLKPG_DEL_PARTITION:
if (!disk->part[part-1])
@@ -71,9 +71,9 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
bdevp = bdget_disk(disk, part);
if (!bdevp)
return -ENOMEM;
- down(&bdevp->bd_sem);
+ mutex_lock(&bdevp->bd_mutex);
if (bdevp->bd_openers) {
- up(&bdevp->bd_sem);
+ mutex_unlock(&bdevp->bd_mutex);
bdput(bdevp);
return -EBUSY;
}
@@ -81,10 +81,10 @@ static int blkpg_ioctl(struct block_device *bdev, struct blkpg_ioctl_arg __user
fsync_bdev(bdevp);
invalidate_bdev(bdevp, 0);
- down(&bdev->bd_sem);
+ mutex_lock(&bdev->bd_mutex);
delete_partition(disk, part);
- up(&bdev->bd_sem);
- up(&bdevp->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
+ mutex_unlock(&bdevp->bd_mutex);
bdput(bdevp);
return 0;
@@ -102,10 +102,10 @@ static int blkdev_reread_part(struct block_device *bdev)
return -EINVAL;
if (!capable(CAP_SYS_ADMIN))
return -EACCES;
- if (down_trylock(&bdev->bd_sem))
+ if (!mutex_trylock(&bdev->bd_mutex))
return -EBUSY;
res = rescan_partitions(disk, bdev);
- up(&bdev->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
return res;
}
diff --git a/drivers/base/power/suspend.c b/drivers/base/power/suspend.c
index 8660779fb288..bdb60663f2ef 100644
--- a/drivers/base/power/suspend.c
+++ b/drivers/base/power/suspend.c
@@ -8,6 +8,7 @@
*
*/
+#include <linux/vt_kern.h>
#include <linux/device.h>
#include "../base.h"
#include "power.h"
@@ -62,7 +63,6 @@ int suspend_device(struct device * dev, pm_message_t state)
return error;
}
-
/**
* device_suspend - Save state and stop all devices in system.
* @state: Power state to put each device in.
@@ -82,6 +82,9 @@ int device_suspend(pm_message_t state)
{
int error = 0;
+ if (!is_console_suspend_safe())
+ return -EINVAL;
+
down(&dpm_sem);
down(&dpm_list_sem);
while (!list_empty(&dpm_active) && error == 0) {
diff --git a/drivers/block/cciss.c b/drivers/block/cciss.c
index cf39cf9aac25..e29b8926f80e 100644
--- a/drivers/block/cciss.c
+++ b/drivers/block/cciss.c
@@ -3268,8 +3268,8 @@ clean2:
unregister_blkdev(hba[i]->major, hba[i]->devname);
clean1:
release_io_mem(hba[i]);
- free_hba(i);
hba[i]->busy_initializing = 0;
+ free_hba(i);
return(-1);
}
diff --git a/drivers/block/floppy.c b/drivers/block/floppy.c
index d23b54332d7e..fb2d0be7cdeb 100644
--- a/drivers/block/floppy.c
+++ b/drivers/block/floppy.c
@@ -179,6 +179,7 @@ static int print_unex = 1;
#include <linux/devfs_fs_kernel.h>
#include <linux/platform_device.h>
#include <linux/buffer_head.h> /* for invalidate_buffers() */
+#include <linux/mutex.h>
/*
* PS/2 floppies have much slower step rates than regular floppies.
@@ -413,7 +414,7 @@ static struct floppy_write_errors write_errors[N_DRIVE];
static struct timer_list motor_off_timer[N_DRIVE];
static struct gendisk *disks[N_DRIVE];
static struct block_device *opened_bdev[N_DRIVE];
-static DECLARE_MUTEX(open_lock);
+static DEFINE_MUTEX(open_lock);
static struct floppy_raw_cmd *raw_cmd, default_raw_cmd;
/*
@@ -3333,7 +3334,7 @@ static inline int set_geometry(unsigned int cmd, struct floppy_struct *g,
if (type) {
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
- down(&open_lock);
+ mutex_lock(&open_lock);
LOCK_FDC(drive, 1);
floppy_type[type] = *g;
floppy_type[type].name = "user format";
@@ -3347,7 +3348,7 @@ static inline int set_geometry(unsigned int cmd, struct floppy_struct *g,
continue;
__invalidate_device(bdev);
}
- up(&open_lock);
+ mutex_unlock(&open_lock);
} else {
int oldStretch;
LOCK_FDC(drive, 1);
@@ -3674,7 +3675,7 @@ static int floppy_release(struct inode *inode, struct file *filp)
{
int drive = (long)inode->i_bdev->bd_disk->private_data;
- down(&open_lock);
+ mutex_lock(&open_lock);
if (UDRS->fd_ref < 0)
UDRS->fd_ref = 0;
else if (!UDRS->fd_ref--) {
@@ -3684,7 +3685,7 @@ static int floppy_release(struct inode *inode, struct file *filp)
if (!UDRS->fd_ref)
opened_bdev[drive] = NULL;
floppy_release_irq_and_dma();
- up(&open_lock);
+ mutex_unlock(&open_lock);
return 0;
}
@@ -3702,7 +3703,7 @@ static int floppy_open(struct inode *inode, struct file *filp)
char *tmp;
filp->private_data = (void *)0;
- down(&open_lock);
+ mutex_lock(&open_lock);
old_dev = UDRS->fd_device;
if (opened_bdev[drive] && opened_bdev[drive] != inode->i_bdev)
goto out2;
@@ -3785,7 +3786,7 @@ static int floppy_open(struct inode *inode, struct file *filp)
if ((filp->f_mode & 2) && !(UTESTF(FD_DISK_WRITABLE)))
goto out;
}
- up(&open_lock);
+ mutex_unlock(&open_lock);
return 0;
out:
if (UDRS->fd_ref < 0)
@@ -3796,7 +3797,7 @@ out:
opened_bdev[drive] = NULL;
floppy_release_irq_and_dma();
out2:
- up(&open_lock);
+ mutex_unlock(&open_lock);
return res;
}
diff --git a/drivers/block/loop.c b/drivers/block/loop.c
index 0010704739e3..74bf0255e98f 100644
--- a/drivers/block/loop.c
+++ b/drivers/block/loop.c
@@ -1144,7 +1144,7 @@ static int lo_ioctl(struct inode * inode, struct file * file,
struct loop_device *lo = inode->i_bdev->bd_disk->private_data;
int err;
- down(&lo->lo_ctl_mutex);
+ mutex_lock(&lo->lo_ctl_mutex);
switch (cmd) {
case LOOP_SET_FD:
err = loop_set_fd(lo, file, inode->i_bdev, arg);
@@ -1170,7 +1170,7 @@ static int lo_ioctl(struct inode * inode, struct file * file,
default:
err = lo->ioctl ? lo->ioctl(lo, cmd, arg) : -EINVAL;
}
- up(&lo->lo_ctl_mutex);
+ mutex_unlock(&lo->lo_ctl_mutex);
return err;
}
@@ -1178,9 +1178,9 @@ static int lo_open(struct inode *inode, struct file *file)
{
struct loop_device *lo = inode->i_bdev->bd_disk->private_data;
- down(&lo->lo_ctl_mutex);
+ mutex_lock(&lo->lo_ctl_mutex);
lo->lo_refcnt++;
- up(&lo->lo_ctl_mutex);
+ mutex_unlock(&lo->lo_ctl_mutex);
return 0;
}
@@ -1189,9 +1189,9 @@ static int lo_release(struct inode *inode, struct file *file)
{
struct loop_device *lo = inode->i_bdev->bd_disk->private_data;
- down(&lo->lo_ctl_mutex);
+ mutex_lock(&lo->lo_ctl_mutex);
--lo->lo_refcnt;
- up(&lo->lo_ctl_mutex);
+ mutex_unlock(&lo->lo_ctl_mutex);
return 0;
}
@@ -1233,12 +1233,12 @@ int loop_unregister_transfer(int number)
xfer_funcs[n] = NULL;
for (lo = &loop_dev[0]; lo < &loop_dev[max_loop]; lo++) {
- down(&lo->lo_ctl_mutex);
+ mutex_lock(&lo->lo_ctl_mutex);
if (lo->lo_encryption == xfer)
loop_release_xfer(lo);
- up(&lo->lo_ctl_mutex);
+ mutex_unlock(&lo->lo_ctl_mutex);
}
return 0;
@@ -1285,7 +1285,7 @@ static int __init loop_init(void)
lo->lo_queue = blk_alloc_queue(GFP_KERNEL);
if (!lo->lo_queue)
goto out_mem4;
- init_MUTEX(&lo->lo_ctl_mutex);
+ mutex_init(&lo->lo_ctl_mutex);
init_completion(&lo->lo_done);
init_completion(&lo->lo_bh_done);
lo->lo_number = i;
diff --git a/drivers/block/nbd.c b/drivers/block/nbd.c
index 6997d8e6bfb5..a9bde30dadad 100644
--- a/drivers/block/nbd.c
+++ b/drivers/block/nbd.c
@@ -459,9 +459,9 @@ static void do_nbd_request(request_queue_t * q)
req->errors = 0;
spin_unlock_irq(q->queue_lock);
- down(&lo->tx_lock);
+ mutex_lock(&lo->tx_lock);
if (unlikely(!lo->sock)) {
- up(&lo->tx_lock);
+ mutex_unlock(&lo->tx_lock);
printk(KERN_ERR "%s: Attempted send on closed socket\n",
lo->disk->disk_name);
req->errors++;
@@ -484,7 +484,7 @@ static void do_nbd_request(request_queue_t * q)
}
lo->active_req = NULL;
- up(&lo->tx_lock);
+ mutex_unlock(&lo->tx_lock);
wake_up_all(&lo->active_wq);
spin_lock_irq(q->queue_lock);
@@ -534,9 +534,9 @@ static int nbd_ioctl(struct inode *inode, struct file *file,
case NBD_CLEAR_SOCK:
error = 0;
- down(&lo->tx_lock);
+ mutex_lock(&lo->tx_lock);
lo->sock = NULL;
- up(&lo->tx_lock);
+ mutex_unlock(&lo->tx_lock);
file = lo->file;
lo->file = NULL;
nbd_clear_que(lo);
@@ -590,7 +590,7 @@ static int nbd_ioctl(struct inode *inode, struct file *file,
* FIXME: This code is duplicated from sys_shutdown, but
* there should be a more generic interface rather than
* calling socket ops directly here */
- down(&lo->tx_lock);
+ mutex_lock(&lo->tx_lock);
if (lo->sock) {
printk(KERN_WARNING "%s: shutting down socket\n",
lo->disk->disk_name);
@@ -598,7 +598,7 @@ static int nbd_ioctl(struct inode *inode, struct file *file,
SEND_SHUTDOWN|RCV_SHUTDOWN);
lo->sock = NULL;
}
- up(&lo->tx_lock);
+ mutex_unlock(&lo->tx_lock);
file = lo->file;
lo->file = NULL;
nbd_clear_que(lo);
@@ -683,7 +683,7 @@ static int __init nbd_init(void)
nbd_dev[i].flags = 0;
spin_lock_init(&nbd_dev[i].queue_lock);
INIT_LIST_HEAD(&nbd_dev[i].queue_head);
- init_MUTEX(&nbd_dev[i].tx_lock);
+ mutex_init(&nbd_dev[i].tx_lock);
init_waitqueue_head(&nbd_dev[i].active_wq);
nbd_dev[i].blksize = 1024;
nbd_dev[i].bytesize = 0x7ffffc00ULL << 10; /* 2TB */
diff --git a/drivers/block/pktcdvd.c b/drivers/block/pktcdvd.c
index 476a5b553f34..1d261f985f31 100644
--- a/drivers/block/pktcdvd.c
+++ b/drivers/block/pktcdvd.c
@@ -56,6 +56,7 @@
#include <linux/seq_file.h>
#include <linux/miscdevice.h>
#include <linux/suspend.h>
+#include <linux/mutex.h>
#include <scsi/scsi_cmnd.h>
#include <scsi/scsi_ioctl.h>
#include <scsi/scsi.h>
@@ -81,7 +82,7 @@
static struct pktcdvd_device *pkt_devs[MAX_WRITERS];
static struct proc_dir_entry *pkt_proc;
static int pkt_major;
-static struct semaphore ctl_mutex; /* Serialize open/close/setup/teardown */
+static struct mutex ctl_mutex; /* Serialize open/close/setup/teardown */
static mempool_t *psd_pool;
@@ -2018,7 +2019,7 @@ static int pkt_open(struct inode *inode, struct file *file)
VPRINTK("pktcdvd: entering open\n");
- down(&ctl_mutex);
+ mutex_lock(&ctl_mutex);
pd = pkt_find_dev_from_minor(iminor(inode));
if (!pd) {
ret = -ENODEV;
@@ -2044,14 +2045,14 @@ static int pkt_open(struct inode *inode, struct file *file)
set_blocksize(inode->i_bdev, CD_FRAMESIZE);
}
- up(&ctl_mutex);
+ mutex_unlock(&ctl_mutex);
return 0;
out_dec:
pd->refcnt--;
out:
VPRINTK("pktcdvd: failed open (%d)\n", ret);
- up(&ctl_mutex);
+ mutex_unlock(&ctl_mutex);
return ret;
}
@@ -2060,14 +2061,14 @@ static int pkt_close(struct inode *inode, struct file *file)
struct pktcdvd_device *pd = inode->i_bdev->bd_disk->private_data;
int ret = 0;
- down(&ctl_mutex);
+ mutex_lock(&ctl_mutex);
pd->refcnt--;
BUG_ON(pd->refcnt < 0);
if (pd->refcnt == 0) {
int flush = test_bit(PACKET_WRITABLE, &pd->flags);
pkt_release_dev(pd, flush);
}
- up(&ctl_mutex);
+ mutex_unlock(&ctl_mutex);
return ret;
}
@@ -2596,21 +2597,21 @@ static int pkt_ctl_ioctl(struct inode *inode, struct file *file, unsigned int cm
case PKT_CTRL_CMD_SETUP:
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
- down(&ctl_mutex);
+ mutex_lock(&ctl_mutex);
ret = pkt_setup_dev(&ctrl_cmd);
- up(&ctl_mutex);
+ mutex_unlock(&ctl_mutex);
break;
case PKT_CTRL_CMD_TEARDOWN:
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
- down(&ctl_mutex);
+ mutex_lock(&ctl_mutex);
ret = pkt_remove_dev(&ctrl_cmd);
- up(&ctl_mutex);
+ mutex_unlock(&ctl_mutex);
break;
case PKT_CTRL_CMD_STATUS:
- down(&ctl_mutex);
+ mutex_lock(&ctl_mutex);
pkt_get_status(&ctrl_cmd);
- up(&ctl_mutex);
+ mutex_unlock(&ctl_mutex);
break;
default:
return -ENOTTY;
@@ -2656,7 +2657,7 @@ static int __init pkt_init(void)
goto out;
}
- init_MUTEX(&ctl_mutex);
+ mutex_init(&ctl_mutex);
pkt_proc = proc_mkdir("pktcdvd", proc_root_driver);
diff --git a/drivers/block/rd.c b/drivers/block/rd.c
index ffd6abd6d5a0..1c54f46d3f70 100644
--- a/drivers/block/rd.c
+++ b/drivers/block/rd.c
@@ -310,12 +310,12 @@ static int rd_ioctl(struct inode *inode, struct file *file,
* cache
*/
error = -EBUSY;
- down(&bdev->bd_sem);
+ mutex_lock(&bdev->bd_mutex);
if (bdev->bd_openers <= 2) {
truncate_inode_pages(bdev->bd_inode->i_mapping, 0);
error = 0;
}
- up(&bdev->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
return error;
}
diff --git a/drivers/cdrom/cdrom.c b/drivers/cdrom/cdrom.c
index 879bbc26ce96..a59876a0bfa1 100644
--- a/drivers/cdrom/cdrom.c
+++ b/drivers/cdrom/cdrom.c
@@ -407,7 +407,6 @@ int register_cdrom(struct cdrom_device_info *cdi)
ENSURE(get_mcn, CDC_MCN);
ENSURE(reset, CDC_RESET);
ENSURE(audio_ioctl, CDC_PLAY_AUDIO);
- ENSURE(dev_ioctl, CDC_IOCTLS);
ENSURE(generic_packet, CDC_GENERIC_PACKET);
cdi->mc_flags = 0;
cdo->n_minors = 0;
@@ -2196,395 +2195,586 @@ retry:
return cdrom_read_cdda_old(cdi, ubuf, lba, nframes);
}
-/* Just about every imaginable ioctl is supported in the Uniform layer
- * these days. ATAPI / SCSI specific code now mainly resides in
- * mmc_ioct().
- */
-int cdrom_ioctl(struct file * file, struct cdrom_device_info *cdi,
- struct inode *ip, unsigned int cmd, unsigned long arg)
+static int cdrom_ioctl_multisession(struct cdrom_device_info *cdi,
+ void __user *argp)
{
- struct cdrom_device_ops *cdo = cdi->ops;
+ struct cdrom_multisession ms_info;
+ u8 requested_format;
int ret;
- /* Try the generic SCSI command ioctl's first.. */
- ret = scsi_cmd_ioctl(file, ip->i_bdev->bd_disk, cmd, (void __user *)arg);
- if (ret != -ENOTTY)
+ cdinfo(CD_DO_IOCTL, "entering CDROMMULTISESSION\n");
+
+ if (!(cdi->ops->capability & CDC_MULTI_SESSION))
+ return -ENOSYS;
+
+ if (copy_from_user(&ms_info, argp, sizeof(ms_info)))
+ return -EFAULT;
+
+ requested_format = ms_info.addr_format;
+ if (requested_format != CDROM_MSF && requested_format != CDROM_LBA)
+ return -EINVAL;
+ ms_info.addr_format = CDROM_LBA;
+
+ ret = cdi->ops->get_last_session(cdi, &ms_info);
+ if (ret)
return ret;
- /* the first few commands do not deal with audio drive_info, but
- only with routines in cdrom device operations. */
- switch (cmd) {
- case CDROMMULTISESSION: {
- struct cdrom_multisession ms_info;
- u_char requested_format;
- cdinfo(CD_DO_IOCTL, "entering CDROMMULTISESSION\n");
- if (!(cdo->capability & CDC_MULTI_SESSION))
- return -ENOSYS;
- IOCTL_IN(arg, struct cdrom_multisession, ms_info);
- requested_format = ms_info.addr_format;
- if (!((requested_format == CDROM_MSF) ||
- (requested_format == CDROM_LBA)))
- return -EINVAL;
- ms_info.addr_format = CDROM_LBA;
- if ((ret=cdo->get_last_session(cdi, &ms_info)))
+ sanitize_format(&ms_info.addr, &ms_info.addr_format, requested_format);
+
+ if (copy_to_user(argp, &ms_info, sizeof(ms_info)))
+ return -EFAULT;
+
+ cdinfo(CD_DO_IOCTL, "CDROMMULTISESSION successful\n");
+ return 0;
+}
+
+static int cdrom_ioctl_eject(struct cdrom_device_info *cdi)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROMEJECT\n");
+
+ if (!CDROM_CAN(CDC_OPEN_TRAY))
+ return -ENOSYS;
+ if (cdi->use_count != 1 || keeplocked)
+ return -EBUSY;
+ if (CDROM_CAN(CDC_LOCK)) {
+ int ret = cdi->ops->lock_door(cdi, 0);
+ if (ret)
return ret;
- sanitize_format(&ms_info.addr, &ms_info.addr_format,
- requested_format);
- IOCTL_OUT(arg, struct cdrom_multisession, ms_info);
- cdinfo(CD_DO_IOCTL, "CDROMMULTISESSION successful\n");
- return 0;
- }
+ }
- case CDROMEJECT: {
- cdinfo(CD_DO_IOCTL, "entering CDROMEJECT\n");
- if (!CDROM_CAN(CDC_OPEN_TRAY))
- return -ENOSYS;
- if (cdi->use_count != 1 || keeplocked)
- return -EBUSY;
- if (CDROM_CAN(CDC_LOCK))
- if ((ret=cdo->lock_door(cdi, 0)))
- return ret;
+ return cdi->ops->tray_move(cdi, 1);
+}
- return cdo->tray_move(cdi, 1);
- }
+static int cdrom_ioctl_closetray(struct cdrom_device_info *cdi)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROMCLOSETRAY\n");
- case CDROMCLOSETRAY: {
- cdinfo(CD_DO_IOCTL, "entering CDROMCLOSETRAY\n");
- if (!CDROM_CAN(CDC_CLOSE_TRAY))
- return -ENOSYS;
- return cdo->tray_move(cdi, 0);
- }
+ if (!CDROM_CAN(CDC_CLOSE_TRAY))
+ return -ENOSYS;
+ return cdi->ops->tray_move(cdi, 0);
+}
- case CDROMEJECT_SW: {
- cdinfo(CD_DO_IOCTL, "entering CDROMEJECT_SW\n");
- if (!CDROM_CAN(CDC_OPEN_TRAY))
- return -ENOSYS;
- if (keeplocked)
- return -EBUSY;
- cdi->options &= ~(CDO_AUTO_CLOSE | CDO_AUTO_EJECT);
- if (arg)
- cdi->options |= CDO_AUTO_CLOSE | CDO_AUTO_EJECT;
- return 0;
- }
+static int cdrom_ioctl_eject_sw(struct cdrom_device_info *cdi,
+ unsigned long arg)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROMEJECT_SW\n");
- case CDROM_MEDIA_CHANGED: {
- struct cdrom_changer_info *info;
- int changed;
+ if (!CDROM_CAN(CDC_OPEN_TRAY))
+ return -ENOSYS;
+ if (keeplocked)
+ return -EBUSY;
- cdinfo(CD_DO_IOCTL, "entering CDROM_MEDIA_CHANGED\n");
- if (!CDROM_CAN(CDC_MEDIA_CHANGED))
- return -ENOSYS;
+ cdi->options &= ~(CDO_AUTO_CLOSE | CDO_AUTO_EJECT);
+ if (arg)
+ cdi->options |= CDO_AUTO_CLOSE | CDO_AUTO_EJECT;
+ return 0;
+}
- /* cannot select disc or select current disc */
- if (!CDROM_CAN(CDC_SELECT_DISC) || arg == CDSL_CURRENT)
- return media_changed(cdi, 1);
+static int cdrom_ioctl_media_changed(struct cdrom_device_info *cdi,
+ unsigned long arg)
+{
+ struct cdrom_changer_info *info;
+ int ret;
- if ((unsigned int)arg >= cdi->capacity)
- return -EINVAL;
+ cdinfo(CD_DO_IOCTL, "entering CDROM_MEDIA_CHANGED\n");
- info = kmalloc(sizeof(*info), GFP_KERNEL);
- if (!info)
- return -ENOMEM;
+ if (!CDROM_CAN(CDC_MEDIA_CHANGED))
+ return -ENOSYS;
- if ((ret = cdrom_read_mech_status(cdi, info))) {
- kfree(info);
- return ret;
- }
+ /* cannot select disc or select current disc */
+ if (!CDROM_CAN(CDC_SELECT_DISC) || arg == CDSL_CURRENT)
+ return media_changed(cdi, 1);
- changed = info->slots[arg].change;
- kfree(info);
- return changed;
- }
+ if ((unsigned int)arg >= cdi->capacity)
+ return -EINVAL;
- case CDROM_SET_OPTIONS: {
- cdinfo(CD_DO_IOCTL, "entering CDROM_SET_OPTIONS\n");
- /* options need to be in sync with capability. too late for
- that, so we have to check each one separately... */
- switch (arg) {
- case CDO_USE_FFLAGS:
- case CDO_CHECK_TYPE:
- break;
- case CDO_LOCK:
- if (!CDROM_CAN(CDC_LOCK))
- return -ENOSYS;
- break;
- case 0:
- return cdi->options;
- /* default is basically CDO_[AUTO_CLOSE|AUTO_EJECT] */
- default:
- if (!CDROM_CAN(arg))
- return -ENOSYS;
- }
- cdi->options |= (int) arg;
- return cdi->options;
- }
+ info = kmalloc(sizeof(*info), GFP_KERNEL);
+ if (!info)
+ return -ENOMEM;
- case CDROM_CLEAR_OPTIONS: {
- cdinfo(CD_DO_IOCTL, "entering CDROM_CLEAR_OPTIONS\n");
- cdi->options &= ~(int) arg;
- return cdi->options;
- }
+ ret = cdrom_read_mech_status(cdi, info);
+ if (!ret)
+ ret = info->slots[arg].change;
+ kfree(info);
+ return ret;
+}
- case CDROM_SELECT_SPEED: {
- cdinfo(CD_DO_IOCTL, "entering CDROM_SELECT_SPEED\n");
- if (!CDROM_CAN(CDC_SELECT_SPEED))
- return -ENOSYS;
- return cdo->select_speed(cdi, arg);
- }
+static int cdrom_ioctl_set_options(struct cdrom_device_info *cdi,
+ unsigned long arg)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROM_SET_OPTIONS\n");
- case CDROM_SELECT_DISC: {
- cdinfo(CD_DO_IOCTL, "entering CDROM_SELECT_DISC\n");
- if (!CDROM_CAN(CDC_SELECT_DISC))
+ /*
+ * Options need to be in sync with capability.
+ * Too late for that, so we have to check each one separately.
+ */
+ switch (arg) {
+ case CDO_USE_FFLAGS:
+ case CDO_CHECK_TYPE:
+ break;
+ case CDO_LOCK:
+ if (!CDROM_CAN(CDC_LOCK))
+ return -ENOSYS;
+ break;
+ case 0:
+ return cdi->options;
+ /* default is basically CDO_[AUTO_CLOSE|AUTO_EJECT] */
+ default:
+ if (!CDROM_CAN(arg))
return -ENOSYS;
+ }
+ cdi->options |= (int) arg;
+ return cdi->options;
+}
- if ((arg != CDSL_CURRENT) && (arg != CDSL_NONE))
- if ((int)arg >= cdi->capacity)
- return -EINVAL;
-
- /* cdo->select_disc is a hook to allow a driver-specific
- * way of seleting disc. However, since there is no
- * equiv hook for cdrom_slot_status this may not
- * actually be useful...
- */
- if (cdo->select_disc != NULL)
- return cdo->select_disc(cdi, arg);
-
- /* no driver specific select_disc(), call our own */
- cdinfo(CD_CHANGER, "Using generic cdrom_select_disc()\n");
- return cdrom_select_disc(cdi, arg);
- }
+static int cdrom_ioctl_clear_options(struct cdrom_device_info *cdi,
+ unsigned long arg)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROM_CLEAR_OPTIONS\n");
- case CDROMRESET: {
- if (!capable(CAP_SYS_ADMIN))
- return -EACCES;
- cdinfo(CD_DO_IOCTL, "entering CDROM_RESET\n");
- if (!CDROM_CAN(CDC_RESET))
- return -ENOSYS;
- invalidate_bdev(ip->i_bdev, 0);
- return cdo->reset(cdi);
- }
+ cdi->options &= ~(int) arg;
+ return cdi->options;
+}
- case CDROM_LOCKDOOR: {
- cdinfo(CD_DO_IOCTL, "%socking door.\n", arg ? "L" : "Unl");
- if (!CDROM_CAN(CDC_LOCK))
- return -EDRIVE_CANT_DO_THIS;
- keeplocked = arg ? 1 : 0;
- /* don't unlock the door on multiple opens,but allow root
- * to do so */
- if ((cdi->use_count != 1) && !arg && !capable(CAP_SYS_ADMIN))
- return -EBUSY;
- return cdo->lock_door(cdi, arg);
- }
+static int cdrom_ioctl_select_speed(struct cdrom_device_info *cdi,
+ unsigned long arg)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROM_SELECT_SPEED\n");
- case CDROM_DEBUG: {
- if (!capable(CAP_SYS_ADMIN))
- return -EACCES;
- cdinfo(CD_DO_IOCTL, "%sabling debug.\n", arg ? "En" : "Dis");
- debug = arg ? 1 : 0;
- return debug;
- }
+ if (!CDROM_CAN(CDC_SELECT_SPEED))
+ return -ENOSYS;
+ return cdi->ops->select_speed(cdi, arg);
+}
- case CDROM_GET_CAPABILITY: {
- cdinfo(CD_DO_IOCTL, "entering CDROM_GET_CAPABILITY\n");
- return (cdo->capability & ~cdi->mask);
- }
+static int cdrom_ioctl_select_disc(struct cdrom_device_info *cdi,
+ unsigned long arg)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROM_SELECT_DISC\n");
+
+ if (!CDROM_CAN(CDC_SELECT_DISC))
+ return -ENOSYS;
+
+ if (arg != CDSL_CURRENT && arg != CDSL_NONE) {
+ if ((int)arg >= cdi->capacity)
+ return -EINVAL;
+ }
+
+ /*
+ * ->select_disc is a hook to allow a driver-specific way of
+ * seleting disc. However, since there is no equivalent hook for
+ * cdrom_slot_status this may not actually be useful...
+ */
+ if (cdi->ops->select_disc)
+ return cdi->ops->select_disc(cdi, arg);
+
+ cdinfo(CD_CHANGER, "Using generic cdrom_select_disc()\n");
+ return cdrom_select_disc(cdi, arg);
+}
+
+static int cdrom_ioctl_reset(struct cdrom_device_info *cdi,
+ struct block_device *bdev)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROM_RESET\n");
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -EACCES;
+ if (!CDROM_CAN(CDC_RESET))
+ return -ENOSYS;
+ invalidate_bdev(bdev, 0);
+ return cdi->ops->reset(cdi);
+}
+
+static int cdrom_ioctl_lock_door(struct cdrom_device_info *cdi,
+ unsigned long arg)
+{
+ cdinfo(CD_DO_IOCTL, "%socking door.\n", arg ? "L" : "Unl");
+
+ if (!CDROM_CAN(CDC_LOCK))
+ return -EDRIVE_CANT_DO_THIS;
+
+ keeplocked = arg ? 1 : 0;
+
+ /*
+ * Don't unlock the door on multiple opens by default, but allow
+ * root to do so.
+ */
+ if (cdi->use_count != 1 && !arg && !capable(CAP_SYS_ADMIN))
+ return -EBUSY;
+ return cdi->ops->lock_door(cdi, arg);
+}
+
+static int cdrom_ioctl_debug(struct cdrom_device_info *cdi,
+ unsigned long arg)
+{
+ cdinfo(CD_DO_IOCTL, "%sabling debug.\n", arg ? "En" : "Dis");
+
+ if (!capable(CAP_SYS_ADMIN))
+ return -EACCES;
+ debug = arg ? 1 : 0;
+ return debug;
+}
-/* The following function is implemented, although very few audio
+static int cdrom_ioctl_get_capability(struct cdrom_device_info *cdi)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROM_GET_CAPABILITY\n");
+ return (cdi->ops->capability & ~cdi->mask);
+}
+
+/*
+ * The following function is implemented, although very few audio
* discs give Universal Product Code information, which should just be
* the Medium Catalog Number on the box. Note, that the way the code
* is written on the CD is /not/ uniform across all discs!
*/
- case CDROM_GET_MCN: {
- struct cdrom_mcn mcn;
- cdinfo(CD_DO_IOCTL, "entering CDROM_GET_MCN\n");
- if (!(cdo->capability & CDC_MCN))
- return -ENOSYS;
- if ((ret=cdo->get_mcn(cdi, &mcn)))
- return ret;
- IOCTL_OUT(arg, struct cdrom_mcn, mcn);
- cdinfo(CD_DO_IOCTL, "CDROM_GET_MCN successful\n");
- return 0;
- }
+static int cdrom_ioctl_get_mcn(struct cdrom_device_info *cdi,
+ void __user *argp)
+{
+ struct cdrom_mcn mcn;
+ int ret;
- case CDROM_DRIVE_STATUS: {
- cdinfo(CD_DO_IOCTL, "entering CDROM_DRIVE_STATUS\n");
- if (!(cdo->capability & CDC_DRIVE_STATUS))
- return -ENOSYS;
- if (!CDROM_CAN(CDC_SELECT_DISC))
- return cdo->drive_status(cdi, CDSL_CURRENT);
- if ((arg == CDSL_CURRENT) || (arg == CDSL_NONE))
- return cdo->drive_status(cdi, CDSL_CURRENT);
- if (((int)arg >= cdi->capacity))
- return -EINVAL;
- return cdrom_slot_status(cdi, arg);
- }
+ cdinfo(CD_DO_IOCTL, "entering CDROM_GET_MCN\n");
- /* Ok, this is where problems start. The current interface for the
- CDROM_DISC_STATUS ioctl is flawed. It makes the false assumption
- that CDs are all CDS_DATA_1 or all CDS_AUDIO, etc. Unfortunatly,
- while this is often the case, it is also very common for CDs to
- have some tracks with data, and some tracks with audio. Just
- because I feel like it, I declare the following to be the best
- way to cope. If the CD has ANY data tracks on it, it will be
- returned as a data CD. If it has any XA tracks, I will return
- it as that. Now I could simplify this interface by combining these
- returns with the above, but this more clearly demonstrates
- the problem with the current interface. Too bad this wasn't
- designed to use bitmasks... -Erik
-
- Well, now we have the option CDS_MIXED: a mixed-type CD.
- User level programmers might feel the ioctl is not very useful.
- ---david
- */
- case CDROM_DISC_STATUS: {
- tracktype tracks;
- cdinfo(CD_DO_IOCTL, "entering CDROM_DISC_STATUS\n");
- cdrom_count_tracks(cdi, &tracks);
- if (tracks.error)
- return(tracks.error);
-
- /* Policy mode on */
- if (tracks.audio > 0) {
- if (tracks.data==0 && tracks.cdi==0 && tracks.xa==0)
- return CDS_AUDIO;
- else
- return CDS_MIXED;
- }
- if (tracks.cdi > 0) return CDS_XA_2_2;
- if (tracks.xa > 0) return CDS_XA_2_1;
- if (tracks.data > 0) return CDS_DATA_1;
- /* Policy mode off */
+ if (!(cdi->ops->capability & CDC_MCN))
+ return -ENOSYS;
+ ret = cdi->ops->get_mcn(cdi, &mcn);
+ if (ret)
+ return ret;
- cdinfo(CD_WARNING,"This disc doesn't have any tracks I recognize!\n");
- return CDS_NO_INFO;
- }
+ if (copy_to_user(argp, &mcn, sizeof(mcn)))
+ return -EFAULT;
+ cdinfo(CD_DO_IOCTL, "CDROM_GET_MCN successful\n");
+ return 0;
+}
- case CDROM_CHANGER_NSLOTS: {
- cdinfo(CD_DO_IOCTL, "entering CDROM_CHANGER_NSLOTS\n");
- return cdi->capacity;
- }
+static int cdrom_ioctl_drive_status(struct cdrom_device_info *cdi,
+ unsigned long arg)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROM_DRIVE_STATUS\n");
+
+ if (!(cdi->ops->capability & CDC_DRIVE_STATUS))
+ return -ENOSYS;
+ if (!CDROM_CAN(CDC_SELECT_DISC) ||
+ (arg == CDSL_CURRENT || arg == CDSL_NONE))
+ return cdi->ops->drive_status(cdi, CDSL_CURRENT);
+ if (((int)arg >= cdi->capacity))
+ return -EINVAL;
+ return cdrom_slot_status(cdi, arg);
+}
+
+/*
+ * Ok, this is where problems start. The current interface for the
+ * CDROM_DISC_STATUS ioctl is flawed. It makes the false assumption that
+ * CDs are all CDS_DATA_1 or all CDS_AUDIO, etc. Unfortunatly, while this
+ * is often the case, it is also very common for CDs to have some tracks
+ * with data, and some tracks with audio. Just because I feel like it,
+ * I declare the following to be the best way to cope. If the CD has ANY
+ * data tracks on it, it will be returned as a data CD. If it has any XA
+ * tracks, I will return it as that. Now I could simplify this interface
+ * by combining these returns with the above, but this more clearly
+ * demonstrates the problem with the current interface. Too bad this
+ * wasn't designed to use bitmasks... -Erik
+ *
+ * Well, now we have the option CDS_MIXED: a mixed-type CD.
+ * User level programmers might feel the ioctl is not very useful.
+ * ---david
+ */
+static int cdrom_ioctl_disc_status(struct cdrom_device_info *cdi)
+{
+ tracktype tracks;
+
+ cdinfo(CD_DO_IOCTL, "entering CDROM_DISC_STATUS\n");
+
+ cdrom_count_tracks(cdi, &tracks);
+ if (tracks.error)
+ return tracks.error;
+
+ /* Policy mode on */
+ if (tracks.audio > 0) {
+ if (!tracks.data && !tracks.cdi && !tracks.xa)
+ return CDS_AUDIO;
+ else
+ return CDS_MIXED;
}
- /* use the ioctls that are implemented through the generic_packet()
- interface. this may look at bit funny, but if -ENOTTY is
- returned that particular ioctl is not implemented and we
- let it go through the device specific ones. */
+ if (tracks.cdi > 0)
+ return CDS_XA_2_2;
+ if (tracks.xa > 0)
+ return CDS_XA_2_1;
+ if (tracks.data > 0)
+ return CDS_DATA_1;
+ /* Policy mode off */
+
+ cdinfo(CD_WARNING,"This disc doesn't have any tracks I recognize!\n");
+ return CDS_NO_INFO;
+}
+
+static int cdrom_ioctl_changer_nslots(struct cdrom_device_info *cdi)
+{
+ cdinfo(CD_DO_IOCTL, "entering CDROM_CHANGER_NSLOTS\n");
+ return cdi->capacity;
+}
+
+static int cdrom_ioctl_get_subchnl(struct cdrom_device_info *cdi,
+ void __user *argp)
+{
+ struct cdrom_subchnl q;
+ u8 requested, back;
+ int ret;
+
+ /* cdinfo(CD_DO_IOCTL,"entering CDROMSUBCHNL\n");*/
+
+ if (!CDROM_CAN(CDC_PLAY_AUDIO))
+ return -ENOSYS;
+ if (copy_from_user(&q, argp, sizeof(q)))
+ return -EFAULT;
+
+ requested = q.cdsc_format;
+ if (requested != CDROM_MSF && requested != CDROM_LBA)
+ return -EINVAL;
+ q.cdsc_format = CDROM_MSF;
+
+ ret = cdi->ops->audio_ioctl(cdi, CDROMSUBCHNL, &q);
+ if (ret)
+ return ret;
+
+ back = q.cdsc_format; /* local copy */
+ sanitize_format(&q.cdsc_absaddr, &back, requested);
+ sanitize_format(&q.cdsc_reladdr, &q.cdsc_format, requested);
+
+ if (copy_to_user(argp, &q, sizeof(q)))
+ return -EFAULT;
+ /* cdinfo(CD_DO_IOCTL, "CDROMSUBCHNL successful\n"); */
+ return 0;
+}
+
+static int cdrom_ioctl_read_tochdr(struct cdrom_device_info *cdi,
+ void __user *argp)
+{
+ struct cdrom_tochdr header;
+ int ret;
+
+ /* cdinfo(CD_DO_IOCTL, "entering CDROMREADTOCHDR\n"); */
+
+ if (!CDROM_CAN(CDC_PLAY_AUDIO))
+ return -ENOSYS;
+ if (copy_from_user(&header, argp, sizeof(header)))
+ return -EFAULT;
+
+ ret = cdi->ops->audio_ioctl(cdi, CDROMREADTOCHDR, &header);
+ if (ret)
+ return ret;
+
+ if (copy_to_user(argp, &header, sizeof(header)))
+ return -EFAULT;
+ /* cdinfo(CD_DO_IOCTL, "CDROMREADTOCHDR successful\n"); */
+ return 0;
+}
+
+static int cdrom_ioctl_read_tocentry(struct cdrom_device_info *cdi,
+ void __user *argp)
+{
+ struct cdrom_tocentry entry;
+ u8 requested_format;
+ int ret;
+
+ /* cdinfo(CD_DO_IOCTL, "entering CDROMREADTOCENTRY\n"); */
+
+ if (!CDROM_CAN(CDC_PLAY_AUDIO))
+ return -ENOSYS;
+ if (copy_from_user(&entry, argp, sizeof(entry)))
+ return -EFAULT;
+
+ requested_format = entry.cdte_format;
+ if (requested_format != CDROM_MSF && requested_format != CDROM_LBA)
+ return -EINVAL;
+ /* make interface to low-level uniform */
+ entry.cdte_format = CDROM_MSF;
+ ret = cdi->ops->audio_ioctl(cdi, CDROMREADTOCENTRY, &entry);
+ if (ret)
+ return ret;
+ sanitize_format(&entry.cdte_addr, &entry.cdte_format, requested_format);
+
+ if (copy_to_user(argp, &entry, sizeof(entry)))
+ return -EFAULT;
+ /* cdinfo(CD_DO_IOCTL, "CDROMREADTOCENTRY successful\n"); */
+ return 0;
+}
+
+static int cdrom_ioctl_play_msf(struct cdrom_device_info *cdi,
+ void __user *argp)
+{
+ struct cdrom_msf msf;
+
+ cdinfo(CD_DO_IOCTL, "entering CDROMPLAYMSF\n");
+
+ if (!CDROM_CAN(CDC_PLAY_AUDIO))
+ return -ENOSYS;
+ if (copy_from_user(&msf, argp, sizeof(msf)))
+ return -EFAULT;
+ return cdi->ops->audio_ioctl(cdi, CDROMPLAYMSF, &msf);
+}
+
+static int cdrom_ioctl_play_trkind(struct cdrom_device_info *cdi,
+ void __user *argp)
+{
+ struct cdrom_ti ti;
+ int ret;
+
+ cdinfo(CD_DO_IOCTL, "entering CDROMPLAYTRKIND\n");
+
+ if (!CDROM_CAN(CDC_PLAY_AUDIO))
+ return -ENOSYS;
+ if (copy_from_user(&ti, argp, sizeof(ti)))
+ return -EFAULT;
+
+ ret = check_for_audio_disc(cdi, cdi->ops);
+ if (ret)
+ return ret;
+ return cdi->ops->audio_ioctl(cdi, CDROMPLAYTRKIND, &ti);
+}
+static int cdrom_ioctl_volctrl(struct cdrom_device_info *cdi,
+ void __user *argp)
+{
+ struct cdrom_volctrl volume;
+
+ cdinfo(CD_DO_IOCTL, "entering CDROMVOLCTRL\n");
+
+ if (!CDROM_CAN(CDC_PLAY_AUDIO))
+ return -ENOSYS;
+ if (copy_from_user(&volume, argp, sizeof(volume)))
+ return -EFAULT;
+ return cdi->ops->audio_ioctl(cdi, CDROMVOLCTRL, &volume);
+}
+
+static int cdrom_ioctl_volread(struct cdrom_device_info *cdi,
+ void __user *argp)
+{
+ struct cdrom_volctrl volume;
+ int ret;
+
+ cdinfo(CD_DO_IOCTL, "entering CDROMVOLREAD\n");
+
+ if (!CDROM_CAN(CDC_PLAY_AUDIO))
+ return -ENOSYS;
+
+ ret = cdi->ops->audio_ioctl(cdi, CDROMVOLREAD, &volume);
+ if (ret)
+ return ret;
+
+ if (copy_to_user(argp, &volume, sizeof(volume)))
+ return -EFAULT;
+ return 0;
+}
+
+static int cdrom_ioctl_audioctl(struct cdrom_device_info *cdi,
+ unsigned int cmd)
+{
+ int ret;
+
+ cdinfo(CD_DO_IOCTL, "doing audio ioctl (start/stop/pause/resume)\n");
+
+ if (!CDROM_CAN(CDC_PLAY_AUDIO))
+ return -ENOSYS;
+ ret = check_for_audio_disc(cdi, cdi->ops);
+ if (ret)
+ return ret;
+ return cdi->ops->audio_ioctl(cdi, cmd, NULL);
+}
+
+/*
+ * Just about every imaginable ioctl is supported in the Uniform layer
+ * these days.
+ * ATAPI / SCSI specific code now mainly resides in mmc_ioctl().
+ */
+int cdrom_ioctl(struct file * file, struct cdrom_device_info *cdi,
+ struct inode *ip, unsigned int cmd, unsigned long arg)
+{
+ void __user *argp = (void __user *)arg;
+ int ret;
+
+ /*
+ * Try the generic SCSI command ioctl's first.
+ */
+ ret = scsi_cmd_ioctl(file, ip->i_bdev->bd_disk, cmd, argp);
+ if (ret != -ENOTTY)
+ return ret;
+
+ switch (cmd) {
+ case CDROMMULTISESSION:
+ return cdrom_ioctl_multisession(cdi, argp);
+ case CDROMEJECT:
+ return cdrom_ioctl_eject(cdi);
+ case CDROMCLOSETRAY:
+ return cdrom_ioctl_closetray(cdi);
+ case CDROMEJECT_SW:
+ return cdrom_ioctl_eject_sw(cdi, arg);
+ case CDROM_MEDIA_CHANGED:
+ return cdrom_ioctl_media_changed(cdi, arg);
+ case CDROM_SET_OPTIONS:
+ return cdrom_ioctl_set_options(cdi, arg);
+ case CDROM_CLEAR_OPTIONS:
+ return cdrom_ioctl_clear_options(cdi, arg);
+ case CDROM_SELECT_SPEED:
+ return cdrom_ioctl_select_speed(cdi, arg);
+ case CDROM_SELECT_DISC:
+ return cdrom_ioctl_select_disc(cdi, arg);
+ case CDROMRESET:
+ return cdrom_ioctl_reset(cdi, ip->i_bdev);
+ case CDROM_LOCKDOOR:
+ return cdrom_ioctl_lock_door(cdi, arg);
+ case CDROM_DEBUG:
+ return cdrom_ioctl_debug(cdi, arg);
+ case CDROM_GET_CAPABILITY:
+ return cdrom_ioctl_get_capability(cdi);
+ case CDROM_GET_MCN:
+ return cdrom_ioctl_get_mcn(cdi, argp);
+ case CDROM_DRIVE_STATUS:
+ return cdrom_ioctl_drive_status(cdi, arg);
+ case CDROM_DISC_STATUS:
+ return cdrom_ioctl_disc_status(cdi);
+ case CDROM_CHANGER_NSLOTS:
+ return cdrom_ioctl_changer_nslots(cdi);
+ }
+
+ /*
+ * Use the ioctls that are implemented through the generic_packet()
+ * interface. this may look at bit funny, but if -ENOTTY is
+ * returned that particular ioctl is not implemented and we
+ * let it go through the device specific ones.
+ */
if (CDROM_CAN(CDC_GENERIC_PACKET)) {
ret = mmc_ioctl(cdi, cmd, arg);
- if (ret != -ENOTTY) {
+ if (ret != -ENOTTY)
return ret;
- }
}
- /* note: most of the cdinfo() calls are commented out here,
- because they fill up the sys log when CD players poll
- the drive. */
+ /*
+ * Note: most of the cdinfo() calls are commented out here,
+ * because they fill up the sys log when CD players poll
+ * the drive.
+ */
switch (cmd) {
- case CDROMSUBCHNL: {
- struct cdrom_subchnl q;
- u_char requested, back;
- if (!CDROM_CAN(CDC_PLAY_AUDIO))
- return -ENOSYS;
- /* cdinfo(CD_DO_IOCTL,"entering CDROMSUBCHNL\n");*/
- IOCTL_IN(arg, struct cdrom_subchnl, q);
- requested = q.cdsc_format;
- if (!((requested == CDROM_MSF) ||
- (requested == CDROM_LBA)))
- return -EINVAL;
- q.cdsc_format = CDROM_MSF;
- if ((ret=cdo->audio_ioctl(cdi, cmd, &q)))
- return ret;
- back = q.cdsc_format; /* local copy */
- sanitize_format(&q.cdsc_absaddr, &back, requested);
- sanitize_format(&q.cdsc_reladdr, &q.cdsc_format, requested);
- IOCTL_OUT(arg, struct cdrom_subchnl, q);
- /* cdinfo(CD_DO_IOCTL, "CDROMSUBCHNL successful\n"); */
- return 0;
- }
- case CDROMREADTOCHDR: {
- struct cdrom_tochdr header;
- if (!CDROM_CAN(CDC_PLAY_AUDIO))
- return -ENOSYS;
- /* cdinfo(CD_DO_IOCTL, "entering CDROMREADTOCHDR\n"); */
- IOCTL_IN(arg, struct cdrom_tochdr, header);
- if ((ret=cdo->audio_ioctl(cdi, cmd, &header)))
- return ret;
- IOCTL_OUT(arg, struct cdrom_tochdr, header);
- /* cdinfo(CD_DO_IOCTL, "CDROMREADTOCHDR successful\n"); */
- return 0;
- }
- case CDROMREADTOCENTRY: {
- struct cdrom_tocentry entry;
- u_char requested_format;
- if (!CDROM_CAN(CDC_PLAY_AUDIO))
- return -ENOSYS;
- /* cdinfo(CD_DO_IOCTL, "entering CDROMREADTOCENTRY\n"); */
- IOCTL_IN(arg, struct cdrom_tocentry, entry);
- requested_format = entry.cdte_format;
- if (!((requested_format == CDROM_MSF) ||
- (requested_format == CDROM_LBA)))
- return -EINVAL;
- /* make interface to low-level uniform */
- entry.cdte_format = CDROM_MSF;
- if ((ret=cdo->audio_ioctl(cdi, cmd, &entry)))
- return ret;
- sanitize_format(&entry.cdte_addr,
- &entry.cdte_format, requested_format);
- IOCTL_OUT(arg, struct cdrom_tocentry, entry);
- /* cdinfo(CD_DO_IOCTL, "CDROMREADTOCENTRY successful\n"); */
- return 0;
- }
- case CDROMPLAYMSF: {
- struct cdrom_msf msf;
- if (!CDROM_CAN(CDC_PLAY_AUDIO))
- return -ENOSYS;
- cdinfo(CD_DO_IOCTL, "entering CDROMPLAYMSF\n");
- IOCTL_IN(arg, struct cdrom_msf, msf);
- return cdo->audio_ioctl(cdi, cmd, &msf);
- }
- case CDROMPLAYTRKIND: {
- struct cdrom_ti ti;
- if (!CDROM_CAN(CDC_PLAY_AUDIO))
- return -ENOSYS;
- cdinfo(CD_DO_IOCTL, "entering CDROMPLAYTRKIND\n");
- IOCTL_IN(arg, struct cdrom_ti, ti);
- CHECKAUDIO;
- return cdo->audio_ioctl(cdi, cmd, &ti);
- }
- case CDROMVOLCTRL: {
- struct cdrom_volctrl volume;
- if (!CDROM_CAN(CDC_PLAY_AUDIO))
- return -ENOSYS;
- cdinfo(CD_DO_IOCTL, "entering CDROMVOLCTRL\n");
- IOCTL_IN(arg, struct cdrom_volctrl, volume);
- return cdo->audio_ioctl(cdi, cmd, &volume);
- }
- case CDROMVOLREAD: {
- struct cdrom_volctrl volume;
- if (!CDROM_CAN(CDC_PLAY_AUDIO))
- return -ENOSYS;
- cdinfo(CD_DO_IOCTL, "entering CDROMVOLREAD\n");
- if ((ret=cdo->audio_ioctl(cdi, cmd, &volume)))
- return ret;
- IOCTL_OUT(arg, struct cdrom_volctrl, volume);
- return 0;
- }
+ case CDROMSUBCHNL:
+ return cdrom_ioctl_get_subchnl(cdi, argp);
+ case CDROMREADTOCHDR:
+ return cdrom_ioctl_read_tochdr(cdi, argp);
+ case CDROMREADTOCENTRY:
+ return cdrom_ioctl_read_tocentry(cdi, argp);
+ case CDROMPLAYMSF:
+ return cdrom_ioctl_play_msf(cdi, argp);
+ case CDROMPLAYTRKIND:
+ return cdrom_ioctl_play_trkind(cdi, argp);
+ case CDROMVOLCTRL:
+ return cdrom_ioctl_volctrl(cdi, argp);
+ case CDROMVOLREAD:
+ return cdrom_ioctl_volread(cdi, argp);
case CDROMSTART:
case CDROMSTOP:
case CDROMPAUSE:
- case CDROMRESUME: {
- if (!CDROM_CAN(CDC_PLAY_AUDIO))
- return -ENOSYS;
- cdinfo(CD_DO_IOCTL, "doing audio ioctl (start/stop/pause/resume)\n");
- CHECKAUDIO;
- return cdo->audio_ioctl(cdi, cmd, NULL);
- }
- } /* switch */
+ case CDROMRESUME:
+ return cdrom_ioctl_audioctl(cdi, cmd);
+ }
- /* do the device specific ioctls */
- if (CDROM_CAN(CDC_IOCTLS))
- return cdo->dev_ioctl(cdi, cmd, arg);
-
return -ENOSYS;
}
diff --git a/drivers/cdrom/cdu31a.c b/drivers/cdrom/cdu31a.c
index 378e88d20757..72ffd64e8b1e 100644
--- a/drivers/cdrom/cdu31a.c
+++ b/drivers/cdrom/cdu31a.c
@@ -2668,7 +2668,7 @@ static int scd_audio_ioctl(struct cdrom_device_info *cdi,
return retval;
}
-static int scd_dev_ioctl(struct cdrom_device_info *cdi,
+static int scd_read_audio(struct cdrom_device_info *cdi,
unsigned int cmd, unsigned long arg)
{
void __user *argp = (void __user *)arg;
@@ -2894,11 +2894,10 @@ static struct cdrom_device_ops scd_dops = {
.get_mcn = scd_get_mcn,
.reset = scd_reset,
.audio_ioctl = scd_audio_ioctl,
- .dev_ioctl = scd_dev_ioctl,
.capability = CDC_OPEN_TRAY | CDC_CLOSE_TRAY | CDC_LOCK |
CDC_SELECT_SPEED | CDC_MULTI_SESSION |
CDC_MCN | CDC_MEDIA_CHANGED | CDC_PLAY_AUDIO |
- CDC_RESET | CDC_IOCTLS | CDC_DRIVE_STATUS,
+ CDC_RESET | CDC_DRIVE_STATUS,
.n_minors = 1,
};
@@ -2936,6 +2935,9 @@ static int scd_block_ioctl(struct inode *inode, struct file *file,
case CDROMCLOSETRAY:
retval = scd_tray_move(&scd_info, 0);
break;
+ case CDROMREADAUDIO:
+ retval = scd_read_audio(&scd_info, CDROMREADAUDIO, arg);
+ break;
default:
retval = cdrom_ioctl(file, &scd_info, inode, cmd, arg);
}
diff --git a/drivers/cdrom/cm206.c b/drivers/cdrom/cm206.c
index ce127f7ec0f6..fad27a87ce35 100644
--- a/drivers/cdrom/cm206.c
+++ b/drivers/cdrom/cm206.c
@@ -1157,32 +1157,6 @@ static int cm206_audio_ioctl(struct cdrom_device_info *cdi, unsigned int cmd,
}
}
-/* Ioctl. These ioctls are specific to the cm206 driver. I have made
- some driver statistics accessible through ioctl calls.
- */
-
-static int cm206_ioctl(struct cdrom_device_info *cdi, unsigned int cmd,
- unsigned long arg)
-{
- switch (cmd) {
-#ifdef STATISTICS
- case CM206CTL_GET_STAT:
- if (arg >= NR_STATS)
- return -EINVAL;
- else
- return cd->stats[arg];
- case CM206CTL_GET_LAST_STAT:
- if (arg >= NR_STATS)
- return -EINVAL;
- else
- return cd->last_stat[arg];
-#endif
- default:
- debug(("Unknown ioctl call 0x%x\n", cmd));
- return -EINVAL;
- }
-}
-
static int cm206_media_changed(struct cdrom_device_info *cdi, int disc_nr)
{
if (cd != NULL) {
@@ -1321,11 +1295,10 @@ static struct cdrom_device_ops cm206_dops = {
.get_mcn = cm206_get_upc,
.reset = cm206_reset,
.audio_ioctl = cm206_audio_ioctl,
- .dev_ioctl = cm206_ioctl,
.capability = CDC_CLOSE_TRAY | CDC_OPEN_TRAY | CDC_LOCK |
CDC_MULTI_SESSION | CDC_MEDIA_CHANGED |
CDC_MCN | CDC_PLAY_AUDIO | CDC_SELECT_SPEED |
- CDC_IOCTLS | CDC_DRIVE_STATUS,
+ CDC_DRIVE_STATUS,
.n_minors = 1,
};
@@ -1350,6 +1323,21 @@ static int cm206_block_release(struct inode *inode, struct file *file)
static int cm206_block_ioctl(struct inode *inode, struct file *file,
unsigned cmd, unsigned long arg)
{
+ switch (cmd) {
+#ifdef STATISTICS
+ case CM206CTL_GET_STAT:
+ if (arg >= NR_STATS)
+ return -EINVAL;
+ return cd->stats[arg];
+ case CM206CTL_GET_LAST_STAT:
+ if (arg >= NR_STATS)
+ return -EINVAL;
+ return cd->last_stat[arg];
+#endif
+ default:
+ break;
+ }
+
return cdrom_ioctl(file, &cm206_info, inode, cmd, arg);
}
diff --git a/drivers/cdrom/sbpcd.c b/drivers/cdrom/sbpcd.c
index 466e9c2974bd..4760f515f591 100644
--- a/drivers/cdrom/sbpcd.c
+++ b/drivers/cdrom/sbpcd.c
@@ -4160,332 +4160,6 @@ static int sbpcd_get_last_session(struct cdrom_device_info *cdi, struct cdrom_mu
return 0;
}
-/*==========================================================================*/
-/*==========================================================================*/
-/*
- * ioctl support
- */
-static int sbpcd_dev_ioctl(struct cdrom_device_info *cdi, u_int cmd,
- u_long arg)
-{
- struct sbpcd_drive *p = cdi->handle;
- int i;
-
- msg(DBG_IO2,"ioctl(%s, 0x%08lX, 0x%08lX)\n", cdi->name, cmd, arg);
- if (p->drv_id==-1) {
- msg(DBG_INF, "ioctl: bad device: %s\n", cdi->name);
- return (-ENXIO); /* no such drive */
- }
- down(&ioctl_read_sem);
- if (p != current_drive)
- switch_drive(p);
-
- msg(DBG_IO2,"ioctl: device %s, request %04X\n",cdi->name,cmd);
- switch (cmd) /* Sun-compatible */
- {
- case DDIOCSDBG: /* DDI Debug */
- if (!capable(CAP_SYS_ADMIN)) RETURN_UP(-EPERM);
- i=sbpcd_dbg_ioctl(arg,1);
- RETURN_UP(i);
- case CDROMRESET: /* hard reset the drive */
- msg(DBG_IOC,"ioctl: CDROMRESET entered.\n");
- i=DriveReset();
- current_drive->audio_state=0;
- RETURN_UP(i);
-
- case CDROMREADMODE1:
- msg(DBG_IOC,"ioctl: CDROMREADMODE1 requested.\n");
-#ifdef SAFE_MIXED
- if (current_drive->has_data>1) RETURN_UP(-EBUSY);
-#endif /* SAFE_MIXED */
- cc_ModeSelect(CD_FRAMESIZE);
- cc_ModeSense();
- current_drive->mode=READ_M1;
- RETURN_UP(0);
-
- case CDROMREADMODE2: /* not usable at the moment */
- msg(DBG_IOC,"ioctl: CDROMREADMODE2 requested.\n");
-#ifdef SAFE_MIXED
- if (current_drive->has_data>1) RETURN_UP(-EBUSY);
-#endif /* SAFE_MIXED */
- cc_ModeSelect(CD_FRAMESIZE_RAW1);
- cc_ModeSense();
- current_drive->mode=READ_M2;
- RETURN_UP(0);
-
- case CDROMAUDIOBUFSIZ: /* configure the audio buffer size */
- msg(DBG_IOC,"ioctl: CDROMAUDIOBUFSIZ entered.\n");
- if (current_drive->sbp_audsiz>0)
- vfree(current_drive->aud_buf);
- current_drive->aud_buf=NULL;
- current_drive->sbp_audsiz=arg;
-
- if (current_drive->sbp_audsiz>16)
- {
- current_drive->sbp_audsiz = 0;
- RETURN_UP(current_drive->sbp_audsiz);
- }
-
- if (current_drive->sbp_audsiz>0)
- {
- current_drive->aud_buf=(u_char *) vmalloc(current_drive->sbp_audsiz*CD_FRAMESIZE_RAW);
- if (current_drive->aud_buf==NULL)
- {
- msg(DBG_INF,"audio buffer (%d frames) not available.\n",current_drive->sbp_audsiz);
- current_drive->sbp_audsiz=0;
- }
- else msg(DBG_INF,"audio buffer size: %d frames.\n",current_drive->sbp_audsiz);
- }
- RETURN_UP(current_drive->sbp_audsiz);
-
- case CDROMREADAUDIO:
- { /* start of CDROMREADAUDIO */
- int i=0, j=0, frame, block=0;
- u_int try=0;
- u_long timeout;
- u_char *p;
- u_int data_tries = 0;
- u_int data_waits = 0;
- u_int data_retrying = 0;
- int status_tries;
- int error_flag;
-
- msg(DBG_IOC,"ioctl: CDROMREADAUDIO entered.\n");
- if (fam0_drive) RETURN_UP(-EINVAL);
- if (famL_drive) RETURN_UP(-EINVAL);
- if (famV_drive) RETURN_UP(-EINVAL);
- if (famT_drive) RETURN_UP(-EINVAL);
-#ifdef SAFE_MIXED
- if (current_drive->has_data>1) RETURN_UP(-EBUSY);
-#endif /* SAFE_MIXED */
- if (current_drive->aud_buf==NULL) RETURN_UP(-EINVAL);
- if (copy_from_user(&read_audio, (void __user *)arg,
- sizeof(struct cdrom_read_audio)))
- RETURN_UP(-EFAULT);
- if (read_audio.nframes < 0 || read_audio.nframes>current_drive->sbp_audsiz) RETURN_UP(-EINVAL);
- if (!access_ok(VERIFY_WRITE, read_audio.buf,
- read_audio.nframes*CD_FRAMESIZE_RAW))
- RETURN_UP(-EFAULT);
-
- if (read_audio.addr_format==CDROM_MSF) /* MSF-bin specification of where to start */
- block=msf2lba(&read_audio.addr.msf.minute);
- else if (read_audio.addr_format==CDROM_LBA) /* lba specification of where to start */
- block=read_audio.addr.lba;
- else RETURN_UP(-EINVAL);
-#if 000
- i=cc_SetSpeed(speed_150,0,0);
- if (i) msg(DBG_AUD,"read_audio: SetSpeed error %d\n", i);
-#endif
- msg(DBG_AUD,"read_audio: lba: %d, msf: %06X\n",
- block, blk2msf(block));
- msg(DBG_AUD,"read_audio: before cc_ReadStatus.\n");
-#if OLD_BUSY
- while (busy_data) sbp_sleep(HZ/10); /* wait a bit */
- busy_audio=1;
-#endif /* OLD_BUSY */
- error_flag=0;
- for (data_tries=5; data_tries>0; data_tries--)
- {
- msg(DBG_AUD,"data_tries=%d ...\n", data_tries);
- current_drive->mode=READ_AU;
- cc_ModeSelect(CD_FRAMESIZE_RAW);
- cc_ModeSense();
- for (status_tries=3; status_tries > 0; status_tries--)
- {
- flags_cmd_out |= f_respo3;
- cc_ReadStatus();
- if (sbp_status() != 0) break;
- if (st_check) cc_ReadError();
- sbp_sleep(1); /* wait a bit, try again */
- }
- if (status_tries == 0)
- {
- msg(DBG_AUD,"read_audio: sbp_status: failed after 3 tries in line %d.\n", __LINE__);
- continue;
- }
- msg(DBG_AUD,"read_audio: sbp_status: ok.\n");
-
- flags_cmd_out = f_putcmd | f_respo2 | f_ResponseStatus | f_obey_p_check;
- if (fam0L_drive)
- {
- flags_cmd_out |= f_lopsta | f_getsta | f_bit1;
- cmd_type=READ_M2;
- drvcmd[0]=CMD0_READ_XA; /* "read XA frames", old drives */
- drvcmd[1]=(block>>16)&0x000000ff;
- drvcmd[2]=(block>>8)&0x000000ff;
- drvcmd[3]=block&0x000000ff;
- drvcmd[4]=0;
- drvcmd[5]=read_audio.nframes; /* # of frames */
- drvcmd[6]=0;
- }
- else if (fam1_drive)
- {
- drvcmd[0]=CMD1_READ; /* "read frames", new drives */
- lba2msf(block,&drvcmd[1]); /* msf-bin format required */
- drvcmd[4]=0;
- drvcmd[5]=0;
- drvcmd[6]=read_audio.nframes; /* # of frames */
- }
- else if (fam2_drive)
- {
- drvcmd[0]=CMD2_READ_XA2;
- lba2msf(block,&drvcmd[1]); /* msf-bin format required */
- drvcmd[4]=0;
- drvcmd[5]=read_audio.nframes; /* # of frames */
- drvcmd[6]=0x11; /* raw mode */
- }
- else if (famT_drive) /* CD-55A: not tested yet */
- {
- }
- msg(DBG_AUD,"read_audio: before giving \"read\" command.\n");
- flags_cmd_out=f_putcmd;
- response_count=0;
- i=cmd_out();
- if (i<0) msg(DBG_INF,"error giving READ AUDIO command: %0d\n", i);
- sbp_sleep(0);
- msg(DBG_AUD,"read_audio: after giving \"read\" command.\n");
- for (frame=1;frame<2 && !error_flag; frame++)
- {
- try=maxtim_data;
- for (timeout=jiffies+9*HZ; ; )
- {
- for ( ; try!=0;try--)
- {
- j=inb(CDi_status);
- if (!(j&s_not_data_ready)) break;
- if (!(j&s_not_result_ready)) break;
- if (fam0L_drive) if (j&s_attention) break;
- }
- if (try != 0 || time_after_eq(jiffies, timeout)) break;
- if (data_retrying == 0) data_waits++;
- data_retrying = 1;
- sbp_sleep(1);
- try = 1;
- }
- if (try==0)
- {
- msg(DBG_INF,"read_audio: sbp_data: CDi_status timeout.\n");
- error_flag++;
- break;
- }
- msg(DBG_AUD,"read_audio: sbp_data: CDi_status ok.\n");
- if (j&s_not_data_ready)
- {
- msg(DBG_INF, "read_audio: sbp_data: DATA_READY timeout.\n");
- error_flag++;
- break;
- }
- msg(DBG_AUD,"read_audio: before reading data.\n");
- error_flag=0;
- p = current_drive->aud_buf;
- if (sbpro_type==1) OUT(CDo_sel_i_d,1);
- if (do_16bit)
- {
- u_short *p2 = (u_short *) p;
-
- for (; (u_char *) p2 < current_drive->aud_buf + read_audio.nframes*CD_FRAMESIZE_RAW;)
- {
- if ((inb_p(CDi_status)&s_not_data_ready)) continue;
-
- /* get one sample */
- *p2++ = inw_p(CDi_data);
- *p2++ = inw_p(CDi_data);
- }
- } else {
- for (; p < current_drive->aud_buf + read_audio.nframes*CD_FRAMESIZE_RAW;)
- {
- if ((inb_p(CDi_status)&s_not_data_ready)) continue;
-
- /* get one sample */
- *p++ = inb_p(CDi_data);
- *p++ = inb_p(CDi_data);
- *p++ = inb_p(CDi_data);
- *p++ = inb_p(CDi_data);
- }
- }
- if (sbpro_type==1) OUT(CDo_sel_i_d,0);
- data_retrying = 0;
- }
- msg(DBG_AUD,"read_audio: after reading data.\n");
- if (error_flag) /* must have been spurious D_RDY or (ATTN&&!D_RDY) */
- {
- msg(DBG_AUD,"read_audio: read aborted by drive\n");
-#if 0000
- i=cc_DriveReset(); /* ugly fix to prevent a hang */
-#else
- i=cc_ReadError();
-#endif
- continue;
- }
- if (fam0L_drive)
- {
- i=maxtim_data;
- for (timeout=jiffies+9*HZ; time_before(jiffies, timeout); timeout--)
- {
- for ( ;i!=0;i--)
- {
- j=inb(CDi_status);
- if (!(j&s_not_data_ready)) break;
- if (!(j&s_not_result_ready)) break;
- if (j&s_attention) break;
- }
- if (i != 0 || time_after_eq(jiffies, timeout)) break;
- sbp_sleep(0);
- i = 1;
- }
- if (i==0) msg(DBG_AUD,"read_audio: STATUS TIMEOUT AFTER READ");
- if (!(j&s_attention))
- {
- msg(DBG_AUD,"read_audio: sbp_data: timeout waiting DRV_ATTN - retrying\n");
- i=cc_DriveReset(); /* ugly fix to prevent a hang */
- continue;
- }
- }
- do
- {
- if (fam0L_drive) cc_ReadStatus();
- i=ResponseStatus(); /* builds status_bits, returns orig. status (old) or faked p_success (new) */
- if (i<0) { msg(DBG_AUD,
- "read_audio: cc_ReadStatus error after read: %02X\n",
- current_drive->status_bits);
- continue; /* FIXME */
- }
- }
- while ((fam0L_drive)&&(!st_check)&&(!(i&p_success)));
- if (st_check)
- {
- i=cc_ReadError();
- msg(DBG_AUD,"read_audio: cc_ReadError was necessary after read: %02X\n",i);
- continue;
- }
- if (copy_to_user(read_audio.buf,
- current_drive->aud_buf,
- read_audio.nframes * CD_FRAMESIZE_RAW))
- RETURN_UP(-EFAULT);
- msg(DBG_AUD,"read_audio: copy_to_user done.\n");
- break;
- }
- cc_ModeSelect(CD_FRAMESIZE);
- cc_ModeSense();
- current_drive->mode=READ_M1;
-#if OLD_BUSY
- busy_audio=0;
-#endif /* OLD_BUSY */
- if (data_tries == 0)
- {
- msg(DBG_AUD,"read_audio: failed after 5 tries in line %d.\n", __LINE__);
- RETURN_UP(-EIO);
- }
- msg(DBG_AUD,"read_audio: successful return.\n");
- RETURN_UP(0);
- } /* end of CDROMREADAUDIO */
-
- default:
- msg(DBG_IOC,"ioctl: unknown function request %04X\n", cmd);
- RETURN_UP(-EINVAL);
- } /* end switch(cmd) */
-}
-
static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
void * arg)
{
@@ -4530,7 +4204,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
default:
RETURN_UP(-EINVAL);
}
-
+
case CDROMRESUME: /* resume paused audio play */
msg(DBG_IOC,"ioctl: CDROMRESUME entered.\n");
/* resume playing audio tracks when a previous PLAY AUDIO call has */
@@ -4544,12 +4218,12 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
if (i<0) RETURN_UP(-EIO);
current_drive->audio_state=audio_playing;
RETURN_UP(0);
-
+
case CDROMPLAYMSF:
msg(DBG_IOC,"ioctl: CDROMPLAYMSF entered.\n");
#ifdef SAFE_MIXED
if (current_drive->has_data>1) RETURN_UP(-EBUSY);
-#endif /* SAFE_MIXED */
+#endif /* SAFE_MIXED */
if (current_drive->audio_state==audio_playing)
{
i=cc_Pause_Resume(1);
@@ -4584,7 +4258,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
msg(DBG_IOC,"ioctl: CDROMPLAYTRKIND entered.\n");
#ifdef SAFE_MIXED
if (current_drive->has_data>1) RETURN_UP(-EBUSY);
-#endif /* SAFE_MIXED */
+#endif /* SAFE_MIXED */
if (current_drive->audio_state==audio_playing)
{
msg(DBG_IOX,"CDROMPLAYTRKIND: already audio_playing.\n");
@@ -4654,13 +4328,13 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
cc_DriveReset();
#endif
RETURN_UP(i);
-
+
case CDROMSTART: /* Spin up the drive */
msg(DBG_IOC,"ioctl: CDROMSTART entered.\n");
cc_SpinUp();
current_drive->audio_state=0;
RETURN_UP(0);
-
+
case CDROMVOLCTRL: /* Volume control */
msg(DBG_IOC,"ioctl: CDROMVOLCTRL entered.\n");
memcpy(&volctrl,(char *) arg,sizeof(volctrl));
@@ -4670,7 +4344,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
current_drive->vol_ctrl1=volctrl.channel1;
i=cc_SetVolume();
RETURN_UP(0);
-
+
case CDROMVOLREAD: /* read Volume settings from drive */
msg(DBG_IOC,"ioctl: CDROMVOLREAD entered.\n");
st=cc_GetVolume();
@@ -4694,7 +4368,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
if (i<0) {
j=cc_ReadError(); /* clear out error status from drive */
current_drive->audio_state=CDROM_AUDIO_NO_STATUS;
- /* get and set the disk state here,
+ /* get and set the disk state here,
probably not the right place, but who cares!
It makes it work properly! --AJK */
if (current_drive->CD_changed==0xFF) {
@@ -4715,8 +4389,8 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
}
}
memcpy(&SC, (void *) arg, sizeof(struct cdrom_subchnl));
- /*
- This virtual crap is very bogus!
+ /*
+ This virtual crap is very bogus!
It doesn't detect when the cd is done playing audio!
Lets do this right with proper hardware register reading!
*/
@@ -4775,7 +4449,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
SC.cdsc_trk,SC.cdsc_ind,
SC.cdsc_absaddr,SC.cdsc_reladdr);
RETURN_UP(0);
-
+
default:
msg(DBG_IOC,"ioctl: unknown function request %04X\n", cmd);
RETURN_UP(-EINVAL);
@@ -4788,7 +4462,7 @@ static int sbpcd_audio_ioctl(struct cdrom_device_info *cdi, u_int cmd,
static void sbp_transfer(struct request *req)
{
long offs;
-
+
while ( (req->nr_sectors > 0) &&
(req->sector/4 >= current_drive->sbp_first_frame) &&
(req->sector/4 <= current_drive->sbp_last_frame) )
@@ -4807,11 +4481,11 @@ static void sbp_transfer(struct request *req)
*
* This is a kludge so we don't need to modify end_request.
* We put the req we take out after INIT_REQUEST in the requests list,
- * so that end_request will discard it.
+ * so that end_request will discard it.
*
* The bug could be present in other block devices, perhaps we
* should modify INIT_REQUEST and end_request instead, and
- * change every block device..
+ * change every block device..
*
* Could be a race here?? Could e.g. a timer interrupt schedule() us?
* If so, we should copy end_request here, and do it right.. (or
@@ -4883,19 +4557,19 @@ static void do_sbpcd_request(request_queue_t * q)
while (busy_audio) sbp_sleep(HZ); /* wait a bit */
busy_data=1;
#endif /* OLD_BUSY */
-
+
if (p->audio_state==audio_playing) goto err_done;
if (p != current_drive)
switch_drive(p);
block = req->sector; /* always numbered as 512-byte-pieces */
nsect = req->nr_sectors; /* always counted as 512-byte-pieces */
-
+
msg(DBG_BSZ,"read sector %d (%d sectors)\n", block, nsect);
#if 0
msg(DBG_MUL,"read LBA %d\n", block/4);
#endif
-
+
sbp_transfer(req);
/* if we satisfied the request from the buffer, we're done. */
if (req->nr_sectors == 0)
@@ -4914,10 +4588,10 @@ static void do_sbpcd_request(request_queue_t * q)
i=prepare(0,0); /* at moment not really a hassle check, but ... */
if (i!=0)
msg(DBG_INF,"\"prepare\" tells error %d -- ignored\n", i);
-#endif /* FUTURE */
-
+#endif /* FUTURE */
+
if (!st_spinning) cc_SpinUp();
-
+
for (data_tries=n_retries; data_tries > 0; data_tries--)
{
for (status_tries=3; status_tries > 0; status_tries--)
@@ -4940,7 +4614,7 @@ static void do_sbpcd_request(request_queue_t * q)
{
#ifdef SAFE_MIXED
current_drive->has_data=2; /* is really a data disk */
-#endif /* SAFE_MIXED */
+#endif /* SAFE_MIXED */
#ifdef DEBUG_GTL
printk(" do_sbpcd_request[%do](%p:%ld+%ld) end 3, Time:%li\n",
xnr, req, req->sector, req->nr_sectors, jiffies);
@@ -4951,7 +4625,7 @@ static void do_sbpcd_request(request_queue_t * q)
goto request_loop;
}
}
-
+
err_done:
#if OLD_BUSY
busy_data=0;
@@ -4976,7 +4650,7 @@ static void sbp_read_cmd(struct request *req)
int i;
int block;
-
+
current_drive->sbp_first_frame=current_drive->sbp_last_frame=-1; /* purge buffer */
current_drive->sbp_current = 0;
block=req->sector/4;
@@ -4993,7 +4667,7 @@ static void sbp_read_cmd(struct request *req)
current_drive->sbp_read_frames=1;
}
}
-
+
flags_cmd_out = f_putcmd | f_respo2 | f_ResponseStatus | f_obey_p_check;
clr_cmdbuf();
if (famV_drive)
@@ -5092,7 +4766,7 @@ static int sbp_data(struct request *req)
int success;
int wait;
int duration;
-
+
error_flag=0;
success=0;
#if LONG_TIMING
@@ -5105,12 +4779,12 @@ static int sbp_data(struct request *req)
for (frame=0;frame<current_drive->sbp_read_frames&&!error_flag; frame++)
{
SBPCD_CLI;
-
+
del_timer(&data_timer);
data_timer.expires=jiffies+max_latency;
timed_out_data=0;
add_timer(&data_timer);
- while (!timed_out_data)
+ while (!timed_out_data)
{
if (current_drive->f_multisession) try=maxtim_data*4;
else try=maxtim_data;
@@ -5207,9 +4881,9 @@ static int sbp_data(struct request *req)
else
{
sbp_sleep(1);
- OUT(CDo_sel_i_d,0);
+ OUT(CDo_sel_i_d,0);
i=inb(CDi_status);
- }
+ }
if (!(i&s_not_data_ready))
{
OUT(CDo_sel_i_d,1);
@@ -5311,7 +4985,7 @@ static int sbp_data(struct request *req)
}
SBPCD_STI;
}
-
+
#if 0
if (!success)
#endif
@@ -5370,7 +5044,326 @@ static int sbpcd_block_ioctl(struct inode *inode, struct file *file,
unsigned cmd, unsigned long arg)
{
struct sbpcd_drive *p = inode->i_bdev->bd_disk->private_data;
- return cdrom_ioctl(file, p->sbpcd_infop, inode, cmd, arg);
+ struct cdrom_device_info *cdi = p->sbpcd_infop;
+ int ret, i;
+
+ ret = cdrom_ioctl(file, p->sbpcd_infop, inode, cmd, arg);
+ if (ret != -ENOSYS)
+ return ret;
+
+ msg(DBG_IO2,"ioctl(%s, 0x%08lX, 0x%08lX)\n", cdi->name, cmd, arg);
+ if (p->drv_id==-1) {
+ msg(DBG_INF, "ioctl: bad device: %s\n", cdi->name);
+ return (-ENXIO); /* no such drive */
+ }
+ down(&ioctl_read_sem);
+ if (p != current_drive)
+ switch_drive(p);
+
+ msg(DBG_IO2,"ioctl: device %s, request %04X\n",cdi->name,cmd);
+ switch (cmd) /* Sun-compatible */
+ {
+ case DDIOCSDBG: /* DDI Debug */
+ if (!capable(CAP_SYS_ADMIN)) RETURN_UP(-EPERM);
+ i=sbpcd_dbg_ioctl(arg,1);
+ RETURN_UP(i);
+ case CDROMRESET: /* hard reset the drive */
+ msg(DBG_IOC,"ioctl: CDROMRESET entered.\n");
+ i=DriveReset();
+ current_drive->audio_state=0;
+ RETURN_UP(i);
+
+ case CDROMREADMODE1:
+ msg(DBG_IOC,"ioctl: CDROMREADMODE1 requested.\n");
+#ifdef SAFE_MIXED
+ if (current_drive->has_data>1) RETURN_UP(-EBUSY);
+#endif /* SAFE_MIXED */
+ cc_ModeSelect(CD_FRAMESIZE);
+ cc_ModeSense();
+ current_drive->mode=READ_M1;
+ RETURN_UP(0);
+
+ case CDROMREADMODE2: /* not usable at the moment */
+ msg(DBG_IOC,"ioctl: CDROMREADMODE2 requested.\n");
+#ifdef SAFE_MIXED
+ if (current_drive->has_data>1) RETURN_UP(-EBUSY);
+#endif /* SAFE_MIXED */
+ cc_ModeSelect(CD_FRAMESIZE_RAW1);
+ cc_ModeSense();
+ current_drive->mode=READ_M2;
+ RETURN_UP(0);
+
+ case CDROMAUDIOBUFSIZ: /* configure the audio buffer size */
+ msg(DBG_IOC,"ioctl: CDROMAUDIOBUFSIZ entered.\n");
+ if (current_drive->sbp_audsiz>0)
+ vfree(current_drive->aud_buf);
+ current_drive->aud_buf=NULL;
+ current_drive->sbp_audsiz=arg;
+
+ if (current_drive->sbp_audsiz>16)
+ {
+ current_drive->sbp_audsiz = 0;
+ RETURN_UP(current_drive->sbp_audsiz);
+ }
+
+ if (current_drive->sbp_audsiz>0)
+ {
+ current_drive->aud_buf=(u_char *) vmalloc(current_drive->sbp_audsiz*CD_FRAMESIZE_RAW);
+ if (current_drive->aud_buf==NULL)
+ {
+ msg(DBG_INF,"audio buffer (%d frames) not available.\n",current_drive->sbp_audsiz);
+ current_drive->sbp_audsiz=0;
+ }
+ else msg(DBG_INF,"audio buffer size: %d frames.\n",current_drive->sbp_audsiz);
+ }
+ RETURN_UP(current_drive->sbp_audsiz);
+
+ case CDROMREADAUDIO:
+ { /* start of CDROMREADAUDIO */
+ int i=0, j=0, frame, block=0;
+ u_int try=0;
+ u_long timeout;
+ u_char *p;
+ u_int data_tries = 0;
+ u_int data_waits = 0;
+ u_int data_retrying = 0;
+ int status_tries;
+ int error_flag;
+
+ msg(DBG_IOC,"ioctl: CDROMREADAUDIO entered.\n");
+ if (fam0_drive) RETURN_UP(-EINVAL);
+ if (famL_drive) RETURN_UP(-EINVAL);
+ if (famV_drive) RETURN_UP(-EINVAL);
+ if (famT_drive) RETURN_UP(-EINVAL);
+#ifdef SAFE_MIXED
+ if (current_drive->has_data>1) RETURN_UP(-EBUSY);
+#endif /* SAFE_MIXED */
+ if (current_drive->aud_buf==NULL) RETURN_UP(-EINVAL);
+ if (copy_from_user(&read_audio, (void __user *)arg,
+ sizeof(struct cdrom_read_audio)))
+ RETURN_UP(-EFAULT);
+ if (read_audio.nframes < 0 || read_audio.nframes>current_drive->sbp_audsiz) RETURN_UP(-EINVAL);
+ if (!access_ok(VERIFY_WRITE, read_audio.buf,
+ read_audio.nframes*CD_FRAMESIZE_RAW))
+ RETURN_UP(-EFAULT);
+
+ if (read_audio.addr_format==CDROM_MSF) /* MSF-bin specification of where to start */
+ block=msf2lba(&read_audio.addr.msf.minute);
+ else if (read_audio.addr_format==CDROM_LBA) /* lba specification of where to start */
+ block=read_audio.addr.lba;
+ else RETURN_UP(-EINVAL);
+#if 000
+ i=cc_SetSpeed(speed_150,0,0);
+ if (i) msg(DBG_AUD,"read_audio: SetSpeed error %d\n", i);
+#endif
+ msg(DBG_AUD,"read_audio: lba: %d, msf: %06X\n",
+ block, blk2msf(block));
+ msg(DBG_AUD,"read_audio: before cc_ReadStatus.\n");
+#if OLD_BUSY
+ while (busy_data) sbp_sleep(HZ/10); /* wait a bit */
+ busy_audio=1;
+#endif /* OLD_BUSY */
+ error_flag=0;
+ for (data_tries=5; data_tries>0; data_tries--)
+ {
+ msg(DBG_AUD,"data_tries=%d ...\n", data_tries);
+ current_drive->mode=READ_AU;
+ cc_ModeSelect(CD_FRAMESIZE_RAW);
+ cc_ModeSense();
+ for (status_tries=3; status_tries > 0; status_tries--)
+ {
+ flags_cmd_out |= f_respo3;
+ cc_ReadStatus();
+ if (sbp_status() != 0) break;
+ if (st_check) cc_ReadError();
+ sbp_sleep(1); /* wait a bit, try again */
+ }
+ if (status_tries == 0)
+ {
+ msg(DBG_AUD,"read_audio: sbp_status: failed after 3 tries in line %d.\n", __LINE__);
+ continue;
+ }
+ msg(DBG_AUD,"read_audio: sbp_status: ok.\n");
+
+ flags_cmd_out = f_putcmd | f_respo2 | f_ResponseStatus | f_obey_p_check;
+ if (fam0L_drive)
+ {
+ flags_cmd_out |= f_lopsta | f_getsta | f_bit1;
+ cmd_type=READ_M2;
+ drvcmd[0]=CMD0_READ_XA; /* "read XA frames", old drives */
+ drvcmd[1]=(block>>16)&0x000000ff;
+ drvcmd[2]=(block>>8)&0x000000ff;
+ drvcmd[3]=block&0x000000ff;
+ drvcmd[4]=0;
+ drvcmd[5]=read_audio.nframes; /* # of frames */
+ drvcmd[6]=0;
+ }
+ else if (fam1_drive)
+ {
+ drvcmd[0]=CMD1_READ; /* "read frames", new drives */
+ lba2msf(block,&drvcmd[1]); /* msf-bin format required */
+ drvcmd[4]=0;
+ drvcmd[5]=0;
+ drvcmd[6]=read_audio.nframes; /* # of frames */
+ }
+ else if (fam2_drive)
+ {
+ drvcmd[0]=CMD2_READ_XA2;
+ lba2msf(block,&drvcmd[1]); /* msf-bin format required */
+ drvcmd[4]=0;
+ drvcmd[5]=read_audio.nframes; /* # of frames */
+ drvcmd[6]=0x11; /* raw mode */
+ }
+ else if (famT_drive) /* CD-55A: not tested yet */
+ {
+ }
+ msg(DBG_AUD,"read_audio: before giving \"read\" command.\n");
+ flags_cmd_out=f_putcmd;
+ response_count=0;
+ i=cmd_out();
+ if (i<0) msg(DBG_INF,"error giving READ AUDIO command: %0d\n", i);
+ sbp_sleep(0);
+ msg(DBG_AUD,"read_audio: after giving \"read\" command.\n");
+ for (frame=1;frame<2 && !error_flag; frame++)
+ {
+ try=maxtim_data;
+ for (timeout=jiffies+9*HZ; ; )
+ {
+ for ( ; try!=0;try--)
+ {
+ j=inb(CDi_status);
+ if (!(j&s_not_data_ready)) break;
+ if (!(j&s_not_result_ready)) break;
+ if (fam0L_drive) if (j&s_attention) break;
+ }
+ if (try != 0 || time_after_eq(jiffies, timeout)) break;
+ if (data_retrying == 0) data_waits++;
+ data_retrying = 1;
+ sbp_sleep(1);
+ try = 1;
+ }
+ if (try==0)
+ {
+ msg(DBG_INF,"read_audio: sbp_data: CDi_status timeout.\n");
+ error_flag++;
+ break;
+ }
+ msg(DBG_AUD,"read_audio: sbp_data: CDi_status ok.\n");
+ if (j&s_not_data_ready)
+ {
+ msg(DBG_INF, "read_audio: sbp_data: DATA_READY timeout.\n");
+ error_flag++;
+ break;
+ }
+ msg(DBG_AUD,"read_audio: before reading data.\n");
+ error_flag=0;
+ p = current_drive->aud_buf;
+ if (sbpro_type==1) OUT(CDo_sel_i_d,1);
+ if (do_16bit)
+ {
+ u_short *p2 = (u_short *) p;
+
+ for (; (u_char *) p2 < current_drive->aud_buf + read_audio.nframes*CD_FRAMESIZE_RAW;)
+ {
+ if ((inb_p(CDi_status)&s_not_data_ready)) continue;
+
+ /* get one sample */
+ *p2++ = inw_p(CDi_data);
+ *p2++ = inw_p(CDi_data);
+ }
+ } else {
+ for (; p < current_drive->aud_buf + read_audio.nframes*CD_FRAMESIZE_RAW;)
+ {
+ if ((inb_p(CDi_status)&s_not_data_ready)) continue;
+
+ /* get one sample */
+ *p++ = inb_p(CDi_data);
+ *p++ = inb_p(CDi_data);
+ *p++ = inb_p(CDi_data);
+ *p++ = inb_p(CDi_data);
+ }
+ }
+ if (sbpro_type==1) OUT(CDo_sel_i_d,0);
+ data_retrying = 0;
+ }
+ msg(DBG_AUD,"read_audio: after reading data.\n");
+ if (error_flag) /* must have been spurious D_RDY or (ATTN&&!D_RDY) */
+ {
+ msg(DBG_AUD,"read_audio: read aborted by drive\n");
+#if 0000
+ i=cc_DriveReset(); /* ugly fix to prevent a hang */
+#else
+ i=cc_ReadError();
+#endif
+ continue;
+ }
+ if (fam0L_drive)
+ {
+ i=maxtim_data;
+ for (timeout=jiffies+9*HZ; time_before(jiffies, timeout); timeout--)
+ {
+ for ( ;i!=0;i--)
+ {
+ j=inb(CDi_status);
+ if (!(j&s_not_data_ready)) break;
+ if (!(j&s_not_result_ready)) break;
+ if (j&s_attention) break;
+ }
+ if (i != 0 || time_after_eq(jiffies, timeout)) break;
+ sbp_sleep(0);
+ i = 1;
+ }
+ if (i==0) msg(DBG_AUD,"read_audio: STATUS TIMEOUT AFTER READ");
+ if (!(j&s_attention))
+ {
+ msg(DBG_AUD,"read_audio: sbp_data: timeout waiting DRV_ATTN - retrying\n");
+ i=cc_DriveReset(); /* ugly fix to prevent a hang */
+ continue;
+ }
+ }
+ do
+ {
+ if (fam0L_drive) cc_ReadStatus();
+ i=ResponseStatus(); /* builds status_bits, returns orig. status (old) or faked p_success (new) */
+ if (i<0) { msg(DBG_AUD,
+ "read_audio: cc_ReadStatus error after read: %02X\n",
+ current_drive->status_bits);
+ continue; /* FIXME */
+ }
+ }
+ while ((fam0L_drive)&&(!st_check)&&(!(i&p_success)));
+ if (st_check)
+ {
+ i=cc_ReadError();
+ msg(DBG_AUD,"read_audio: cc_ReadError was necessary after read: %02X\n",i);
+ continue;
+ }
+ if (copy_to_user(read_audio.buf,
+ current_drive->aud_buf,
+ read_audio.nframes * CD_FRAMESIZE_RAW))
+ RETURN_UP(-EFAULT);
+ msg(DBG_AUD,"read_audio: copy_to_user done.\n");
+ break;
+ }
+ cc_ModeSelect(CD_FRAMESIZE);
+ cc_ModeSense();
+ current_drive->mode=READ_M1;
+#if OLD_BUSY
+ busy_audio=0;
+#endif /* OLD_BUSY */
+ if (data_tries == 0)
+ {
+ msg(DBG_AUD,"read_audio: failed after 5 tries in line %d.\n", __LINE__);
+ RETURN_UP(-EIO);
+ }
+ msg(DBG_AUD,"read_audio: successful return.\n");
+ RETURN_UP(0);
+ } /* end of CDROMREADAUDIO */
+
+ default:
+ msg(DBG_IOC,"ioctl: unknown function request %04X\n", cmd);
+ RETURN_UP(-EINVAL);
+ } /* end switch(cmd) */
}
static int sbpcd_block_media_changed(struct gendisk *disk)
@@ -5478,10 +5471,9 @@ static struct cdrom_device_ops sbpcd_dops = {
.get_mcn = sbpcd_get_mcn,
.reset = sbpcd_reset,
.audio_ioctl = sbpcd_audio_ioctl,
- .dev_ioctl = sbpcd_dev_ioctl,
.capability = CDC_CLOSE_TRAY | CDC_OPEN_TRAY | CDC_LOCK |
CDC_MULTI_SESSION | CDC_MEDIA_CHANGED |
- CDC_MCN | CDC_PLAY_AUDIO | CDC_IOCTLS,
+ CDC_MCN | CDC_PLAY_AUDIO,
.n_minors = 1,
};
diff --git a/drivers/cdrom/viocd.c b/drivers/cdrom/viocd.c
index e27617259552..c0f817ba7adb 100644
--- a/drivers/cdrom/viocd.c
+++ b/drivers/cdrom/viocd.c
@@ -627,7 +627,7 @@ static struct cdrom_device_ops viocd_dops = {
.media_changed = viocd_media_changed,
.lock_door = viocd_lock_door,
.generic_packet = viocd_packet,
- .capability = CDC_CLOSE_TRAY | CDC_OPEN_TRAY | CDC_LOCK | CDC_SELECT_SPEED | CDC_SELECT_DISC | CDC_MULTI_SESSION | CDC_MCN | CDC_MEDIA_CHANGED | CDC_PLAY_AUDIO | CDC_RESET | CDC_IOCTLS | CDC_DRIVE_STATUS | CDC_GENERIC_PACKET | CDC_CD_R | CDC_CD_RW | CDC_DVD | CDC_DVD_R | CDC_DVD_RAM | CDC_RAM
+ .capability = CDC_CLOSE_TRAY | CDC_OPEN_TRAY | CDC_LOCK | CDC_SELECT_SPEED | CDC_SELECT_DISC | CDC_MULTI_SESSION | CDC_MCN | CDC_MEDIA_CHANGED | CDC_PLAY_AUDIO | CDC_RESET | CDC_DRIVE_STATUS | CDC_GENERIC_PACKET | CDC_CD_R | CDC_CD_RW | CDC_DVD | CDC_DVD_R | CDC_DVD_RAM | CDC_RAM
};
static int __init find_capability(const char *type)
diff --git a/drivers/char/amiserial.c b/drivers/char/amiserial.c
index 7ac365b5d9ec..6602b3156df5 100644
--- a/drivers/char/amiserial.c
+++ b/drivers/char/amiserial.c
@@ -46,8 +46,6 @@
/* Sanity checks */
-#define SERIAL_INLINE
-
#if defined(MODULE) && defined(SERIAL_DEBUG_MCOUNT)
#define DBG_CNT(s) printk("(%s): [%x] refc=%d, serc=%d, ttyc=%d -> %s\n", \
tty->name, (info->flags), serial_driver->refcount,info->count,tty->count,s)
@@ -95,10 +93,6 @@ static char *serial_version = "4.30";
#include <asm/amigahw.h>
#include <asm/amigaints.h>
-#ifdef SERIAL_INLINE
-#define _INLINE_ inline
-#endif
-
#define custom amiga_custom
static char *serial_name = "Amiga-builtin serial driver";
@@ -253,14 +247,14 @@ static void rs_start(struct tty_struct *tty)
* This routine is used by the interrupt handler to schedule
* processing in the software interrupt portion of the driver.
*/
-static _INLINE_ void rs_sched_event(struct async_struct *info,
- int event)
+static void rs_sched_event(struct async_struct *info,
+ int event)
{
info->event |= 1 << event;
tasklet_schedule(&info->tlet);
}
-static _INLINE_ void receive_chars(struct async_struct *info)
+static void receive_chars(struct async_struct *info)
{
int status;
int serdatr;
@@ -349,7 +343,7 @@ out:
return;
}
-static _INLINE_ void transmit_chars(struct async_struct *info)
+static void transmit_chars(struct async_struct *info)
{
custom.intreq = IF_TBE;
mb();
@@ -389,7 +383,7 @@ static _INLINE_ void transmit_chars(struct async_struct *info)
}
}
-static _INLINE_ void check_modem_status(struct async_struct *info)
+static void check_modem_status(struct async_struct *info)
{
unsigned char status = ciab.pra & (SER_DCD | SER_CTS | SER_DSR);
unsigned char dstatus;
@@ -1959,7 +1953,7 @@ done:
* number, and identifies which options were configured into this
* driver.
*/
-static _INLINE_ void show_serial_version(void)
+static void show_serial_version(void)
{
printk(KERN_INFO "%s version %s\n", serial_name, serial_version);
}
diff --git a/drivers/char/generic_serial.c b/drivers/char/generic_serial.c
index e38a5f0e07bb..5e59c0b42731 100644
--- a/drivers/char/generic_serial.c
+++ b/drivers/char/generic_serial.c
@@ -48,8 +48,8 @@ static int gs_debug;
#define NEW_WRITE_LOCKING 1
#if NEW_WRITE_LOCKING
#define DECL /* Nothing */
-#define LOCKIT down (& port->port_write_sem);
-#define RELEASEIT up (&port->port_write_sem);
+#define LOCKIT mutex_lock(& port->port_write_mutex);
+#define RELEASEIT mutex_unlock(&port->port_write_mutex);
#else
#define DECL unsigned long flags;
#define LOCKIT save_flags (flags);cli ()
@@ -124,14 +124,14 @@ int gs_write(struct tty_struct * tty,
/* get exclusive "write" access to this port (problem 3) */
/* This is not a spinlock because we can have a disk access (page
fault) in copy_from_user */
- down (& port->port_write_sem);
+ mutex_lock(& port->port_write_mutex);
while (1) {
c = count;
/* This is safe because we "OWN" the "head". Noone else can
- change the "head": we own the port_write_sem. */
+ change the "head": we own the port_write_mutex. */
/* Don't overrun the end of the buffer */
t = SERIAL_XMIT_SIZE - port->xmit_head;
if (t < c) c = t;
@@ -153,7 +153,7 @@ int gs_write(struct tty_struct * tty,
count -= c;
total += c;
}
- up (& port->port_write_sem);
+ mutex_unlock(& port->port_write_mutex);
gs_dprintk (GS_DEBUG_WRITE, "write: interrupts are %s\n",
(port->flags & GS_TX_INTEN)?"enabled": "disabled");
@@ -214,7 +214,7 @@ int gs_write(struct tty_struct * tty,
c = count;
/* This is safe because we "OWN" the "head". Noone else can
- change the "head": we own the port_write_sem. */
+ change the "head": we own the port_write_mutex. */
/* Don't overrun the end of the buffer */
t = SERIAL_XMIT_SIZE - port->xmit_head;
if (t < c) c = t;
@@ -888,7 +888,7 @@ int gs_init_port(struct gs_port *port)
spin_lock_irqsave (&port->driver_lock, flags);
if (port->tty)
clear_bit(TTY_IO_ERROR, &port->tty->flags);
- init_MUTEX(&port->port_write_sem);
+ mutex_init(&port->port_write_mutex);
port->xmit_cnt = port->xmit_head = port->xmit_tail = 0;
spin_unlock_irqrestore(&port->driver_lock, flags);
gs_set_termios(port->tty, NULL);
diff --git a/drivers/char/istallion.c b/drivers/char/istallion.c
index 28c5a3193b81..ede128356af2 100644
--- a/drivers/char/istallion.c
+++ b/drivers/char/istallion.c
@@ -181,7 +181,6 @@ static struct tty_driver *stli_serial;
* is already swapping a shared buffer won't make things any worse.
*/
static char *stli_tmpwritebuf;
-static DECLARE_MUTEX(stli_tmpwritesem);
#define STLI_TXBUFSIZE 4096
diff --git a/drivers/char/n_tty.c b/drivers/char/n_tty.c
index ccad7ae94541..ede365d05387 100644
--- a/drivers/char/n_tty.c
+++ b/drivers/char/n_tty.c
@@ -132,7 +132,7 @@ static void put_tty_queue(unsigned char c, struct tty_struct *tty)
* We test the TTY_THROTTLED bit first so that it always
* indicates the current state. The decision about whether
* it is worth allowing more input has been taken by the caller.
- * Can sleep, may be called under the atomic_read semaphore but
+ * Can sleep, may be called under the atomic_read_lock mutex but
* this is not guaranteed.
*/
@@ -1132,7 +1132,7 @@ static inline int input_available_p(struct tty_struct *tty, int amt)
* buffer, and once to drain the space from the (physical) beginning of
* the buffer to head pointer.
*
- * Called under the tty->atomic_read sem and with TTY_DONT_FLIP set
+ * Called under the tty->atomic_read_lock sem and with TTY_DONT_FLIP set
*
*/
@@ -1262,11 +1262,11 @@ do_it_again:
* Internal serialization of reads.
*/
if (file->f_flags & O_NONBLOCK) {
- if (down_trylock(&tty->atomic_read))
+ if (!mutex_trylock(&tty->atomic_read_lock))
return -EAGAIN;
}
else {
- if (down_interruptible(&tty->atomic_read))
+ if (mutex_lock_interruptible(&tty->atomic_read_lock))
return -ERESTARTSYS;
}
@@ -1393,7 +1393,7 @@ do_it_again:
timeout = time;
}
clear_bit(TTY_DONT_FLIP, &tty->flags);
- up(&tty->atomic_read);
+ mutex_unlock(&tty->atomic_read_lock);
remove_wait_queue(&tty->read_wait, &wait);
if (!waitqueue_active(&tty->read_wait))
diff --git a/drivers/char/nwflash.c b/drivers/char/nwflash.c
index ca41d62b1d9d..8865387d3448 100644
--- a/drivers/char/nwflash.c
+++ b/drivers/char/nwflash.c
@@ -27,6 +27,7 @@
#include <linux/rwsem.h>
#include <linux/init.h>
#include <linux/smp_lock.h>
+#include <linux/mutex.h>
#include <asm/hardware/dec21285.h>
#include <asm/io.h>
@@ -56,7 +57,7 @@ static int gbWriteEnable;
static int gbWriteBase64Enable;
static volatile unsigned char *FLASH_BASE;
static int gbFlashSize = KFLASH_SIZE;
-static DECLARE_MUTEX(nwflash_sem);
+static DEFINE_MUTEX(nwflash_mutex);
extern spinlock_t gpio_lock;
@@ -140,7 +141,7 @@ static ssize_t flash_read(struct file *file, char __user *buf, size_t size,
/*
* We now lock against reads and writes. --rmk
*/
- if (down_interruptible(&nwflash_sem))
+ if (mutex_lock_interruptible(&nwflash_mutex))
return -ERESTARTSYS;
ret = copy_to_user(buf, (void *)(FLASH_BASE + p), count);
@@ -149,7 +150,7 @@ static ssize_t flash_read(struct file *file, char __user *buf, size_t size,
*ppos += count;
} else
ret = -EFAULT;
- up(&nwflash_sem);
+ mutex_unlock(&nwflash_mutex);
}
return ret;
}
@@ -188,7 +189,7 @@ static ssize_t flash_write(struct file *file, const char __user *buf,
/*
* We now lock against reads and writes. --rmk
*/
- if (down_interruptible(&nwflash_sem))
+ if (mutex_lock_interruptible(&nwflash_mutex))
return -ERESTARTSYS;
written = 0;
@@ -277,7 +278,7 @@ static ssize_t flash_write(struct file *file, const char __user *buf,
*/
leds_event(led_release);
- up(&nwflash_sem);
+ mutex_unlock(&nwflash_mutex);
return written;
}
diff --git a/drivers/char/raw.c b/drivers/char/raw.c
index 30e4cbe16bb0..15a7b4086524 100644
--- a/drivers/char/raw.c
+++ b/drivers/char/raw.c
@@ -19,6 +19,7 @@
#include <linux/uio.h>
#include <linux/cdev.h>
#include <linux/device.h>
+#include <linux/mutex.h>
#include <asm/uaccess.h>
@@ -29,7 +30,7 @@ struct raw_device_data {
static struct class *raw_class;
static struct raw_device_data raw_devices[MAX_RAW_MINORS];
-static DECLARE_MUTEX(raw_mutex);
+static DEFINE_MUTEX(raw_mutex);
static struct file_operations raw_ctl_fops; /* forward declaration */
/*
@@ -53,7 +54,7 @@ static int raw_open(struct inode *inode, struct file *filp)
return 0;
}
- down(&raw_mutex);
+ mutex_lock(&raw_mutex);
/*
* All we need to do on open is check that the device is bound.
@@ -78,7 +79,7 @@ static int raw_open(struct inode *inode, struct file *filp)
filp->f_dentry->d_inode->i_mapping =
bdev->bd_inode->i_mapping;
filp->private_data = bdev;
- up(&raw_mutex);
+ mutex_unlock(&raw_mutex);
return 0;
out2:
@@ -86,7 +87,7 @@ out2:
out1:
blkdev_put(bdev);
out:
- up(&raw_mutex);
+ mutex_unlock(&raw_mutex);
return err;
}
@@ -99,14 +100,14 @@ static int raw_release(struct inode *inode, struct file *filp)
const int minor= iminor(inode);
struct block_device *bdev;
- down(&raw_mutex);
+ mutex_lock(&raw_mutex);
bdev = raw_devices[minor].binding;
if (--raw_devices[minor].inuse == 0) {
/* Here inode->i_mapping == bdev->bd_inode->i_mapping */
inode->i_mapping = &inode->i_data;
inode->i_mapping->backing_dev_info = &default_backing_dev_info;
}
- up(&raw_mutex);
+ mutex_unlock(&raw_mutex);
bd_release(bdev);
blkdev_put(bdev);
@@ -187,9 +188,9 @@ static int raw_ctl_ioctl(struct inode *inode, struct file *filp,
goto out;
}
- down(&raw_mutex);
+ mutex_lock(&raw_mutex);
if (rawdev->inuse) {
- up(&raw_mutex);
+ mutex_unlock(&raw_mutex);
err = -EBUSY;
goto out;
}
@@ -211,11 +212,11 @@ static int raw_ctl_ioctl(struct inode *inode, struct file *filp,
bind_device(&rq);
}
}
- up(&raw_mutex);
+ mutex_unlock(&raw_mutex);
} else {
struct block_device *bdev;
- down(&raw_mutex);
+ mutex_lock(&raw_mutex);
bdev = rawdev->binding;
if (bdev) {
rq.block_major = MAJOR(bdev->bd_dev);
@@ -223,7 +224,7 @@ static int raw_ctl_ioctl(struct inode *inode, struct file *filp,
} else {
rq.block_major = rq.block_minor = 0;
}
- up(&raw_mutex);
+ mutex_unlock(&raw_mutex);
if (copy_to_user((void __user *)arg, &rq, sizeof(rq))) {
err = -EFAULT;
goto out;
diff --git a/drivers/char/ser_a2232.c b/drivers/char/ser_a2232.c
index fee68cc895f8..510bd3e0e88b 100644
--- a/drivers/char/ser_a2232.c
+++ b/drivers/char/ser_a2232.c
@@ -97,7 +97,7 @@
#include <asm/amigahw.h>
#include <linux/zorro.h>
#include <asm/irq.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
#include <linux/delay.h>
@@ -654,7 +654,7 @@ static void a2232_init_portstructs(void)
port->gs.closing_wait = 30 * HZ;
port->gs.rd = &a2232_real_driver;
#ifdef NEW_WRITE_LOCKING
- init_MUTEX(&(port->gs.port_write_sem));
+ init_MUTEX(&(port->gs.port_write_mutex));
#endif
init_waitqueue_head(&port->gs.open_wait);
init_waitqueue_head(&port->gs.close_wait);
diff --git a/drivers/char/snsc.c b/drivers/char/snsc.c
index 0e7d216e7eb0..b543821d8cb4 100644
--- a/drivers/char/snsc.c
+++ b/drivers/char/snsc.c
@@ -5,7 +5,7 @@
* License. See the file "COPYING" in the main directory of this archive
* for more details.
*
- * Copyright (C) 2004 Silicon Graphics, Inc. All rights reserved.
+ * Copyright (C) 2004, 2006 Silicon Graphics, Inc. All rights reserved.
*/
/*
@@ -77,7 +77,7 @@ scdrv_open(struct inode *inode, struct file *file)
scd = container_of(inode->i_cdev, struct sysctl_data_s, scd_cdev);
/* allocate memory for subchannel data */
- sd = kmalloc(sizeof (struct subch_data_s), GFP_KERNEL);
+ sd = kzalloc(sizeof (struct subch_data_s), GFP_KERNEL);
if (sd == NULL) {
printk("%s: couldn't allocate subchannel data\n",
__FUNCTION__);
@@ -85,7 +85,6 @@ scdrv_open(struct inode *inode, struct file *file)
}
/* initialize subch_data_s fields */
- memset(sd, 0, sizeof (struct subch_data_s));
sd->sd_nasid = scd->scd_nasid;
sd->sd_subch = ia64_sn_irtr_open(scd->scd_nasid);
@@ -394,7 +393,7 @@ scdrv_init(void)
sprintf(devnamep, "#%d", geo_slab(geoid));
/* allocate sysctl device data */
- scd = kmalloc(sizeof (struct sysctl_data_s),
+ scd = kzalloc(sizeof (struct sysctl_data_s),
GFP_KERNEL);
if (!scd) {
printk("%s: failed to allocate device info"
@@ -402,7 +401,6 @@ scdrv_init(void)
SYSCTL_BASENAME, devname);
continue;
}
- memset(scd, 0, sizeof (struct sysctl_data_s));
/* initialize sysctl device data fields */
scd->scd_nasid = cnodeid_to_nasid(cnode);
diff --git a/drivers/char/snsc_event.c b/drivers/char/snsc_event.c
index a4fa507eed9e..e234d50e142a 100644
--- a/drivers/char/snsc_event.c
+++ b/drivers/char/snsc_event.c
@@ -287,7 +287,7 @@ scdrv_event_init(struct sysctl_data_s *scd)
{
int rv;
- event_sd = kmalloc(sizeof (struct subch_data_s), GFP_KERNEL);
+ event_sd = kzalloc(sizeof (struct subch_data_s), GFP_KERNEL);
if (event_sd == NULL) {
printk(KERN_WARNING "%s: couldn't allocate subchannel info"
" for event monitoring\n", __FUNCTION__);
@@ -295,7 +295,6 @@ scdrv_event_init(struct sysctl_data_s *scd)
}
/* initialize subch_data_s fields */
- memset(event_sd, 0, sizeof (struct subch_data_s));
event_sd->sd_nasid = scd->scd_nasid;
spin_lock_init(&event_sd->sd_rlock);
@@ -321,5 +320,3 @@ scdrv_event_init(struct sysctl_data_s *scd)
return;
}
}
-
-
diff --git a/drivers/char/stallion.c b/drivers/char/stallion.c
index bdaab6992109..3f5d6077f39c 100644
--- a/drivers/char/stallion.c
+++ b/drivers/char/stallion.c
@@ -148,7 +148,6 @@ static struct tty_driver *stl_serial;
* is already swapping a shared buffer won't make things any worse.
*/
static char *stl_tmpwritebuf;
-static DECLARE_MUTEX(stl_tmpwritesem);
/*
* Define a local default termios struct. All ports will be created
diff --git a/drivers/char/sx.c b/drivers/char/sx.c
index a6b4f02bdceb..3b4747230270 100644
--- a/drivers/char/sx.c
+++ b/drivers/char/sx.c
@@ -2318,7 +2318,7 @@ static int sx_init_portstructs (int nboards, int nports)
port->board = board;
port->gs.rd = &sx_real_driver;
#ifdef NEW_WRITE_LOCKING
- port->gs.port_write_sem = MUTEX;
+ port->gs.port_write_mutex = MUTEX;
#endif
port->gs.driver_lock = SPIN_LOCK_UNLOCKED;
/*
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 53d3d066554e..76592ee1fb38 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -130,7 +130,7 @@ LIST_HEAD(tty_drivers); /* linked list of tty drivers */
/* Semaphore to protect creating and releasing a tty. This is shared with
vt.c for deeply disgusting hack reasons */
-DECLARE_MUTEX(tty_sem);
+DEFINE_MUTEX(tty_mutex);
#ifdef CONFIG_UNIX98_PTYS
extern struct tty_driver *ptm_driver; /* Unix98 pty masters; for /dev/ptmx */
@@ -1188,11 +1188,11 @@ void disassociate_ctty(int on_exit)
lock_kernel();
- down(&tty_sem);
+ mutex_lock(&tty_mutex);
tty = current->signal->tty;
if (tty) {
tty_pgrp = tty->pgrp;
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
if (on_exit && tty->driver->type != TTY_DRIVER_TYPE_PTY)
tty_vhangup(tty);
} else {
@@ -1200,7 +1200,7 @@ void disassociate_ctty(int on_exit)
kill_pg(current->signal->tty_old_pgrp, SIGHUP, on_exit);
kill_pg(current->signal->tty_old_pgrp, SIGCONT, on_exit);
}
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
unlock_kernel();
return;
}
@@ -1211,7 +1211,7 @@ void disassociate_ctty(int on_exit)
}
/* Must lock changes to tty_old_pgrp */
- down(&tty_sem);
+ mutex_lock(&tty_mutex);
current->signal->tty_old_pgrp = 0;
tty->session = 0;
tty->pgrp = -1;
@@ -1222,7 +1222,7 @@ void disassociate_ctty(int on_exit)
p->signal->tty = NULL;
} while_each_task_pid(current->signal->session, PIDTYPE_SID, p);
read_unlock(&tasklist_lock);
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
unlock_kernel();
}
@@ -1306,7 +1306,7 @@ static inline ssize_t do_tty_write(
ssize_t ret = 0, written = 0;
unsigned int chunk;
- if (down_interruptible(&tty->atomic_write)) {
+ if (mutex_lock_interruptible(&tty->atomic_write_lock)) {
return -ERESTARTSYS;
}
@@ -1329,7 +1329,7 @@ static inline ssize_t do_tty_write(
if (count < chunk)
chunk = count;
- /* write_buf/write_cnt is protected by the atomic_write semaphore */
+ /* write_buf/write_cnt is protected by the atomic_write_lock mutex */
if (tty->write_cnt < chunk) {
unsigned char *buf;
@@ -1338,7 +1338,7 @@ static inline ssize_t do_tty_write(
buf = kmalloc(chunk, GFP_KERNEL);
if (!buf) {
- up(&tty->atomic_write);
+ mutex_unlock(&tty->atomic_write_lock);
return -ENOMEM;
}
kfree(tty->write_buf);
@@ -1374,7 +1374,7 @@ static inline ssize_t do_tty_write(
inode->i_mtime = current_fs_time(inode->i_sb);
ret = written;
}
- up(&tty->atomic_write);
+ mutex_unlock(&tty->atomic_write_lock);
return ret;
}
@@ -1442,8 +1442,8 @@ static inline void tty_line_name(struct tty_driver *driver, int index, char *p)
/*
* WSH 06/09/97: Rewritten to remove races and properly clean up after a
- * failed open. The new code protects the open with a semaphore, so it's
- * really quite straightforward. The semaphore locking can probably be
+ * failed open. The new code protects the open with a mutex, so it's
+ * really quite straightforward. The mutex locking can probably be
* relaxed for the (most common) case of reopening a tty.
*/
static int init_dev(struct tty_driver *driver, int idx,
@@ -1640,7 +1640,7 @@ fast_track:
success:
*ret_tty = tty;
- /* All paths come through here to release the semaphore */
+ /* All paths come through here to release the mutex */
end_init:
return retval;
@@ -1837,7 +1837,7 @@ static void release_dev(struct file * filp)
/* Guard against races with tty->count changes elsewhere and
opens on /dev/tty */
- down(&tty_sem);
+ mutex_lock(&tty_mutex);
tty_closing = tty->count <= 1;
o_tty_closing = o_tty &&
(o_tty->count <= (pty_master ? 1 : 0));
@@ -1868,7 +1868,7 @@ static void release_dev(struct file * filp)
printk(KERN_WARNING "release_dev: %s: read/write wait queue "
"active!\n", tty_name(tty, buf));
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
schedule();
}
@@ -1934,7 +1934,7 @@ static void release_dev(struct file * filp)
read_unlock(&tasklist_lock);
}
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
/* check whether both sides are closing ... */
if (!tty_closing || (o_tty && !o_tty_closing))
@@ -2040,11 +2040,11 @@ retry_open:
index = -1;
retval = 0;
- down(&tty_sem);
+ mutex_lock(&tty_mutex);
if (device == MKDEV(TTYAUX_MAJOR,0)) {
if (!current->signal->tty) {
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
return -ENXIO;
}
driver = current->signal->tty->driver;
@@ -2070,18 +2070,18 @@ retry_open:
noctty = 1;
goto got_driver;
}
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
return -ENODEV;
}
driver = get_tty_driver(device, &index);
if (!driver) {
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
return -ENODEV;
}
got_driver:
retval = init_dev(driver, index, &tty);
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
if (retval)
return retval;
@@ -2167,9 +2167,9 @@ static int ptmx_open(struct inode * inode, struct file * filp)
}
up(&allocated_ptys_lock);
- down(&tty_sem);
+ mutex_lock(&tty_mutex);
retval = init_dev(ptm_driver, index, &tty);
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
if (retval)
goto out;
@@ -2915,8 +2915,8 @@ static void initialize_tty_struct(struct tty_struct *tty)
init_waitqueue_head(&tty->write_wait);
init_waitqueue_head(&tty->read_wait);
INIT_WORK(&tty->hangup_work, do_tty_hangup, tty);
- sema_init(&tty->atomic_read, 1);
- sema_init(&tty->atomic_write, 1);
+ mutex_init(&tty->atomic_read_lock);
+ mutex_init(&tty->atomic_write_lock);
spin_lock_init(&tty->read_lock);
INIT_LIST_HEAD(&tty->tty_files);
INIT_WORK(&tty->SAK_work, NULL, NULL);
diff --git a/drivers/char/vme_scc.c b/drivers/char/vme_scc.c
index d9325281e482..fd00822ac145 100644
--- a/drivers/char/vme_scc.c
+++ b/drivers/char/vme_scc.c
@@ -184,7 +184,7 @@ static void scc_init_portstructs(void)
port->gs.closing_wait = 30 * HZ;
port->gs.rd = &scc_real_driver;
#ifdef NEW_WRITE_LOCKING
- port->gs.port_write_sem = MUTEX;
+ port->gs.port_write_mutex = MUTEX;
#endif
init_waitqueue_head(&port->gs.open_wait);
init_waitqueue_head(&port->gs.close_wait);
diff --git a/drivers/char/vt.c b/drivers/char/vt.c
index 0900d1dbee59..ca4844c527da 100644
--- a/drivers/char/vt.c
+++ b/drivers/char/vt.c
@@ -2489,7 +2489,7 @@ static int con_open(struct tty_struct *tty, struct file *filp)
}
/*
- * We take tty_sem in here to prevent another thread from coming in via init_dev
+ * We take tty_mutex in here to prevent another thread from coming in via init_dev
* and taking a ref against the tty while we're in the process of forgetting
* about it and cleaning things up.
*
@@ -2497,7 +2497,7 @@ static int con_open(struct tty_struct *tty, struct file *filp)
*/
static void con_close(struct tty_struct *tty, struct file *filp)
{
- down(&tty_sem);
+ mutex_lock(&tty_mutex);
acquire_console_sem();
if (tty && tty->count == 1) {
struct vc_data *vc = tty->driver_data;
@@ -2507,15 +2507,15 @@ static void con_close(struct tty_struct *tty, struct file *filp)
tty->driver_data = NULL;
release_console_sem();
vcs_remove_devfs(tty);
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
/*
- * tty_sem is released, but we still hold BKL, so there is
+ * tty_mutex is released, but we still hold BKL, so there is
* still exclusion against init_dev()
*/
return;
}
release_console_sem();
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
}
static void vc_init(struct vc_data *vc, unsigned int rows,
@@ -2869,9 +2869,9 @@ void unblank_screen(void)
}
/*
- * We defer the timer blanking to work queue so it can take the console semaphore
+ * We defer the timer blanking to work queue so it can take the console mutex
* (console operations can still happen at irq time, but only from printk which
- * has the console semaphore. Not perfect yet, but better than no locking
+ * has the console mutex. Not perfect yet, but better than no locking
*/
static void blank_screen_t(unsigned long dummy)
{
@@ -3234,6 +3234,14 @@ void vcs_scr_writew(struct vc_data *vc, u16 val, u16 *org)
}
}
+int is_console_suspend_safe(void)
+{
+ /* It is unsafe to suspend devices while X has control of the
+ * hardware. Make sure we are running on a kernel-controlled console.
+ */
+ return vc_cons[fg_console].d->vc_mode == KD_TEXT;
+}
+
/*
* Visible symbols for modules
*/
diff --git a/drivers/char/watchdog/pcwd_usb.c b/drivers/char/watchdog/pcwd_usb.c
index 1533f56baa42..2700c5c45b8a 100644
--- a/drivers/char/watchdog/pcwd_usb.c
+++ b/drivers/char/watchdog/pcwd_usb.c
@@ -42,6 +42,7 @@
#include <linux/completion.h>
#include <asm/uaccess.h>
#include <linux/usb.h>
+#include <linux/mutex.h>
#ifdef CONFIG_USB_DEBUG
@@ -143,7 +144,7 @@ struct usb_pcwd_private {
static struct usb_pcwd_private *usb_pcwd_device;
/* prevent races between open() and disconnect() */
-static DECLARE_MUTEX (disconnect_sem);
+static DEFINE_MUTEX(disconnect_mutex);
/* local function prototypes */
static int usb_pcwd_probe (struct usb_interface *interface, const struct usb_device_id *id);
@@ -723,7 +724,7 @@ static void usb_pcwd_disconnect(struct usb_interface *interface)
struct usb_pcwd_private *usb_pcwd;
/* prevent races with open() */
- down (&disconnect_sem);
+ mutex_lock(&disconnect_mutex);
usb_pcwd = usb_get_intfdata (interface);
usb_set_intfdata (interface, NULL);
@@ -749,7 +750,7 @@ static void usb_pcwd_disconnect(struct usb_interface *interface)
cards_found--;
- up (&disconnect_sem);
+ mutex_unlock(&disconnect_mutex);
printk(KERN_INFO PFX "USB PC Watchdog disconnected\n");
}
diff --git a/drivers/connector/connector.c b/drivers/connector/connector.c
index d7125f4d9113..35897079a78d 100644
--- a/drivers/connector/connector.c
+++ b/drivers/connector/connector.c
@@ -26,6 +26,7 @@
#include <linux/netlink.h>
#include <linux/moduleparam.h>
#include <linux/connector.h>
+#include <linux/mutex.h>
#include <net/sock.h>
@@ -41,7 +42,7 @@ module_param(cn_val, uint, 0);
MODULE_PARM_DESC(cn_idx, "Connector's main device idx.");
MODULE_PARM_DESC(cn_val, "Connector's main device val.");
-static DECLARE_MUTEX(notify_lock);
+static DEFINE_MUTEX(notify_lock);
static LIST_HEAD(notify_list);
static struct cn_dev cdev;
@@ -260,7 +261,7 @@ static void cn_notify(struct cb_id *id, u32 notify_event)
{
struct cn_ctl_entry *ent;
- down(&notify_lock);
+ mutex_lock(&notify_lock);
list_for_each_entry(ent, &notify_list, notify_entry) {
int i;
struct cn_notify_req *req;
@@ -293,7 +294,7 @@ static void cn_notify(struct cb_id *id, u32 notify_event)
cn_netlink_send(&m, ctl->group, GFP_KERNEL);
}
}
- up(&notify_lock);
+ mutex_unlock(&notify_lock);
}
/*
@@ -407,14 +408,14 @@ static void cn_callback(void *data)
if (ctl->group == 0) {
struct cn_ctl_entry *n;
- down(&notify_lock);
+ mutex_lock(&notify_lock);
list_for_each_entry_safe(ent, n, &notify_list, notify_entry) {
if (cn_ctl_msg_equals(ent->msg, ctl)) {
list_del(&ent->notify_entry);
kfree(ent);
}
}
- up(&notify_lock);
+ mutex_unlock(&notify_lock);
return;
}
@@ -429,9 +430,9 @@ static void cn_callback(void *data)
memcpy(ent->msg, ctl, size - sizeof(*ent));
- down(&notify_lock);
+ mutex_lock(&notify_lock);
list_add(&ent->notify_entry, &notify_list);
- up(&notify_lock);
+ mutex_unlock(&notify_lock);
}
static int __init cn_init(void)
diff --git a/drivers/firmware/dcdbas.c b/drivers/firmware/dcdbas.c
index 3a4e5c5b4e1f..d6543fc4a923 100644
--- a/drivers/firmware/dcdbas.c
+++ b/drivers/firmware/dcdbas.c
@@ -33,6 +33,7 @@
#include <linux/spinlock.h>
#include <linux/string.h>
#include <linux/types.h>
+#include <linux/mutex.h>
#include <asm/io.h>
#include <asm/semaphore.h>
@@ -48,7 +49,7 @@ static u8 *smi_data_buf;
static dma_addr_t smi_data_buf_handle;
static unsigned long smi_data_buf_size;
static u32 smi_data_buf_phys_addr;
-static DECLARE_MUTEX(smi_data_lock);
+static DEFINE_MUTEX(smi_data_lock);
static unsigned int host_control_action;
static unsigned int host_control_smi_type;
@@ -139,9 +140,9 @@ static ssize_t smi_data_buf_size_store(struct device *dev,
buf_size = simple_strtoul(buf, NULL, 10);
/* make sure SMI data buffer is at least buf_size */
- down(&smi_data_lock);
+ mutex_lock(&smi_data_lock);
ret = smi_data_buf_realloc(buf_size);
- up(&smi_data_lock);
+ mutex_unlock(&smi_data_lock);
if (ret)
return ret;
@@ -154,7 +155,7 @@ static ssize_t smi_data_read(struct kobject *kobj, char *buf, loff_t pos,
size_t max_read;
ssize_t ret;
- down(&smi_data_lock);
+ mutex_lock(&smi_data_lock);
if (pos >= smi_data_buf_size) {
ret = 0;
@@ -165,7 +166,7 @@ static ssize_t smi_data_read(struct kobject *kobj, char *buf, loff_t pos,
ret = min(max_read, count);
memcpy(buf, smi_data_buf + pos, ret);
out:
- up(&smi_data_lock);
+ mutex_unlock(&smi_data_lock);
return ret;
}
@@ -174,7 +175,7 @@ static ssize_t smi_data_write(struct kobject *kobj, char *buf, loff_t pos,
{
ssize_t ret;
- down(&smi_data_lock);
+ mutex_lock(&smi_data_lock);
ret = smi_data_buf_realloc(pos + count);
if (ret)
@@ -183,7 +184,7 @@ static ssize_t smi_data_write(struct kobject *kobj, char *buf, loff_t pos,
memcpy(smi_data_buf + pos, buf, count);
ret = count;
out:
- up(&smi_data_lock);
+ mutex_unlock(&smi_data_lock);
return ret;
}
@@ -201,9 +202,9 @@ static ssize_t host_control_action_store(struct device *dev,
ssize_t ret;
/* make sure buffer is available for host control command */
- down(&smi_data_lock);
+ mutex_lock(&smi_data_lock);
ret = smi_data_buf_realloc(sizeof(struct apm_cmd));
- up(&smi_data_lock);
+ mutex_unlock(&smi_data_lock);
if (ret)
return ret;
@@ -302,7 +303,7 @@ static ssize_t smi_request_store(struct device *dev,
unsigned long val = simple_strtoul(buf, NULL, 10);
ssize_t ret;
- down(&smi_data_lock);
+ mutex_lock(&smi_data_lock);
if (smi_data_buf_size < sizeof(struct smi_cmd)) {
ret = -ENODEV;
@@ -334,7 +335,7 @@ static ssize_t smi_request_store(struct device *dev,
}
out:
- up(&smi_data_lock);
+ mutex_unlock(&smi_data_lock);
return ret;
}
diff --git a/drivers/ide/ide-cd.c b/drivers/ide/ide-cd.c
index 3325660f7248..c7671e188017 100644
--- a/drivers/ide/ide-cd.c
+++ b/drivers/ide/ide-cd.c
@@ -313,6 +313,7 @@
#include <linux/cdrom.h>
#include <linux/ide.h>
#include <linux/completion.h>
+#include <linux/mutex.h>
#include <scsi/scsi.h> /* For SCSI -> ATAPI command conversion */
@@ -324,7 +325,7 @@
#include "ide-cd.h"
-static DECLARE_MUTEX(idecd_ref_sem);
+static DEFINE_MUTEX(idecd_ref_mutex);
#define to_ide_cd(obj) container_of(obj, struct cdrom_info, kref)
@@ -335,11 +336,11 @@ static struct cdrom_info *ide_cd_get(struct gendisk *disk)
{
struct cdrom_info *cd = NULL;
- down(&idecd_ref_sem);
+ mutex_lock(&idecd_ref_mutex);
cd = ide_cd_g(disk);
if (cd)
kref_get(&cd->kref);
- up(&idecd_ref_sem);
+ mutex_unlock(&idecd_ref_mutex);
return cd;
}
@@ -347,9 +348,9 @@ static void ide_cd_release(struct kref *);
static void ide_cd_put(struct cdrom_info *cd)
{
- down(&idecd_ref_sem);
+ mutex_lock(&idecd_ref_mutex);
kref_put(&cd->kref, ide_cd_release);
- up(&idecd_ref_sem);
+ mutex_unlock(&idecd_ref_mutex);
}
/****************************************************************************
@@ -2471,52 +2472,6 @@ static int ide_cdrom_packet(struct cdrom_device_info *cdi,
}
static
-int ide_cdrom_dev_ioctl (struct cdrom_device_info *cdi,
- unsigned int cmd, unsigned long arg)
-{
- struct packet_command cgc;
- char buffer[16];
- int stat;
-
- init_cdrom_command(&cgc, buffer, sizeof(buffer), CGC_DATA_UNKNOWN);
-
- /* These will be moved into the Uniform layer shortly... */
- switch (cmd) {
- case CDROMSETSPINDOWN: {
- char spindown;
-
- if (copy_from_user(&spindown, (void __user *) arg, sizeof(char)))
- return -EFAULT;
-
- if ((stat = cdrom_mode_sense(cdi, &cgc, GPMODE_CDROM_PAGE, 0)))
- return stat;
-
- buffer[11] = (buffer[11] & 0xf0) | (spindown & 0x0f);
-
- return cdrom_mode_select(cdi, &cgc);
- }
-
- case CDROMGETSPINDOWN: {
- char spindown;
-
- if ((stat = cdrom_mode_sense(cdi, &cgc, GPMODE_CDROM_PAGE, 0)))
- return stat;
-
- spindown = buffer[11] & 0x0f;
-
- if (copy_to_user((void __user *) arg, &spindown, sizeof (char)))
- return -EFAULT;
-
- return 0;
- }
-
- default:
- return -EINVAL;
- }
-
-}
-
-static
int ide_cdrom_audio_ioctl (struct cdrom_device_info *cdi,
unsigned int cmd, void *arg)
@@ -2852,12 +2807,11 @@ static struct cdrom_device_ops ide_cdrom_dops = {
.get_mcn = ide_cdrom_get_mcn,
.reset = ide_cdrom_reset,
.audio_ioctl = ide_cdrom_audio_ioctl,
- .dev_ioctl = ide_cdrom_dev_ioctl,
.capability = CDC_CLOSE_TRAY | CDC_OPEN_TRAY | CDC_LOCK |
CDC_SELECT_SPEED | CDC_SELECT_DISC |
CDC_MULTI_SESSION | CDC_MCN |
CDC_MEDIA_CHANGED | CDC_PLAY_AUDIO | CDC_RESET |
- CDC_IOCTLS | CDC_DRIVE_STATUS | CDC_CD_R |
+ CDC_DRIVE_STATUS | CDC_CD_R |
CDC_CD_RW | CDC_DVD | CDC_DVD_R| CDC_DVD_RAM |
CDC_GENERIC_PACKET | CDC_MO_DRIVE | CDC_MRW |
CDC_MRW_W | CDC_RAM,
@@ -3367,6 +3321,45 @@ static int idecd_release(struct inode * inode, struct file * file)
return 0;
}
+static int idecd_set_spindown(struct cdrom_device_info *cdi, unsigned long arg)
+{
+ struct packet_command cgc;
+ char buffer[16];
+ int stat;
+ char spindown;
+
+ if (copy_from_user(&spindown, (void __user *)arg, sizeof(char)))
+ return -EFAULT;
+
+ init_cdrom_command(&cgc, buffer, sizeof(buffer), CGC_DATA_UNKNOWN);
+
+ stat = cdrom_mode_sense(cdi, &cgc, GPMODE_CDROM_PAGE, 0);
+ if (stat)
+ return stat;
+
+ buffer[11] = (buffer[11] & 0xf0) | (spindown & 0x0f);
+ return cdrom_mode_select(cdi, &cgc);
+}
+
+static int idecd_get_spindown(struct cdrom_device_info *cdi, unsigned long arg)
+{
+ struct packet_command cgc;
+ char buffer[16];
+ int stat;
+ char spindown;
+
+ init_cdrom_command(&cgc, buffer, sizeof(buffer), CGC_DATA_UNKNOWN);
+
+ stat = cdrom_mode_sense(cdi, &cgc, GPMODE_CDROM_PAGE, 0);
+ if (stat)
+ return stat;
+
+ spindown = buffer[11] & 0x0f;
+ if (copy_to_user((void __user *)arg, &spindown, sizeof (char)))
+ return -EFAULT;
+ return 0;
+}
+
static int idecd_ioctl (struct inode *inode, struct file *file,
unsigned int cmd, unsigned long arg)
{
@@ -3374,7 +3367,16 @@ static int idecd_ioctl (struct inode *inode, struct file *file,
struct cdrom_info *info = ide_cd_g(bdev->bd_disk);
int err;
- err = generic_ide_ioctl(info->drive, file, bdev, cmd, arg);
+ switch (cmd) {
+ case CDROMSETSPINDOWN:
+ return idecd_set_spindown(&info->devinfo, arg);
+ case CDROMGETSPINDOWN:
+ return idecd_get_spindown(&info->devinfo, arg);
+ default:
+ break;
+ }
+
+ err = generic_ide_ioctl(info->drive, file, bdev, cmd, arg);
if (err == -EINVAL)
err = cdrom_ioctl(file, &info->devinfo, inode, cmd, arg);
diff --git a/drivers/ide/ide-disk.c b/drivers/ide/ide-disk.c
index 09086b8b6486..e238b7da824b 100644
--- a/drivers/ide/ide-disk.c
+++ b/drivers/ide/ide-disk.c
@@ -60,6 +60,7 @@
#include <linux/genhd.h>
#include <linux/slab.h>
#include <linux/delay.h>
+#include <linux/mutex.h>
#define _IDE_DISK
@@ -78,7 +79,7 @@ struct ide_disk_obj {
struct kref kref;
};
-static DECLARE_MUTEX(idedisk_ref_sem);
+static DEFINE_MUTEX(idedisk_ref_mutex);
#define to_ide_disk(obj) container_of(obj, struct ide_disk_obj, kref)
@@ -89,11 +90,11 @@ static struct ide_disk_obj *ide_disk_get(struct gendisk *disk)
{
struct ide_disk_obj *idkp = NULL;
- down(&idedisk_ref_sem);
+ mutex_lock(&idedisk_ref_mutex);
idkp = ide_disk_g(disk);
if (idkp)
kref_get(&idkp->kref);
- up(&idedisk_ref_sem);
+ mutex_unlock(&idedisk_ref_mutex);
return idkp;
}
@@ -101,9 +102,9 @@ static void ide_disk_release(struct kref *);
static void ide_disk_put(struct ide_disk_obj *idkp)
{
- down(&idedisk_ref_sem);
+ mutex_lock(&idedisk_ref_mutex);
kref_put(&idkp->kref, ide_disk_release);
- up(&idedisk_ref_sem);
+ mutex_unlock(&idedisk_ref_mutex);
}
/*
diff --git a/drivers/ide/ide-floppy.c b/drivers/ide/ide-floppy.c
index 1f8db9ac05d1..a53e3ce4a142 100644
--- a/drivers/ide/ide-floppy.c
+++ b/drivers/ide/ide-floppy.c
@@ -98,6 +98,7 @@
#include <linux/cdrom.h>
#include <linux/ide.h>
#include <linux/bitops.h>
+#include <linux/mutex.h>
#include <asm/byteorder.h>
#include <asm/irq.h>
@@ -517,7 +518,7 @@ typedef struct {
u8 reserved[4];
} idefloppy_mode_parameter_header_t;
-static DECLARE_MUTEX(idefloppy_ref_sem);
+static DEFINE_MUTEX(idefloppy_ref_mutex);
#define to_ide_floppy(obj) container_of(obj, struct ide_floppy_obj, kref)
@@ -528,11 +529,11 @@ static struct ide_floppy_obj *ide_floppy_get(struct gendisk *disk)
{
struct ide_floppy_obj *floppy = NULL;
- down(&idefloppy_ref_sem);
+ mutex_lock(&idefloppy_ref_mutex);
floppy = ide_floppy_g(disk);
if (floppy)
kref_get(&floppy->kref);
- up(&idefloppy_ref_sem);
+ mutex_unlock(&idefloppy_ref_mutex);
return floppy;
}
@@ -540,9 +541,9 @@ static void ide_floppy_release(struct kref *);
static void ide_floppy_put(struct ide_floppy_obj *floppy)
{
- down(&idefloppy_ref_sem);
+ mutex_lock(&idefloppy_ref_mutex);
kref_put(&floppy->kref, ide_floppy_release);
- up(&idefloppy_ref_sem);
+ mutex_unlock(&idefloppy_ref_mutex);
}
/*
diff --git a/drivers/ide/ide-tape.c b/drivers/ide/ide-tape.c
index 0101d0def7c5..ebc59064b475 100644
--- a/drivers/ide/ide-tape.c
+++ b/drivers/ide/ide-tape.c
@@ -443,6 +443,7 @@
#include <linux/smp_lock.h>
#include <linux/completion.h>
#include <linux/bitops.h>
+#include <linux/mutex.h>
#include <asm/byteorder.h>
#include <asm/irq.h>
@@ -1011,7 +1012,7 @@ typedef struct ide_tape_obj {
int debug_level;
} idetape_tape_t;
-static DECLARE_MUTEX(idetape_ref_sem);
+static DEFINE_MUTEX(idetape_ref_mutex);
static struct class *idetape_sysfs_class;
@@ -1024,11 +1025,11 @@ static struct ide_tape_obj *ide_tape_get(struct gendisk *disk)
{
struct ide_tape_obj *tape = NULL;
- down(&idetape_ref_sem);
+ mutex_lock(&idetape_ref_mutex);
tape = ide_tape_g(disk);
if (tape)
kref_get(&tape->kref);
- up(&idetape_ref_sem);
+ mutex_unlock(&idetape_ref_mutex);
return tape;
}
@@ -1036,9 +1037,9 @@ static void ide_tape_release(struct kref *);
static void ide_tape_put(struct ide_tape_obj *tape)
{
- down(&idetape_ref_sem);
+ mutex_lock(&idetape_ref_mutex);
kref_put(&tape->kref, ide_tape_release);
- up(&idetape_ref_sem);
+ mutex_unlock(&idetape_ref_mutex);
}
/*
@@ -1290,11 +1291,11 @@ static struct ide_tape_obj *ide_tape_chrdev_get(unsigned int i)
{
struct ide_tape_obj *tape = NULL;
- down(&idetape_ref_sem);
+ mutex_lock(&idetape_ref_mutex);
tape = idetape_devs[i];
if (tape)
kref_get(&tape->kref);
- up(&idetape_ref_sem);
+ mutex_unlock(&idetape_ref_mutex);
return tape;
}
@@ -4870,11 +4871,11 @@ static int ide_tape_probe(ide_drive_t *drive)
drive->driver_data = tape;
- down(&idetape_ref_sem);
+ mutex_lock(&idetape_ref_mutex);
for (minor = 0; idetape_devs[minor]; minor++)
;
idetape_devs[minor] = tape;
- up(&idetape_ref_sem);
+ mutex_unlock(&idetape_ref_mutex);
idetape_setup(drive, tape, minor);
diff --git a/drivers/isdn/capi/kcapi.c b/drivers/isdn/capi/kcapi.c
index feec40cf5900..8c4fcb9027b3 100644
--- a/drivers/isdn/capi/kcapi.c
+++ b/drivers/isdn/capi/kcapi.c
@@ -32,6 +32,7 @@
#ifdef CONFIG_AVMB1_COMPAT
#include <linux/b1lli.h>
#endif
+#include <linux/mutex.h>
static char *revision = "$Revision: 1.1.2.8 $";
@@ -66,7 +67,7 @@ LIST_HEAD(capi_drivers);
DEFINE_RWLOCK(capi_drivers_list_lock);
static DEFINE_RWLOCK(application_lock);
-static DECLARE_MUTEX(controller_sem);
+static DEFINE_MUTEX(controller_mutex);
struct capi20_appl *capi_applications[CAPI_MAXAPPL];
struct capi_ctr *capi_cards[CAPI_MAXCONTR];
@@ -395,20 +396,20 @@ attach_capi_ctr(struct capi_ctr *card)
{
int i;
- down(&controller_sem);
+ mutex_lock(&controller_mutex);
for (i = 0; i < CAPI_MAXCONTR; i++) {
if (capi_cards[i] == NULL)
break;
}
if (i == CAPI_MAXCONTR) {
- up(&controller_sem);
+ mutex_unlock(&controller_mutex);
printk(KERN_ERR "kcapi: out of controller slots\n");
return -EBUSY;
}
capi_cards[i] = card;
- up(&controller_sem);
+ mutex_unlock(&controller_mutex);
card->nrecvctlpkt = 0;
card->nrecvdatapkt = 0;
@@ -531,13 +532,13 @@ u16 capi20_register(struct capi20_appl *ap)
write_unlock_irqrestore(&application_lock, flags);
- down(&controller_sem);
+ mutex_lock(&controller_mutex);
for (i = 0; i < CAPI_MAXCONTR; i++) {
if (!capi_cards[i] || capi_cards[i]->cardstate != CARD_RUNNING)
continue;
register_appl(capi_cards[i], applid, &ap->rparam);
}
- up(&controller_sem);
+ mutex_unlock(&controller_mutex);
if (showcapimsgs & 1) {
printk(KERN_DEBUG "kcapi: appl %d up\n", applid);
@@ -560,13 +561,13 @@ u16 capi20_release(struct capi20_appl *ap)
capi_applications[ap->applid - 1] = NULL;
write_unlock_irqrestore(&application_lock, flags);
- down(&controller_sem);
+ mutex_lock(&controller_mutex);
for (i = 0; i < CAPI_MAXCONTR; i++) {
if (!capi_cards[i] || capi_cards[i]->cardstate != CARD_RUNNING)
continue;
release_appl(capi_cards[i], ap->applid);
}
- up(&controller_sem);
+ mutex_unlock(&controller_mutex);
flush_scheduled_work();
skb_queue_purge(&ap->recv_queue);
diff --git a/drivers/isdn/hisax/config.c b/drivers/isdn/hisax/config.c
index df9d65201819..27332506f9f7 100644
--- a/drivers/isdn/hisax/config.c
+++ b/drivers/isdn/hisax/config.c
@@ -25,7 +25,6 @@
#include <linux/workqueue.h>
#include <linux/interrupt.h>
#define HISAX_STATUS_BUFSIZE 4096
-#define INCLUDE_INLINE_FUNCS
/*
* This structure array contains one entry per card. An entry looks
diff --git a/drivers/isdn/hisax/elsa.c b/drivers/isdn/hisax/elsa.c
index 110e9fd669c5..f8ca4b323331 100644
--- a/drivers/isdn/hisax/elsa.c
+++ b/drivers/isdn/hisax/elsa.c
@@ -108,7 +108,6 @@ static const char *ITACVer[] =
#define ELSA_ASSIGN 4
#define RS_ISR_PASS_LIMIT 256
-#define _INLINE_ inline
#define FLG_MODEM_ACTIVE 1
/* IPAC AUX */
#define ELSA_IPAC_LINE_LED 0x40 /* Bit 6 Gelbe LED */
diff --git a/drivers/net/loopback.c b/drivers/net/loopback.c
index 690a1aae0b34..0c13795dca38 100644
--- a/drivers/net/loopback.c
+++ b/drivers/net/loopback.c
@@ -172,11 +172,9 @@ static struct net_device_stats *get_stats(struct net_device *dev)
memset(stats, 0, sizeof(struct net_device_stats));
- for (i=0; i < NR_CPUS; i++) {
+ for_each_cpu(i) {
struct net_device_stats *lb_stats;
- if (!cpu_possible(i))
- continue;
lb_stats = &per_cpu(loopback_stats, i);
stats->rx_bytes += lb_stats->rx_bytes;
stats->tx_bytes += lb_stats->tx_bytes;
diff --git a/drivers/net/ppp_generic.c b/drivers/net/ppp_generic.c
index f608c12e3e8b..b2073fce8216 100644
--- a/drivers/net/ppp_generic.c
+++ b/drivers/net/ppp_generic.c
@@ -46,6 +46,7 @@
#include <linux/rwsem.h>
#include <linux/stddef.h>
#include <linux/device.h>
+#include <linux/mutex.h>
#include <net/slhc_vj.h>
#include <asm/atomic.h>
@@ -198,11 +199,11 @@ static unsigned int cardmap_find_first_free(struct cardmap *map);
static void cardmap_destroy(struct cardmap **map);
/*
- * all_ppp_sem protects the all_ppp_units mapping.
+ * all_ppp_mutex protects the all_ppp_units mapping.
* It also ensures that finding a ppp unit in the all_ppp_units map
* and updating its file.refcnt field is atomic.
*/
-static DECLARE_MUTEX(all_ppp_sem);
+static DEFINE_MUTEX(all_ppp_mutex);
static struct cardmap *all_ppp_units;
static atomic_t ppp_unit_count = ATOMIC_INIT(0);
@@ -804,7 +805,7 @@ static int ppp_unattached_ioctl(struct ppp_file *pf, struct file *file,
/* Attach to an existing ppp unit */
if (get_user(unit, p))
break;
- down(&all_ppp_sem);
+ mutex_lock(&all_ppp_mutex);
err = -ENXIO;
ppp = ppp_find_unit(unit);
if (ppp != 0) {
@@ -812,7 +813,7 @@ static int ppp_unattached_ioctl(struct ppp_file *pf, struct file *file,
file->private_data = &ppp->file;
err = 0;
}
- up(&all_ppp_sem);
+ mutex_unlock(&all_ppp_mutex);
break;
case PPPIOCATTCHAN:
@@ -2446,7 +2447,7 @@ ppp_create_interface(int unit, int *retp)
dev->do_ioctl = ppp_net_ioctl;
ret = -EEXIST;
- down(&all_ppp_sem);
+ mutex_lock(&all_ppp_mutex);
if (unit < 0)
unit = cardmap_find_first_free(all_ppp_units);
else if (cardmap_get(all_ppp_units, unit) != NULL)
@@ -2465,12 +2466,12 @@ ppp_create_interface(int unit, int *retp)
atomic_inc(&ppp_unit_count);
cardmap_set(&all_ppp_units, unit, ppp);
- up(&all_ppp_sem);
+ mutex_unlock(&all_ppp_mutex);
*retp = 0;
return ppp;
out2:
- up(&all_ppp_sem);
+ mutex_unlock(&all_ppp_mutex);
free_netdev(dev);
out1:
kfree(ppp);
@@ -2500,7 +2501,7 @@ static void ppp_shutdown_interface(struct ppp *ppp)
{
struct net_device *dev;
- down(&all_ppp_sem);
+ mutex_lock(&all_ppp_mutex);
ppp_lock(ppp);
dev = ppp->dev;
ppp->dev = NULL;
@@ -2514,7 +2515,7 @@ static void ppp_shutdown_interface(struct ppp *ppp)
ppp->file.dead = 1;
ppp->owner = NULL;
wake_up_interruptible(&ppp->file.rwait);
- up(&all_ppp_sem);
+ mutex_unlock(&all_ppp_mutex);
}
/*
@@ -2556,7 +2557,7 @@ static void ppp_destroy_interface(struct ppp *ppp)
/*
* Locate an existing ppp unit.
- * The caller should have locked the all_ppp_sem.
+ * The caller should have locked the all_ppp_mutex.
*/
static struct ppp *
ppp_find_unit(int unit)
@@ -2601,7 +2602,7 @@ ppp_connect_channel(struct channel *pch, int unit)
int ret = -ENXIO;
int hdrlen;
- down(&all_ppp_sem);
+ mutex_lock(&all_ppp_mutex);
ppp = ppp_find_unit(unit);
if (ppp == 0)
goto out;
@@ -2626,7 +2627,7 @@ ppp_connect_channel(struct channel *pch, int unit)
outl:
write_unlock_bh(&pch->upl);
out:
- up(&all_ppp_sem);
+ mutex_unlock(&all_ppp_mutex);
return ret;
}
diff --git a/drivers/oprofile/cpu_buffer.c b/drivers/oprofile/cpu_buffer.c
index 78193e4bbdb5..330d3869b41e 100644
--- a/drivers/oprofile/cpu_buffer.c
+++ b/drivers/oprofile/cpu_buffer.c
@@ -38,9 +38,8 @@ void free_cpu_buffers(void)
{
int i;
- for_each_online_cpu(i) {
+ for_each_online_cpu(i)
vfree(cpu_buffer[i].buffer);
- }
}
int alloc_cpu_buffers(void)
diff --git a/drivers/pnp/pnpbios/rsparser.c b/drivers/pnp/pnpbios/rsparser.c
index 5e38cd7335f7..c89c98a2cca8 100644
--- a/drivers/pnp/pnpbios/rsparser.c
+++ b/drivers/pnp/pnpbios/rsparser.c
@@ -448,11 +448,7 @@ pnpbios_parse_resource_option_data(unsigned char * p, unsigned char * end, struc
break;
case SMALL_TAG_END:
- if (option_independent != option)
- printk(KERN_WARNING "PnPBIOS: Missing SMALL_TAG_ENDDEP tag\n");
- p = p + 2;
- return (unsigned char *)p;
- break;
+ return p + 2;
default: /* an unkown tag */
len_err:
diff --git a/drivers/s390/block/dasd_ioctl.c b/drivers/s390/block/dasd_ioctl.c
index fafeeae52675..f9930552ab54 100644
--- a/drivers/s390/block/dasd_ioctl.c
+++ b/drivers/s390/block/dasd_ioctl.c
@@ -151,9 +151,9 @@ dasd_ioctl_enable(struct block_device *bdev, int no, long args)
return -ENODEV;
dasd_enable_device(device);
/* Formatting the dasd device can change the capacity. */
- down(&bdev->bd_sem);
+ mutex_lock(&bdev->bd_mutex);
i_size_write(bdev->bd_inode, (loff_t)get_capacity(device->gdp) << 9);
- up(&bdev->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
return 0;
}
@@ -184,9 +184,9 @@ dasd_ioctl_disable(struct block_device *bdev, int no, long args)
* Set i_size to zero, since read, write, etc. check against this
* value.
*/
- down(&bdev->bd_sem);
+ mutex_lock(&bdev->bd_mutex);
i_size_write(bdev->bd_inode, 0);
- up(&bdev->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
return 0;
}
diff --git a/drivers/scsi/ide-scsi.c b/drivers/scsi/ide-scsi.c
index 0cf0e4c7ac0c..39b760a24241 100644
--- a/drivers/scsi/ide-scsi.c
+++ b/drivers/scsi/ide-scsi.c
@@ -47,6 +47,7 @@
#include <linux/ide.h>
#include <linux/scatterlist.h>
#include <linux/delay.h>
+#include <linux/mutex.h>
#include <asm/io.h>
#include <asm/bitops.h>
@@ -109,7 +110,7 @@ typedef struct ide_scsi_obj {
unsigned long log; /* log flags */
} idescsi_scsi_t;
-static DECLARE_MUTEX(idescsi_ref_sem);
+static DEFINE_MUTEX(idescsi_ref_mutex);
#define ide_scsi_g(disk) \
container_of((disk)->private_data, struct ide_scsi_obj, driver)
@@ -118,19 +119,19 @@ static struct ide_scsi_obj *ide_scsi_get(struct gendisk *disk)
{
struct ide_scsi_obj *scsi = NULL;
- down(&idescsi_ref_sem);
+ mutex_lock(&idescsi_ref_mutex);
scsi = ide_scsi_g(disk);
if (scsi)
scsi_host_get(scsi->host);
- up(&idescsi_ref_sem);
+ mutex_unlock(&idescsi_ref_mutex);
return scsi;
}
static void ide_scsi_put(struct ide_scsi_obj *scsi)
{
- down(&idescsi_ref_sem);
+ mutex_lock(&idescsi_ref_mutex);
scsi_host_put(scsi->host);
- up(&idescsi_ref_sem);
+ mutex_unlock(&idescsi_ref_mutex);
}
static inline idescsi_scsi_t *scsihost_to_idescsi(struct Scsi_Host *host)
diff --git a/drivers/scsi/sr.c b/drivers/scsi/sr.c
index f9c1192dc15e..7c80711e18ed 100644
--- a/drivers/scsi/sr.c
+++ b/drivers/scsi/sr.c
@@ -71,7 +71,7 @@ MODULE_ALIAS_BLOCKDEV_MAJOR(SCSI_CDROM_MAJOR);
#define SR_CAPABILITIES \
(CDC_CLOSE_TRAY|CDC_OPEN_TRAY|CDC_LOCK|CDC_SELECT_SPEED| \
CDC_SELECT_DISC|CDC_MULTI_SESSION|CDC_MCN|CDC_MEDIA_CHANGED| \
- CDC_PLAY_AUDIO|CDC_RESET|CDC_IOCTLS|CDC_DRIVE_STATUS| \
+ CDC_PLAY_AUDIO|CDC_RESET|CDC_DRIVE_STATUS| \
CDC_CD_R|CDC_CD_RW|CDC_DVD|CDC_DVD_R|CDC_DVD_RAM|CDC_GENERIC_PACKET| \
CDC_MRW|CDC_MRW_W|CDC_RAM)
@@ -118,7 +118,6 @@ static struct cdrom_device_ops sr_dops = {
.get_mcn = sr_get_mcn,
.reset = sr_reset,
.audio_ioctl = sr_audio_ioctl,
- .dev_ioctl = sr_dev_ioctl,
.capability = SR_CAPABILITIES,
.generic_packet = sr_packet,
};
@@ -456,17 +455,33 @@ static int sr_block_ioctl(struct inode *inode, struct file *file, unsigned cmd,
{
struct scsi_cd *cd = scsi_cd(inode->i_bdev->bd_disk);
struct scsi_device *sdev = cd->device;
+ void __user *argp = (void __user *)arg;
+ int ret;
- /*
- * Send SCSI addressing ioctls directly to mid level, send other
- * ioctls to cdrom/block level.
- */
- switch (cmd) {
- case SCSI_IOCTL_GET_IDLUN:
- case SCSI_IOCTL_GET_BUS_NUMBER:
- return scsi_ioctl(sdev, cmd, (void __user *)arg);
+ /*
+ * Send SCSI addressing ioctls directly to mid level, send other
+ * ioctls to cdrom/block level.
+ */
+ switch (cmd) {
+ case SCSI_IOCTL_GET_IDLUN:
+ case SCSI_IOCTL_GET_BUS_NUMBER:
+ return scsi_ioctl(sdev, cmd, argp);
}
- return cdrom_ioctl(file, &cd->cdi, inode, cmd, arg);
+
+ ret = cdrom_ioctl(file, &cd->cdi, inode, cmd, arg);
+ if (ret != ENOSYS)
+ return ret;
+
+ /*
+ * ENODEV means that we didn't recognise the ioctl, or that we
+ * cannot execute it in the current device state. In either
+ * case fall through to scsi_ioctl, which will return ENDOEV again
+ * if it doesn't recognise the ioctl
+ */
+ ret = scsi_nonblockable_ioctl(sdev, cmd, argp, NULL);
+ if (ret != -ENODEV)
+ return ret;
+ return scsi_ioctl(sdev, cmd, argp);
}
static int sr_block_media_changed(struct gendisk *disk)
diff --git a/drivers/scsi/sr.h b/drivers/scsi/sr.h
index d2bcd99c272f..d65de9621b27 100644
--- a/drivers/scsi/sr.h
+++ b/drivers/scsi/sr.h
@@ -55,7 +55,6 @@ int sr_get_mcn(struct cdrom_device_info *, struct cdrom_mcn *);
int sr_reset(struct cdrom_device_info *);
int sr_select_speed(struct cdrom_device_info *cdi, int speed);
int sr_audio_ioctl(struct cdrom_device_info *, unsigned int, void *);
-int sr_dev_ioctl(struct cdrom_device_info *, unsigned int, unsigned long);
int sr_is_xa(Scsi_CD *);
diff --git a/drivers/scsi/sr_ioctl.c b/drivers/scsi/sr_ioctl.c
index b65462f76484..d1268cb46837 100644
--- a/drivers/scsi/sr_ioctl.c
+++ b/drivers/scsi/sr_ioctl.c
@@ -562,22 +562,3 @@ int sr_is_xa(Scsi_CD *cd)
#endif
return is_xa;
}
-
-int sr_dev_ioctl(struct cdrom_device_info *cdi,
- unsigned int cmd, unsigned long arg)
-{
- Scsi_CD *cd = cdi->handle;
- int ret;
-
- ret = scsi_nonblockable_ioctl(cd->device, cmd,
- (void __user *)arg, NULL);
- /*
- * ENODEV means that we didn't recognise the ioctl, or that we
- * cannot execute it in the current device state. In either
- * case fall through to scsi_ioctl, which will return ENDOEV again
- * if it doesn't recognise the ioctl
- */
- if (ret != -ENODEV)
- return ret;
- return scsi_ioctl(cd->device, cmd, (void __user *)arg);
-}
diff --git a/drivers/serial/68328serial.c b/drivers/serial/68328serial.c
index 7f0f35a05dca..b88a7c1158af 100644
--- a/drivers/serial/68328serial.c
+++ b/drivers/serial/68328serial.c
@@ -101,8 +101,6 @@ struct tty_driver *serial_driver;
#define RS_ISR_PASS_LIMIT 256
-#define _INLINE_ inline
-
static void change_speed(struct m68k_serial *info);
/*
@@ -262,7 +260,7 @@ static void batten_down_hatches(void)
/* Drop into the debugger */
}
-static _INLINE_ void status_handle(struct m68k_serial *info, unsigned short status)
+static void status_handle(struct m68k_serial *info, unsigned short status)
{
#if 0
if(status & DCD) {
@@ -289,7 +287,8 @@ static _INLINE_ void status_handle(struct m68k_serial *info, unsigned short stat
return;
}
-static _INLINE_ void receive_chars(struct m68k_serial *info, struct pt_regs *regs, unsigned short rx)
+static void receive_chars(struct m68k_serial *info, struct pt_regs *regs,
+ unsigned short rx)
{
struct tty_struct *tty = info->tty;
m68328_uart *uart = &uart_addr[info->line];
@@ -359,7 +358,7 @@ clear_and_exit:
return;
}
-static _INLINE_ void transmit_chars(struct m68k_serial *info)
+static void transmit_chars(struct m68k_serial *info)
{
m68328_uart *uart = &uart_addr[info->line];
diff --git a/drivers/serial/au1x00_uart.c b/drivers/serial/au1x00_uart.c
index 29f94bbb79be..948880ac5878 100644
--- a/drivers/serial/au1x00_uart.c
+++ b/drivers/serial/au1x00_uart.c
@@ -133,13 +133,12 @@ static const struct serial_uart_config uart_config[PORT_MAX_8250+1] = {
{ "AU1X00_UART",16, UART_CLEAR_FIFO | UART_USE_FIFO },
};
-static _INLINE_ unsigned int serial_in(struct uart_8250_port *up, int offset)
+static unsigned int serial_in(struct uart_8250_port *up, int offset)
{
return au_readl((unsigned long)up->port.membase + offset);
}
-static _INLINE_ void
-serial_out(struct uart_8250_port *up, int offset, int value)
+static void serial_out(struct uart_8250_port *up, int offset, int value)
{
au_writel(value, (unsigned long)up->port.membase + offset);
}
@@ -237,7 +236,7 @@ static void serial8250_enable_ms(struct uart_port *port)
serial_out(up, UART_IER, up->ier);
}
-static _INLINE_ void
+static void
receive_chars(struct uart_8250_port *up, int *status, struct pt_regs *regs)
{
struct tty_struct *tty = up->port.info->tty;
@@ -312,7 +311,7 @@ receive_chars(struct uart_8250_port *up, int *status, struct pt_regs *regs)
spin_lock(&up->port.lock);
}
-static _INLINE_ void transmit_chars(struct uart_8250_port *up)
+static void transmit_chars(struct uart_8250_port *up)
{
struct circ_buf *xmit = &up->port.info->xmit;
int count;
@@ -346,7 +345,7 @@ static _INLINE_ void transmit_chars(struct uart_8250_port *up)
serial8250_stop_tx(&up->port);
}
-static _INLINE_ void check_modem_status(struct uart_8250_port *up)
+static void check_modem_status(struct uart_8250_port *up)
{
int status;
diff --git a/drivers/serial/crisv10.c b/drivers/serial/crisv10.c
index be12623d8544..89700141f87e 100644
--- a/drivers/serial/crisv10.c
+++ b/drivers/serial/crisv10.c
@@ -481,8 +481,6 @@ static char *serial_version = "$Revision: 1.25 $";
#include "serial_compat.h"
#endif
-#define _INLINE_ inline
-
struct tty_driver *serial_driver;
/* serial subtype definitions */
@@ -591,8 +589,6 @@ static void rs_throttle(struct tty_struct * tty);
static void rs_wait_until_sent(struct tty_struct *tty, int timeout);
static int rs_write(struct tty_struct * tty, int from_user,
const unsigned char *buf, int count);
-extern _INLINE_ int rs_raw_write(struct tty_struct * tty, int from_user,
- const unsigned char *buf, int count);
#ifdef CONFIG_ETRAX_RS485
static int e100_write_rs485(struct tty_struct * tty, int from_user,
const unsigned char *buf, int count);
@@ -1538,8 +1534,7 @@ e100_enable_rxdma_irq(struct e100_serial *info)
/* the tx DMA uses only dma_descr interrupt */
-static _INLINE_ void
-e100_disable_txdma_irq(struct e100_serial *info)
+static void e100_disable_txdma_irq(struct e100_serial *info)
{
#ifdef SERIAL_DEBUG_INTR
printk("txdma_irq(%d): 0\n",info->line);
@@ -1548,8 +1543,7 @@ e100_disable_txdma_irq(struct e100_serial *info)
*R_IRQ_MASK2_CLR = info->irq;
}
-static _INLINE_ void
-e100_enable_txdma_irq(struct e100_serial *info)
+static void e100_enable_txdma_irq(struct e100_serial *info)
{
#ifdef SERIAL_DEBUG_INTR
printk("txdma_irq(%d): 1\n",info->line);
@@ -1558,8 +1552,7 @@ e100_enable_txdma_irq(struct e100_serial *info)
*R_IRQ_MASK2_SET = info->irq;
}
-static _INLINE_ void
-e100_disable_txdma_channel(struct e100_serial *info)
+static void e100_disable_txdma_channel(struct e100_serial *info)
{
unsigned long flags;
@@ -1599,8 +1592,7 @@ e100_disable_txdma_channel(struct e100_serial *info)
}
-static _INLINE_ void
-e100_enable_txdma_channel(struct e100_serial *info)
+static void e100_enable_txdma_channel(struct e100_serial *info)
{
unsigned long flags;
@@ -1625,8 +1617,7 @@ e100_enable_txdma_channel(struct e100_serial *info)
restore_flags(flags);
}
-static _INLINE_ void
-e100_disable_rxdma_channel(struct e100_serial *info)
+static void e100_disable_rxdma_channel(struct e100_serial *info)
{
unsigned long flags;
@@ -1665,8 +1656,7 @@ e100_disable_rxdma_channel(struct e100_serial *info)
}
-static _INLINE_ void
-e100_enable_rxdma_channel(struct e100_serial *info)
+static void e100_enable_rxdma_channel(struct e100_serial *info)
{
unsigned long flags;
@@ -1913,9 +1903,7 @@ rs_start(struct tty_struct *tty)
* This routine is used by the interrupt handler to schedule
* processing in the software interrupt portion of the driver.
*/
-static _INLINE_ void
-rs_sched_event(struct e100_serial *info,
- int event)
+static void rs_sched_event(struct e100_serial *info, int event)
{
if (info->event & (1 << event))
return;
@@ -2155,8 +2143,9 @@ add_char_and_flag(struct e100_serial *info, unsigned char data, unsigned char fl
return 1;
}
-extern _INLINE_ unsigned int
-handle_descr_data(struct e100_serial *info, struct etrax_dma_descr *descr, unsigned int recvl)
+static unsigned int handle_descr_data(struct e100_serial *info,
+ struct etrax_dma_descr *descr,
+ unsigned int recvl)
{
struct etrax_recv_buffer *buffer = phys_to_virt(descr->buf) - sizeof *buffer;
@@ -2182,8 +2171,7 @@ handle_descr_data(struct e100_serial *info, struct etrax_dma_descr *descr, unsig
return recvl;
}
-static _INLINE_ unsigned int
-handle_all_descr_data(struct e100_serial *info)
+static unsigned int handle_all_descr_data(struct e100_serial *info)
{
struct etrax_dma_descr *descr;
unsigned int recvl;
@@ -2230,8 +2218,7 @@ handle_all_descr_data(struct e100_serial *info)
return ret;
}
-static _INLINE_ void
-receive_chars_dma(struct e100_serial *info)
+static void receive_chars_dma(struct e100_serial *info)
{
struct tty_struct *tty;
unsigned char rstat;
@@ -2292,8 +2279,7 @@ receive_chars_dma(struct e100_serial *info)
*info->icmdadr = IO_STATE(R_DMA_CH6_CMD, cmd, restart);
}
-static _INLINE_ int
-start_recv_dma(struct e100_serial *info)
+static int start_recv_dma(struct e100_serial *info)
{
struct etrax_dma_descr *descr = info->rec_descr;
struct etrax_recv_buffer *buffer;
@@ -2348,11 +2334,6 @@ start_receive(struct e100_serial *info)
}
-static _INLINE_ void
-status_handle(struct e100_serial *info, unsigned short status)
-{
-}
-
/* the bits in the MASK2 register are laid out like this:
DMAI_EOP DMAI_DESCR DMAO_EOP DMAO_DESCR
where I is the input channel and O is the output channel for the port.
@@ -2454,8 +2435,7 @@ rec_interrupt(int irq, void *dev_id, struct pt_regs * regs)
return IRQ_RETVAL(handled);
} /* rec_interrupt */
-static _INLINE_ int
-force_eop_if_needed(struct e100_serial *info)
+static int force_eop_if_needed(struct e100_serial *info)
{
/* We check data_avail bit to determine if data has
* arrived since last time
@@ -2499,8 +2479,7 @@ force_eop_if_needed(struct e100_serial *info)
return 1;
}
-extern _INLINE_ void
-flush_to_flip_buffer(struct e100_serial *info)
+static void flush_to_flip_buffer(struct e100_serial *info)
{
struct tty_struct *tty;
struct etrax_recv_buffer *buffer;
@@ -2611,8 +2590,7 @@ flush_to_flip_buffer(struct e100_serial *info)
tty_flip_buffer_push(tty);
}
-static _INLINE_ void
-check_flush_timeout(struct e100_serial *info)
+static void check_flush_timeout(struct e100_serial *info)
{
/* Flip what we've got (if we can) */
flush_to_flip_buffer(info);
@@ -2741,7 +2719,7 @@ TODO: The break will be delayed until an F or V character is received.
*/
-extern _INLINE_
+static
struct e100_serial * handle_ser_rx_interrupt_no_dma(struct e100_serial *info)
{
unsigned long data_read;
@@ -2875,8 +2853,7 @@ more_data:
return info;
}
-extern _INLINE_
-struct e100_serial* handle_ser_rx_interrupt(struct e100_serial *info)
+static struct e100_serial* handle_ser_rx_interrupt(struct e100_serial *info)
{
unsigned char rstat;
@@ -2995,7 +2972,7 @@ struct e100_serial* handle_ser_rx_interrupt(struct e100_serial *info)
return info;
} /* handle_ser_rx_interrupt */
-extern _INLINE_ void handle_ser_tx_interrupt(struct e100_serial *info)
+static void handle_ser_tx_interrupt(struct e100_serial *info)
{
unsigned long flags;
@@ -3621,9 +3598,8 @@ rs_flush_chars(struct tty_struct *tty)
restore_flags(flags);
}
-extern _INLINE_ int
-rs_raw_write(struct tty_struct * tty, int from_user,
- const unsigned char *buf, int count)
+static int rs_raw_write(struct tty_struct * tty, int from_user,
+ const unsigned char *buf, int count)
{
int c, ret = 0;
struct e100_serial *info = (struct e100_serial *)tty->driver_data;
@@ -4710,7 +4686,7 @@ rs_open(struct tty_struct *tty, struct file * filp)
* /proc fs routines....
*/
-extern _INLINE_ int line_info(char *buf, struct e100_serial *info)
+static int line_info(char *buf, struct e100_serial *info)
{
char stat_buf[30];
int ret;
diff --git a/drivers/serial/m32r_sio.c b/drivers/serial/m32r_sio.c
index 876bc5e027bb..e9c10c0a30fc 100644
--- a/drivers/serial/m32r_sio.c
+++ b/drivers/serial/m32r_sio.c
@@ -248,17 +248,17 @@ static void sio_error(int *status)
#endif /* CONFIG_SERIAL_M32R_PLDSIO */
-static _INLINE_ unsigned int sio_in(struct uart_sio_port *up, int offset)
+static unsigned int sio_in(struct uart_sio_port *up, int offset)
{
return __sio_in(up->port.iobase + offset);
}
-static _INLINE_ void sio_out(struct uart_sio_port *up, int offset, int value)
+static void sio_out(struct uart_sio_port *up, int offset, int value)
{
__sio_out(value, up->port.iobase + offset);
}
-static _INLINE_ unsigned int serial_in(struct uart_sio_port *up, int offset)
+static unsigned int serial_in(struct uart_sio_port *up, int offset)
{
if (!offset)
return 0;
@@ -266,8 +266,7 @@ static _INLINE_ unsigned int serial_in(struct uart_sio_port *up, int offset)
return __sio_in(offset);
}
-static _INLINE_ void
-serial_out(struct uart_sio_port *up, int offset, int value)
+static void serial_out(struct uart_sio_port *up, int offset, int value)
{
if (!offset)
return;
@@ -326,8 +325,8 @@ static void m32r_sio_enable_ms(struct uart_port *port)
serial_out(up, UART_IER, up->ier);
}
-static _INLINE_ void receive_chars(struct uart_sio_port *up, int *status,
- struct pt_regs *regs)
+static void receive_chars(struct uart_sio_port *up, int *status,
+ struct pt_regs *regs)
{
struct tty_struct *tty = up->port.info->tty;
unsigned char ch;
@@ -400,7 +399,7 @@ static _INLINE_ void receive_chars(struct uart_sio_port *up, int *status,
tty_flip_buffer_push(tty);
}
-static _INLINE_ void transmit_chars(struct uart_sio_port *up)
+static void transmit_chars(struct uart_sio_port *up)
{
struct circ_buf *xmit = &up->port.info->xmit;
int count;
diff --git a/drivers/serial/sunsu.c b/drivers/serial/sunsu.c
index 7fc3d3b41d18..9fe2283d91e5 100644
--- a/drivers/serial/sunsu.c
+++ b/drivers/serial/sunsu.c
@@ -102,9 +102,7 @@ struct uart_sunsu_port {
#endif
};
-#define _INLINE_
-
-static _INLINE_ unsigned int serial_in(struct uart_sunsu_port *up, int offset)
+static unsigned int serial_in(struct uart_sunsu_port *up, int offset)
{
offset <<= up->port.regshift;
@@ -121,8 +119,7 @@ static _INLINE_ unsigned int serial_in(struct uart_sunsu_port *up, int offset)
}
}
-static _INLINE_ void
-serial_out(struct uart_sunsu_port *up, int offset, int value)
+static void serial_out(struct uart_sunsu_port *up, int offset, int value)
{
#ifndef CONFIG_SPARC64
/*
@@ -316,7 +313,7 @@ static void sunsu_enable_ms(struct uart_port *port)
spin_unlock_irqrestore(&up->port.lock, flags);
}
-static _INLINE_ struct tty_struct *
+static struct tty_struct *
receive_chars(struct uart_sunsu_port *up, unsigned char *status, struct pt_regs *regs)
{
struct tty_struct *tty = up->port.info->tty;
@@ -395,7 +392,7 @@ receive_chars(struct uart_sunsu_port *up, unsigned char *status, struct pt_regs
return tty;
}
-static _INLINE_ void transmit_chars(struct uart_sunsu_port *up)
+static void transmit_chars(struct uart_sunsu_port *up)
{
struct circ_buf *xmit = &up->port.info->xmit;
int count;
@@ -431,7 +428,7 @@ static _INLINE_ void transmit_chars(struct uart_sunsu_port *up)
__stop_tx(up);
}
-static _INLINE_ void check_modem_status(struct uart_sunsu_port *up)
+static void check_modem_status(struct uart_sunsu_port *up)
{
int status;
diff --git a/drivers/tc/zs.c b/drivers/tc/zs.c
index 6756d0fab6fe..2dffa8e303b2 100644
--- a/drivers/tc/zs.c
+++ b/drivers/tc/zs.c
@@ -186,8 +186,6 @@ static struct tty_driver *serial_driver;
#define RS_STROBE_TIME 10
#define RS_ISR_PASS_LIMIT 256
-#define _INLINE_ inline
-
static void probe_sccs(void);
static void change_speed(struct dec_serial *info);
static void rs_wait_until_sent(struct tty_struct *tty, int timeout);
@@ -344,14 +342,13 @@ static inline void rs_recv_clear(struct dec_zschannel *zsc)
* This routine is used by the interrupt handler to schedule
* processing in the software interrupt portion of the driver.
*/
-static _INLINE_ void rs_sched_event(struct dec_serial *info, int event)
+static void rs_sched_event(struct dec_serial *info, int event)
{
info->event |= 1 << event;
tasklet_schedule(&info->tlet);
}
-static _INLINE_ void receive_chars(struct dec_serial *info,
- struct pt_regs *regs)
+static void receive_chars(struct dec_serial *info, struct pt_regs *regs)
{
struct tty_struct *tty = info->tty;
unsigned char ch, stat, flag;
@@ -441,7 +438,7 @@ static void transmit_chars(struct dec_serial *info)
rs_sched_event(info, RS_EVENT_WRITE_WAKEUP);
}
-static _INLINE_ void status_handle(struct dec_serial *info)
+static void status_handle(struct dec_serial *info)
{
unsigned char stat;
diff --git a/fs/9p/mux.c b/fs/9p/mux.c
index ea1134eb47c8..8e8356c1c229 100644
--- a/fs/9p/mux.c
+++ b/fs/9p/mux.c
@@ -31,6 +31,7 @@
#include <linux/poll.h>
#include <linux/kthread.h>
#include <linux/idr.h>
+#include <linux/mutex.h>
#include "debug.h"
#include "v9fs.h"
@@ -110,7 +111,7 @@ static void v9fs_pollwait(struct file *filp, wait_queue_head_t * wait_address,
static u16 v9fs_mux_get_tag(struct v9fs_mux_data *);
static void v9fs_mux_put_tag(struct v9fs_mux_data *, u16);
-static DECLARE_MUTEX(v9fs_mux_task_lock);
+static DEFINE_MUTEX(v9fs_mux_task_lock);
static struct workqueue_struct *v9fs_mux_wq;
static int v9fs_mux_num;
@@ -166,7 +167,7 @@ static int v9fs_mux_poll_start(struct v9fs_mux_data *m)
dprintk(DEBUG_MUX, "mux %p muxnum %d procnum %d\n", m, v9fs_mux_num,
v9fs_mux_poll_task_num);
- up(&v9fs_mux_task_lock);
+ mutex_lock(&v9fs_mux_task_lock);
n = v9fs_mux_calc_poll_procs(v9fs_mux_num + 1);
if (n > v9fs_mux_poll_task_num) {
@@ -225,7 +226,7 @@ static int v9fs_mux_poll_start(struct v9fs_mux_data *m)
}
v9fs_mux_num++;
- down(&v9fs_mux_task_lock);
+ mutex_unlock(&v9fs_mux_task_lock);
return 0;
}
@@ -235,7 +236,7 @@ static void v9fs_mux_poll_stop(struct v9fs_mux_data *m)
int i;
struct v9fs_mux_poll_task *vpt;
- up(&v9fs_mux_task_lock);
+ mutex_lock(&v9fs_mux_task_lock);
vpt = m->poll_task;
list_del(&m->mux_list);
for(i = 0; i < ARRAY_SIZE(m->poll_waddr); i++) {
@@ -252,7 +253,7 @@ static void v9fs_mux_poll_stop(struct v9fs_mux_data *m)
v9fs_mux_poll_task_num--;
}
v9fs_mux_num--;
- down(&v9fs_mux_task_lock);
+ mutex_unlock(&v9fs_mux_task_lock);
}
/**
diff --git a/fs/adfs/file.c b/fs/adfs/file.c
index afebbfde6968..6af10885f9d6 100644
--- a/fs/adfs/file.c
+++ b/fs/adfs/file.c
@@ -19,11 +19,7 @@
*
* adfs regular file handling primitives
*/
-#include <linux/errno.h>
#include <linux/fs.h>
-#include <linux/fcntl.h>
-#include <linux/time.h>
-#include <linux/stat.h>
#include <linux/buffer_head.h> /* for file_fsync() */
#include <linux/adfs_fs.h>
diff --git a/fs/autofs4/autofs_i.h b/fs/autofs4/autofs_i.h
index 385bed09b0d8..f54c5b21f876 100644
--- a/fs/autofs4/autofs_i.h
+++ b/fs/autofs4/autofs_i.h
@@ -13,6 +13,7 @@
/* Internal header file for autofs */
#include <linux/auto_fs4.h>
+#include <linux/mutex.h>
#include <linux/list.h>
/* This is the range of ioctl() numbers we claim as ours */
@@ -102,7 +103,7 @@ struct autofs_sb_info {
int reghost_enabled;
int needs_reghost;
struct super_block *sb;
- struct semaphore wq_sem;
+ struct mutex wq_mutex;
spinlock_t fs_lock;
struct autofs_wait_queue *queues; /* Wait queue pointer */
};
diff --git a/fs/autofs4/inode.c b/fs/autofs4/inode.c
index 2d3082854a29..1ad98d48e550 100644
--- a/fs/autofs4/inode.c
+++ b/fs/autofs4/inode.c
@@ -269,7 +269,7 @@ int autofs4_fill_super(struct super_block *s, void *data, int silent)
sbi->sb = s;
sbi->version = 0;
sbi->sub_version = 0;
- init_MUTEX(&sbi->wq_sem);
+ mutex_init(&sbi->wq_mutex);
spin_lock_init(&sbi->fs_lock);
sbi->queues = NULL;
s->s_blocksize = 1024;
diff --git a/fs/autofs4/waitq.c b/fs/autofs4/waitq.c
index 394ff36ef8f1..be78e9378c03 100644
--- a/fs/autofs4/waitq.c
+++ b/fs/autofs4/waitq.c
@@ -178,7 +178,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
return -ENOENT;
}
- if (down_interruptible(&sbi->wq_sem)) {
+ if (mutex_lock_interruptible(&sbi->wq_mutex)) {
kfree(name);
return -EINTR;
}
@@ -194,7 +194,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
/* Can't wait for an expire if there's no mount */
if (notify == NFY_NONE && !d_mountpoint(dentry)) {
kfree(name);
- up(&sbi->wq_sem);
+ mutex_unlock(&sbi->wq_mutex);
return -ENOENT;
}
@@ -202,7 +202,7 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
wq = kmalloc(sizeof(struct autofs_wait_queue),GFP_KERNEL);
if ( !wq ) {
kfree(name);
- up(&sbi->wq_sem);
+ mutex_unlock(&sbi->wq_mutex);
return -ENOMEM;
}
@@ -218,10 +218,10 @@ int autofs4_wait(struct autofs_sb_info *sbi, struct dentry *dentry,
wq->status = -EINTR; /* Status return if interrupted */
atomic_set(&wq->wait_ctr, 2);
atomic_set(&wq->notified, 1);
- up(&sbi->wq_sem);
+ mutex_unlock(&sbi->wq_mutex);
} else {
atomic_inc(&wq->wait_ctr);
- up(&sbi->wq_sem);
+ mutex_unlock(&sbi->wq_mutex);
kfree(name);
DPRINTK("existing wait id = 0x%08lx, name = %.*s, nfy=%d",
(unsigned long) wq->wait_queue_token, wq->len, wq->name, notify);
@@ -282,19 +282,19 @@ int autofs4_wait_release(struct autofs_sb_info *sbi, autofs_wqt_t wait_queue_tok
{
struct autofs_wait_queue *wq, **wql;
- down(&sbi->wq_sem);
+ mutex_lock(&sbi->wq_mutex);
for ( wql = &sbi->queues ; (wq = *wql) != 0 ; wql = &wq->next ) {
if ( wq->wait_queue_token == wait_queue_token )
break;
}
if ( !wq ) {
- up(&sbi->wq_sem);
+ mutex_unlock(&sbi->wq_mutex);
return -EINVAL;
}
*wql = wq->next; /* Unlink from chain */
- up(&sbi->wq_sem);
+ mutex_unlock(&sbi->wq_mutex);
kfree(wq->name);
wq->name = NULL; /* Do not wait on this queue */
diff --git a/fs/bio.c b/fs/bio.c
index 1f3bb501c262..8f1d2e815c96 100644
--- a/fs/bio.c
+++ b/fs/bio.c
@@ -1243,11 +1243,11 @@ static int __init init_bio(void)
scale = 4;
/*
- * scale number of entries
+ * Limit number of entries reserved -- mempools are only used when
+ * the system is completely unable to allocate memory, so we only
+ * need enough to make progress.
*/
- bvec_pool_entries = megabytes * 2;
- if (bvec_pool_entries > 256)
- bvec_pool_entries = 256;
+ bvec_pool_entries = 1 + scale;
fs_bio_set = bioset_create(BIO_POOL_SIZE, bvec_pool_entries, scale);
if (!fs_bio_set)
diff --git a/fs/block_dev.c b/fs/block_dev.c
index 6e50346fb1ee..44d05e6e34db 100644
--- a/fs/block_dev.c
+++ b/fs/block_dev.c
@@ -265,8 +265,8 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
SLAB_CTOR_CONSTRUCTOR)
{
memset(bdev, 0, sizeof(*bdev));
- sema_init(&bdev->bd_sem, 1);
- sema_init(&bdev->bd_mount_sem, 1);
+ mutex_init(&bdev->bd_mutex);
+ mutex_init(&bdev->bd_mount_mutex);
INIT_LIST_HEAD(&bdev->bd_inodes);
INIT_LIST_HEAD(&bdev->bd_list);
inode_init_once(&ei->vfs_inode);
@@ -574,7 +574,7 @@ static int do_open(struct block_device *bdev, struct file *file)
}
owner = disk->fops->owner;
- down(&bdev->bd_sem);
+ mutex_lock(&bdev->bd_mutex);
if (!bdev->bd_openers) {
bdev->bd_disk = disk;
bdev->bd_contains = bdev;
@@ -605,21 +605,21 @@ static int do_open(struct block_device *bdev, struct file *file)
if (ret)
goto out_first;
bdev->bd_contains = whole;
- down(&whole->bd_sem);
+ mutex_lock(&whole->bd_mutex);
whole->bd_part_count++;
p = disk->part[part - 1];
bdev->bd_inode->i_data.backing_dev_info =
whole->bd_inode->i_data.backing_dev_info;
if (!(disk->flags & GENHD_FL_UP) || !p || !p->nr_sects) {
whole->bd_part_count--;
- up(&whole->bd_sem);
+ mutex_unlock(&whole->bd_mutex);
ret = -ENXIO;
goto out_first;
}
kobject_get(&p->kobj);
bdev->bd_part = p;
bd_set_size(bdev, (loff_t) p->nr_sects << 9);
- up(&whole->bd_sem);
+ mutex_unlock(&whole->bd_mutex);
}
} else {
put_disk(disk);
@@ -633,13 +633,13 @@ static int do_open(struct block_device *bdev, struct file *file)
if (bdev->bd_invalidated)
rescan_partitions(bdev->bd_disk, bdev);
} else {
- down(&bdev->bd_contains->bd_sem);
+ mutex_lock(&bdev->bd_contains->bd_mutex);
bdev->bd_contains->bd_part_count++;
- up(&bdev->bd_contains->bd_sem);
+ mutex_unlock(&bdev->bd_contains->bd_mutex);
}
}
bdev->bd_openers++;
- up(&bdev->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
unlock_kernel();
return 0;
@@ -652,7 +652,7 @@ out_first:
put_disk(disk);
module_put(owner);
out:
- up(&bdev->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
unlock_kernel();
if (ret)
bdput(bdev);
@@ -714,7 +714,7 @@ int blkdev_put(struct block_device *bdev)
struct inode *bd_inode = bdev->bd_inode;
struct gendisk *disk = bdev->bd_disk;
- down(&bdev->bd_sem);
+ mutex_lock(&bdev->bd_mutex);
lock_kernel();
if (!--bdev->bd_openers) {
sync_blockdev(bdev);
@@ -724,9 +724,9 @@ int blkdev_put(struct block_device *bdev)
if (disk->fops->release)
ret = disk->fops->release(bd_inode, NULL);
} else {
- down(&bdev->bd_contains->bd_sem);
+ mutex_lock(&bdev->bd_contains->bd_mutex);
bdev->bd_contains->bd_part_count--;
- up(&bdev->bd_contains->bd_sem);
+ mutex_unlock(&bdev->bd_contains->bd_mutex);
}
if (!bdev->bd_openers) {
struct module *owner = disk->fops->owner;
@@ -746,7 +746,7 @@ int blkdev_put(struct block_device *bdev)
bdev->bd_contains = NULL;
}
unlock_kernel();
- up(&bdev->bd_sem);
+ mutex_unlock(&bdev->bd_mutex);
bdput(bdev);
return ret;
}
diff --git a/fs/buffer.c b/fs/buffer.c
index 1d3683d496f8..0d6ca7bac6c8 100644
--- a/fs/buffer.c
+++ b/fs/buffer.c
@@ -201,7 +201,7 @@ int fsync_bdev(struct block_device *bdev)
* freeze_bdev -- lock a filesystem and force it into a consistent state
* @bdev: blockdevice to lock
*
- * This takes the block device bd_mount_sem to make sure no new mounts
+ * This takes the block device bd_mount_mutex to make sure no new mounts
* happen on bdev until thaw_bdev() is called.
* If a superblock is found on this device, we take the s_umount semaphore
* on it to make sure nobody unmounts until the snapshot creation is done.
@@ -210,7 +210,7 @@ struct super_block *freeze_bdev(struct block_device *bdev)
{
struct super_block *sb;
- down(&bdev->bd_mount_sem);
+ mutex_lock(&bdev->bd_mount_mutex);
sb = get_super(bdev);
if (sb && !(sb->s_flags & MS_RDONLY)) {
sb->s_frozen = SB_FREEZE_WRITE;
@@ -264,7 +264,7 @@ void thaw_bdev(struct block_device *bdev, struct super_block *sb)
drop_super(sb);
}
- up(&bdev->bd_mount_sem);
+ mutex_unlock(&bdev->bd_mount_mutex);
}
EXPORT_SYMBOL(thaw_bdev);
diff --git a/fs/cifs/dir.c b/fs/cifs/dir.c
index fed55e3c53df..632561dd9c50 100644
--- a/fs/cifs/dir.c
+++ b/fs/cifs/dir.c
@@ -138,9 +138,9 @@ cifs_create(struct inode *inode, struct dentry *direntry, int mode,
cifs_sb = CIFS_SB(inode->i_sb);
pTcon = cifs_sb->tcon;
- down(&direntry->d_sb->s_vfs_rename_sem);
+ mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&direntry->d_sb->s_vfs_rename_sem);
+ mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);
if(full_path == NULL) {
FreeXid(xid);
return -ENOMEM;
@@ -317,9 +317,9 @@ int cifs_mknod(struct inode *inode, struct dentry *direntry, int mode,
cifs_sb = CIFS_SB(inode->i_sb);
pTcon = cifs_sb->tcon;
- down(&direntry->d_sb->s_vfs_rename_sem);
+ mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&direntry->d_sb->s_vfs_rename_sem);
+ mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);
if(full_path == NULL)
rc = -ENOMEM;
else if (pTcon->ses->capabilities & CAP_UNIX) {
diff --git a/fs/cifs/fcntl.c b/fs/cifs/fcntl.c
index a7a47bb36bf3..ec4dfe9bf5ef 100644
--- a/fs/cifs/fcntl.c
+++ b/fs/cifs/fcntl.c
@@ -86,9 +86,9 @@ int cifs_dir_notify(struct file * file, unsigned long arg)
cifs_sb = CIFS_SB(file->f_dentry->d_sb);
pTcon = cifs_sb->tcon;
- down(&file->f_dentry->d_sb->s_vfs_rename_sem);
+ mutex_lock(&file->f_dentry->d_sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(file->f_dentry);
- up(&file->f_dentry->d_sb->s_vfs_rename_sem);
+ mutex_unlock(&file->f_dentry->d_sb->s_vfs_rename_mutex);
if(full_path == NULL) {
rc = -ENOMEM;
diff --git a/fs/cifs/file.c b/fs/cifs/file.c
index 675bd2568297..165d67426381 100644
--- a/fs/cifs/file.c
+++ b/fs/cifs/file.c
@@ -203,9 +203,9 @@ int cifs_open(struct inode *inode, struct file *file)
}
}
- down(&inode->i_sb->s_vfs_rename_sem);
+ mutex_lock(&inode->i_sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(file->f_dentry);
- up(&inode->i_sb->s_vfs_rename_sem);
+ mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);
if (full_path == NULL) {
FreeXid(xid);
return -ENOMEM;
diff --git a/fs/cifs/inode.c b/fs/cifs/inode.c
index 59359911f481..ff93a9f81d1c 100644
--- a/fs/cifs/inode.c
+++ b/fs/cifs/inode.c
@@ -574,9 +574,9 @@ int cifs_unlink(struct inode *inode, struct dentry *direntry)
/* Unlink can be called from rename so we can not grab the sem here
since we deadlock otherwise */
-/* down(&direntry->d_sb->s_vfs_rename_sem);*/
+/* mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);*/
full_path = build_path_from_dentry(direntry);
-/* up(&direntry->d_sb->s_vfs_rename_sem);*/
+/* mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);*/
if (full_path == NULL) {
FreeXid(xid);
return -ENOMEM;
@@ -718,9 +718,9 @@ int cifs_mkdir(struct inode *inode, struct dentry *direntry, int mode)
cifs_sb = CIFS_SB(inode->i_sb);
pTcon = cifs_sb->tcon;
- down(&inode->i_sb->s_vfs_rename_sem);
+ mutex_lock(&inode->i_sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&inode->i_sb->s_vfs_rename_sem);
+ mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);
if (full_path == NULL) {
FreeXid(xid);
return -ENOMEM;
@@ -803,9 +803,9 @@ int cifs_rmdir(struct inode *inode, struct dentry *direntry)
cifs_sb = CIFS_SB(inode->i_sb);
pTcon = cifs_sb->tcon;
- down(&inode->i_sb->s_vfs_rename_sem);
+ mutex_lock(&inode->i_sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&inode->i_sb->s_vfs_rename_sem);
+ mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);
if (full_path == NULL) {
FreeXid(xid);
return -ENOMEM;
@@ -1137,9 +1137,9 @@ int cifs_setattr(struct dentry *direntry, struct iattr *attrs)
rc = 0;
}
- down(&direntry->d_sb->s_vfs_rename_sem);
+ mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&direntry->d_sb->s_vfs_rename_sem);
+ mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);
if (full_path == NULL) {
FreeXid(xid);
return -ENOMEM;
diff --git a/fs/cifs/link.c b/fs/cifs/link.c
index 0f99aae33162..8d0da7c87c7b 100644
--- a/fs/cifs/link.c
+++ b/fs/cifs/link.c
@@ -48,10 +48,10 @@ cifs_hardlink(struct dentry *old_file, struct inode *inode,
/* No need to check for cross device links since server will do that
BB note DFS case in future though (when we may have to check) */
- down(&inode->i_sb->s_vfs_rename_sem);
+ mutex_lock(&inode->i_sb->s_vfs_rename_mutex);
fromName = build_path_from_dentry(old_file);
toName = build_path_from_dentry(direntry);
- up(&inode->i_sb->s_vfs_rename_sem);
+ mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);
if((fromName == NULL) || (toName == NULL)) {
rc = -ENOMEM;
goto cifs_hl_exit;
@@ -103,9 +103,9 @@ cifs_follow_link(struct dentry *direntry, struct nameidata *nd)
xid = GetXid();
- down(&direntry->d_sb->s_vfs_rename_sem);
+ mutex_lock(&direntry->d_sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&direntry->d_sb->s_vfs_rename_sem);
+ mutex_unlock(&direntry->d_sb->s_vfs_rename_mutex);
if (!full_path)
goto out_no_free;
@@ -164,9 +164,9 @@ cifs_symlink(struct inode *inode, struct dentry *direntry, const char *symname)
cifs_sb = CIFS_SB(inode->i_sb);
pTcon = cifs_sb->tcon;
- down(&inode->i_sb->s_vfs_rename_sem);
+ mutex_lock(&inode->i_sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&inode->i_sb->s_vfs_rename_sem);
+ mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);
if(full_path == NULL) {
FreeXid(xid);
@@ -232,9 +232,9 @@ cifs_readlink(struct dentry *direntry, char __user *pBuffer, int buflen)
/* BB would it be safe against deadlock to grab this sem
even though rename itself grabs the sem and calls lookup? */
-/* down(&inode->i_sb->s_vfs_rename_sem);*/
+/* mutex_lock(&inode->i_sb->s_vfs_rename_mutex);*/
full_path = build_path_from_dentry(direntry);
-/* up(&inode->i_sb->s_vfs_rename_sem);*/
+/* mutex_unlock(&inode->i_sb->s_vfs_rename_mutex);*/
if(full_path == NULL) {
FreeXid(xid);
diff --git a/fs/cifs/readdir.c b/fs/cifs/readdir.c
index 288cc048d37f..edb3b6eb34bc 100644
--- a/fs/cifs/readdir.c
+++ b/fs/cifs/readdir.c
@@ -404,9 +404,9 @@ static int initiate_cifs_search(const int xid, struct file *file)
if(pTcon == NULL)
return -EINVAL;
- down(&file->f_dentry->d_sb->s_vfs_rename_sem);
+ mutex_lock(&file->f_dentry->d_sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(file->f_dentry);
- up(&file->f_dentry->d_sb->s_vfs_rename_sem);
+ mutex_unlock(&file->f_dentry->d_sb->s_vfs_rename_mutex);
if(full_path == NULL) {
return -ENOMEM;
diff --git a/fs/cifs/xattr.c b/fs/cifs/xattr.c
index 777e3363c2a4..3938444d87b2 100644
--- a/fs/cifs/xattr.c
+++ b/fs/cifs/xattr.c
@@ -62,9 +62,9 @@ int cifs_removexattr(struct dentry * direntry, const char * ea_name)
cifs_sb = CIFS_SB(sb);
pTcon = cifs_sb->tcon;
- down(&sb->s_vfs_rename_sem);
+ mutex_lock(&sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&sb->s_vfs_rename_sem);
+ mutex_unlock(&sb->s_vfs_rename_mutex);
if(full_path == NULL) {
FreeXid(xid);
return -ENOMEM;
@@ -116,9 +116,9 @@ int cifs_setxattr(struct dentry * direntry, const char * ea_name,
cifs_sb = CIFS_SB(sb);
pTcon = cifs_sb->tcon;
- down(&sb->s_vfs_rename_sem);
+ mutex_lock(&sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&sb->s_vfs_rename_sem);
+ mutex_unlock(&sb->s_vfs_rename_mutex);
if(full_path == NULL) {
FreeXid(xid);
return -ENOMEM;
@@ -223,9 +223,9 @@ ssize_t cifs_getxattr(struct dentry * direntry, const char * ea_name,
cifs_sb = CIFS_SB(sb);
pTcon = cifs_sb->tcon;
- down(&sb->s_vfs_rename_sem);
+ mutex_lock(&sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&sb->s_vfs_rename_sem);
+ mutex_unlock(&sb->s_vfs_rename_mutex);
if(full_path == NULL) {
FreeXid(xid);
return -ENOMEM;
@@ -341,9 +341,9 @@ ssize_t cifs_listxattr(struct dentry * direntry, char * data, size_t buf_size)
cifs_sb = CIFS_SB(sb);
pTcon = cifs_sb->tcon;
- down(&sb->s_vfs_rename_sem);
+ mutex_lock(&sb->s_vfs_rename_mutex);
full_path = build_path_from_dentry(direntry);
- up(&sb->s_vfs_rename_sem);
+ mutex_unlock(&sb->s_vfs_rename_mutex);
if(full_path == NULL) {
FreeXid(xid);
return -ENOMEM;
diff --git a/fs/devpts/inode.c b/fs/devpts/inode.c
index bfb8a230bac9..14c5620b5cab 100644
--- a/fs/devpts/inode.c
+++ b/fs/devpts/inode.c
@@ -18,6 +18,7 @@
#include <linux/mount.h>
#include <linux/tty.h>
#include <linux/devpts_fs.h>
+#include <linux/parser.h>
#define DEVPTS_SUPER_MAGIC 0x1cd1
@@ -32,39 +33,60 @@ static struct {
umode_t mode;
} config = {.mode = 0600};
+enum {
+ Opt_uid, Opt_gid, Opt_mode,
+ Opt_err
+};
+
+static match_table_t tokens = {
+ {Opt_uid, "uid=%u"},
+ {Opt_gid, "gid=%u"},
+ {Opt_mode, "mode=%o"},
+ {Opt_err, NULL}
+};
+
static int devpts_remount(struct super_block *sb, int *flags, char *data)
{
- int setuid = 0;
- int setgid = 0;
- uid_t uid = 0;
- gid_t gid = 0;
- umode_t mode = 0600;
- char *this_char;
-
- this_char = NULL;
- while ((this_char = strsep(&data, ",")) != NULL) {
- int n;
- char dummy;
- if (!*this_char)
+ char *p;
+
+ config.setuid = 0;
+ config.setgid = 0;
+ config.uid = 0;
+ config.gid = 0;
+ config.mode = 0600;
+
+ while ((p = strsep(&data, ",")) != NULL) {
+ substring_t args[MAX_OPT_ARGS];
+ int token;
+ int option;
+
+ if (!*p)
continue;
- if (sscanf(this_char, "uid=%i%c", &n, &dummy) == 1) {
- setuid = 1;
- uid = n;
- } else if (sscanf(this_char, "gid=%i%c", &n, &dummy) == 1) {
- setgid = 1;
- gid = n;
- } else if (sscanf(this_char, "mode=%o%c", &n, &dummy) == 1)
- mode = n & ~S_IFMT;
- else {
- printk("devpts: called with bogus options\n");
+
+ token = match_token(p, tokens, args);
+ switch (token) {
+ case Opt_uid:
+ if (match_int(&args[0], &option))
+ return -EINVAL;
+ config.uid = option;
+ config.setuid = 1;
+ break;
+ case Opt_gid:
+ if (match_int(&args[0], &option))
+ return -EINVAL;
+ config.gid = option;
+ config.setgid = 1;
+ break;
+ case Opt_mode:
+ if (match_octal(&args[0], &option))
+ return -EINVAL;
+ config.mode = option & ~S_IFMT;
+ break;
+ default:
+ printk(KERN_ERR "devpts: called with bogus options\n");
return -EINVAL;
}
}
- config.setuid = setuid;
- config.setgid = setgid;
- config.uid = uid;
- config.gid = gid;
- config.mode = mode;
return 0;
}
diff --git a/fs/dquot.c b/fs/dquot.c
index 1966c890b48d..acf07e581f8c 100644
--- a/fs/dquot.c
+++ b/fs/dquot.c
@@ -103,12 +103,12 @@
* (these locking rules also apply for S_NOQUOTA flag in the inode - note that
* for altering the flag i_mutex is also needed). If operation is holding
* reference to dquot in other way (e.g. quotactl ops) it must be guarded by
- * dqonoff_sem.
+ * dqonoff_mutex.
* This locking assures that:
* a) update/access to dquot pointers in inode is serialized
* b) everyone is guarded against invalidate_dquots()
*
- * Each dquot has its dq_lock semaphore. Locked dquots might not be referenced
+ * Each dquot has its dq_lock mutex. Locked dquots might not be referenced
* from inodes (dquot_alloc_space() and such don't check the dq_lock).
* Currently dquot is locked only when it is being read to memory (or space for
* it is being allocated) on the first dqget() and when it is being released on
@@ -118,9 +118,9 @@
* spinlock to internal buffers before writing.
*
* Lock ordering (including related VFS locks) is the following:
- * i_mutex > dqonoff_sem > iprune_sem > journal_lock > dqptr_sem >
- * > dquot->dq_lock > dqio_sem
- * i_mutex on quota files is special (it's below dqio_sem)
+ * i_mutex > dqonoff_sem > journal_lock > dqptr_sem > dquot->dq_lock >
+ * dqio_mutex
+ * i_mutex on quota files is special (it's below dqio_mutex)
*/
static DEFINE_SPINLOCK(dq_list_lock);
@@ -281,8 +281,8 @@ static inline void remove_inuse(struct dquot *dquot)
static void wait_on_dquot(struct dquot *dquot)
{
- down(&dquot->dq_lock);
- up(&dquot->dq_lock);
+ mutex_lock(&dquot->dq_lock);
+ mutex_unlock(&dquot->dq_lock);
}
#define mark_dquot_dirty(dquot) ((dquot)->dq_sb->dq_op->mark_dirty(dquot))
@@ -321,8 +321,8 @@ int dquot_acquire(struct dquot *dquot)
int ret = 0, ret2 = 0;
struct quota_info *dqopt = sb_dqopt(dquot->dq_sb);
- down(&dquot->dq_lock);
- down(&dqopt->dqio_sem);
+ mutex_lock(&dquot->dq_lock);
+ mutex_lock(&dqopt->dqio_mutex);
if (!test_bit(DQ_READ_B, &dquot->dq_flags))
ret = dqopt->ops[dquot->dq_type]->read_dqblk(dquot);
if (ret < 0)
@@ -343,8 +343,8 @@ int dquot_acquire(struct dquot *dquot)
}
set_bit(DQ_ACTIVE_B, &dquot->dq_flags);
out_iolock:
- up(&dqopt->dqio_sem);
- up(&dquot->dq_lock);
+ mutex_unlock(&dqopt->dqio_mutex);
+ mutex_unlock(&dquot->dq_lock);
return ret;
}
@@ -356,7 +356,7 @@ int dquot_commit(struct dquot *dquot)
int ret = 0, ret2 = 0;
struct quota_info *dqopt = sb_dqopt(dquot->dq_sb);
- down(&dqopt->dqio_sem);
+ mutex_lock(&dqopt->dqio_mutex);
spin_lock(&dq_list_lock);
if (!clear_dquot_dirty(dquot)) {
spin_unlock(&dq_list_lock);
@@ -373,7 +373,7 @@ int dquot_commit(struct dquot *dquot)
ret = ret2;
}
out_sem:
- up(&dqopt->dqio_sem);
+ mutex_unlock(&dqopt->dqio_mutex);
return ret;
}
@@ -385,11 +385,11 @@ int dquot_release(struct dquot *dquot)
int ret = 0, ret2 = 0;
struct quota_info *dqopt = sb_dqopt(dquot->dq_sb);
- down(&dquot->dq_lock);
+ mutex_lock(&dquot->dq_lock);
/* Check whether we are not racing with some other dqget() */
if (atomic_read(&dquot->dq_count) > 1)
goto out_dqlock;
- down(&dqopt->dqio_sem);
+ mutex_lock(&dqopt->dqio_mutex);
if (dqopt->ops[dquot->dq_type]->release_dqblk) {
ret = dqopt->ops[dquot->dq_type]->release_dqblk(dquot);
/* Write the info */
@@ -399,31 +399,57 @@ int dquot_release(struct dquot *dquot)
ret = ret2;
}
clear_bit(DQ_ACTIVE_B, &dquot->dq_flags);
- up(&dqopt->dqio_sem);
+ mutex_unlock(&dqopt->dqio_mutex);
out_dqlock:
- up(&dquot->dq_lock);
+ mutex_unlock(&dquot->dq_lock);
return ret;
}
/* Invalidate all dquots on the list. Note that this function is called after
* quota is disabled and pointers from inodes removed so there cannot be new
- * quota users. Also because we hold dqonoff_sem there can be no quota users
- * for this sb+type at all. */
+ * quota users. There can still be some users of quotas due to inodes being
+ * just deleted or pruned by prune_icache() (those are not attached to any
+ * list). We have to wait for such users.
+ */
static void invalidate_dquots(struct super_block *sb, int type)
{
struct dquot *dquot, *tmp;
+restart:
spin_lock(&dq_list_lock);
list_for_each_entry_safe(dquot, tmp, &inuse_list, dq_inuse) {
if (dquot->dq_sb != sb)
continue;
if (dquot->dq_type != type)
continue;
-#ifdef __DQUOT_PARANOIA
- if (atomic_read(&dquot->dq_count))
- BUG();
-#endif
- /* Quota now has no users and it has been written on last dqput() */
+ /* Wait for dquot users */
+ if (atomic_read(&dquot->dq_count)) {
+ DEFINE_WAIT(wait);
+
+ atomic_inc(&dquot->dq_count);
+ prepare_to_wait(&dquot->dq_wait_unused, &wait,
+ TASK_UNINTERRUPTIBLE);
+ spin_unlock(&dq_list_lock);
+ /* Once dqput() wakes us up, we know it's time to free
+ * the dquot.
+ * IMPORTANT: we rely on the fact that there is always
+ * at most one process waiting for dquot to free.
+ * Otherwise dq_count would be > 1 and we would never
+ * wake up.
+ */
+ if (atomic_read(&dquot->dq_count) > 1)
+ schedule();
+ finish_wait(&dquot->dq_wait_unused, &wait);
+ dqput(dquot);
+ /* At this moment dquot() need not exist (it could be
+ * reclaimed by prune_dqcache(). Hence we must
+ * restart. */
+ goto restart;
+ }
+ /*
+ * Quota now has no users and it has been written on last
+ * dqput()
+ */
remove_dquot_hash(dquot);
remove_free_dquot(dquot);
remove_inuse(dquot);
@@ -439,7 +465,7 @@ int vfs_quota_sync(struct super_block *sb, int type)
struct quota_info *dqopt = sb_dqopt(sb);
int cnt;
- down(&dqopt->dqonoff_sem);
+ mutex_lock(&dqopt->dqonoff_mutex);
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
if (type != -1 && cnt != type)
continue;
@@ -474,7 +500,7 @@ int vfs_quota_sync(struct super_block *sb, int type)
spin_lock(&dq_list_lock);
dqstats.syncs++;
spin_unlock(&dq_list_lock);
- up(&dqopt->dqonoff_sem);
+ mutex_unlock(&dqopt->dqonoff_mutex);
return 0;
}
@@ -515,7 +541,7 @@ static int shrink_dqcache_memory(int nr, gfp_t gfp_mask)
/*
* Put reference to dquot
* NOTE: If you change this function please check whether dqput_blocks() works right...
- * MUST be called with either dqptr_sem or dqonoff_sem held
+ * MUST be called with either dqptr_sem or dqonoff_mutex held
*/
static void dqput(struct dquot *dquot)
{
@@ -540,6 +566,10 @@ we_slept:
if (atomic_read(&dquot->dq_count) > 1) {
/* We have more than one user... nothing to do */
atomic_dec(&dquot->dq_count);
+ /* Releasing dquot during quotaoff phase? */
+ if (!sb_has_quota_enabled(dquot->dq_sb, dquot->dq_type) &&
+ atomic_read(&dquot->dq_count) == 1)
+ wake_up(&dquot->dq_wait_unused);
spin_unlock(&dq_list_lock);
return;
}
@@ -576,11 +606,12 @@ static struct dquot *get_empty_dquot(struct super_block *sb, int type)
return NODQUOT;
memset((caddr_t)dquot, 0, sizeof(struct dquot));
- sema_init(&dquot->dq_lock, 1);
+ mutex_init(&dquot->dq_lock);
INIT_LIST_HEAD(&dquot->dq_free);
INIT_LIST_HEAD(&dquot->dq_inuse);
INIT_HLIST_NODE(&dquot->dq_hash);
INIT_LIST_HEAD(&dquot->dq_dirty);
+ init_waitqueue_head(&dquot->dq_wait_unused);
dquot->dq_sb = sb;
dquot->dq_type = type;
atomic_set(&dquot->dq_count, 1);
@@ -590,7 +621,7 @@ static struct dquot *get_empty_dquot(struct super_block *sb, int type)
/*
* Get reference to dquot
- * MUST be called with either dqptr_sem or dqonoff_sem held
+ * MUST be called with either dqptr_sem or dqonoff_mutex held
*/
static struct dquot *dqget(struct super_block *sb, unsigned int id, int type)
{
@@ -656,7 +687,7 @@ static int dqinit_needed(struct inode *inode, int type)
return 0;
}
-/* This routine is guarded by dqonoff_sem semaphore */
+/* This routine is guarded by dqonoff_mutex mutex */
static void add_dquot_ref(struct super_block *sb, int type)
{
struct list_head *p;
@@ -732,13 +763,9 @@ static void drop_dquot_ref(struct super_block *sb, int type)
{
LIST_HEAD(tofree_head);
- /* We need to be guarded against prune_icache to reach all the
- * inodes - otherwise some can be on the local list of prune_icache */
- down(&iprune_sem);
down_write(&sb_dqopt(sb)->dqptr_sem);
remove_dquot_ref(sb, type, &tofree_head);
up_write(&sb_dqopt(sb)->dqptr_sem);
- up(&iprune_sem);
put_dquot_list(&tofree_head);
}
@@ -938,8 +965,8 @@ int dquot_initialize(struct inode *inode, int type)
unsigned int id = 0;
int cnt, ret = 0;
- /* First test before acquiring semaphore - solves deadlocks when we
- * re-enter the quota code and are already holding the semaphore */
+ /* First test before acquiring mutex - solves deadlocks when we
+ * re-enter the quota code and are already holding the mutex */
if (IS_NOQUOTA(inode))
return 0;
down_write(&sb_dqopt(inode->i_sb)->dqptr_sem);
@@ -1002,8 +1029,8 @@ int dquot_alloc_space(struct inode *inode, qsize_t number, int warn)
int cnt, ret = NO_QUOTA;
char warntype[MAXQUOTAS];
- /* First test before acquiring semaphore - solves deadlocks when we
- * re-enter the quota code and are already holding the semaphore */
+ /* First test before acquiring mutex - solves deadlocks when we
+ * re-enter the quota code and are already holding the mutex */
if (IS_NOQUOTA(inode)) {
out_add:
inode_add_bytes(inode, number);
@@ -1051,8 +1078,8 @@ int dquot_alloc_inode(const struct inode *inode, unsigned long number)
int cnt, ret = NO_QUOTA;
char warntype[MAXQUOTAS];
- /* First test before acquiring semaphore - solves deadlocks when we
- * re-enter the quota code and are already holding the semaphore */
+ /* First test before acquiring mutex - solves deadlocks when we
+ * re-enter the quota code and are already holding the mutex */
if (IS_NOQUOTA(inode))
return QUOTA_OK;
for (cnt = 0; cnt < MAXQUOTAS; cnt++)
@@ -1095,8 +1122,8 @@ int dquot_free_space(struct inode *inode, qsize_t number)
{
unsigned int cnt;
- /* First test before acquiring semaphore - solves deadlocks when we
- * re-enter the quota code and are already holding the semaphore */
+ /* First test before acquiring mutex - solves deadlocks when we
+ * re-enter the quota code and are already holding the mutex */
if (IS_NOQUOTA(inode)) {
out_sub:
inode_sub_bytes(inode, number);
@@ -1131,8 +1158,8 @@ int dquot_free_inode(const struct inode *inode, unsigned long number)
{
unsigned int cnt;
- /* First test before acquiring semaphore - solves deadlocks when we
- * re-enter the quota code and are already holding the semaphore */
+ /* First test before acquiring mutex - solves deadlocks when we
+ * re-enter the quota code and are already holding the mutex */
if (IS_NOQUOTA(inode))
return QUOTA_OK;
down_read(&sb_dqopt(inode->i_sb)->dqptr_sem);
@@ -1171,8 +1198,8 @@ int dquot_transfer(struct inode *inode, struct iattr *iattr)
chgid = (iattr->ia_valid & ATTR_GID) && inode->i_gid != iattr->ia_gid;
char warntype[MAXQUOTAS];
- /* First test before acquiring semaphore - solves deadlocks when we
- * re-enter the quota code and are already holding the semaphore */
+ /* First test before acquiring mutex - solves deadlocks when we
+ * re-enter the quota code and are already holding the mutex */
if (IS_NOQUOTA(inode))
return QUOTA_OK;
/* Clear the arrays */
@@ -1266,9 +1293,9 @@ int dquot_commit_info(struct super_block *sb, int type)
int ret;
struct quota_info *dqopt = sb_dqopt(sb);
- down(&dqopt->dqio_sem);
+ mutex_lock(&dqopt->dqio_mutex);
ret = dqopt->ops[type]->write_file_info(sb, type);
- up(&dqopt->dqio_sem);
+ mutex_unlock(&dqopt->dqio_mutex);
return ret;
}
@@ -1324,7 +1351,7 @@ int vfs_quota_off(struct super_block *sb, int type)
struct inode *toputinode[MAXQUOTAS];
/* We need to serialize quota_off() for device */
- down(&dqopt->dqonoff_sem);
+ mutex_lock(&dqopt->dqonoff_mutex);
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
toputinode[cnt] = NULL;
if (type != -1 && cnt != type)
@@ -1353,7 +1380,7 @@ int vfs_quota_off(struct super_block *sb, int type)
dqopt->info[cnt].dqi_bgrace = 0;
dqopt->ops[cnt] = NULL;
}
- up(&dqopt->dqonoff_sem);
+ mutex_unlock(&dqopt->dqonoff_mutex);
/* Sync the superblock so that buffers with quota data are written to
* disk (and so userspace sees correct data afterwards). */
if (sb->s_op->sync_fs)
@@ -1366,7 +1393,7 @@ int vfs_quota_off(struct super_block *sb, int type)
* changes done by userspace on the next quotaon() */
for (cnt = 0; cnt < MAXQUOTAS; cnt++)
if (toputinode[cnt]) {
- down(&dqopt->dqonoff_sem);
+ mutex_lock(&dqopt->dqonoff_mutex);
/* If quota was reenabled in the meantime, we have
* nothing to do */
if (!sb_has_quota_enabled(sb, cnt)) {
@@ -1378,7 +1405,7 @@ int vfs_quota_off(struct super_block *sb, int type)
mark_inode_dirty(toputinode[cnt]);
iput(toputinode[cnt]);
}
- up(&dqopt->dqonoff_sem);
+ mutex_unlock(&dqopt->dqonoff_mutex);
}
if (sb->s_bdev)
invalidate_bdev(sb->s_bdev, 0);
@@ -1419,7 +1446,7 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id)
/* And now flush the block cache so that kernel sees the changes */
invalidate_bdev(sb->s_bdev, 0);
mutex_lock(&inode->i_mutex);
- down(&dqopt->dqonoff_sem);
+ mutex_lock(&dqopt->dqonoff_mutex);
if (sb_has_quota_enabled(sb, type)) {
error = -EBUSY;
goto out_lock;
@@ -1444,17 +1471,17 @@ static int vfs_quota_on_inode(struct inode *inode, int type, int format_id)
dqopt->ops[type] = fmt->qf_ops;
dqopt->info[type].dqi_format = fmt;
INIT_LIST_HEAD(&dqopt->info[type].dqi_dirty_list);
- down(&dqopt->dqio_sem);
+ mutex_lock(&dqopt->dqio_mutex);
if ((error = dqopt->ops[type]->read_file_info(sb, type)) < 0) {
- up(&dqopt->dqio_sem);
+ mutex_unlock(&dqopt->dqio_mutex);
goto out_file_init;
}
- up(&dqopt->dqio_sem);
+ mutex_unlock(&dqopt->dqio_mutex);
mutex_unlock(&inode->i_mutex);
set_enable_flags(dqopt, type);
add_dquot_ref(sb, type);
- up(&dqopt->dqonoff_sem);
+ mutex_unlock(&dqopt->dqonoff_mutex);
return 0;
@@ -1462,7 +1489,7 @@ out_file_init:
dqopt->files[type] = NULL;
iput(inode);
out_lock:
- up(&dqopt->dqonoff_sem);
+ mutex_unlock(&dqopt->dqonoff_mutex);
if (oldflags != -1) {
down_write(&dqopt->dqptr_sem);
/* Set the flags back (in the case of accidental quotaon()
@@ -1550,14 +1577,14 @@ int vfs_get_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *d
{
struct dquot *dquot;
- down(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
if (!(dquot = dqget(sb, id, type))) {
- up(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
return -ESRCH;
}
do_get_dqblk(dquot, di);
dqput(dquot);
- up(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
return 0;
}
@@ -1619,14 +1646,14 @@ int vfs_set_dqblk(struct super_block *sb, int type, qid_t id, struct if_dqblk *d
{
struct dquot *dquot;
- down(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
if (!(dquot = dqget(sb, id, type))) {
- up(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
return -ESRCH;
}
do_set_dqblk(dquot, di);
dqput(dquot);
- up(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
return 0;
}
@@ -1635,9 +1662,9 @@ int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
{
struct mem_dqinfo *mi;
- down(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
if (!sb_has_quota_enabled(sb, type)) {
- up(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
return -ESRCH;
}
mi = sb_dqopt(sb)->info + type;
@@ -1647,7 +1674,7 @@ int vfs_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
ii->dqi_flags = mi->dqi_flags & DQF_MASK;
ii->dqi_valid = IIF_ALL;
spin_unlock(&dq_data_lock);
- up(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
return 0;
}
@@ -1656,9 +1683,9 @@ int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
{
struct mem_dqinfo *mi;
- down(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
if (!sb_has_quota_enabled(sb, type)) {
- up(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
return -ESRCH;
}
mi = sb_dqopt(sb)->info + type;
@@ -1673,7 +1700,7 @@ int vfs_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii)
mark_info_dirty(sb, type);
/* Force write to disk */
sb->dq_op->write_info(sb, type);
- up(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
return 0;
}
diff --git a/fs/eventpoll.c b/fs/eventpoll.c
index 4284cd31eba6..1c2b16fda13a 100644
--- a/fs/eventpoll.c
+++ b/fs/eventpoll.c
@@ -34,6 +34,7 @@
#include <linux/eventpoll.h>
#include <linux/mount.h>
#include <linux/bitops.h>
+#include <linux/mutex.h>
#include <asm/uaccess.h>
#include <asm/system.h>
#include <asm/io.h>
@@ -46,7 +47,7 @@
* LOCKING:
* There are three level of locking required by epoll :
*
- * 1) epsem (semaphore)
+ * 1) epmutex (mutex)
* 2) ep->sem (rw_semaphore)
* 3) ep->lock (rw_lock)
*
@@ -67,9 +68,9 @@
* if a file has been pushed inside an epoll set and it is then
* close()d without a previous call toepoll_ctl(EPOLL_CTL_DEL).
* It is possible to drop the "ep->sem" and to use the global
- * semaphore "epsem" (together with "ep->lock") to have it working,
+ * semaphore "epmutex" (together with "ep->lock") to have it working,
* but having "ep->sem" will make the interface more scalable.
- * Events that require holding "epsem" are very rare, while for
+ * Events that require holding "epmutex" are very rare, while for
* normal operations the epoll private "ep->sem" will guarantee
* a greater scalability.
*/
@@ -274,7 +275,7 @@ static struct super_block *eventpollfs_get_sb(struct file_system_type *fs_type,
/*
* This semaphore is used to serialize ep_free() and eventpoll_release_file().
*/
-static struct semaphore epsem;
+static struct mutex epmutex;
/* Safe wake up implementation */
static struct poll_safewake psw;
@@ -451,15 +452,6 @@ static void ep_poll_safewake(struct poll_safewake *psw, wait_queue_head_t *wq)
}
-/* Used to initialize the epoll bits inside the "struct file" */
-void eventpoll_init_file(struct file *file)
-{
-
- INIT_LIST_HEAD(&file->f_ep_links);
- spin_lock_init(&file->f_ep_lock);
-}
-
-
/*
* This is called from eventpoll_release() to unlink files from the eventpoll
* interface. We need to have this facility to cleanup correctly files that are
@@ -477,10 +469,10 @@ void eventpoll_release_file(struct file *file)
* cleanup path, and this means that noone is using this file anymore.
* The only hit might come from ep_free() but by holding the semaphore
* will correctly serialize the operation. We do need to acquire
- * "ep->sem" after "epsem" because ep_remove() requires it when called
+ * "ep->sem" after "epmutex" because ep_remove() requires it when called
* from anywhere but ep_free().
*/
- down(&epsem);
+ mutex_lock(&epmutex);
while (!list_empty(lsthead)) {
epi = list_entry(lsthead->next, struct epitem, fllink);
@@ -492,7 +484,7 @@ void eventpoll_release_file(struct file *file)
up_write(&ep->sem);
}
- up(&epsem);
+ mutex_unlock(&epmutex);
}
@@ -819,9 +811,9 @@ static void ep_free(struct eventpoll *ep)
* We do not need to hold "ep->sem" here because the epoll file
* is on the way to be removed and no one has references to it
* anymore. The only hit might come from eventpoll_release_file() but
- * holding "epsem" is sufficent here.
+ * holding "epmutex" is sufficent here.
*/
- down(&epsem);
+ mutex_lock(&epmutex);
/*
* Walks through the whole tree by unregistering poll callbacks.
@@ -843,7 +835,7 @@ static void ep_free(struct eventpoll *ep)
ep_remove(ep, epi);
}
- up(&epsem);
+ mutex_unlock(&epmutex);
}
@@ -1615,7 +1607,7 @@ static int __init eventpoll_init(void)
{
int error;
- init_MUTEX(&epsem);
+ mutex_init(&epmutex);
/* Initialize the structure used to perform safe poll wait head wake ups */
ep_poll_safewake_init(&psw);
diff --git a/fs/ext2/namei.c b/fs/ext2/namei.c
index ad1432a2a62e..4ca824985321 100644
--- a/fs/ext2/namei.c
+++ b/fs/ext2/namei.c
@@ -36,22 +36,6 @@
#include "acl.h"
#include "xip.h"
-/*
- * Couple of helper functions - make the code slightly cleaner.
- */
-
-static inline void ext2_inc_count(struct inode *inode)
-{
- inode->i_nlink++;
- mark_inode_dirty(inode);
-}
-
-static inline void ext2_dec_count(struct inode *inode)
-{
- inode->i_nlink--;
- mark_inode_dirty(inode);
-}
-
static inline int ext2_add_nondir(struct dentry *dentry, struct inode *inode)
{
int err = ext2_add_link(dentry, inode);
@@ -59,7 +43,7 @@ static inline int ext2_add_nondir(struct dentry *dentry, struct inode *inode)
d_instantiate(dentry, inode);
return 0;
}
- ext2_dec_count(inode);
+ inode_dec_link_count(inode);
iput(inode);
return err;
}
@@ -201,7 +185,7 @@ out:
return err;
out_fail:
- ext2_dec_count(inode);
+ inode_dec_link_count(inode);
iput (inode);
goto out;
}
@@ -215,7 +199,7 @@ static int ext2_link (struct dentry * old_dentry, struct inode * dir,
return -EMLINK;
inode->i_ctime = CURRENT_TIME_SEC;
- ext2_inc_count(inode);
+ inode_inc_link_count(inode);
atomic_inc(&inode->i_count);
return ext2_add_nondir(dentry, inode);
@@ -229,7 +213,7 @@ static int ext2_mkdir(struct inode * dir, struct dentry * dentry, int mode)
if (dir->i_nlink >= EXT2_LINK_MAX)
goto out;
- ext2_inc_count(dir);
+ inode_inc_link_count(dir);
inode = ext2_new_inode (dir, S_IFDIR | mode);
err = PTR_ERR(inode);
@@ -243,7 +227,7 @@ static int ext2_mkdir(struct inode * dir, struct dentry * dentry, int mode)
else
inode->i_mapping->a_ops = &ext2_aops;
- ext2_inc_count(inode);
+ inode_inc_link_count(inode);
err = ext2_make_empty(inode, dir);
if (err)
@@ -258,11 +242,11 @@ out:
return err;
out_fail:
- ext2_dec_count(inode);
- ext2_dec_count(inode);
+ inode_dec_link_count(inode);
+ inode_dec_link_count(inode);
iput(inode);
out_dir:
- ext2_dec_count(dir);
+ inode_dec_link_count(dir);
goto out;
}
@@ -282,7 +266,7 @@ static int ext2_unlink(struct inode * dir, struct dentry *dentry)
goto out;
inode->i_ctime = dir->i_ctime;
- ext2_dec_count(inode);
+ inode_dec_link_count(inode);
err = 0;
out:
return err;
@@ -297,8 +281,8 @@ static int ext2_rmdir (struct inode * dir, struct dentry *dentry)
err = ext2_unlink(dir, dentry);
if (!err) {
inode->i_size = 0;
- ext2_dec_count(inode);
- ext2_dec_count(dir);
+ inode_dec_link_count(inode);
+ inode_dec_link_count(dir);
}
}
return err;
@@ -338,41 +322,41 @@ static int ext2_rename (struct inode * old_dir, struct dentry * old_dentry,
new_de = ext2_find_entry (new_dir, new_dentry, &new_page);
if (!new_de)
goto out_dir;
- ext2_inc_count(old_inode);
+ inode_inc_link_count(old_inode);
ext2_set_link(new_dir, new_de, new_page, old_inode);
new_inode->i_ctime = CURRENT_TIME_SEC;
if (dir_de)
new_inode->i_nlink--;
- ext2_dec_count(new_inode);
+ inode_dec_link_count(new_inode);
} else {
if (dir_de) {
err = -EMLINK;
if (new_dir->i_nlink >= EXT2_LINK_MAX)
goto out_dir;
}
- ext2_inc_count(old_inode);
+ inode_inc_link_count(old_inode);
err = ext2_add_link(new_dentry, old_inode);
if (err) {
- ext2_dec_count(old_inode);
+ inode_dec_link_count(old_inode);
goto out_dir;
}
if (dir_de)
- ext2_inc_count(new_dir);
+ inode_inc_link_count(new_dir);
}
/*
* Like most other Unix systems, set the ctime for inodes on a
* rename.
- * ext2_dec_count() will mark the inode dirty.
+ * inode_dec_link_count() will mark the inode dirty.
*/
old_inode->i_ctime = CURRENT_TIME_SEC;
ext2_delete_entry (old_de, old_page);
- ext2_dec_count(old_inode);
+ inode_dec_link_count(old_inode);
if (dir_de) {
ext2_set_link(old_inode, dir_de, dir_page, new_dir);
- ext2_dec_count(old_dir);
+ inode_dec_link_count(old_dir);
}
return 0;
diff --git a/fs/ext3/dir.c b/fs/ext3/dir.c
index 832867aef3dc..773459164bb2 100644
--- a/fs/ext3/dir.c
+++ b/fs/ext3/dir.c
@@ -95,11 +95,10 @@ static int ext3_readdir(struct file * filp,
void * dirent, filldir_t filldir)
{
int error = 0;
- unsigned long offset, blk;
- int i, num, stored;
- struct buffer_head * bh, * tmp, * bha[16];
- struct ext3_dir_entry_2 * de;
- struct super_block * sb;
+ unsigned long offset;
+ int i, stored;
+ struct ext3_dir_entry_2 *de;
+ struct super_block *sb;
int err;
struct inode *inode = filp->f_dentry->d_inode;
int ret = 0;
@@ -124,12 +123,29 @@ static int ext3_readdir(struct file * filp,
}
#endif
stored = 0;
- bh = NULL;
offset = filp->f_pos & (sb->s_blocksize - 1);
while (!error && !stored && filp->f_pos < inode->i_size) {
- blk = (filp->f_pos) >> EXT3_BLOCK_SIZE_BITS(sb);
- bh = ext3_bread(NULL, inode, blk, 0, &err);
+ unsigned long blk = filp->f_pos >> EXT3_BLOCK_SIZE_BITS(sb);
+ struct buffer_head map_bh;
+ struct buffer_head *bh = NULL;
+
+ map_bh.b_state = 0;
+ err = ext3_get_block_handle(NULL, inode, blk, &map_bh, 0, 0);
+ if (!err) {
+ page_cache_readahead(sb->s_bdev->bd_inode->i_mapping,
+ &filp->f_ra,
+ filp,
+ map_bh.b_blocknr >>
+ (PAGE_CACHE_SHIFT - inode->i_blkbits),
+ 1);
+ bh = ext3_bread(NULL, inode, blk, 0, &err);
+ }
+
+ /*
+ * We ignore I/O errors on directories so users have a chance
+ * of recovering data when there's a bad sector
+ */
if (!bh) {
ext3_error (sb, "ext3_readdir",
"directory #%lu contains a hole at offset %lu",
@@ -138,26 +154,6 @@ static int ext3_readdir(struct file * filp,
continue;
}
- /*
- * Do the readahead
- */
- if (!offset) {
- for (i = 16 >> (EXT3_BLOCK_SIZE_BITS(sb) - 9), num = 0;
- i > 0; i--) {
- tmp = ext3_getblk (NULL, inode, ++blk, 0, &err);
- if (tmp && !buffer_uptodate(tmp) &&
- !buffer_locked(tmp))
- bha[num++] = tmp;
- else
- brelse (tmp);
- }
- if (num) {
- ll_rw_block (READA, num, bha);
- for (i = 0; i < num; i++)
- brelse (bha[i]);
- }
- }
-
revalidate:
/* If the dir block has changed since the last call to
* readdir(2), then we might be pointing to an invalid
diff --git a/fs/ext3/file.c b/fs/ext3/file.c
index 98e78345ead9..59098ea56711 100644
--- a/fs/ext3/file.c
+++ b/fs/ext3/file.c
@@ -37,9 +37,9 @@ static int ext3_release_file (struct inode * inode, struct file * filp)
if ((filp->f_mode & FMODE_WRITE) &&
(atomic_read(&inode->i_writecount) == 1))
{
- down(&EXT3_I(inode)->truncate_sem);
+ mutex_lock(&EXT3_I(inode)->truncate_mutex);
ext3_discard_reservation(inode);
- up(&EXT3_I(inode)->truncate_sem);
+ mutex_unlock(&EXT3_I(inode)->truncate_mutex);
}
if (is_dx(inode) && filp->private_data)
ext3_htree_free_dir_info(filp->private_data);
diff --git a/fs/ext3/inode.c b/fs/ext3/inode.c
index 0384e539b88f..2c361377e0a5 100644
--- a/fs/ext3/inode.c
+++ b/fs/ext3/inode.c
@@ -671,7 +671,7 @@ err_out:
* The BKL may not be held on entry here. Be sure to take it early.
*/
-static int
+int
ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock,
struct buffer_head *bh_result, int create, int extend_disksize)
{
@@ -702,7 +702,7 @@ ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock,
if (!create || err == -EIO)
goto cleanup;
- down(&ei->truncate_sem);
+ mutex_lock(&ei->truncate_mutex);
/*
* If the indirect block is missing while we are reading
@@ -723,7 +723,7 @@ ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock,
}
partial = ext3_get_branch(inode, depth, offsets, chain, &err);
if (!partial) {
- up(&ei->truncate_sem);
+ mutex_unlock(&ei->truncate_mutex);
if (err)
goto cleanup;
clear_buffer_new(bh_result);
@@ -759,13 +759,13 @@ ext3_get_block_handle(handle_t *handle, struct inode *inode, sector_t iblock,
err = ext3_splice_branch(handle, inode, iblock, chain,
partial, left);
/*
- * i_disksize growing is protected by truncate_sem. Don't forget to
+ * i_disksize growing is protected by truncate_mutex. Don't forget to
* protect it if you're about to implement concurrent
* ext3_get_block() -bzzz
*/
if (!err && extend_disksize && inode->i_size > ei->i_disksize)
ei->i_disksize = inode->i_size;
- up(&ei->truncate_sem);
+ mutex_unlock(&ei->truncate_mutex);
if (err)
goto cleanup;
@@ -1227,7 +1227,7 @@ static int journal_dirty_data_fn(handle_t *handle, struct buffer_head *bh)
* ext3_file_write() -> generic_file_write() -> __alloc_pages() -> ...
*
* Same applies to ext3_get_block(). We will deadlock on various things like
- * lock_journal and i_truncate_sem.
+ * lock_journal and i_truncate_mutex.
*
* Setting PF_MEMALLOC here doesn't work - too many internal memory
* allocations fail.
@@ -2161,7 +2161,7 @@ void ext3_truncate(struct inode * inode)
* From here we block out all ext3_get_block() callers who want to
* modify the block allocation tree.
*/
- down(&ei->truncate_sem);
+ mutex_lock(&ei->truncate_mutex);
if (n == 1) { /* direct blocks */
ext3_free_data(handle, inode, NULL, i_data+offsets[0],
@@ -2228,7 +2228,7 @@ do_indirects:
ext3_discard_reservation(inode);
- up(&ei->truncate_sem);
+ mutex_unlock(&ei->truncate_mutex);
inode->i_mtime = inode->i_ctime = CURRENT_TIME_SEC;
ext3_mark_inode_dirty(handle, inode);
diff --git a/fs/ext3/ioctl.c b/fs/ext3/ioctl.c
index 556cd5510078..aaf1da17b6d4 100644
--- a/fs/ext3/ioctl.c
+++ b/fs/ext3/ioctl.c
@@ -182,7 +182,7 @@ flags_err:
* need to allocate reservation structure for this inode
* before set the window size
*/
- down(&ei->truncate_sem);
+ mutex_lock(&ei->truncate_mutex);
if (!ei->i_block_alloc_info)
ext3_init_block_alloc_info(inode);
@@ -190,7 +190,7 @@ flags_err:
struct ext3_reserve_window_node *rsv = &ei->i_block_alloc_info->rsv_window_node;
rsv->rsv_goal_size = rsv_window_size;
}
- up(&ei->truncate_sem);
+ mutex_unlock(&ei->truncate_mutex);
return 0;
}
case EXT3_IOC_GROUP_EXTEND: {
diff --git a/fs/ext3/super.c b/fs/ext3/super.c
index 56bf76586019..efe5b20d7a5a 100644
--- a/fs/ext3/super.c
+++ b/fs/ext3/super.c
@@ -472,7 +472,7 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
#ifdef CONFIG_EXT3_FS_XATTR
init_rwsem(&ei->xattr_sem);
#endif
- init_MUTEX(&ei->truncate_sem);
+ mutex_init(&ei->truncate_mutex);
inode_init_once(&ei->vfs_inode);
}
}
@@ -2382,8 +2382,8 @@ static int ext3_statfs (struct super_block * sb, struct kstatfs * buf)
* Process 1 Process 2
* ext3_create() quota_sync()
* journal_start() write_dquot()
- * DQUOT_INIT() down(dqio_sem)
- * down(dqio_sem) journal_start()
+ * DQUOT_INIT() down(dqio_mutex)
+ * down(dqio_mutex) journal_start()
*
*/
diff --git a/fs/fat/fatent.c b/fs/fat/fatent.c
index a1a9e0451217..ab171ea8e869 100644
--- a/fs/fat/fatent.c
+++ b/fs/fat/fatent.c
@@ -267,19 +267,19 @@ static struct fatent_operations fat32_ops = {
static inline void lock_fat(struct msdos_sb_info *sbi)
{
- down(&sbi->fat_lock);
+ mutex_lock(&sbi->fat_lock);
}
static inline void unlock_fat(struct msdos_sb_info *sbi)
{
- up(&sbi->fat_lock);
+ mutex_unlock(&sbi->fat_lock);
}
void fat_ent_access_init(struct super_block *sb)
{
struct msdos_sb_info *sbi = MSDOS_SB(sb);
- init_MUTEX(&sbi->fat_lock);
+ mutex_init(&sbi->fat_lock);
switch (sbi->fat_bits) {
case 32:
diff --git a/fs/fcntl.c b/fs/fcntl.c
index dc4a7007f4e7..03c789560fb8 100644
--- a/fs/fcntl.c
+++ b/fs/fcntl.c
@@ -73,8 +73,8 @@ repeat:
* orig_start..fdt->next_fd
*/
start = orig_start;
- if (start < fdt->next_fd)
- start = fdt->next_fd;
+ if (start < files->next_fd)
+ start = files->next_fd;
newfd = start;
if (start < fdt->max_fdset) {
@@ -102,9 +102,8 @@ repeat:
* we reacquire the fdtable pointer and use it while holding
* the lock, no one can free it during that time.
*/
- fdt = files_fdtable(files);
- if (start <= fdt->next_fd)
- fdt->next_fd = newfd + 1;
+ if (start <= files->next_fd)
+ files->next_fd = newfd + 1;
error = newfd;
diff --git a/fs/file.c b/fs/file.c
index cea7cbea11d0..bbc743314730 100644
--- a/fs/file.c
+++ b/fs/file.c
@@ -125,7 +125,8 @@ static void free_fdtable_rcu(struct rcu_head *rcu)
kmem_cache_free(files_cachep, fdt->free_files);
return;
}
- if (fdt->max_fdset <= __FD_SETSIZE && fdt->max_fds <= NR_OPEN_DEFAULT) {
+ if (fdt->max_fdset <= EMBEDDED_FD_SET_SIZE &&
+ fdt->max_fds <= NR_OPEN_DEFAULT) {
/*
* The fdtable was embedded
*/
@@ -155,8 +156,9 @@ static void free_fdtable_rcu(struct rcu_head *rcu)
void free_fdtable(struct fdtable *fdt)
{
- if (fdt->free_files || fdt->max_fdset > __FD_SETSIZE ||
- fdt->max_fds > NR_OPEN_DEFAULT)
+ if (fdt->free_files ||
+ fdt->max_fdset > EMBEDDED_FD_SET_SIZE ||
+ fdt->max_fds > NR_OPEN_DEFAULT)
call_rcu(&fdt->rcu, free_fdtable_rcu);
}
@@ -199,7 +201,6 @@ static void copy_fdtable(struct fdtable *nfdt, struct fdtable *fdt)
(nfdt->max_fds - fdt->max_fds) *
sizeof(struct file *));
}
- nfdt->next_fd = fdt->next_fd;
}
/*
@@ -220,11 +221,9 @@ fd_set * alloc_fdset(int num)
void free_fdset(fd_set *array, int num)
{
- int size = num / 8;
-
- if (num <= __FD_SETSIZE) /* Don't free an embedded fdset */
+ if (num <= EMBEDDED_FD_SET_SIZE) /* Don't free an embedded fdset */
return;
- else if (size <= PAGE_SIZE)
+ else if (num <= 8 * PAGE_SIZE)
kfree(array);
else
vfree(array);
@@ -237,22 +236,17 @@ static struct fdtable *alloc_fdtable(int nr)
fd_set *new_openset = NULL, *new_execset = NULL;
struct file **new_fds;
- fdt = kmalloc(sizeof(*fdt), GFP_KERNEL);
+ fdt = kzalloc(sizeof(*fdt), GFP_KERNEL);
if (!fdt)
goto out;
- memset(fdt, 0, sizeof(*fdt));
- nfds = __FD_SETSIZE;
+ nfds = 8 * L1_CACHE_BYTES;
/* Expand to the max in easy steps */
- do {
- if (nfds < (PAGE_SIZE * 8))
- nfds = PAGE_SIZE * 8;
- else {
- nfds = nfds * 2;
- if (nfds > NR_OPEN)
- nfds = NR_OPEN;
- }
- } while (nfds <= nr);
+ while (nfds <= nr) {
+ nfds = nfds * 2;
+ if (nfds > NR_OPEN)
+ nfds = NR_OPEN;
+ }
new_openset = alloc_fdset(nfds);
new_execset = alloc_fdset(nfds);
diff --git a/fs/file_table.c b/fs/file_table.c
index 44fabeaa9415..bcea1998b4de 100644
--- a/fs/file_table.c
+++ b/fs/file_table.c
@@ -88,6 +88,7 @@ int proc_nr_files(ctl_table *table, int write, struct file *filp,
*/
struct file *get_empty_filp(void)
{
+ struct task_struct *tsk;
static int old_max;
struct file * f;
@@ -112,13 +113,14 @@ struct file *get_empty_filp(void)
if (security_file_alloc(f))
goto fail_sec;
- eventpoll_init_file(f);
+ tsk = current;
+ INIT_LIST_HEAD(&f->f_u.fu_list);
atomic_set(&f->f_count, 1);
- f->f_uid = current->fsuid;
- f->f_gid = current->fsgid;
rwlock_init(&f->f_owner.lock);
+ f->f_uid = tsk->fsuid;
+ f->f_gid = tsk->fsgid;
+ eventpoll_init_file(f);
/* f->f_version: 0 */
- INIT_LIST_HEAD(&f->f_u.fu_list);
return f;
over:
diff --git a/fs/hpfs/hpfs_fn.h b/fs/hpfs/hpfs_fn.h
index 6628c3b352cb..4c6473ab3b34 100644
--- a/fs/hpfs/hpfs_fn.h
+++ b/fs/hpfs/hpfs_fn.h
@@ -9,6 +9,7 @@
//#define DBG
//#define DEBUG_LOCKS
+#include <linux/mutex.h>
#include <linux/pagemap.h>
#include <linux/buffer_head.h>
#include <linux/hpfs_fs.h>
@@ -57,8 +58,8 @@ struct hpfs_inode_info {
unsigned i_ea_uid : 1; /* file's uid is stored in ea */
unsigned i_ea_gid : 1; /* file's gid is stored in ea */
unsigned i_dirty : 1;
- struct semaphore i_sem;
- struct semaphore i_parent;
+ struct mutex i_mutex;
+ struct mutex i_parent_mutex;
loff_t **i_rddir_off;
struct inode vfs_inode;
};
diff --git a/fs/hpfs/inode.c b/fs/hpfs/inode.c
index e3d17e9ea6c1..56f2c338c4d9 100644
--- a/fs/hpfs/inode.c
+++ b/fs/hpfs/inode.c
@@ -186,9 +186,9 @@ void hpfs_write_inode(struct inode *i)
kfree(hpfs_inode->i_rddir_off);
hpfs_inode->i_rddir_off = NULL;
}
- down(&hpfs_inode->i_parent);
+ mutex_lock(&hpfs_inode->i_parent_mutex);
if (!i->i_nlink) {
- up(&hpfs_inode->i_parent);
+ mutex_unlock(&hpfs_inode->i_parent_mutex);
return;
}
parent = iget_locked(i->i_sb, hpfs_inode->i_parent_dir);
@@ -199,14 +199,14 @@ void hpfs_write_inode(struct inode *i)
hpfs_read_inode(parent);
unlock_new_inode(parent);
}
- down(&hpfs_inode->i_sem);
+ mutex_lock(&hpfs_inode->i_mutex);
hpfs_write_inode_nolock(i);
- up(&hpfs_inode->i_sem);
+ mutex_unlock(&hpfs_inode->i_mutex);
iput(parent);
} else {
mark_inode_dirty(i);
}
- up(&hpfs_inode->i_parent);
+ mutex_unlock(&hpfs_inode->i_parent_mutex);
}
void hpfs_write_inode_nolock(struct inode *i)
diff --git a/fs/hpfs/namei.c b/fs/hpfs/namei.c
index 8ff8fc433fc1..a03abb12c610 100644
--- a/fs/hpfs/namei.c
+++ b/fs/hpfs/namei.c
@@ -60,7 +60,7 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
if (dee.read_only)
result->i_mode &= ~0222;
- down(&hpfs_i(dir)->i_sem);
+ mutex_lock(&hpfs_i(dir)->i_mutex);
r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
if (r == 1)
goto bail3;
@@ -101,11 +101,11 @@ static int hpfs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
hpfs_write_inode_nolock(result);
}
d_instantiate(dentry, result);
- up(&hpfs_i(dir)->i_sem);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
unlock_kernel();
return 0;
bail3:
- up(&hpfs_i(dir)->i_sem);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
iput(result);
bail2:
hpfs_brelse4(&qbh0);
@@ -168,7 +168,7 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
result->i_data.a_ops = &hpfs_aops;
hpfs_i(result)->mmu_private = 0;
- down(&hpfs_i(dir)->i_sem);
+ mutex_lock(&hpfs_i(dir)->i_mutex);
r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
if (r == 1)
goto bail2;
@@ -193,12 +193,12 @@ static int hpfs_create(struct inode *dir, struct dentry *dentry, int mode, struc
hpfs_write_inode_nolock(result);
}
d_instantiate(dentry, result);
- up(&hpfs_i(dir)->i_sem);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
unlock_kernel();
return 0;
bail2:
- up(&hpfs_i(dir)->i_sem);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
iput(result);
bail1:
brelse(bh);
@@ -254,7 +254,7 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t
result->i_blocks = 1;
init_special_inode(result, mode, rdev);
- down(&hpfs_i(dir)->i_sem);
+ mutex_lock(&hpfs_i(dir)->i_mutex);
r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
if (r == 1)
goto bail2;
@@ -271,12 +271,12 @@ static int hpfs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t
hpfs_write_inode_nolock(result);
d_instantiate(dentry, result);
- up(&hpfs_i(dir)->i_sem);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
brelse(bh);
unlock_kernel();
return 0;
bail2:
- up(&hpfs_i(dir)->i_sem);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
iput(result);
bail1:
brelse(bh);
@@ -333,7 +333,7 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
result->i_op = &page_symlink_inode_operations;
result->i_data.a_ops = &hpfs_symlink_aops;
- down(&hpfs_i(dir)->i_sem);
+ mutex_lock(&hpfs_i(dir)->i_mutex);
r = hpfs_add_dirent(dir, (char *)name, len, &dee, 0);
if (r == 1)
goto bail2;
@@ -352,11 +352,11 @@ static int hpfs_symlink(struct inode *dir, struct dentry *dentry, const char *sy
hpfs_write_inode_nolock(result);
d_instantiate(dentry, result);
- up(&hpfs_i(dir)->i_sem);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
unlock_kernel();
return 0;
bail2:
- up(&hpfs_i(dir)->i_sem);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
iput(result);
bail1:
brelse(bh);
@@ -382,8 +382,8 @@ static int hpfs_unlink(struct inode *dir, struct dentry *dentry)
lock_kernel();
hpfs_adjust_length((char *)name, &len);
again:
- down(&hpfs_i(inode)->i_parent);
- down(&hpfs_i(dir)->i_sem);
+ mutex_lock(&hpfs_i(inode)->i_parent_mutex);
+ mutex_lock(&hpfs_i(dir)->i_mutex);
err = -ENOENT;
de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *)name, len, &dno, &qbh);
if (!de)
@@ -410,8 +410,8 @@ again:
if (rep++)
break;
- up(&hpfs_i(dir)->i_sem);
- up(&hpfs_i(inode)->i_parent);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
+ mutex_unlock(&hpfs_i(inode)->i_parent_mutex);
d_drop(dentry);
spin_lock(&dentry->d_lock);
if (atomic_read(&dentry->d_count) > 1 ||
@@ -442,8 +442,8 @@ again:
out1:
hpfs_brelse4(&qbh);
out:
- up(&hpfs_i(dir)->i_sem);
- up(&hpfs_i(inode)->i_parent);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
+ mutex_unlock(&hpfs_i(inode)->i_parent_mutex);
unlock_kernel();
return err;
}
@@ -463,8 +463,8 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
hpfs_adjust_length((char *)name, &len);
lock_kernel();
- down(&hpfs_i(inode)->i_parent);
- down(&hpfs_i(dir)->i_sem);
+ mutex_lock(&hpfs_i(inode)->i_parent_mutex);
+ mutex_lock(&hpfs_i(dir)->i_mutex);
err = -ENOENT;
de = map_dirent(dir, hpfs_i(dir)->i_dno, (char *)name, len, &dno, &qbh);
if (!de)
@@ -502,8 +502,8 @@ static int hpfs_rmdir(struct inode *dir, struct dentry *dentry)
out1:
hpfs_brelse4(&qbh);
out:
- up(&hpfs_i(dir)->i_sem);
- up(&hpfs_i(inode)->i_parent);
+ mutex_unlock(&hpfs_i(dir)->i_mutex);
+ mutex_unlock(&hpfs_i(inode)->i_parent_mutex);
unlock_kernel();
return err;
}
@@ -565,12 +565,12 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
lock_kernel();
/* order doesn't matter, due to VFS exclusion */
- down(&hpfs_i(i)->i_parent);
+ mutex_lock(&hpfs_i(i)->i_parent_mutex);
if (new_inode)
- down(&hpfs_i(new_inode)->i_parent);
- down(&hpfs_i(old_dir)->i_sem);
+ mutex_lock(&hpfs_i(new_inode)->i_parent_mutex);
+ mutex_lock(&hpfs_i(old_dir)->i_mutex);
if (new_dir != old_dir)
- down(&hpfs_i(new_dir)->i_sem);
+ mutex_lock(&hpfs_i(new_dir)->i_mutex);
/* Erm? Moving over the empty non-busy directory is perfectly legal */
if (new_inode && S_ISDIR(new_inode->i_mode)) {
@@ -650,11 +650,11 @@ static int hpfs_rename(struct inode *old_dir, struct dentry *old_dentry,
hpfs_decide_conv(i, (char *)new_name, new_len);
end1:
if (old_dir != new_dir)
- up(&hpfs_i(new_dir)->i_sem);
- up(&hpfs_i(old_dir)->i_sem);
- up(&hpfs_i(i)->i_parent);
+ mutex_unlock(&hpfs_i(new_dir)->i_mutex);
+ mutex_unlock(&hpfs_i(old_dir)->i_mutex);
+ mutex_unlock(&hpfs_i(i)->i_parent_mutex);
if (new_inode)
- up(&hpfs_i(new_inode)->i_parent);
+ mutex_unlock(&hpfs_i(new_inode)->i_parent_mutex);
unlock_kernel();
return err;
}
diff --git a/fs/hpfs/super.c b/fs/hpfs/super.c
index 63e88d7e2c3b..9488a794076e 100644
--- a/fs/hpfs/super.c
+++ b/fs/hpfs/super.c
@@ -181,8 +181,8 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
SLAB_CTOR_CONSTRUCTOR) {
- init_MUTEX(&ei->i_sem);
- init_MUTEX(&ei->i_parent);
+ mutex_init(&ei->i_mutex);
+ mutex_init(&ei->i_parent_mutex);
inode_init_once(&ei->vfs_inode);
}
}
diff --git a/fs/inode.c b/fs/inode.c
index d0be6159eb7f..25967b67903d 100644
--- a/fs/inode.c
+++ b/fs/inode.c
@@ -84,14 +84,14 @@ static struct hlist_head *inode_hashtable;
DEFINE_SPINLOCK(inode_lock);
/*
- * iprune_sem provides exclusion between the kswapd or try_to_free_pages
+ * iprune_mutex provides exclusion between the kswapd or try_to_free_pages
* icache shrinking path, and the umount path. Without this exclusion,
* by the time prune_icache calls iput for the inode whose pages it has
* been invalidating, or by the time it calls clear_inode & destroy_inode
* from its final dispose_list, the struct super_block they refer to
* (for inode->i_sb->s_op) may already have been freed and reused.
*/
-DECLARE_MUTEX(iprune_sem);
+DEFINE_MUTEX(iprune_mutex);
/*
* Statistics gathering..
@@ -206,7 +206,7 @@ void inode_init_once(struct inode *inode)
i_size_ordered_init(inode);
#ifdef CONFIG_INOTIFY
INIT_LIST_HEAD(&inode->inotify_watches);
- sema_init(&inode->inotify_sem, 1);
+ mutex_init(&inode->inotify_mutex);
#endif
}
@@ -319,7 +319,7 @@ static int invalidate_list(struct list_head *head, struct list_head *dispose)
/*
* We can reschedule here without worrying about the list's
* consistency because the per-sb list of inodes must not
- * change during umount anymore, and because iprune_sem keeps
+ * change during umount anymore, and because iprune_mutex keeps
* shrink_icache_memory() away.
*/
cond_resched_lock(&inode_lock);
@@ -355,14 +355,14 @@ int invalidate_inodes(struct super_block * sb)
int busy;
LIST_HEAD(throw_away);
- down(&iprune_sem);
+ mutex_lock(&iprune_mutex);
spin_lock(&inode_lock);
inotify_unmount_inodes(&sb->s_inodes);
busy = invalidate_list(&sb->s_inodes, &throw_away);
spin_unlock(&inode_lock);
dispose_list(&throw_away);
- up(&iprune_sem);
+ mutex_unlock(&iprune_mutex);
return busy;
}
@@ -377,7 +377,7 @@ int __invalidate_device(struct block_device *bdev)
if (sb) {
/*
* no need to lock the super, get_super holds the
- * read semaphore so the filesystem cannot go away
+ * read mutex so the filesystem cannot go away
* under us (->put_super runs with the write lock
* hold).
*/
@@ -423,7 +423,7 @@ static void prune_icache(int nr_to_scan)
int nr_scanned;
unsigned long reap = 0;
- down(&iprune_sem);
+ mutex_lock(&iprune_mutex);
spin_lock(&inode_lock);
for (nr_scanned = 0; nr_scanned < nr_to_scan; nr_scanned++) {
struct inode *inode;
@@ -459,7 +459,7 @@ static void prune_icache(int nr_to_scan)
spin_unlock(&inode_lock);
dispose_list(&freeable);
- up(&iprune_sem);
+ mutex_unlock(&iprune_mutex);
if (current_is_kswapd())
mod_page_state(kswapd_inodesteal, reap);
diff --git a/fs/inotify.c b/fs/inotify.c
index 3041503bde02..0ee39ef591c6 100644
--- a/fs/inotify.c
+++ b/fs/inotify.c
@@ -54,10 +54,10 @@ int inotify_max_queued_events;
* Lock ordering:
*
* dentry->d_lock (used to keep d_move() away from dentry->d_parent)
- * iprune_sem (synchronize shrink_icache_memory())
+ * iprune_mutex (synchronize shrink_icache_memory())
* inode_lock (protects the super_block->s_inodes list)
- * inode->inotify_sem (protects inode->inotify_watches and watches->i_list)
- * inotify_dev->sem (protects inotify_device and watches->d_list)
+ * inode->inotify_mutex (protects inode->inotify_watches and watches->i_list)
+ * inotify_dev->mutex (protects inotify_device and watches->d_list)
*/
/*
@@ -79,12 +79,12 @@ int inotify_max_queued_events;
/*
* struct inotify_device - represents an inotify instance
*
- * This structure is protected by the semaphore 'sem'.
+ * This structure is protected by the mutex 'mutex'.
*/
struct inotify_device {
wait_queue_head_t wq; /* wait queue for i/o */
struct idr idr; /* idr mapping wd -> watch */
- struct semaphore sem; /* protects this bad boy */
+ struct mutex mutex; /* protects this bad boy */
struct list_head events; /* list of queued events */
struct list_head watches; /* list of watches */
atomic_t count; /* reference count */
@@ -101,7 +101,7 @@ struct inotify_device {
* device. In read(), this list is walked and all events that can fit in the
* buffer are returned.
*
- * Protected by dev->sem of the device in which we are queued.
+ * Protected by dev->mutex of the device in which we are queued.
*/
struct inotify_kernel_event {
struct inotify_event event; /* the user-space event */
@@ -112,8 +112,8 @@ struct inotify_kernel_event {
/*
* struct inotify_watch - represents a watch request on a specific inode
*
- * d_list is protected by dev->sem of the associated watch->dev.
- * i_list and mask are protected by inode->inotify_sem of the associated inode.
+ * d_list is protected by dev->mutex of the associated watch->dev.
+ * i_list and mask are protected by inode->inotify_mutex of the associated inode.
* dev, inode, and wd are never written to once the watch is created.
*/
struct inotify_watch {
@@ -261,7 +261,7 @@ static struct inotify_kernel_event * kernel_event(s32 wd, u32 mask, u32 cookie,
/*
* inotify_dev_get_event - return the next event in the given dev's queue
*
- * Caller must hold dev->sem.
+ * Caller must hold dev->mutex.
*/
static inline struct inotify_kernel_event *
inotify_dev_get_event(struct inotify_device *dev)
@@ -272,7 +272,7 @@ inotify_dev_get_event(struct inotify_device *dev)
/*
* inotify_dev_queue_event - add a new event to the given device
*
- * Caller must hold dev->sem. Can sleep (calls kernel_event()).
+ * Caller must hold dev->mutex. Can sleep (calls kernel_event()).
*/
static void inotify_dev_queue_event(struct inotify_device *dev,
struct inotify_watch *watch, u32 mask,
@@ -315,7 +315,7 @@ static void inotify_dev_queue_event(struct inotify_device *dev,
/*
* remove_kevent - cleans up and ultimately frees the given kevent
*
- * Caller must hold dev->sem.
+ * Caller must hold dev->mutex.
*/
static void remove_kevent(struct inotify_device *dev,
struct inotify_kernel_event *kevent)
@@ -332,7 +332,7 @@ static void remove_kevent(struct inotify_device *dev,
/*
* inotify_dev_event_dequeue - destroy an event on the given device
*
- * Caller must hold dev->sem.
+ * Caller must hold dev->mutex.
*/
static void inotify_dev_event_dequeue(struct inotify_device *dev)
{
@@ -346,7 +346,7 @@ static void inotify_dev_event_dequeue(struct inotify_device *dev)
/*
* inotify_dev_get_wd - returns the next WD for use by the given dev
*
- * Callers must hold dev->sem. This function can sleep.
+ * Callers must hold dev->mutex. This function can sleep.
*/
static int inotify_dev_get_wd(struct inotify_device *dev,
struct inotify_watch *watch)
@@ -383,7 +383,7 @@ static int find_inode(const char __user *dirname, struct nameidata *nd,
/*
* create_watch - creates a watch on the given device.
*
- * Callers must hold dev->sem. Calls inotify_dev_get_wd() so may sleep.
+ * Callers must hold dev->mutex. Calls inotify_dev_get_wd() so may sleep.
* Both 'dev' and 'inode' (by way of nameidata) need to be pinned.
*/
static struct inotify_watch *create_watch(struct inotify_device *dev,
@@ -434,7 +434,7 @@ static struct inotify_watch *create_watch(struct inotify_device *dev,
/*
* inotify_find_dev - find the watch associated with the given inode and dev
*
- * Callers must hold inode->inotify_sem.
+ * Callers must hold inode->inotify_mutex.
*/
static struct inotify_watch *inode_find_dev(struct inode *inode,
struct inotify_device *dev)
@@ -469,7 +469,7 @@ static void remove_watch_no_event(struct inotify_watch *watch,
* the IN_IGNORED event to the given device signifying that the inode is no
* longer watched.
*
- * Callers must hold both inode->inotify_sem and dev->sem. We drop a
+ * Callers must hold both inode->inotify_mutex and dev->mutex. We drop a
* reference to the inode before returning.
*
* The inode is not iput() so as to remain atomic. If the inode needs to be
@@ -507,21 +507,21 @@ void inotify_inode_queue_event(struct inode *inode, u32 mask, u32 cookie,
if (!inotify_inode_watched(inode))
return;
- down(&inode->inotify_sem);
+ mutex_lock(&inode->inotify_mutex);
list_for_each_entry_safe(watch, next, &inode->inotify_watches, i_list) {
u32 watch_mask = watch->mask;
if (watch_mask & mask) {
struct inotify_device *dev = watch->dev;
get_inotify_watch(watch);
- down(&dev->sem);
+ mutex_lock(&dev->mutex);
inotify_dev_queue_event(dev, watch, mask, cookie, name);
if (watch_mask & IN_ONESHOT)
remove_watch_no_event(watch, dev);
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
put_inotify_watch(watch);
}
}
- up(&inode->inotify_sem);
+ mutex_unlock(&inode->inotify_mutex);
}
EXPORT_SYMBOL_GPL(inotify_inode_queue_event);
@@ -569,7 +569,7 @@ EXPORT_SYMBOL_GPL(inotify_get_cookie);
* @list: list of inodes being unmounted (sb->s_inodes)
*
* Called with inode_lock held, protecting the unmounting super block's list
- * of inodes, and with iprune_sem held, keeping shrink_icache_memory() at bay.
+ * of inodes, and with iprune_mutex held, keeping shrink_icache_memory() at bay.
* We temporarily drop inode_lock, however, and CAN block.
*/
void inotify_unmount_inodes(struct list_head *list)
@@ -618,7 +618,7 @@ void inotify_unmount_inodes(struct list_head *list)
* We can safely drop inode_lock here because we hold
* references on both inode and next_i. Also no new inodes
* will be added since the umount has begun. Finally,
- * iprune_sem keeps shrink_icache_memory() away.
+ * iprune_mutex keeps shrink_icache_memory() away.
*/
spin_unlock(&inode_lock);
@@ -626,16 +626,16 @@ void inotify_unmount_inodes(struct list_head *list)
iput(need_iput_tmp);
/* for each watch, send IN_UNMOUNT and then remove it */
- down(&inode->inotify_sem);
+ mutex_lock(&inode->inotify_mutex);
watches = &inode->inotify_watches;
list_for_each_entry_safe(watch, next_w, watches, i_list) {
struct inotify_device *dev = watch->dev;
- down(&dev->sem);
+ mutex_lock(&dev->mutex);
inotify_dev_queue_event(dev, watch, IN_UNMOUNT,0,NULL);
remove_watch(watch, dev);
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
}
- up(&inode->inotify_sem);
+ mutex_unlock(&inode->inotify_mutex);
iput(inode);
spin_lock(&inode_lock);
@@ -651,14 +651,14 @@ void inotify_inode_is_dead(struct inode *inode)
{
struct inotify_watch *watch, *next;
- down(&inode->inotify_sem);
+ mutex_lock(&inode->inotify_mutex);
list_for_each_entry_safe(watch, next, &inode->inotify_watches, i_list) {
struct inotify_device *dev = watch->dev;
- down(&dev->sem);
+ mutex_lock(&dev->mutex);
remove_watch(watch, dev);
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
}
- up(&inode->inotify_sem);
+ mutex_unlock(&inode->inotify_mutex);
}
EXPORT_SYMBOL_GPL(inotify_inode_is_dead);
@@ -670,10 +670,10 @@ static unsigned int inotify_poll(struct file *file, poll_table *wait)
int ret = 0;
poll_wait(file, &dev->wq, wait);
- down(&dev->sem);
+ mutex_lock(&dev->mutex);
if (!list_empty(&dev->events))
ret = POLLIN | POLLRDNORM;
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
return ret;
}
@@ -695,9 +695,9 @@ static ssize_t inotify_read(struct file *file, char __user *buf,
prepare_to_wait(&dev->wq, &wait, TASK_INTERRUPTIBLE);
- down(&dev->sem);
+ mutex_lock(&dev->mutex);
events = !list_empty(&dev->events);
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
if (events) {
ret = 0;
break;
@@ -720,7 +720,7 @@ static ssize_t inotify_read(struct file *file, char __user *buf,
if (ret)
return ret;
- down(&dev->sem);
+ mutex_lock(&dev->mutex);
while (1) {
struct inotify_kernel_event *kevent;
@@ -750,7 +750,7 @@ static ssize_t inotify_read(struct file *file, char __user *buf,
remove_kevent(dev, kevent);
}
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
return ret;
}
@@ -763,37 +763,37 @@ static int inotify_release(struct inode *ignored, struct file *file)
* Destroy all of the watches on this device. Unfortunately, not very
* pretty. We cannot do a simple iteration over the list, because we
* do not know the inode until we iterate to the watch. But we need to
- * hold inode->inotify_sem before dev->sem. The following works.
+ * hold inode->inotify_mutex before dev->mutex. The following works.
*/
while (1) {
struct inotify_watch *watch;
struct list_head *watches;
struct inode *inode;
- down(&dev->sem);
+ mutex_lock(&dev->mutex);
watches = &dev->watches;
if (list_empty(watches)) {
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
break;
}
watch = list_entry(watches->next, struct inotify_watch, d_list);
get_inotify_watch(watch);
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
inode = watch->inode;
- down(&inode->inotify_sem);
- down(&dev->sem);
+ mutex_lock(&inode->inotify_mutex);
+ mutex_lock(&dev->mutex);
remove_watch_no_event(watch, dev);
- up(&dev->sem);
- up(&inode->inotify_sem);
+ mutex_unlock(&dev->mutex);
+ mutex_unlock(&inode->inotify_mutex);
put_inotify_watch(watch);
}
/* destroy all of the events on this device */
- down(&dev->sem);
+ mutex_lock(&dev->mutex);
while (!list_empty(&dev->events))
inotify_dev_event_dequeue(dev);
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
/* free this device: the put matching the get in inotify_init() */
put_inotify_dev(dev);
@@ -811,26 +811,26 @@ static int inotify_ignore(struct inotify_device *dev, s32 wd)
struct inotify_watch *watch;
struct inode *inode;
- down(&dev->sem);
+ mutex_lock(&dev->mutex);
watch = idr_find(&dev->idr, wd);
if (unlikely(!watch)) {
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
return -EINVAL;
}
get_inotify_watch(watch);
inode = watch->inode;
- up(&dev->sem);
+ mutex_unlock(&dev->mutex);
- down(&inode->inotify_sem);
- down(&dev->sem);
+ mutex_lock(&inode->inotify_mutex);
+ mutex_lock(&dev->mutex);
/* make sure that we did not race */
watch = idr_find(&dev->idr, wd);
if (likely(watch))
remove_watch(watch, dev);
- up(&dev->sem);
- up(&inode->inotify_sem);
+ mutex_unlock(&dev->mutex);
+ mutex_unlock(&inode->inotify_mutex);
put_inotify_watch(watch);
return 0;
@@ -905,7 +905,7 @@ asmlinkage long sys_inotify_init(void)
INIT_LIST_HEAD(&dev->events);
INIT_LIST_HEAD(&dev->watches);
init_waitqueue_head(&dev->wq);
- sema_init(&dev->sem, 1);
+ mutex_init(&dev->mutex);
dev->event_count = 0;
dev->queue_size = 0;
dev->max_events = inotify_max_queued_events;
@@ -960,8 +960,8 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask)
inode = nd.dentry->d_inode;
dev = filp->private_data;
- down(&inode->inotify_sem);
- down(&dev->sem);
+ mutex_lock(&inode->inotify_mutex);
+ mutex_lock(&dev->mutex);
if (mask & IN_MASK_ADD)
mask_add = 1;
@@ -998,8 +998,8 @@ asmlinkage long sys_inotify_add_watch(int fd, const char __user *path, u32 mask)
list_add(&watch->i_list, &inode->inotify_watches);
ret = watch->wd;
out:
- up(&dev->sem);
- up(&inode->inotify_sem);
+ mutex_unlock(&dev->mutex);
+ mutex_unlock(&inode->inotify_mutex);
path_release(&nd);
fput_and_out:
fput_light(filp, fput_needed);
diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c
index 543ed543d1e5..3f5102b069db 100644
--- a/fs/jbd/checkpoint.c
+++ b/fs/jbd/checkpoint.c
@@ -85,7 +85,7 @@ void __log_wait_for_space(journal_t *journal)
if (journal->j_flags & JFS_ABORT)
return;
spin_unlock(&journal->j_state_lock);
- down(&journal->j_checkpoint_sem);
+ mutex_lock(&journal->j_checkpoint_mutex);
/*
* Test again, another process may have checkpointed while we
@@ -98,7 +98,7 @@ void __log_wait_for_space(journal_t *journal)
log_do_checkpoint(journal);
spin_lock(&journal->j_state_lock);
}
- up(&journal->j_checkpoint_sem);
+ mutex_unlock(&journal->j_checkpoint_mutex);
}
}
diff --git a/fs/jbd/journal.c b/fs/jbd/journal.c
index e4b516ac4989..95a628d8cac8 100644
--- a/fs/jbd/journal.c
+++ b/fs/jbd/journal.c
@@ -659,8 +659,8 @@ static journal_t * journal_init_common (void)
init_waitqueue_head(&journal->j_wait_checkpoint);
init_waitqueue_head(&journal->j_wait_commit);
init_waitqueue_head(&journal->j_wait_updates);
- init_MUTEX(&journal->j_barrier);
- init_MUTEX(&journal->j_checkpoint_sem);
+ mutex_init(&journal->j_barrier);
+ mutex_init(&journal->j_checkpoint_mutex);
spin_lock_init(&journal->j_revoke_lock);
spin_lock_init(&journal->j_list_lock);
spin_lock_init(&journal->j_state_lock);
diff --git a/fs/jbd/transaction.c b/fs/jbd/transaction.c
index ca917973c2c0..5fc40888f4cf 100644
--- a/fs/jbd/transaction.c
+++ b/fs/jbd/transaction.c
@@ -455,7 +455,7 @@ void journal_lock_updates(journal_t *journal)
* to make sure that we serialise special journal-locked operations
* too.
*/
- down(&journal->j_barrier);
+ mutex_lock(&journal->j_barrier);
}
/**
@@ -470,7 +470,7 @@ void journal_unlock_updates (journal_t *journal)
{
J_ASSERT(journal->j_barrier_count != 0);
- up(&journal->j_barrier);
+ mutex_unlock(&journal->j_barrier);
spin_lock(&journal->j_state_lock);
--journal->j_barrier_count;
spin_unlock(&journal->j_state_lock);
diff --git a/fs/jffs/inode-v23.c b/fs/jffs/inode-v23.c
index fc3855a1aef3..890d7ff7456d 100644
--- a/fs/jffs/inode-v23.c
+++ b/fs/jffs/inode-v23.c
@@ -42,7 +42,7 @@
#include <linux/quotaops.h>
#include <linux/highmem.h>
#include <linux/vfs.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
#include <asm/byteorder.h>
#include <asm/uaccess.h>
@@ -203,7 +203,7 @@ jffs_setattr(struct dentry *dentry, struct iattr *iattr)
fmc = c->fmc;
D3(printk (KERN_NOTICE "notify_change(): down biglock\n"));
- down(&fmc->biglock);
+ mutex_lock(&fmc->biglock);
f = jffs_find_file(c, inode->i_ino);
@@ -211,7 +211,7 @@ jffs_setattr(struct dentry *dentry, struct iattr *iattr)
printk("jffs_setattr(): Invalid inode number: %lu\n",
inode->i_ino);
D3(printk (KERN_NOTICE "notify_change(): up biglock\n"));
- up(&fmc->biglock);
+ mutex_unlock(&fmc->biglock);
res = -EINVAL;
goto out;
});
@@ -232,7 +232,7 @@ jffs_setattr(struct dentry *dentry, struct iattr *iattr)
if (!(new_node = jffs_alloc_node())) {
D(printk("jffs_setattr(): Allocation failed!\n"));
D3(printk (KERN_NOTICE "notify_change(): up biglock\n"));
- up(&fmc->biglock);
+ mutex_unlock(&fmc->biglock);
res = -ENOMEM;
goto out;
}
@@ -319,7 +319,7 @@ jffs_setattr(struct dentry *dentry, struct iattr *iattr)
D(printk("jffs_notify_change(): The write failed!\n"));
jffs_free_node(new_node);
D3(printk (KERN_NOTICE "n_c(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
goto out;
}
@@ -327,7 +327,7 @@ jffs_setattr(struct dentry *dentry, struct iattr *iattr)
mark_inode_dirty(inode);
D3(printk (KERN_NOTICE "n_c(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
out:
unlock_kernel();
return res;
@@ -461,7 +461,7 @@ jffs_rename(struct inode *old_dir, struct dentry *old_dentry,
goto jffs_rename_end;
}
D3(printk (KERN_NOTICE "rename(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
/* Create a node and initialize as much as needed. */
result = -ENOMEM;
if (!(node = jffs_alloc_node())) {
@@ -555,7 +555,7 @@ jffs_rename(struct inode *old_dir, struct dentry *old_dentry,
jffs_rename_end:
D3(printk (KERN_NOTICE "rename(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return result;
} /* jffs_rename() */
@@ -574,14 +574,14 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
int ddino;
lock_kernel();
D3(printk (KERN_NOTICE "readdir(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
D2(printk("jffs_readdir(): inode: 0x%p, filp: 0x%p\n", inode, filp));
if (filp->f_pos == 0) {
D3(printk("jffs_readdir(): \".\" %lu\n", inode->i_ino));
if (filldir(dirent, ".", 1, filp->f_pos, inode->i_ino, DT_DIR) < 0) {
D3(printk (KERN_NOTICE "readdir(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return 0;
}
@@ -598,7 +598,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
D3(printk("jffs_readdir(): \"..\" %u\n", ddino));
if (filldir(dirent, "..", 2, filp->f_pos, ddino, DT_DIR) < 0) {
D3(printk (KERN_NOTICE "readdir(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return 0;
}
@@ -617,7 +617,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
if (filldir(dirent, f->name, f->nsize,
filp->f_pos , f->ino, DT_UNKNOWN) < 0) {
D3(printk (KERN_NOTICE "readdir(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return 0;
}
@@ -627,7 +627,7 @@ jffs_readdir(struct file *filp, void *dirent, filldir_t filldir)
} while(f && f->deleted);
}
D3(printk (KERN_NOTICE "readdir(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return filp->f_pos;
} /* jffs_readdir() */
@@ -660,7 +660,7 @@ jffs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
});
D3(printk (KERN_NOTICE "lookup(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
r = -ENAMETOOLONG;
if (len > JFFS_MAX_NAME_LEN) {
@@ -683,31 +683,31 @@ jffs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
if ((len == 1) && (name[0] == '.')) {
D3(printk (KERN_NOTICE "lookup(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
if (!(inode = iget(dir->i_sb, d->ino))) {
D(printk("jffs_lookup(): . iget() ==> NULL\n"));
goto jffs_lookup_end_no_biglock;
}
D3(printk (KERN_NOTICE "lookup(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
} else if ((len == 2) && (name[0] == '.') && (name[1] == '.')) {
D3(printk (KERN_NOTICE "lookup(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
if (!(inode = iget(dir->i_sb, d->pino))) {
D(printk("jffs_lookup(): .. iget() ==> NULL\n"));
goto jffs_lookup_end_no_biglock;
}
D3(printk (KERN_NOTICE "lookup(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
} else if ((f = jffs_find_child(d, name, len))) {
D3(printk (KERN_NOTICE "lookup(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
if (!(inode = iget(dir->i_sb, f->ino))) {
D(printk("jffs_lookup(): iget() ==> NULL\n"));
goto jffs_lookup_end_no_biglock;
}
D3(printk (KERN_NOTICE "lookup(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
} else {
D3(printk("jffs_lookup(): Couldn't find the file. "
"f = 0x%p, name = \"%s\", d = 0x%p, d->ino = %u\n",
@@ -717,13 +717,13 @@ jffs_lookup(struct inode *dir, struct dentry *dentry, struct nameidata *nd)
d_add(dentry, inode);
D3(printk (KERN_NOTICE "lookup(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return NULL;
jffs_lookup_end:
D3(printk (KERN_NOTICE "lookup(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
jffs_lookup_end_no_biglock:
unlock_kernel();
@@ -753,7 +753,7 @@ jffs_do_readpage_nolock(struct file *file, struct page *page)
ClearPageError(page);
D3(printk (KERN_NOTICE "readpage(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
read_len = 0;
result = 0;
@@ -782,7 +782,7 @@ jffs_do_readpage_nolock(struct file *file, struct page *page)
kunmap(page);
D3(printk (KERN_NOTICE "readpage(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
if (result) {
SetPageError(page);
@@ -839,7 +839,7 @@ jffs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
c = dir_f->c;
D3(printk (KERN_NOTICE "mkdir(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
dir_mode = S_IFDIR | (mode & (S_IRWXUGO|S_ISVTX)
& ~current->fs->umask);
@@ -906,7 +906,7 @@ jffs_mkdir(struct inode *dir, struct dentry *dentry, int mode)
result = 0;
jffs_mkdir_end:
D3(printk (KERN_NOTICE "mkdir(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return result;
} /* jffs_mkdir() */
@@ -921,10 +921,10 @@ jffs_rmdir(struct inode *dir, struct dentry *dentry)
D3(printk("***jffs_rmdir()\n"));
D3(printk (KERN_NOTICE "rmdir(): down biglock\n"));
lock_kernel();
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
ret = jffs_remove(dir, dentry, S_IFDIR);
D3(printk (KERN_NOTICE "rmdir(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return ret;
}
@@ -940,10 +940,10 @@ jffs_unlink(struct inode *dir, struct dentry *dentry)
lock_kernel();
D3(printk("***jffs_unlink()\n"));
D3(printk (KERN_NOTICE "unlink(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
ret = jffs_remove(dir, dentry, 0);
D3(printk (KERN_NOTICE "unlink(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return ret;
}
@@ -1086,7 +1086,7 @@ jffs_mknod(struct inode *dir, struct dentry *dentry, int mode, dev_t rdev)
c = dir_f->c;
D3(printk (KERN_NOTICE "mknod(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
/* Create and initialize a new node. */
if (!(node = jffs_alloc_node())) {
@@ -1152,7 +1152,7 @@ jffs_mknod_err:
jffs_mknod_end:
D3(printk (KERN_NOTICE "mknod(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return result;
} /* jffs_mknod() */
@@ -1203,7 +1203,7 @@ jffs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
return -ENOMEM;
}
D3(printk (KERN_NOTICE "symlink(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
node->data_offset = 0;
node->removed_size = 0;
@@ -1253,7 +1253,7 @@ jffs_symlink(struct inode *dir, struct dentry *dentry, const char *symname)
d_instantiate(dentry, inode);
jffs_symlink_end:
D3(printk (KERN_NOTICE "symlink(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return err;
} /* jffs_symlink() */
@@ -1306,7 +1306,7 @@ jffs_create(struct inode *dir, struct dentry *dentry, int mode,
return -ENOMEM;
}
D3(printk (KERN_NOTICE "create(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
node->data_offset = 0;
node->removed_size = 0;
@@ -1359,7 +1359,7 @@ jffs_create(struct inode *dir, struct dentry *dentry, int mode,
d_instantiate(dentry, inode);
jffs_create_end:
D3(printk (KERN_NOTICE "create(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
unlock_kernel();
return err;
} /* jffs_create() */
@@ -1423,7 +1423,7 @@ jffs_file_write(struct file *filp, const char *buf, size_t count,
thiscount = min(c->fmc->max_chunk_size - sizeof(struct jffs_raw_inode), count);
D3(printk (KERN_NOTICE "file_write(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
/* Urgh. POSIX says we can do short writes if we feel like it.
* In practice, we can't. Nothing will cope. So we loop until
@@ -1511,7 +1511,7 @@ jffs_file_write(struct file *filp, const char *buf, size_t count,
}
out:
D3(printk (KERN_NOTICE "file_write(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
/* Fix things in the real inode. */
if (pos > inode->i_size) {
@@ -1567,7 +1567,7 @@ jffs_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
return -EIO;
}
D3(printk (KERN_NOTICE "ioctl(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
switch (cmd) {
case JFFS_PRINT_HASH:
@@ -1609,7 +1609,7 @@ jffs_ioctl(struct inode *inode, struct file *filp, unsigned int cmd,
ret = -ENOTTY;
}
D3(printk (KERN_NOTICE "ioctl(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
return ret;
} /* jffs_ioctl() */
@@ -1685,12 +1685,12 @@ jffs_read_inode(struct inode *inode)
}
c = (struct jffs_control *)inode->i_sb->s_fs_info;
D3(printk (KERN_NOTICE "read_inode(): down biglock\n"));
- down(&c->fmc->biglock);
+ mutex_lock(&c->fmc->biglock);
if (!(f = jffs_find_file(c, inode->i_ino))) {
D(printk("jffs_read_inode(): No such inode (%lu).\n",
inode->i_ino));
D3(printk (KERN_NOTICE "read_inode(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
return;
}
inode->u.generic_ip = (void *)f;
@@ -1732,7 +1732,7 @@ jffs_read_inode(struct inode *inode)
}
D3(printk (KERN_NOTICE "read_inode(): up biglock\n"));
- up(&c->fmc->biglock);
+ mutex_unlock(&c->fmc->biglock);
}
diff --git a/fs/jffs/intrep.c b/fs/jffs/intrep.c
index ce7b54b0b2b7..0ef207dfaf6f 100644
--- a/fs/jffs/intrep.c
+++ b/fs/jffs/intrep.c
@@ -62,7 +62,7 @@
#include <linux/fs.h>
#include <linux/stat.h>
#include <linux/pagemap.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
#include <asm/byteorder.h>
#include <linux/smp_lock.h>
#include <linux/time.h>
@@ -3416,7 +3416,7 @@ jffs_garbage_collect_thread(void *ptr)
D1(printk (KERN_NOTICE "jffs_garbage_collect_thread(): collecting.\n"));
D3(printk (KERN_NOTICE "g_c_thread(): down biglock\n"));
- down(&fmc->biglock);
+ mutex_lock(&fmc->biglock);
D1(printk("***jffs_garbage_collect_thread(): round #%u, "
"fmc->dirty_size = %u\n", i++, fmc->dirty_size));
@@ -3447,6 +3447,6 @@ jffs_garbage_collect_thread(void *ptr)
gc_end:
D3(printk (KERN_NOTICE "g_c_thread(): up biglock\n"));
- up(&fmc->biglock);
+ mutex_unlock(&fmc->biglock);
} /* for (;;) */
} /* jffs_garbage_collect_thread() */
diff --git a/fs/jffs/jffs_fm.c b/fs/jffs/jffs_fm.c
index 6da13b309bd1..7d8ca1aeace2 100644
--- a/fs/jffs/jffs_fm.c
+++ b/fs/jffs/jffs_fm.c
@@ -139,7 +139,7 @@ jffs_build_begin(struct jffs_control *c, int unit)
fmc->tail = NULL;
fmc->head_extra = NULL;
fmc->tail_extra = NULL;
- init_MUTEX(&fmc->biglock);
+ mutex_init(&fmc->biglock);
return fmc;
}
diff --git a/fs/jffs/jffs_fm.h b/fs/jffs/jffs_fm.h
index f64151e74122..c794d923df2a 100644
--- a/fs/jffs/jffs_fm.h
+++ b/fs/jffs/jffs_fm.h
@@ -20,10 +20,11 @@
#ifndef __LINUX_JFFS_FM_H__
#define __LINUX_JFFS_FM_H__
+#include <linux/config.h>
#include <linux/types.h>
#include <linux/jffs.h>
#include <linux/mtd/mtd.h>
-#include <linux/config.h>
+#include <linux/mutex.h>
/* The alignment between two nodes in the flash memory. */
#define JFFS_ALIGN_SIZE 4
@@ -97,7 +98,7 @@ struct jffs_fmcontrol
struct jffs_fm *tail;
struct jffs_fm *head_extra;
struct jffs_fm *tail_extra;
- struct semaphore biglock;
+ struct mutex biglock;
};
/* Notice the two members head_extra and tail_extra in the jffs_control
diff --git a/fs/libfs.c b/fs/libfs.c
index 71fd08fa4103..4fdeaceb892c 100644
--- a/fs/libfs.c
+++ b/fs/libfs.c
@@ -7,6 +7,8 @@
#include <linux/pagemap.h>
#include <linux/mount.h>
#include <linux/vfs.h>
+#include <linux/mutex.h>
+
#include <asm/uaccess.h>
int simple_getattr(struct vfsmount *mnt, struct dentry *dentry,
@@ -530,7 +532,7 @@ struct simple_attr {
char set_buf[24];
void *data;
const char *fmt; /* format for read operation */
- struct semaphore sem; /* protects access to these buffers */
+ struct mutex mutex; /* protects access to these buffers */
};
/* simple_attr_open is called by an actual attribute open file operation
@@ -549,7 +551,7 @@ int simple_attr_open(struct inode *inode, struct file *file,
attr->set = set;
attr->data = inode->u.generic_ip;
attr->fmt = fmt;
- init_MUTEX(&attr->sem);
+ mutex_init(&attr->mutex);
file->private_data = attr;
@@ -575,7 +577,7 @@ ssize_t simple_attr_read(struct file *file, char __user *buf,
if (!attr->get)
return -EACCES;
- down(&attr->sem);
+ mutex_lock(&attr->mutex);
if (*ppos) /* continued read */
size = strlen(attr->get_buf);
else /* first read */
@@ -584,7 +586,7 @@ ssize_t simple_attr_read(struct file *file, char __user *buf,
(unsigned long long)attr->get(attr->data));
ret = simple_read_from_buffer(buf, len, ppos, attr->get_buf, size);
- up(&attr->sem);
+ mutex_unlock(&attr->mutex);
return ret;
}
@@ -602,7 +604,7 @@ ssize_t simple_attr_write(struct file *file, const char __user *buf,
if (!attr->set)
return -EACCES;
- down(&attr->sem);
+ mutex_lock(&attr->mutex);
ret = -EFAULT;
size = min(sizeof(attr->set_buf) - 1, len);
if (copy_from_user(attr->set_buf, buf, size))
@@ -613,7 +615,7 @@ ssize_t simple_attr_write(struct file *file, const char __user *buf,
val = simple_strtol(attr->set_buf, NULL, 0);
attr->set(attr->data, val);
out:
- up(&attr->sem);
+ mutex_unlock(&attr->mutex);
return ret;
}
diff --git a/fs/minix/namei.c b/fs/minix/namei.c
index b25bca5bdb57..5b6a4540a05b 100644
--- a/fs/minix/namei.c
+++ b/fs/minix/namei.c
@@ -6,18 +6,6 @@
#include "minix.h"
-static inline void inc_count(struct inode *inode)
-{
- inode->i_nlink++;
- mark_inode_dirty(inode);
-}
-
-static inline void dec_count(struct inode *inode)
-{
- inode->i_nlink--;
- mark_inode_dirty(inode);
-}
-
static int add_nondir(struct dentry *dentry, struct inode *inode)
{
int err = minix_add_link(dentry, inode);
@@ -25,7 +13,7 @@ static int add_nondir(struct dentry *dentry, struct inode *inode)
d_instantiate(dentry, inode);
return 0;
}
- dec_count(inode);
+ inode_dec_link_count(inode);
iput(inode);
return err;
}
@@ -125,7 +113,7 @@ out:
return err;
out_fail:
- dec_count(inode);
+ inode_dec_link_count(inode);
iput(inode);
goto out;
}
@@ -139,7 +127,7 @@ static int minix_link(struct dentry * old_dentry, struct inode * dir,
return -EMLINK;
inode->i_ctime = CURRENT_TIME_SEC;
- inc_count(inode);
+ inode_inc_link_count(inode);
atomic_inc(&inode->i_count);
return add_nondir(dentry, inode);
}
@@ -152,7 +140,7 @@ static int minix_mkdir(struct inode * dir, struct dentry *dentry, int mode)
if (dir->i_nlink >= minix_sb(dir->i_sb)->s_link_max)
goto out;
- inc_count(dir);
+ inode_inc_link_count(dir);
inode = minix_new_inode(dir, &err);
if (!inode)
@@ -163,7 +151,7 @@ static int minix_mkdir(struct inode * dir, struct dentry *dentry, int mode)
inode->i_mode |= S_ISGID;
minix_set_inode(inode, 0);
- inc_count(inode);
+ inode_inc_link_count(inode);
err = minix_make_empty(inode, dir);
if (err)
@@ -178,11 +166,11 @@ out:
return err;
out_fail:
- dec_count(inode);
- dec_count(inode);
+ inode_dec_link_count(inode);
+ inode_dec_link_count(inode);
iput(inode);
out_dir:
- dec_count(dir);
+ inode_dec_link_count(dir);
goto out;
}
@@ -202,7 +190,7 @@ static int minix_unlink(struct inode * dir, struct dentry *dentry)
goto end_unlink;
inode->i_ctime = dir->i_ctime;
- dec_count(inode);
+ inode_dec_link_count(inode);
end_unlink:
return err;
}
@@ -215,8 +203,8 @@ static int minix_rmdir(struct inode * dir, struct dentry *dentry)
if (minix_empty_dir(inode)) {
err = minix_unlink(dir, dentry);
if (!err) {
- dec_count(dir);
- dec_count(inode);
+ inode_dec_link_count(dir);
+ inode_dec_link_count(inode);
}
}
return err;
@@ -257,34 +245,34 @@ static int minix_rename(struct inode * old_dir, struct dentry *old_dentry,
new_de = minix_find_entry(new_dentry, &new_page);
if (!new_de)
goto out_dir;
- inc_count(old_inode);
+ inode_inc_link_count(old_inode);
minix_set_link(new_de, new_page, old_inode);
new_inode->i_ctime = CURRENT_TIME_SEC;
if (dir_de)
new_inode->i_nlink--;
- dec_count(new_inode);
+ inode_dec_link_count(new_inode);
} else {
if (dir_de) {
err = -EMLINK;
if (new_dir->i_nlink >= info->s_link_max)
goto out_dir;
}
- inc_count(old_inode);
+ inode_inc_link_count(old_inode);
err = minix_add_link(new_dentry, old_inode);
if (err) {
- dec_count(old_inode);
+ inode_dec_link_count(old_inode);
goto out_dir;
}
if (dir_de)
- inc_count(new_dir);
+ inode_inc_link_count(new_dir);
}
minix_delete_entry(old_de, old_page);
- dec_count(old_inode);
+ inode_dec_link_count(old_inode);
if (dir_de) {
minix_set_link(dir_de, dir_page, new_dir);
- dec_count(old_dir);
+ inode_dec_link_count(old_dir);
}
return 0;
diff --git a/fs/namei.c b/fs/namei.c
index 8dc2b038d5d9..c72b940797fc 100644
--- a/fs/namei.c
+++ b/fs/namei.c
@@ -104,7 +104,7 @@
*/
/*
* [Sep 2001 AV] Single-semaphore locking scheme (kudos to David Holland)
- * implemented. Let's see if raised priority of ->s_vfs_rename_sem gives
+ * implemented. Let's see if raised priority of ->s_vfs_rename_mutex gives
* any extra contention...
*/
@@ -1422,7 +1422,7 @@ struct dentry *lock_rename(struct dentry *p1, struct dentry *p2)
return NULL;
}
- down(&p1->d_inode->i_sb->s_vfs_rename_sem);
+ mutex_lock(&p1->d_inode->i_sb->s_vfs_rename_mutex);
for (p = p1; p->d_parent != p; p = p->d_parent) {
if (p->d_parent == p2) {
@@ -1450,7 +1450,7 @@ void unlock_rename(struct dentry *p1, struct dentry *p2)
mutex_unlock(&p1->d_inode->i_mutex);
if (p1 != p2) {
mutex_unlock(&p2->d_inode->i_mutex);
- up(&p1->d_inode->i_sb->s_vfs_rename_sem);
+ mutex_unlock(&p1->d_inode->i_sb->s_vfs_rename_mutex);
}
}
@@ -2277,17 +2277,17 @@ asmlinkage long sys_link(const char __user *oldname, const char __user *newname)
* a) we can get into loop creation. Check is done in is_subdir().
* b) race potential - two innocent renames can create a loop together.
* That's where 4.4 screws up. Current fix: serialization on
- * sb->s_vfs_rename_sem. We might be more accurate, but that's another
+ * sb->s_vfs_rename_mutex. We might be more accurate, but that's another
* story.
* c) we have to lock _three_ objects - parents and victim (if it exists).
* And that - after we got ->i_mutex on parents (until then we don't know
* whether the target exists). Solution: try to be smart with locking
* order for inodes. We rely on the fact that tree topology may change
- * only under ->s_vfs_rename_sem _and_ that parent of the object we
+ * only under ->s_vfs_rename_mutex _and_ that parent of the object we
* move will be locked. Thus we can rank directories by the tree
* (ancestors first) and rank all non-directories after them.
* That works since everybody except rename does "lock parent, lookup,
- * lock child" and rename is under ->s_vfs_rename_sem.
+ * lock child" and rename is under ->s_vfs_rename_mutex.
* HOWEVER, it relies on the assumption that any object with ->lookup()
* has no more than 1 dentry. If "hybrid" objects will ever appear,
* we'd better make sure that there's no link(2) for them.
diff --git a/fs/ncpfs/file.c b/fs/ncpfs/file.c
index 973b444d6914..ebdad8f6398f 100644
--- a/fs/ncpfs/file.c
+++ b/fs/ncpfs/file.c
@@ -46,7 +46,7 @@ int ncp_make_open(struct inode *inode, int right)
NCP_FINFO(inode)->volNumber,
NCP_FINFO(inode)->dirEntNum);
error = -EACCES;
- down(&NCP_FINFO(inode)->open_sem);
+ mutex_lock(&NCP_FINFO(inode)->open_mutex);
if (!atomic_read(&NCP_FINFO(inode)->opened)) {
struct ncp_entry_info finfo;
int result;
@@ -93,7 +93,7 @@ int ncp_make_open(struct inode *inode, int right)
}
out_unlock:
- up(&NCP_FINFO(inode)->open_sem);
+ mutex_unlock(&NCP_FINFO(inode)->open_mutex);
out:
return error;
}
diff --git a/fs/ncpfs/inode.c b/fs/ncpfs/inode.c
index d277a58bd128..0b521d3d97ce 100644
--- a/fs/ncpfs/inode.c
+++ b/fs/ncpfs/inode.c
@@ -63,7 +63,7 @@ static void init_once(void * foo, kmem_cache_t * cachep, unsigned long flags)
if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
SLAB_CTOR_CONSTRUCTOR) {
- init_MUTEX(&ei->open_sem);
+ mutex_init(&ei->open_mutex);
inode_init_once(&ei->vfs_inode);
}
}
@@ -520,7 +520,7 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent)
}
/* server->lock = 0; */
- init_MUTEX(&server->sem);
+ mutex_init(&server->mutex);
server->packet = NULL;
/* server->buffer_size = 0; */
/* server->conn_status = 0; */
@@ -557,7 +557,7 @@ static int ncp_fill_super(struct super_block *sb, void *raw_data, int silent)
server->dentry_ttl = 0; /* no caching */
INIT_LIST_HEAD(&server->tx.requests);
- init_MUTEX(&server->rcv.creq_sem);
+ mutex_init(&server->rcv.creq_mutex);
server->tx.creq = NULL;
server->rcv.creq = NULL;
server->data_ready = sock->sk->sk_data_ready;
diff --git a/fs/ncpfs/ncplib_kernel.c b/fs/ncpfs/ncplib_kernel.c
index c755e1848a42..d9ebf6439f59 100644
--- a/fs/ncpfs/ncplib_kernel.c
+++ b/fs/ncpfs/ncplib_kernel.c
@@ -291,7 +291,7 @@ ncp_make_closed(struct inode *inode)
int err;
err = 0;
- down(&NCP_FINFO(inode)->open_sem);
+ mutex_lock(&NCP_FINFO(inode)->open_mutex);
if (atomic_read(&NCP_FINFO(inode)->opened) == 1) {
atomic_set(&NCP_FINFO(inode)->opened, 0);
err = ncp_close_file(NCP_SERVER(inode), NCP_FINFO(inode)->file_handle);
@@ -301,7 +301,7 @@ ncp_make_closed(struct inode *inode)
NCP_FINFO(inode)->volNumber,
NCP_FINFO(inode)->dirEntNum, err);
}
- up(&NCP_FINFO(inode)->open_sem);
+ mutex_unlock(&NCP_FINFO(inode)->open_mutex);
return err;
}
diff --git a/fs/ncpfs/sock.c b/fs/ncpfs/sock.c
index 6593a5ca88ba..8783eb7ec641 100644
--- a/fs/ncpfs/sock.c
+++ b/fs/ncpfs/sock.c
@@ -171,9 +171,9 @@ static inline void __ncp_abort_request(struct ncp_server *server, struct ncp_req
static inline void ncp_abort_request(struct ncp_server *server, struct ncp_request_reply *req, int err)
{
- down(&server->rcv.creq_sem);
+ mutex_lock(&server->rcv.creq_mutex);
__ncp_abort_request(server, req, err);
- up(&server->rcv.creq_sem);
+ mutex_unlock(&server->rcv.creq_mutex);
}
static inline void __ncptcp_abort(struct ncp_server *server)
@@ -303,20 +303,20 @@ static inline void __ncp_start_request(struct ncp_server *server, struct ncp_req
static int ncp_add_request(struct ncp_server *server, struct ncp_request_reply *req)
{
- down(&server->rcv.creq_sem);
+ mutex_lock(&server->rcv.creq_mutex);
if (!ncp_conn_valid(server)) {
- up(&server->rcv.creq_sem);
+ mutex_unlock(&server->rcv.creq_mutex);
printk(KERN_ERR "ncpfs: tcp: Server died\n");
return -EIO;
}
if (server->tx.creq || server->rcv.creq) {
req->status = RQ_QUEUED;
list_add_tail(&req->req, &server->tx.requests);
- up(&server->rcv.creq_sem);
+ mutex_unlock(&server->rcv.creq_mutex);
return 0;
}
__ncp_start_request(server, req);
- up(&server->rcv.creq_sem);
+ mutex_unlock(&server->rcv.creq_mutex);
return 0;
}
@@ -400,7 +400,7 @@ void ncpdgram_rcv_proc(void *s)
info_server(server, 0, server->unexpected_packet.data, result);
continue;
}
- down(&server->rcv.creq_sem);
+ mutex_lock(&server->rcv.creq_mutex);
req = server->rcv.creq;
if (req && (req->tx_type == NCP_ALLOC_SLOT_REQUEST || (server->sequence == reply.sequence &&
server->connection == get_conn_number(&reply)))) {
@@ -430,11 +430,11 @@ void ncpdgram_rcv_proc(void *s)
server->rcv.creq = NULL;
ncp_finish_request(req, result);
__ncp_next_request(server);
- up(&server->rcv.creq_sem);
+ mutex_unlock(&server->rcv.creq_mutex);
continue;
}
}
- up(&server->rcv.creq_sem);
+ mutex_unlock(&server->rcv.creq_mutex);
}
drop:;
_recv(sock, &reply, sizeof(reply), MSG_DONTWAIT);
@@ -472,9 +472,9 @@ static void __ncpdgram_timeout_proc(struct ncp_server *server)
void ncpdgram_timeout_proc(void *s)
{
struct ncp_server *server = s;
- down(&server->rcv.creq_sem);
+ mutex_lock(&server->rcv.creq_mutex);
__ncpdgram_timeout_proc(server);
- up(&server->rcv.creq_sem);
+ mutex_unlock(&server->rcv.creq_mutex);
}
static inline void ncp_init_req(struct ncp_request_reply* req)
@@ -657,18 +657,18 @@ void ncp_tcp_rcv_proc(void *s)
{
struct ncp_server *server = s;
- down(&server->rcv.creq_sem);
+ mutex_lock(&server->rcv.creq_mutex);
__ncptcp_rcv_proc(server);
- up(&server->rcv.creq_sem);
+ mutex_unlock(&server->rcv.creq_mutex);
}
void ncp_tcp_tx_proc(void *s)
{
struct ncp_server *server = s;
- down(&server->rcv.creq_sem);
+ mutex_lock(&server->rcv.creq_mutex);
__ncptcp_try_send(server);
- up(&server->rcv.creq_sem);
+ mutex_unlock(&server->rcv.creq_mutex);
}
static int do_ncp_rpc_call(struct ncp_server *server, int size,
@@ -833,7 +833,7 @@ int ncp_disconnect(struct ncp_server *server)
void ncp_lock_server(struct ncp_server *server)
{
- down(&server->sem);
+ mutex_lock(&server->mutex);
if (server->lock)
printk(KERN_WARNING "ncp_lock_server: was locked!\n");
server->lock = 1;
@@ -846,5 +846,5 @@ void ncp_unlock_server(struct ncp_server *server)
return;
}
server->lock = 0;
- up(&server->sem);
+ mutex_unlock(&server->mutex);
}
diff --git a/fs/open.c b/fs/open.c
index 70e0230d8e77..1091dadd6c38 100644
--- a/fs/open.c
+++ b/fs/open.c
@@ -973,7 +973,7 @@ repeat:
fdt = files_fdtable(files);
fd = find_next_zero_bit(fdt->open_fds->fds_bits,
fdt->max_fdset,
- fdt->next_fd);
+ files->next_fd);
/*
* N.B. For clone tasks sharing a files structure, this test
@@ -998,7 +998,7 @@ repeat:
FD_SET(fd, fdt->open_fds);
FD_CLR(fd, fdt->close_on_exec);
- fdt->next_fd = fd + 1;
+ files->next_fd = fd + 1;
#if 1
/* Sanity check */
if (fdt->fd[fd] != NULL) {
@@ -1019,8 +1019,8 @@ static void __put_unused_fd(struct files_struct *files, unsigned int fd)
{
struct fdtable *fdt = files_fdtable(files);
__FD_CLR(fd, fdt->open_fds);
- if (fd < fdt->next_fd)
- fdt->next_fd = fd;
+ if (fd < files->next_fd)
+ files->next_fd = fd;
}
void fastcall put_unused_fd(unsigned int fd)
diff --git a/fs/proc/proc_misc.c b/fs/proc/proc_misc.c
index 1d24fead51a6..826c131994c3 100644
--- a/fs/proc/proc_misc.c
+++ b/fs/proc/proc_misc.c
@@ -312,7 +312,7 @@ static void *devinfo_next(struct seq_file *f, void *v, loff_t *pos)
case BLK_HDR:
info->state = BLK_LIST;
(*pos)++;
- break;
+ /*fallthrough*/
case BLK_LIST:
if (get_blkdev_info(info->blkdev,&idummy,&ndummy)) {
/*
diff --git a/fs/qnx4/file.c b/fs/qnx4/file.c
index b471315e24ef..c33963fded9e 100644
--- a/fs/qnx4/file.c
+++ b/fs/qnx4/file.c
@@ -12,10 +12,7 @@
* 27-06-1998 by Frank Denis : file overwriting.
*/
-#include <linux/config.h>
-#include <linux/types.h>
#include <linux/fs.h>
-#include <linux/time.h>
#include <linux/qnx4_fs.h>
/*
diff --git a/fs/quota.c b/fs/quota.c
index ba9e0bf32f67..d6a2be826e29 100644
--- a/fs/quota.c
+++ b/fs/quota.c
@@ -170,10 +170,10 @@ static void quota_sync_sb(struct super_block *sb, int type)
/* Now when everything is written we can discard the pagecache so
* that userspace sees the changes. We need i_mutex and so we could
- * not do it inside dqonoff_sem. Moreover we need to be carefull
+ * not do it inside dqonoff_mutex. Moreover we need to be carefull
* about races with quotaoff() (that is the reason why we have own
* reference to inode). */
- down(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_lock(&sb_dqopt(sb)->dqonoff_mutex);
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
discard[cnt] = NULL;
if (type != -1 && cnt != type)
@@ -182,7 +182,7 @@ static void quota_sync_sb(struct super_block *sb, int type)
continue;
discard[cnt] = igrab(sb_dqopt(sb)->files[cnt]);
}
- up(&sb_dqopt(sb)->dqonoff_sem);
+ mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex);
for (cnt = 0; cnt < MAXQUOTAS; cnt++) {
if (discard[cnt]) {
mutex_lock(&discard[cnt]->i_mutex);
diff --git a/fs/quota_v2.c b/fs/quota_v2.c
index b4199ec3ece4..c519a583e681 100644
--- a/fs/quota_v2.c
+++ b/fs/quota_v2.c
@@ -394,7 +394,7 @@ static int v2_write_dquot(struct dquot *dquot)
ssize_t ret;
struct v2_disk_dqblk ddquot, empty;
- /* dq_off is guarded by dqio_sem */
+ /* dq_off is guarded by dqio_mutex */
if (!dquot->dq_off)
if ((ret = dq_insert_tree(dquot)) < 0) {
printk(KERN_ERR "VFS: Error %zd occurred while creating quota.\n", ret);
diff --git a/fs/ramfs/file-mmu.c b/fs/ramfs/file-mmu.c
index 2115383dcc8d..6ada2095b9ac 100644
--- a/fs/ramfs/file-mmu.c
+++ b/fs/ramfs/file-mmu.c
@@ -24,18 +24,7 @@
* caches is sufficient.
*/
-#include <linux/module.h>
#include <linux/fs.h>
-#include <linux/pagemap.h>
-#include <linux/highmem.h>
-#include <linux/init.h>
-#include <linux/string.h>
-#include <linux/smp_lock.h>
-#include <linux/backing-dev.h>
-#include <linux/ramfs.h>
-
-#include <asm/uaccess.h>
-#include "internal.h"
struct address_space_operations ramfs_aops = {
.readpage = simple_readpage,
diff --git a/fs/seq_file.c b/fs/seq_file.c
index 7c40570b71dc..555b9ac04c25 100644
--- a/fs/seq_file.c
+++ b/fs/seq_file.c
@@ -37,7 +37,7 @@ int seq_open(struct file *file, struct seq_operations *op)
file->private_data = p;
}
memset(p, 0, sizeof(*p));
- sema_init(&p->sem, 1);
+ mutex_init(&p->lock);
p->op = op;
/*
@@ -71,7 +71,7 @@ ssize_t seq_read(struct file *file, char __user *buf, size_t size, loff_t *ppos)
void *p;
int err = 0;
- down(&m->sem);
+ mutex_lock(&m->lock);
/*
* seq_file->op->..m_start/m_stop/m_next may do special actions
* or optimisations based on the file->f_version, so we want to
@@ -164,7 +164,7 @@ Done:
else
*ppos += copied;
file->f_version = m->version;
- up(&m->sem);
+ mutex_unlock(&m->lock);
return copied;
Enomem:
err = -ENOMEM;
@@ -237,7 +237,7 @@ loff_t seq_lseek(struct file *file, loff_t offset, int origin)
struct seq_file *m = (struct seq_file *)file->private_data;
long long retval = -EINVAL;
- down(&m->sem);
+ mutex_lock(&m->lock);
m->version = file->f_version;
switch (origin) {
case 1:
@@ -260,7 +260,7 @@ loff_t seq_lseek(struct file *file, loff_t offset, int origin)
}
}
}
- up(&m->sem);
+ mutex_unlock(&m->lock);
file->f_version = m->version;
return retval;
}
diff --git a/fs/super.c b/fs/super.c
index e20b5580afd5..425861cb1caa 100644
--- a/fs/super.c
+++ b/fs/super.c
@@ -76,9 +76,9 @@ static struct super_block *alloc_super(void)
down_write(&s->s_umount);
s->s_count = S_BIAS;
atomic_set(&s->s_active, 1);
- sema_init(&s->s_vfs_rename_sem,1);
- sema_init(&s->s_dquot.dqio_sem, 1);
- sema_init(&s->s_dquot.dqonoff_sem, 1);
+ mutex_init(&s->s_vfs_rename_mutex);
+ mutex_init(&s->s_dquot.dqio_mutex);
+ mutex_init(&s->s_dquot.dqonoff_mutex);
init_rwsem(&s->s_dquot.dqptr_sem);
init_waitqueue_head(&s->s_wait_unfrozen);
s->s_maxbytes = MAX_NON_LFS;
@@ -693,9 +693,9 @@ struct super_block *get_sb_bdev(struct file_system_type *fs_type,
* will protect the lockfs code from trying to start a snapshot
* while we are mounting
*/
- down(&bdev->bd_mount_sem);
+ mutex_lock(&bdev->bd_mount_mutex);
s = sget(fs_type, test_bdev_super, set_bdev_super, bdev);
- up(&bdev->bd_mount_sem);
+ mutex_unlock(&bdev->bd_mount_mutex);
if (IS_ERR(s))
goto out;
diff --git a/fs/sysv/namei.c b/fs/sysv/namei.c
index 7f0e4b53085e..b8a73f716fbe 100644
--- a/fs/sysv/namei.c
+++ b/fs/sysv/namei.c
@@ -16,18 +16,6 @@
#include <linux/smp_lock.h>
#include "sysv.h"
-static inline void inc_count(struct inode *inode)
-{
- inode->i_nlink++;
- mark_inode_dirty(inode);
-}
-
-static inline void dec_count(struct inode *inode)
-{
- inode->i_nlink--;
- mark_inode_dirty(inode);
-}
-
static int add_nondir(struct dentry *dentry, struct inode *inode)
{
int err = sysv_add_link(dentry, inode);
@@ -35,7 +23,7 @@ static int add_nondir(struct dentry *dentry, struct inode *inode)
d_instantiate(dentry, inode);
return 0;
}
- dec_count(inode);
+ inode_dec_link_count(inode);
iput(inode);
return err;
}
@@ -124,7 +112,7 @@ out:
return err;
out_fail:
- dec_count(inode);
+ inode_dec_link_count(inode);
iput(inode);
goto out;
}
@@ -138,7 +126,7 @@ static int sysv_link(struct dentry * old_dentry, struct inode * dir,
return -EMLINK;
inode->i_ctime = CURRENT_TIME_SEC;
- inc_count(inode);
+ inode_inc_link_count(inode);
atomic_inc(&inode->i_count);
return add_nondir(dentry, inode);
@@ -151,7 +139,7 @@ static int sysv_mkdir(struct inode * dir, struct dentry *dentry, int mode)
if (dir->i_nlink >= SYSV_SB(dir->i_sb)->s_link_max)
goto out;
- inc_count(dir);
+ inode_inc_link_count(dir);
inode = sysv_new_inode(dir, S_IFDIR|mode);
err = PTR_ERR(inode);
@@ -160,7 +148,7 @@ static int sysv_mkdir(struct inode * dir, struct dentry *dentry, int mode)
sysv_set_inode(inode, 0);
- inc_count(inode);
+ inode_inc_link_count(inode);
err = sysv_make_empty(inode, dir);
if (err)
@@ -175,11 +163,11 @@ out:
return err;
out_fail:
- dec_count(inode);
- dec_count(inode);
+ inode_dec_link_count(inode);
+ inode_dec_link_count(inode);
iput(inode);
out_dir:
- dec_count(dir);
+ inode_dec_link_count(dir);
goto out;
}
@@ -199,7 +187,7 @@ static int sysv_unlink(struct inode * dir, struct dentry * dentry)
goto out;
inode->i_ctime = dir->i_ctime;
- dec_count(inode);
+ inode_dec_link_count(inode);
out:
return err;
}
@@ -213,8 +201,8 @@ static int sysv_rmdir(struct inode * dir, struct dentry * dentry)
err = sysv_unlink(dir, dentry);
if (!err) {
inode->i_size = 0;
- dec_count(inode);
- dec_count(dir);
+ inode_dec_link_count(inode);
+ inode_dec_link_count(dir);
}
}
return err;
@@ -258,34 +246,34 @@ static int sysv_rename(struct inode * old_dir, struct dentry * old_dentry,
new_de = sysv_find_entry(new_dentry, &new_page);
if (!new_de)
goto out_dir;
- inc_count(old_inode);
+ inode_inc_link_count(old_inode);
sysv_set_link(new_de, new_page, old_inode);
new_inode->i_ctime = CURRENT_TIME_SEC;
if (dir_de)
new_inode->i_nlink--;
- dec_count(new_inode);
+ inode_dec_link_count(new_inode);
} else {
if (dir_de) {
err = -EMLINK;
if (new_dir->i_nlink >= SYSV_SB(new_dir->i_sb)->s_link_max)
goto out_dir;
}
- inc_count(old_inode);
+ inode_inc_link_count(old_inode);
err = sysv_add_link(new_dentry, old_inode);
if (err) {
- dec_count(old_inode);
+ inode_dec_link_count(old_inode);
goto out_dir;
}
if (dir_de)
- inc_count(new_dir);
+ inode_inc_link_count(new_dir);
}
sysv_delete_entry(old_de, old_page);
- dec_count(old_inode);
+ inode_dec_link_count(old_inode);
if (dir_de) {
sysv_set_link(dir_de, dir_page, new_dir);
- dec_count(old_dir);
+ inode_dec_link_count(old_dir);
}
return 0;
diff --git a/fs/udf/balloc.c b/fs/udf/balloc.c
index 201049ac8a96..ea521f846d97 100644
--- a/fs/udf/balloc.c
+++ b/fs/udf/balloc.c
@@ -152,7 +152,7 @@ static void udf_bitmap_free_blocks(struct super_block * sb,
int bitmap_nr;
unsigned long overflow;
- down(&sbi->s_alloc_sem);
+ mutex_lock(&sbi->s_alloc_mutex);
if (bloc.logicalBlockNum < 0 ||
(bloc.logicalBlockNum + count) > UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum))
{
@@ -211,7 +211,7 @@ error_return:
sb->s_dirt = 1;
if (UDF_SB_LVIDBH(sb))
mark_buffer_dirty(UDF_SB_LVIDBH(sb));
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
return;
}
@@ -226,7 +226,7 @@ static int udf_bitmap_prealloc_blocks(struct super_block * sb,
int nr_groups, bitmap_nr;
struct buffer_head *bh;
- down(&sbi->s_alloc_sem);
+ mutex_lock(&sbi->s_alloc_mutex);
if (first_block < 0 || first_block >= UDF_SB_PARTLEN(sb, partition))
goto out;
@@ -275,7 +275,7 @@ out:
mark_buffer_dirty(UDF_SB_LVIDBH(sb));
}
sb->s_dirt = 1;
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
return alloc_count;
}
@@ -291,7 +291,7 @@ static int udf_bitmap_new_block(struct super_block * sb,
int newblock = 0;
*err = -ENOSPC;
- down(&sbi->s_alloc_sem);
+ mutex_lock(&sbi->s_alloc_mutex);
repeat:
if (goal < 0 || goal >= UDF_SB_PARTLEN(sb, partition))
@@ -364,7 +364,7 @@ repeat:
}
if (i >= (nr_groups*2))
{
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
return newblock;
}
if (bit < sb->s_blocksize << 3)
@@ -373,7 +373,7 @@ repeat:
bit = udf_find_next_one_bit(bh->b_data, sb->s_blocksize << 3, group_start << 3);
if (bit >= sb->s_blocksize << 3)
{
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
return 0;
}
@@ -387,7 +387,7 @@ got_block:
*/
if (inode && DQUOT_ALLOC_BLOCK(inode, 1))
{
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
*err = -EDQUOT;
return 0;
}
@@ -410,13 +410,13 @@ got_block:
mark_buffer_dirty(UDF_SB_LVIDBH(sb));
}
sb->s_dirt = 1;
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
*err = 0;
return newblock;
error_return:
*err = -EIO;
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
return 0;
}
@@ -433,7 +433,7 @@ static void udf_table_free_blocks(struct super_block * sb,
int8_t etype;
int i;
- down(&sbi->s_alloc_sem);
+ mutex_lock(&sbi->s_alloc_mutex);
if (bloc.logicalBlockNum < 0 ||
(bloc.logicalBlockNum + count) > UDF_SB_PARTLEN(sb, bloc.partitionReferenceNum))
{
@@ -666,7 +666,7 @@ static void udf_table_free_blocks(struct super_block * sb,
error_return:
sb->s_dirt = 1;
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
return;
}
@@ -692,7 +692,7 @@ static int udf_table_prealloc_blocks(struct super_block * sb,
else
return 0;
- down(&sbi->s_alloc_sem);
+ mutex_lock(&sbi->s_alloc_mutex);
extoffset = sizeof(struct unallocSpaceEntry);
bloc = UDF_I_LOCATION(table);
@@ -736,7 +736,7 @@ static int udf_table_prealloc_blocks(struct super_block * sb,
mark_buffer_dirty(UDF_SB_LVIDBH(sb));
sb->s_dirt = 1;
}
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
return alloc_count;
}
@@ -761,7 +761,7 @@ static int udf_table_new_block(struct super_block * sb,
else
return newblock;
- down(&sbi->s_alloc_sem);
+ mutex_lock(&sbi->s_alloc_mutex);
if (goal < 0 || goal >= UDF_SB_PARTLEN(sb, partition))
goal = 0;
@@ -811,7 +811,7 @@ static int udf_table_new_block(struct super_block * sb,
if (spread == 0xFFFFFFFF)
{
udf_release_data(goal_bh);
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
return 0;
}
@@ -827,7 +827,7 @@ static int udf_table_new_block(struct super_block * sb,
if (inode && DQUOT_ALLOC_BLOCK(inode, 1))
{
udf_release_data(goal_bh);
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
*err = -EDQUOT;
return 0;
}
@@ -846,7 +846,7 @@ static int udf_table_new_block(struct super_block * sb,
}
sb->s_dirt = 1;
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
*err = 0;
return newblock;
}
diff --git a/fs/udf/ialloc.c b/fs/udf/ialloc.c
index c9b707b470ca..3873c672cb4c 100644
--- a/fs/udf/ialloc.c
+++ b/fs/udf/ialloc.c
@@ -42,7 +42,7 @@ void udf_free_inode(struct inode * inode)
clear_inode(inode);
- down(&sbi->s_alloc_sem);
+ mutex_lock(&sbi->s_alloc_mutex);
if (sbi->s_lvidbh) {
if (S_ISDIR(inode->i_mode))
UDF_SB_LVIDIU(sb)->numDirs =
@@ -53,7 +53,7 @@ void udf_free_inode(struct inode * inode)
mark_buffer_dirty(sbi->s_lvidbh);
}
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
udf_free_blocks(sb, NULL, UDF_I_LOCATION(inode), 0, 1);
}
@@ -83,7 +83,7 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err)
return NULL;
}
- down(&sbi->s_alloc_sem);
+ mutex_lock(&sbi->s_alloc_mutex);
UDF_I_UNIQUE(inode) = 0;
UDF_I_LENEXTENTS(inode) = 0;
UDF_I_NEXT_ALLOC_BLOCK(inode) = 0;
@@ -148,7 +148,7 @@ struct inode * udf_new_inode (struct inode *dir, int mode, int * err)
UDF_I_CRTIME(inode) = current_fs_time(inode->i_sb);
insert_inode_hash(inode);
mark_inode_dirty(inode);
- up(&sbi->s_alloc_sem);
+ mutex_unlock(&sbi->s_alloc_mutex);
if (DQUOT_ALLOC_INODE(inode))
{
diff --git a/fs/udf/super.c b/fs/udf/super.c
index 368d8f81fe54..9303c50c5d55 100644
--- a/fs/udf/super.c
+++ b/fs/udf/super.c
@@ -1515,7 +1515,7 @@ static int udf_fill_super(struct super_block *sb, void *options, int silent)
sb->s_fs_info = sbi;
memset(UDF_SB(sb), 0x00, sizeof(struct udf_sb_info));
- init_MUTEX(&sbi->s_alloc_sem);
+ mutex_init(&sbi->s_alloc_mutex);
if (!udf_parse_options((char *)options, &uopt))
goto error_out;
diff --git a/fs/ufs/file.c b/fs/ufs/file.c
index ed69d7fe1b5d..62ad481810ef 100644
--- a/fs/ufs/file.c
+++ b/fs/ufs/file.c
@@ -23,18 +23,8 @@
* ext2 fs regular file handling primitives
*/
-#include <asm/uaccess.h>
-#include <asm/system.h>
-
-#include <linux/errno.h>
#include <linux/fs.h>
#include <linux/ufs_fs.h>
-#include <linux/fcntl.h>
-#include <linux/time.h>
-#include <linux/stat.h>
-#include <linux/mm.h>
-#include <linux/pagemap.h>
-#include <linux/smp_lock.h>
/*
* We have mostly NULL's here: the current defaults are ok for
diff --git a/fs/ufs/namei.c b/fs/ufs/namei.c
index 2958cde7d3d6..8d5f98a01c74 100644
--- a/fs/ufs/namei.c
+++ b/fs/ufs/namei.c
@@ -43,18 +43,6 @@
#define UFSD(x)
#endif
-static inline void ufs_inc_count(struct inode *inode)
-{
- inode->i_nlink++;
- mark_inode_dirty(inode);
-}
-
-static inline void ufs_dec_count(struct inode *inode)
-{
- inode->i_nlink--;
- mark_inode_dirty(inode);
-}
-
static inline int ufs_add_nondir(struct dentry *dentry, struct inode *inode)
{
int err = ufs_add_link(dentry, inode);
@@ -62,7 +50,7 @@ static inline int ufs_add_nondir(struct dentry *dentry, struct inode *inode)
d_instantiate(dentry, inode);
return 0;
}
- ufs_dec_count(inode);
+ inode_dec_link_count(inode);
iput(inode);
return err;
}
@@ -173,7 +161,7 @@ out:
return err;
out_fail:
- ufs_dec_count(inode);
+ inode_dec_link_count(inode);
iput(inode);
goto out;
}
@@ -191,7 +179,7 @@ static int ufs_link (struct dentry * old_dentry, struct inode * dir,
}
inode->i_ctime = CURRENT_TIME_SEC;
- ufs_inc_count(inode);
+ inode_inc_link_count(inode);
atomic_inc(&inode->i_count);
error = ufs_add_nondir(dentry, inode);
@@ -208,7 +196,7 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, int mode)
goto out;
lock_kernel();
- ufs_inc_count(dir);
+ inode_inc_link_count(dir);
inode = ufs_new_inode(dir, S_IFDIR|mode);
err = PTR_ERR(inode);
@@ -218,7 +206,7 @@ static int ufs_mkdir(struct inode * dir, struct dentry * dentry, int mode)
inode->i_op = &ufs_dir_inode_operations;
inode->i_fop = &ufs_dir_operations;
- ufs_inc_count(inode);
+ inode_inc_link_count(inode);
err = ufs_make_empty(inode, dir);
if (err)
@@ -234,11 +222,11 @@ out:
return err;
out_fail:
- ufs_dec_count(inode);
- ufs_dec_count(inode);
+ inode_dec_link_count(inode);
+ inode_dec_link_count(inode);
iput (inode);
out_dir:
- ufs_dec_count(dir);
+ inode_dec_link_count(dir);
unlock_kernel();
goto out;
}
@@ -260,7 +248,7 @@ static int ufs_unlink(struct inode * dir, struct dentry *dentry)
goto out;
inode->i_ctime = dir->i_ctime;
- ufs_dec_count(inode);
+ inode_dec_link_count(inode);
err = 0;
out:
unlock_kernel();
@@ -277,8 +265,8 @@ static int ufs_rmdir (struct inode * dir, struct dentry *dentry)
err = ufs_unlink(dir, dentry);
if (!err) {
inode->i_size = 0;
- ufs_dec_count(inode);
- ufs_dec_count(dir);
+ inode_dec_link_count(inode);
+ inode_dec_link_count(dir);
}
}
unlock_kernel();
@@ -319,35 +307,35 @@ static int ufs_rename (struct inode * old_dir, struct dentry * old_dentry,
new_de = ufs_find_entry (new_dentry, &new_bh);
if (!new_de)
goto out_dir;
- ufs_inc_count(old_inode);
+ inode_inc_link_count(old_inode);
ufs_set_link(new_dir, new_de, new_bh, old_inode);
new_inode->i_ctime = CURRENT_TIME_SEC;
if (dir_de)
new_inode->i_nlink--;
- ufs_dec_count(new_inode);
+ inode_dec_link_count(new_inode);
} else {
if (dir_de) {
err = -EMLINK;
if (new_dir->i_nlink >= UFS_LINK_MAX)
goto out_dir;
}
- ufs_inc_count(old_inode);
+ inode_inc_link_count(old_inode);
err = ufs_add_link(new_dentry, old_inode);
if (err) {
- ufs_dec_count(old_inode);
+ inode_dec_link_count(old_inode);
goto out_dir;
}
if (dir_de)
- ufs_inc_count(new_dir);
+ inode_inc_link_count(new_dir);
}
ufs_delete_entry (old_dir, old_de, old_bh);
- ufs_dec_count(old_inode);
+ inode_dec_link_count(old_inode);
if (dir_de) {
ufs_set_link(old_inode, dir_de, dir_bh, new_dir);
- ufs_dec_count(old_dir);
+ inode_dec_link_count(old_dir);
}
unlock_kernel();
return 0;
diff --git a/fs/xfs/Makefile-linux-2.6 b/fs/xfs/Makefile-linux-2.6
index 97bd4743b461..5d73eaa1971f 100644
--- a/fs/xfs/Makefile-linux-2.6
+++ b/fs/xfs/Makefile-linux-2.6
@@ -1,33 +1,19 @@
#
-# Copyright (c) 2000-2005 Silicon Graphics, Inc. All Rights Reserved.
+# Copyright (c) 2000-2005 Silicon Graphics, Inc.
+# All Rights Reserved.
#
-# This program is free software; you can redistribute it and/or modify it
-# under the terms of version 2 of the GNU General Public License as
+# This program is free software; you can redistribute it and/or
+# modify it under the terms of the GNU General Public License as
# published by the Free Software Foundation.
#
-# This program is distributed in the hope that it would be useful, but
-# WITHOUT ANY WARRANTY; without even the implied warranty of
-# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
+# This program is distributed in the hope that it would be useful,
+# but WITHOUT ANY WARRANTY; without even the implied warranty of
+# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+# GNU General Public License for more details.
#
-# Further, this software is distributed without any warranty that it is
-# free of the rightful claim of any third person regarding infringement
-# or the like. Any license provided herein, whether implied or
-# otherwise, applies only to this software file. Patent licenses, if
-# any, provided herein do not apply to combinations of this program with
-# other software, or any other product whatsoever.
-#
-# You should have received a copy of the GNU General Public License along
-# with this program; if not, write the Free Software Foundation, Inc., 59
-# Temple Place - Suite 330, Boston MA 02111-1307, USA.
-#
-# Contact information: Silicon Graphics, Inc., 1600 Amphitheatre Pkwy,
-# Mountain View, CA 94043, or:
-#
-# http://www.sgi.com
-#
-# For further information regarding this notice, see:
-#
-# http://oss.sgi.com/projects/GenInfo/SGIGPLNoticeExplan/
+# You should have received a copy of the GNU General Public License
+# along with this program; if not, write the Free Software Foundation,
+# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA
#
EXTRA_CFLAGS += -Ifs/xfs -Ifs/xfs/linux-2.6 -funsigned-char
@@ -36,7 +22,7 @@ XFS_LINUX := linux-2.6
ifeq ($(CONFIG_XFS_DEBUG),y)
EXTRA_CFLAGS += -g -DSTATIC="" -DDEBUG
- EXTRA_CFLAGS += -DPAGEBUF_LOCK_TRACKING
+ EXTRA_CFLAGS += -DXFS_BUF_LOCK_TRACKING
endif
ifeq ($(CONFIG_XFS_TRACE),y)
EXTRA_CFLAGS += -DXFS_ALLOC_TRACE
@@ -50,7 +36,7 @@ ifeq ($(CONFIG_XFS_TRACE),y)
EXTRA_CFLAGS += -DXFS_ILOCK_TRACE
EXTRA_CFLAGS += -DXFS_LOG_TRACE
EXTRA_CFLAGS += -DXFS_RW_TRACE
- EXTRA_CFLAGS += -DPAGEBUF_TRACE
+ EXTRA_CFLAGS += -DXFS_BUF_TRACE
EXTRA_CFLAGS += -DXFS_VNODE_TRACE
endif
diff --git a/fs/xfs/linux-2.6/kmem.h b/fs/xfs/linux-2.6/kmem.h
index c64a29cdfff3..f0268a84e6fd 100644
--- a/fs/xfs/linux-2.6/kmem.h
+++ b/fs/xfs/linux-2.6/kmem.h
@@ -23,17 +23,8 @@
#include <linux/mm.h>
/*
- * memory management routines
+ * Process flags handling
*/
-#define KM_SLEEP 0x0001u
-#define KM_NOSLEEP 0x0002u
-#define KM_NOFS 0x0004u
-#define KM_MAYFAIL 0x0008u
-
-#define kmem_zone kmem_cache
-#define kmem_zone_t struct kmem_cache
-
-typedef unsigned long xfs_pflags_t;
#define PFLAGS_TEST_NOIO() (current->flags & PF_NOIO)
#define PFLAGS_TEST_FSTRANS() (current->flags & PF_FSTRANS)
@@ -67,74 +58,102 @@ typedef unsigned long xfs_pflags_t;
*(NSTATEP) = *(OSTATEP); \
} while (0)
-static __inline gfp_t kmem_flags_convert(unsigned int __nocast flags)
+/*
+ * General memory allocation interfaces
+ */
+
+#define KM_SLEEP 0x0001u
+#define KM_NOSLEEP 0x0002u
+#define KM_NOFS 0x0004u
+#define KM_MAYFAIL 0x0008u
+
+/*
+ * We use a special process flag to avoid recursive callbacks into
+ * the filesystem during transactions. We will also issue our own
+ * warnings, so we explicitly skip any generic ones (silly of us).
+ */
+static inline gfp_t
+kmem_flags_convert(unsigned int __nocast flags)
{
- gfp_t lflags = __GFP_NOWARN; /* we'll report problems, if need be */
+ gfp_t lflags;
-#ifdef DEBUG
- if (unlikely(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL))) {
- printk(KERN_WARNING
- "XFS: memory allocation with wrong flags (%x)\n", flags);
- BUG();
- }
-#endif
+ BUG_ON(flags & ~(KM_SLEEP|KM_NOSLEEP|KM_NOFS|KM_MAYFAIL));
if (flags & KM_NOSLEEP) {
- lflags |= GFP_ATOMIC;
+ lflags = GFP_ATOMIC | __GFP_NOWARN;
} else {
- lflags |= GFP_KERNEL;
-
- /* avoid recusive callbacks to filesystem during transactions */
+ lflags = GFP_KERNEL | __GFP_NOWARN;
if (PFLAGS_TEST_FSTRANS() || (flags & KM_NOFS))
lflags &= ~__GFP_FS;
}
-
- return lflags;
+ return lflags;
}
-static __inline kmem_zone_t *
+extern void *kmem_alloc(size_t, unsigned int __nocast);
+extern void *kmem_realloc(void *, size_t, size_t, unsigned int __nocast);
+extern void *kmem_zalloc(size_t, unsigned int __nocast);
+extern void kmem_free(void *, size_t);
+
+/*
+ * Zone interfaces
+ */
+
+#define KM_ZONE_HWALIGN SLAB_HWCACHE_ALIGN
+#define KM_ZONE_RECLAIM SLAB_RECLAIM_ACCOUNT
+#define KM_ZONE_SPREAD 0
+
+#define kmem_zone kmem_cache
+#define kmem_zone_t struct kmem_cache
+
+static inline kmem_zone_t *
kmem_zone_init(int size, char *zone_name)
{
return kmem_cache_create(zone_name, size, 0, 0, NULL, NULL);
}
-static __inline void
+static inline kmem_zone_t *
+kmem_zone_init_flags(int size, char *zone_name, unsigned long flags,
+ void (*construct)(void *, kmem_zone_t *, unsigned long))
+{
+ return kmem_cache_create(zone_name, size, 0, flags, construct, NULL);
+}
+
+static inline void
kmem_zone_free(kmem_zone_t *zone, void *ptr)
{
kmem_cache_free(zone, ptr);
}
-static __inline void
+static inline void
kmem_zone_destroy(kmem_zone_t *zone)
{
if (zone && kmem_cache_destroy(zone))
BUG();
}
-extern void *kmem_zone_zalloc(kmem_zone_t *, unsigned int __nocast);
extern void *kmem_zone_alloc(kmem_zone_t *, unsigned int __nocast);
+extern void *kmem_zone_zalloc(kmem_zone_t *, unsigned int __nocast);
-extern void *kmem_alloc(size_t, unsigned int __nocast);
-extern void *kmem_realloc(void *, size_t, size_t, unsigned int __nocast);
-extern void *kmem_zalloc(size_t, unsigned int __nocast);
-extern void kmem_free(void *, size_t);
+/*
+ * Low memory cache shrinkers
+ */
typedef struct shrinker *kmem_shaker_t;
typedef int (*kmem_shake_func_t)(int, gfp_t);
-static __inline kmem_shaker_t
+static inline kmem_shaker_t
kmem_shake_register(kmem_shake_func_t sfunc)
{
return set_shrinker(DEFAULT_SEEKS, sfunc);
}
-static __inline void
+static inline void
kmem_shake_deregister(kmem_shaker_t shrinker)
{
remove_shrinker(shrinker);
}
-static __inline int
+static inline int
kmem_shake_allow(gfp_t gfp_mask)
{
return (gfp_mask & __GFP_WAIT);
diff --git a/fs/xfs/linux-2.6/xfs_aops.c b/fs/xfs/linux-2.6/xfs_aops.c
index 74d8be87f983..97fc056130eb 100644
--- a/fs/xfs/linux-2.6/xfs_aops.c
+++ b/fs/xfs/linux-2.6/xfs_aops.c
@@ -43,7 +43,29 @@
#include <linux/pagevec.h>
#include <linux/writeback.h>
-STATIC void xfs_count_page_state(struct page *, int *, int *, int *);
+STATIC void
+xfs_count_page_state(
+ struct page *page,
+ int *delalloc,
+ int *unmapped,
+ int *unwritten)
+{
+ struct buffer_head *bh, *head;
+
+ *delalloc = *unmapped = *unwritten = 0;
+
+ bh = head = page_buffers(page);
+ do {
+ if (buffer_uptodate(bh) && !buffer_mapped(bh))
+ (*unmapped) = 1;
+ else if (buffer_unwritten(bh) && !buffer_delay(bh))
+ clear_buffer_unwritten(bh);
+ else if (buffer_unwritten(bh))
+ (*unwritten) = 1;
+ else if (buffer_delay(bh))
+ (*delalloc) = 1;
+ } while ((bh = bh->b_this_page) != head);
+}
#if defined(XFS_RW_TRACE)
void
@@ -54,7 +76,7 @@ xfs_page_trace(
int mask)
{
xfs_inode_t *ip;
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
loff_t isize = i_size_read(inode);
loff_t offset = page_offset(page);
int delalloc = -1, unmapped = -1, unwritten = -1;
@@ -81,7 +103,7 @@ xfs_page_trace(
(void *)((unsigned long)delalloc),
(void *)((unsigned long)unmapped),
(void *)((unsigned long)unwritten),
- (void *)NULL,
+ (void *)((unsigned long)current_pid()),
(void *)NULL);
}
#else
@@ -192,7 +214,7 @@ xfs_alloc_ioend(
ioend->io_uptodate = 1; /* cleared if any I/O fails */
ioend->io_list = NULL;
ioend->io_type = type;
- ioend->io_vnode = LINVFS_GET_VP(inode);
+ ioend->io_vnode = vn_from_inode(inode);
ioend->io_buffer_head = NULL;
ioend->io_buffer_tail = NULL;
atomic_inc(&ioend->io_vnode->v_iocount);
@@ -217,7 +239,7 @@ xfs_map_blocks(
xfs_iomap_t *mapp,
int flags)
{
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
int error, nmaps = 1;
VOP_BMAP(vp, offset, count, flags, mapp, &nmaps, error);
@@ -462,28 +484,37 @@ xfs_add_to_ioend(
}
STATIC void
+xfs_map_buffer(
+ struct buffer_head *bh,
+ xfs_iomap_t *mp,
+ xfs_off_t offset,
+ uint block_bits)
+{
+ sector_t bn;
+
+ ASSERT(mp->iomap_bn != IOMAP_DADDR_NULL);
+
+ bn = (mp->iomap_bn >> (block_bits - BBSHIFT)) +
+ ((offset - mp->iomap_offset) >> block_bits);
+
+ ASSERT(bn || (mp->iomap_flags & IOMAP_REALTIME));
+
+ bh->b_blocknr = bn;
+ set_buffer_mapped(bh);
+}
+
+STATIC void
xfs_map_at_offset(
struct buffer_head *bh,
loff_t offset,
int block_bits,
xfs_iomap_t *iomapp)
{
- xfs_daddr_t bn;
- int sector_shift;
-
ASSERT(!(iomapp->iomap_flags & IOMAP_HOLE));
ASSERT(!(iomapp->iomap_flags & IOMAP_DELAY));
- ASSERT(iomapp->iomap_bn != IOMAP_DADDR_NULL);
-
- sector_shift = block_bits - BBSHIFT;
- bn = (iomapp->iomap_bn >> sector_shift) +
- ((offset - iomapp->iomap_offset) >> block_bits);
-
- ASSERT(bn || (iomapp->iomap_flags & IOMAP_REALTIME));
- ASSERT((bn << sector_shift) >= iomapp->iomap_bn);
lock_buffer(bh);
- bh->b_blocknr = bn;
+ xfs_map_buffer(bh, iomapp, offset, block_bits);
bh->b_bdev = iomapp->iomap_target->bt_bdev;
set_buffer_mapped(bh);
clear_buffer_delay(bh);
@@ -616,7 +647,7 @@ xfs_is_delayed_page(
acceptable = (type == IOMAP_UNWRITTEN);
else if (buffer_delay(bh))
acceptable = (type == IOMAP_DELAY);
- else if (buffer_mapped(bh))
+ else if (buffer_dirty(bh) && buffer_mapped(bh))
acceptable = (type == 0);
else
break;
@@ -1040,8 +1071,159 @@ error:
return err;
}
+/*
+ * writepage: Called from one of two places:
+ *
+ * 1. we are flushing a delalloc buffer head.
+ *
+ * 2. we are writing out a dirty page. Typically the page dirty
+ * state is cleared before we get here. In this case is it
+ * conceivable we have no buffer heads.
+ *
+ * For delalloc space on the page we need to allocate space and
+ * flush it. For unmapped buffer heads on the page we should
+ * allocate space if the page is uptodate. For any other dirty
+ * buffer heads on the page we should flush them.
+ *
+ * If we detect that a transaction would be required to flush
+ * the page, we have to check the process flags first, if we
+ * are already in a transaction or disk I/O during allocations
+ * is off, we need to fail the writepage and redirty the page.
+ */
+
+STATIC int
+xfs_vm_writepage(
+ struct page *page,
+ struct writeback_control *wbc)
+{
+ int error;
+ int need_trans;
+ int delalloc, unmapped, unwritten;
+ struct inode *inode = page->mapping->host;
+
+ xfs_page_trace(XFS_WRITEPAGE_ENTER, inode, page, 0);
+
+ /*
+ * We need a transaction if:
+ * 1. There are delalloc buffers on the page
+ * 2. The page is uptodate and we have unmapped buffers
+ * 3. The page is uptodate and we have no buffers
+ * 4. There are unwritten buffers on the page
+ */
+
+ if (!page_has_buffers(page)) {
+ unmapped = 1;
+ need_trans = 1;
+ } else {
+ xfs_count_page_state(page, &delalloc, &unmapped, &unwritten);
+ if (!PageUptodate(page))
+ unmapped = 0;
+ need_trans = delalloc + unmapped + unwritten;
+ }
+
+ /*
+ * If we need a transaction and the process flags say
+ * we are already in a transaction, or no IO is allowed
+ * then mark the page dirty again and leave the page
+ * as is.
+ */
+ if (PFLAGS_TEST_FSTRANS() && need_trans)
+ goto out_fail;
+
+ /*
+ * Delay hooking up buffer heads until we have
+ * made our go/no-go decision.
+ */
+ if (!page_has_buffers(page))
+ create_empty_buffers(page, 1 << inode->i_blkbits, 0);
+
+ /*
+ * Convert delayed allocate, unwritten or unmapped space
+ * to real space and flush out to disk.
+ */
+ error = xfs_page_state_convert(inode, page, wbc, 1, unmapped);
+ if (error == -EAGAIN)
+ goto out_fail;
+ if (unlikely(error < 0))
+ goto out_unlock;
+
+ return 0;
+
+out_fail:
+ redirty_page_for_writepage(wbc, page);
+ unlock_page(page);
+ return 0;
+out_unlock:
+ unlock_page(page);
+ return error;
+}
+
+/*
+ * Called to move a page into cleanable state - and from there
+ * to be released. Possibly the page is already clean. We always
+ * have buffer heads in this call.
+ *
+ * Returns 0 if the page is ok to release, 1 otherwise.
+ *
+ * Possible scenarios are:
+ *
+ * 1. We are being called to release a page which has been written
+ * to via regular I/O. buffer heads will be dirty and possibly
+ * delalloc. If no delalloc buffer heads in this case then we
+ * can just return zero.
+ *
+ * 2. We are called to release a page which has been written via
+ * mmap, all we need to do is ensure there is no delalloc
+ * state in the buffer heads, if not we can let the caller
+ * free them and we should come back later via writepage.
+ */
STATIC int
-__linvfs_get_block(
+xfs_vm_releasepage(
+ struct page *page,
+ gfp_t gfp_mask)
+{
+ struct inode *inode = page->mapping->host;
+ int dirty, delalloc, unmapped, unwritten;
+ struct writeback_control wbc = {
+ .sync_mode = WB_SYNC_ALL,
+ .nr_to_write = 1,
+ };
+
+ xfs_page_trace(XFS_RELEASEPAGE_ENTER, inode, page, gfp_mask);
+
+ if (!page_has_buffers(page))
+ return 0;
+
+ xfs_count_page_state(page, &delalloc, &unmapped, &unwritten);
+ if (!delalloc && !unwritten)
+ goto free_buffers;
+
+ if (!(gfp_mask & __GFP_FS))
+ return 0;
+
+ /* If we are already inside a transaction or the thread cannot
+ * do I/O, we cannot release this page.
+ */
+ if (PFLAGS_TEST_FSTRANS())
+ return 0;
+
+ /*
+ * Convert delalloc space to real space, do not flush the
+ * data out to disk, that will be done by the caller.
+ * Never need to allocate space here - we will always
+ * come back to writepage in that case.
+ */
+ dirty = xfs_page_state_convert(inode, page, &wbc, 0, 0);
+ if (dirty == 0 && !unwritten)
+ goto free_buffers;
+ return 0;
+
+free_buffers:
+ return try_to_free_buffers(page);
+}
+
+STATIC int
+__xfs_get_block(
struct inode *inode,
sector_t iblock,
unsigned long blocks,
@@ -1050,7 +1232,7 @@ __linvfs_get_block(
int direct,
bmapi_flags_t flags)
{
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
xfs_iomap_t iomap;
xfs_off_t offset;
ssize_t size;
@@ -1073,21 +1255,13 @@ __linvfs_get_block(
return 0;
if (iomap.iomap_bn != IOMAP_DADDR_NULL) {
- xfs_daddr_t bn;
- xfs_off_t delta;
-
- /* For unwritten extents do not report a disk address on
+ /*
+ * For unwritten extents do not report a disk address on
* the read case (treat as if we're reading into a hole).
*/
if (create || !(iomap.iomap_flags & IOMAP_UNWRITTEN)) {
- delta = offset - iomap.iomap_offset;
- delta >>= inode->i_blkbits;
-
- bn = iomap.iomap_bn >> (inode->i_blkbits - BBSHIFT);
- bn += delta;
- BUG_ON(!bn && !(iomap.iomap_flags & IOMAP_REALTIME));
- bh_result->b_blocknr = bn;
- set_buffer_mapped(bh_result);
+ xfs_map_buffer(bh_result, &iomap, offset,
+ inode->i_blkbits);
}
if (create && (iomap.iomap_flags & IOMAP_UNWRITTEN)) {
if (direct)
@@ -1130,30 +1304,30 @@ __linvfs_get_block(
}
int
-linvfs_get_block(
+xfs_get_block(
struct inode *inode,
sector_t iblock,
struct buffer_head *bh_result,
int create)
{
- return __linvfs_get_block(inode, iblock, 0, bh_result,
+ return __xfs_get_block(inode, iblock, 0, bh_result,
create, 0, BMAPI_WRITE);
}
STATIC int
-linvfs_get_blocks_direct(
+xfs_get_blocks_direct(
struct inode *inode,
sector_t iblock,
unsigned long max_blocks,
struct buffer_head *bh_result,
int create)
{
- return __linvfs_get_block(inode, iblock, max_blocks, bh_result,
+ return __xfs_get_block(inode, iblock, max_blocks, bh_result,
create, 1, BMAPI_WRITE|BMAPI_DIRECT);
}
STATIC void
-linvfs_end_io_direct(
+xfs_end_io_direct(
struct kiocb *iocb,
loff_t offset,
ssize_t size,
@@ -1191,7 +1365,7 @@ linvfs_end_io_direct(
}
STATIC ssize_t
-linvfs_direct_IO(
+xfs_vm_direct_IO(
int rw,
struct kiocb *iocb,
const struct iovec *iov,
@@ -1200,7 +1374,7 @@ linvfs_direct_IO(
{
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
xfs_iomap_t iomap;
int maps = 1;
int error;
@@ -1215,164 +1389,61 @@ linvfs_direct_IO(
ret = blockdev_direct_IO_own_locking(rw, iocb, inode,
iomap.iomap_target->bt_bdev,
iov, offset, nr_segs,
- linvfs_get_blocks_direct,
- linvfs_end_io_direct);
+ xfs_get_blocks_direct,
+ xfs_end_io_direct);
if (unlikely(ret <= 0 && iocb->private))
xfs_destroy_ioend(iocb->private);
return ret;
}
+STATIC int
+xfs_vm_prepare_write(
+ struct file *file,
+ struct page *page,
+ unsigned int from,
+ unsigned int to)
+{
+ return block_prepare_write(page, from, to, xfs_get_block);
+}
STATIC sector_t
-linvfs_bmap(
+xfs_vm_bmap(
struct address_space *mapping,
sector_t block)
{
struct inode *inode = (struct inode *)mapping->host;
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
int error;
- vn_trace_entry(vp, "linvfs_bmap", (inst_t *)__return_address);
+ vn_trace_entry(vp, __FUNCTION__, (inst_t *)__return_address);
VOP_RWLOCK(vp, VRWLOCK_READ);
VOP_FLUSH_PAGES(vp, (xfs_off_t)0, -1, 0, FI_REMAPF, error);
VOP_RWUNLOCK(vp, VRWLOCK_READ);
- return generic_block_bmap(mapping, block, linvfs_get_block);
+ return generic_block_bmap(mapping, block, xfs_get_block);
}
STATIC int
-linvfs_readpage(
+xfs_vm_readpage(
struct file *unused,
struct page *page)
{
- return mpage_readpage(page, linvfs_get_block);
+ return mpage_readpage(page, xfs_get_block);
}
STATIC int
-linvfs_readpages(
+xfs_vm_readpages(
struct file *unused,
struct address_space *mapping,
struct list_head *pages,
unsigned nr_pages)
{
- return mpage_readpages(mapping, pages, nr_pages, linvfs_get_block);
-}
-
-STATIC void
-xfs_count_page_state(
- struct page *page,
- int *delalloc,
- int *unmapped,
- int *unwritten)
-{
- struct buffer_head *bh, *head;
-
- *delalloc = *unmapped = *unwritten = 0;
-
- bh = head = page_buffers(page);
- do {
- if (buffer_uptodate(bh) && !buffer_mapped(bh))
- (*unmapped) = 1;
- else if (buffer_unwritten(bh) && !buffer_delay(bh))
- clear_buffer_unwritten(bh);
- else if (buffer_unwritten(bh))
- (*unwritten) = 1;
- else if (buffer_delay(bh))
- (*delalloc) = 1;
- } while ((bh = bh->b_this_page) != head);
+ return mpage_readpages(mapping, pages, nr_pages, xfs_get_block);
}
-
-/*
- * writepage: Called from one of two places:
- *
- * 1. we are flushing a delalloc buffer head.
- *
- * 2. we are writing out a dirty page. Typically the page dirty
- * state is cleared before we get here. In this case is it
- * conceivable we have no buffer heads.
- *
- * For delalloc space on the page we need to allocate space and
- * flush it. For unmapped buffer heads on the page we should
- * allocate space if the page is uptodate. For any other dirty
- * buffer heads on the page we should flush them.
- *
- * If we detect that a transaction would be required to flush
- * the page, we have to check the process flags first, if we
- * are already in a transaction or disk I/O during allocations
- * is off, we need to fail the writepage and redirty the page.
- */
-
STATIC int
-linvfs_writepage(
- struct page *page,
- struct writeback_control *wbc)
-{
- int error;
- int need_trans;
- int delalloc, unmapped, unwritten;
- struct inode *inode = page->mapping->host;
-
- xfs_page_trace(XFS_WRITEPAGE_ENTER, inode, page, 0);
-
- /*
- * We need a transaction if:
- * 1. There are delalloc buffers on the page
- * 2. The page is uptodate and we have unmapped buffers
- * 3. The page is uptodate and we have no buffers
- * 4. There are unwritten buffers on the page
- */
-
- if (!page_has_buffers(page)) {
- unmapped = 1;
- need_trans = 1;
- } else {
- xfs_count_page_state(page, &delalloc, &unmapped, &unwritten);
- if (!PageUptodate(page))
- unmapped = 0;
- need_trans = delalloc + unmapped + unwritten;
- }
-
- /*
- * If we need a transaction and the process flags say
- * we are already in a transaction, or no IO is allowed
- * then mark the page dirty again and leave the page
- * as is.
- */
- if (PFLAGS_TEST_FSTRANS() && need_trans)
- goto out_fail;
-
- /*
- * Delay hooking up buffer heads until we have
- * made our go/no-go decision.
- */
- if (!page_has_buffers(page))
- create_empty_buffers(page, 1 << inode->i_blkbits, 0);
-
- /*
- * Convert delayed allocate, unwritten or unmapped space
- * to real space and flush out to disk.
- */
- error = xfs_page_state_convert(inode, page, wbc, 1, unmapped);
- if (error == -EAGAIN)
- goto out_fail;
- if (unlikely(error < 0))
- goto out_unlock;
-
- return 0;
-
-out_fail:
- redirty_page_for_writepage(wbc, page);
- unlock_page(page);
- return 0;
-out_unlock:
- unlock_page(page);
- return error;
-}
-
-STATIC int
-linvfs_invalidate_page(
+xfs_vm_invalidatepage(
struct page *page,
unsigned long offset)
{
@@ -1381,87 +1452,16 @@ linvfs_invalidate_page(
return block_invalidatepage(page, offset);
}
-/*
- * Called to move a page into cleanable state - and from there
- * to be released. Possibly the page is already clean. We always
- * have buffer heads in this call.
- *
- * Returns 0 if the page is ok to release, 1 otherwise.
- *
- * Possible scenarios are:
- *
- * 1. We are being called to release a page which has been written
- * to via regular I/O. buffer heads will be dirty and possibly
- * delalloc. If no delalloc buffer heads in this case then we
- * can just return zero.
- *
- * 2. We are called to release a page which has been written via
- * mmap, all we need to do is ensure there is no delalloc
- * state in the buffer heads, if not we can let the caller
- * free them and we should come back later via writepage.
- */
-STATIC int
-linvfs_release_page(
- struct page *page,
- gfp_t gfp_mask)
-{
- struct inode *inode = page->mapping->host;
- int dirty, delalloc, unmapped, unwritten;
- struct writeback_control wbc = {
- .sync_mode = WB_SYNC_ALL,
- .nr_to_write = 1,
- };
-
- xfs_page_trace(XFS_RELEASEPAGE_ENTER, inode, page, gfp_mask);
-
- xfs_count_page_state(page, &delalloc, &unmapped, &unwritten);
- if (!delalloc && !unwritten)
- goto free_buffers;
-
- if (!(gfp_mask & __GFP_FS))
- return 0;
-
- /* If we are already inside a transaction or the thread cannot
- * do I/O, we cannot release this page.
- */
- if (PFLAGS_TEST_FSTRANS())
- return 0;
-
- /*
- * Convert delalloc space to real space, do not flush the
- * data out to disk, that will be done by the caller.
- * Never need to allocate space here - we will always
- * come back to writepage in that case.
- */
- dirty = xfs_page_state_convert(inode, page, &wbc, 0, 0);
- if (dirty == 0 && !unwritten)
- goto free_buffers;
- return 0;
-
-free_buffers:
- return try_to_free_buffers(page);
-}
-
-STATIC int
-linvfs_prepare_write(
- struct file *file,
- struct page *page,
- unsigned int from,
- unsigned int to)
-{
- return block_prepare_write(page, from, to, linvfs_get_block);
-}
-
-struct address_space_operations linvfs_aops = {
- .readpage = linvfs_readpage,
- .readpages = linvfs_readpages,
- .writepage = linvfs_writepage,
+struct address_space_operations xfs_address_space_operations = {
+ .readpage = xfs_vm_readpage,
+ .readpages = xfs_vm_readpages,
+ .writepage = xfs_vm_writepage,
.sync_page = block_sync_page,
- .releasepage = linvfs_release_page,
- .invalidatepage = linvfs_invalidate_page,
- .prepare_write = linvfs_prepare_write,
+ .releasepage = xfs_vm_releasepage,
+ .invalidatepage = xfs_vm_invalidatepage,
+ .prepare_write = xfs_vm_prepare_write,
.commit_write = generic_commit_write,
- .bmap = linvfs_bmap,
- .direct_IO = linvfs_direct_IO,
+ .bmap = xfs_vm_bmap,
+ .direct_IO = xfs_vm_direct_IO,
.migratepage = buffer_migrate_page,
};
diff --git a/fs/xfs/linux-2.6/xfs_aops.h b/fs/xfs/linux-2.6/xfs_aops.h
index 55339dd5a30d..795699f121d2 100644
--- a/fs/xfs/linux-2.6/xfs_aops.h
+++ b/fs/xfs/linux-2.6/xfs_aops.h
@@ -40,7 +40,7 @@ typedef struct xfs_ioend {
struct work_struct io_work; /* xfsdatad work queue */
} xfs_ioend_t;
-extern struct address_space_operations linvfs_aops;
-extern int linvfs_get_block(struct inode *, sector_t, struct buffer_head *, int);
+extern struct address_space_operations xfs_address_space_operations;
+extern int xfs_get_block(struct inode *, sector_t, struct buffer_head *, int);
#endif /* __XFS_IOPS_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_buf.c b/fs/xfs/linux-2.6/xfs_buf.c
index 8cdfa4151659..9fb0312665ca 100644
--- a/fs/xfs/linux-2.6/xfs_buf.c
+++ b/fs/xfs/linux-2.6/xfs_buf.c
@@ -1806,13 +1806,12 @@ xfs_flush_buftarg(
int __init
xfs_buf_init(void)
{
- int error = -ENOMEM;
-
#ifdef XFS_BUF_TRACE
xfs_buf_trace_buf = ktrace_alloc(XFS_BUF_TRACE_SIZE, KM_SLEEP);
#endif
- xfs_buf_zone = kmem_zone_init(sizeof(xfs_buf_t), "xfs_buf");
+ xfs_buf_zone = kmem_zone_init_flags(sizeof(xfs_buf_t), "xfs_buf",
+ KM_ZONE_HWALIGN, NULL);
if (!xfs_buf_zone)
goto out_free_trace_buf;
@@ -1840,7 +1839,7 @@ xfs_buf_init(void)
#ifdef XFS_BUF_TRACE
ktrace_free(xfs_buf_trace_buf);
#endif
- return error;
+ return -ENOMEM;
}
void
diff --git a/fs/xfs/linux-2.6/xfs_export.c b/fs/xfs/linux-2.6/xfs_export.c
index 80eb249f2fa0..b768ea910bbe 100644
--- a/fs/xfs/linux-2.6/xfs_export.c
+++ b/fs/xfs/linux-2.6/xfs_export.c
@@ -25,6 +25,8 @@
#include "xfs_mount.h"
#include "xfs_export.h"
+STATIC struct dentry dotdot = { .d_name.name = "..", .d_name.len = 2, };
+
/*
* XFS encodes and decodes the fileid portion of NFS filehandles
* itself instead of letting the generic NFS code do it. This
@@ -37,7 +39,7 @@
*/
STATIC struct dentry *
-linvfs_decode_fh(
+xfs_fs_decode_fh(
struct super_block *sb,
__u32 *fh,
int fh_len,
@@ -78,12 +80,12 @@ linvfs_decode_fh(
}
fh = (__u32 *)&ifid;
- return find_exported_dentry(sb, fh, parent, acceptable, context);
+ return sb->s_export_op->find_exported_dentry(sb, fh, parent, acceptable, context);
}
STATIC int
-linvfs_encode_fh(
+xfs_fs_encode_fh(
struct dentry *dentry,
__u32 *fh,
int *max_len,
@@ -95,7 +97,7 @@ linvfs_encode_fh(
int len;
int is64 = 0;
#if XFS_BIG_INUMS
- vfs_t *vfs = LINVFS_GET_VFS(inode->i_sb);
+ vfs_t *vfs = vfs_from_sb(inode->i_sb);
if (!(vfs->vfs_flag & VFS_32BITINODES)) {
/* filesystem may contain 64bit inode numbers */
@@ -130,21 +132,21 @@ linvfs_encode_fh(
}
STATIC struct dentry *
-linvfs_get_dentry(
+xfs_fs_get_dentry(
struct super_block *sb,
void *data)
{
vnode_t *vp;
struct inode *inode;
struct dentry *result;
- vfs_t *vfsp = LINVFS_GET_VFS(sb);
+ vfs_t *vfsp = vfs_from_sb(sb);
int error;
VFS_VGET(vfsp, &vp, (fid_t *)data, error);
if (error || vp == NULL)
return ERR_PTR(-ESTALE) ;
- inode = LINVFS_GET_IP(vp);
+ inode = vn_to_inode(vp);
result = d_alloc_anon(inode);
if (!result) {
iput(inode);
@@ -154,25 +156,20 @@ linvfs_get_dentry(
}
STATIC struct dentry *
-linvfs_get_parent(
+xfs_fs_get_parent(
struct dentry *child)
{
int error;
vnode_t *vp, *cvp;
struct dentry *parent;
- struct dentry dotdot;
-
- dotdot.d_name.name = "..";
- dotdot.d_name.len = 2;
- dotdot.d_inode = NULL;
cvp = NULL;
- vp = LINVFS_GET_VP(child->d_inode);
+ vp = vn_from_inode(child->d_inode);
VOP_LOOKUP(vp, &dotdot, &cvp, 0, NULL, NULL, error);
if (unlikely(error))
return ERR_PTR(-error);
- parent = d_alloc_anon(LINVFS_GET_IP(cvp));
+ parent = d_alloc_anon(vn_to_inode(cvp));
if (unlikely(!parent)) {
VN_RELE(cvp);
return ERR_PTR(-ENOMEM);
@@ -180,9 +177,9 @@ linvfs_get_parent(
return parent;
}
-struct export_operations linvfs_export_ops = {
- .decode_fh = linvfs_decode_fh,
- .encode_fh = linvfs_encode_fh,
- .get_parent = linvfs_get_parent,
- .get_dentry = linvfs_get_dentry,
+struct export_operations xfs_export_operations = {
+ .decode_fh = xfs_fs_decode_fh,
+ .encode_fh = xfs_fs_encode_fh,
+ .get_parent = xfs_fs_get_parent,
+ .get_dentry = xfs_fs_get_dentry,
};
diff --git a/fs/xfs/linux-2.6/xfs_file.c b/fs/xfs/linux-2.6/xfs_file.c
index ced4404339c7..185567a6a561 100644
--- a/fs/xfs/linux-2.6/xfs_file.c
+++ b/fs/xfs/linux-2.6/xfs_file.c
@@ -43,13 +43,13 @@
#include <linux/dcache.h>
#include <linux/smp_lock.h>
-static struct vm_operations_struct linvfs_file_vm_ops;
+static struct vm_operations_struct xfs_file_vm_ops;
#ifdef CONFIG_XFS_DMAPI
-static struct vm_operations_struct linvfs_dmapi_file_vm_ops;
+static struct vm_operations_struct xfs_dmapi_file_vm_ops;
#endif
STATIC inline ssize_t
-__linvfs_read(
+__xfs_file_read(
struct kiocb *iocb,
char __user *buf,
int ioflags,
@@ -58,7 +58,7 @@ __linvfs_read(
{
struct iovec iov = {buf, count};
struct file *file = iocb->ki_filp;
- vnode_t *vp = LINVFS_GET_VP(file->f_dentry->d_inode);
+ vnode_t *vp = vn_from_inode(file->f_dentry->d_inode);
ssize_t rval;
BUG_ON(iocb->ki_pos != pos);
@@ -71,28 +71,28 @@ __linvfs_read(
STATIC ssize_t
-linvfs_aio_read(
+xfs_file_aio_read(
struct kiocb *iocb,
char __user *buf,
size_t count,
loff_t pos)
{
- return __linvfs_read(iocb, buf, IO_ISAIO, count, pos);
+ return __xfs_file_read(iocb, buf, IO_ISAIO, count, pos);
}
STATIC ssize_t
-linvfs_aio_read_invis(
+xfs_file_aio_read_invis(
struct kiocb *iocb,
char __user *buf,
size_t count,
loff_t pos)
{
- return __linvfs_read(iocb, buf, IO_ISAIO|IO_INVIS, count, pos);
+ return __xfs_file_read(iocb, buf, IO_ISAIO|IO_INVIS, count, pos);
}
STATIC inline ssize_t
-__linvfs_write(
+__xfs_file_write(
struct kiocb *iocb,
const char __user *buf,
int ioflags,
@@ -102,7 +102,7 @@ __linvfs_write(
struct iovec iov = {(void __user *)buf, count};
struct file *file = iocb->ki_filp;
struct inode *inode = file->f_mapping->host;
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
ssize_t rval;
BUG_ON(iocb->ki_pos != pos);
@@ -115,28 +115,28 @@ __linvfs_write(
STATIC ssize_t
-linvfs_aio_write(
+xfs_file_aio_write(
struct kiocb *iocb,
const char __user *buf,
size_t count,
loff_t pos)
{
- return __linvfs_write(iocb, buf, IO_ISAIO, count, pos);
+ return __xfs_file_write(iocb, buf, IO_ISAIO, count, pos);
}
STATIC ssize_t
-linvfs_aio_write_invis(
+xfs_file_aio_write_invis(
struct kiocb *iocb,
const char __user *buf,
size_t count,
loff_t pos)
{
- return __linvfs_write(iocb, buf, IO_ISAIO|IO_INVIS, count, pos);
+ return __xfs_file_write(iocb, buf, IO_ISAIO|IO_INVIS, count, pos);
}
STATIC inline ssize_t
-__linvfs_readv(
+__xfs_file_readv(
struct file *file,
const struct iovec *iov,
int ioflags,
@@ -144,8 +144,8 @@ __linvfs_readv(
loff_t *ppos)
{
struct inode *inode = file->f_mapping->host;
- vnode_t *vp = LINVFS_GET_VP(inode);
- struct kiocb kiocb;
+ vnode_t *vp = vn_from_inode(inode);
+ struct kiocb kiocb;
ssize_t rval;
init_sync_kiocb(&kiocb, file);
@@ -160,28 +160,28 @@ __linvfs_readv(
}
STATIC ssize_t
-linvfs_readv(
+xfs_file_readv(
struct file *file,
const struct iovec *iov,
unsigned long nr_segs,
loff_t *ppos)
{
- return __linvfs_readv(file, iov, 0, nr_segs, ppos);
+ return __xfs_file_readv(file, iov, 0, nr_segs, ppos);
}
STATIC ssize_t
-linvfs_readv_invis(
+xfs_file_readv_invis(
struct file *file,
const struct iovec *iov,
unsigned long nr_segs,
loff_t *ppos)
{
- return __linvfs_readv(file, iov, IO_INVIS, nr_segs, ppos);
+ return __xfs_file_readv(file, iov, IO_INVIS, nr_segs, ppos);
}
STATIC inline ssize_t
-__linvfs_writev(
+__xfs_file_writev(
struct file *file,
const struct iovec *iov,
int ioflags,
@@ -189,8 +189,8 @@ __linvfs_writev(
loff_t *ppos)
{
struct inode *inode = file->f_mapping->host;
- vnode_t *vp = LINVFS_GET_VP(inode);
- struct kiocb kiocb;
+ vnode_t *vp = vn_from_inode(inode);
+ struct kiocb kiocb;
ssize_t rval;
init_sync_kiocb(&kiocb, file);
@@ -206,34 +206,34 @@ __linvfs_writev(
STATIC ssize_t
-linvfs_writev(
+xfs_file_writev(
struct file *file,
const struct iovec *iov,
unsigned long nr_segs,
loff_t *ppos)
{
- return __linvfs_writev(file, iov, 0, nr_segs, ppos);
+ return __xfs_file_writev(file, iov, 0, nr_segs, ppos);
}
STATIC ssize_t
-linvfs_writev_invis(
+xfs_file_writev_invis(
struct file *file,
const struct iovec *iov,
unsigned long nr_segs,
loff_t *ppos)
{
- return __linvfs_writev(file, iov, IO_INVIS, nr_segs, ppos);
+ return __xfs_file_writev(file, iov, IO_INVIS, nr_segs, ppos);
}
STATIC ssize_t
-linvfs_sendfile(
+xfs_file_sendfile(
struct file *filp,
loff_t *ppos,
size_t count,
read_actor_t actor,
void *target)
{
- vnode_t *vp = LINVFS_GET_VP(filp->f_dentry->d_inode);
+ vnode_t *vp = vn_from_inode(filp->f_dentry->d_inode);
ssize_t rval;
VOP_SENDFILE(vp, filp, ppos, 0, count, actor, target, NULL, rval);
@@ -242,11 +242,11 @@ linvfs_sendfile(
STATIC int
-linvfs_open(
+xfs_file_open(
struct inode *inode,
struct file *filp)
{
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
int error;
if (!(filp->f_flags & O_LARGEFILE) && i_size_read(inode) > MAX_NON_LFS)
@@ -259,11 +259,11 @@ linvfs_open(
STATIC int
-linvfs_release(
+xfs_file_release(
struct inode *inode,
struct file *filp)
{
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
int error = 0;
if (vp)
@@ -273,13 +273,13 @@ linvfs_release(
STATIC int
-linvfs_fsync(
+xfs_file_fsync(
struct file *filp,
struct dentry *dentry,
int datasync)
{
struct inode *inode = dentry->d_inode;
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
int error;
int flags = FSYNC_WAIT;
@@ -292,7 +292,7 @@ linvfs_fsync(
}
/*
- * linvfs_readdir maps to VOP_READDIR().
+ * xfs_file_readdir maps to VOP_READDIR().
* We need to build a uio, cred, ...
*/
@@ -301,13 +301,13 @@ linvfs_fsync(
#ifdef CONFIG_XFS_DMAPI
STATIC struct page *
-linvfs_filemap_nopage(
+xfs_vm_nopage(
struct vm_area_struct *area,
unsigned long address,
int *type)
{
struct inode *inode = area->vm_file->f_dentry->d_inode;
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
xfs_mount_t *mp = XFS_VFSTOM(vp->v_vfsp);
int error;
@@ -324,7 +324,7 @@ linvfs_filemap_nopage(
STATIC int
-linvfs_readdir(
+xfs_file_readdir(
struct file *filp,
void *dirent,
filldir_t filldir)
@@ -340,7 +340,7 @@ linvfs_readdir(
xfs_off_t start_offset, curr_offset;
xfs_dirent_t *dbp = NULL;
- vp = LINVFS_GET_VP(filp->f_dentry->d_inode);
+ vp = vn_from_inode(filp->f_dentry->d_inode);
ASSERT(vp);
/* Try fairly hard to get memory */
@@ -404,39 +404,40 @@ done:
STATIC int
-linvfs_file_mmap(
+xfs_file_mmap(
struct file *filp,
struct vm_area_struct *vma)
{
struct inode *ip = filp->f_dentry->d_inode;
- vnode_t *vp = LINVFS_GET_VP(ip);
- vattr_t va = { .va_mask = XFS_AT_UPDATIME };
+ vnode_t *vp = vn_from_inode(ip);
+ vattr_t vattr;
int error;
- vma->vm_ops = &linvfs_file_vm_ops;
+ vma->vm_ops = &xfs_file_vm_ops;
#ifdef CONFIG_XFS_DMAPI
if (vp->v_vfsp->vfs_flag & VFS_DMI) {
- vma->vm_ops = &linvfs_dmapi_file_vm_ops;
+ vma->vm_ops = &xfs_dmapi_file_vm_ops;
}
#endif /* CONFIG_XFS_DMAPI */
- VOP_SETATTR(vp, &va, XFS_AT_UPDATIME, NULL, error);
- if (!error)
- vn_revalidate(vp); /* update Linux inode flags */
+ vattr.va_mask = XFS_AT_UPDATIME;
+ VOP_SETATTR(vp, &vattr, XFS_AT_UPDATIME, NULL, error);
+ if (likely(!error))
+ __vn_revalidate(vp, &vattr); /* update flags */
return 0;
}
STATIC long
-linvfs_ioctl(
+xfs_file_ioctl(
struct file *filp,
unsigned int cmd,
unsigned long arg)
{
int error;
- struct inode *inode = filp->f_dentry->d_inode;
- vnode_t *vp = LINVFS_GET_VP(inode);
+ struct inode *inode = filp->f_dentry->d_inode;
+ vnode_t *vp = vn_from_inode(inode);
VOP_IOCTL(vp, inode, filp, 0, cmd, (void __user *)arg, error);
VMODIFY(vp);
@@ -451,14 +452,14 @@ linvfs_ioctl(
}
STATIC long
-linvfs_ioctl_invis(
+xfs_file_ioctl_invis(
struct file *filp,
unsigned int cmd,
unsigned long arg)
{
int error;
- struct inode *inode = filp->f_dentry->d_inode;
- vnode_t *vp = LINVFS_GET_VP(inode);
+ struct inode *inode = filp->f_dentry->d_inode;
+ vnode_t *vp = vn_from_inode(inode);
ASSERT(vp);
VOP_IOCTL(vp, inode, filp, IO_INVIS, cmd, (void __user *)arg, error);
@@ -476,11 +477,11 @@ linvfs_ioctl_invis(
#ifdef CONFIG_XFS_DMAPI
#ifdef HAVE_VMOP_MPROTECT
STATIC int
-linvfs_mprotect(
+xfs_vm_mprotect(
struct vm_area_struct *vma,
unsigned int newflags)
{
- vnode_t *vp = LINVFS_GET_VP(vma->vm_file->f_dentry->d_inode);
+ vnode_t *vp = vn_from_inode(vma->vm_file->f_dentry->d_inode);
int error = 0;
if (vp->v_vfsp->vfs_flag & VFS_DMI) {
@@ -503,10 +504,10 @@ linvfs_mprotect(
* it back online.
*/
STATIC int
-linvfs_open_exec(
+xfs_file_open_exec(
struct inode *inode)
{
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
xfs_mount_t *mp = XFS_VFSTOM(vp->v_vfsp);
int error = 0;
xfs_inode_t *ip;
@@ -527,69 +528,69 @@ open_exec_out:
}
#endif /* HAVE_FOP_OPEN_EXEC */
-struct file_operations linvfs_file_operations = {
+struct file_operations xfs_file_operations = {
.llseek = generic_file_llseek,
.read = do_sync_read,
.write = do_sync_write,
- .readv = linvfs_readv,
- .writev = linvfs_writev,
- .aio_read = linvfs_aio_read,
- .aio_write = linvfs_aio_write,
- .sendfile = linvfs_sendfile,
- .unlocked_ioctl = linvfs_ioctl,
+ .readv = xfs_file_readv,
+ .writev = xfs_file_writev,
+ .aio_read = xfs_file_aio_read,
+ .aio_write = xfs_file_aio_write,
+ .sendfile = xfs_file_sendfile,
+ .unlocked_ioctl = xfs_file_ioctl,
#ifdef CONFIG_COMPAT
- .compat_ioctl = linvfs_compat_ioctl,
+ .compat_ioctl = xfs_file_compat_ioctl,
#endif
- .mmap = linvfs_file_mmap,
- .open = linvfs_open,
- .release = linvfs_release,
- .fsync = linvfs_fsync,
+ .mmap = xfs_file_mmap,
+ .open = xfs_file_open,
+ .release = xfs_file_release,
+ .fsync = xfs_file_fsync,
#ifdef HAVE_FOP_OPEN_EXEC
- .open_exec = linvfs_open_exec,
+ .open_exec = xfs_file_open_exec,
#endif
};
-struct file_operations linvfs_invis_file_operations = {
+struct file_operations xfs_invis_file_operations = {
.llseek = generic_file_llseek,
.read = do_sync_read,
.write = do_sync_write,
- .readv = linvfs_readv_invis,
- .writev = linvfs_writev_invis,
- .aio_read = linvfs_aio_read_invis,
- .aio_write = linvfs_aio_write_invis,
- .sendfile = linvfs_sendfile,
- .unlocked_ioctl = linvfs_ioctl_invis,
+ .readv = xfs_file_readv_invis,
+ .writev = xfs_file_writev_invis,
+ .aio_read = xfs_file_aio_read_invis,
+ .aio_write = xfs_file_aio_write_invis,
+ .sendfile = xfs_file_sendfile,
+ .unlocked_ioctl = xfs_file_ioctl_invis,
#ifdef CONFIG_COMPAT
- .compat_ioctl = linvfs_compat_invis_ioctl,
+ .compat_ioctl = xfs_file_compat_invis_ioctl,
#endif
- .mmap = linvfs_file_mmap,
- .open = linvfs_open,
- .release = linvfs_release,
- .fsync = linvfs_fsync,
+ .mmap = xfs_file_mmap,
+ .open = xfs_file_open,
+ .release = xfs_file_release,
+ .fsync = xfs_file_fsync,
};
-struct file_operations linvfs_dir_operations = {
+struct file_operations xfs_dir_file_operations = {
.read = generic_read_dir,
- .readdir = linvfs_readdir,
- .unlocked_ioctl = linvfs_ioctl,
+ .readdir = xfs_file_readdir,
+ .unlocked_ioctl = xfs_file_ioctl,
#ifdef CONFIG_COMPAT
- .compat_ioctl = linvfs_compat_ioctl,
+ .compat_ioctl = xfs_file_compat_ioctl,
#endif
- .fsync = linvfs_fsync,
+ .fsync = xfs_file_fsync,
};
-static struct vm_operations_struct linvfs_file_vm_ops = {
+static struct vm_operations_struct xfs_file_vm_ops = {
.nopage = filemap_nopage,
.populate = filemap_populate,
};
#ifdef CONFIG_XFS_DMAPI
-static struct vm_operations_struct linvfs_dmapi_file_vm_ops = {
- .nopage = linvfs_filemap_nopage,
+static struct vm_operations_struct xfs_dmapi_file_vm_ops = {
+ .nopage = xfs_vm_nopage,
.populate = filemap_populate,
#ifdef HAVE_VMOP_MPROTECT
- .mprotect = linvfs_mprotect,
+ .mprotect = xfs_vm_mprotect,
#endif
};
#endif /* CONFIG_XFS_DMAPI */
diff --git a/fs/xfs/linux-2.6/xfs_fs_subr.c b/fs/xfs/linux-2.6/xfs_fs_subr.c
index 4fa4b1a5187e..575f2a790f31 100644
--- a/fs/xfs/linux-2.6/xfs_fs_subr.c
+++ b/fs/xfs/linux-2.6/xfs_fs_subr.c
@@ -57,7 +57,7 @@ fs_tosspages(
int fiopt)
{
vnode_t *vp = BHV_TO_VNODE(bdp);
- struct inode *ip = LINVFS_GET_IP(vp);
+ struct inode *ip = vn_to_inode(vp);
if (VN_CACHED(vp))
truncate_inode_pages(ip->i_mapping, first);
@@ -76,7 +76,7 @@ fs_flushinval_pages(
int fiopt)
{
vnode_t *vp = BHV_TO_VNODE(bdp);
- struct inode *ip = LINVFS_GET_IP(vp);
+ struct inode *ip = vn_to_inode(vp);
if (VN_CACHED(vp)) {
filemap_write_and_wait(ip->i_mapping);
@@ -98,7 +98,7 @@ fs_flush_pages(
int fiopt)
{
vnode_t *vp = BHV_TO_VNODE(bdp);
- struct inode *ip = LINVFS_GET_IP(vp);
+ struct inode *ip = vn_to_inode(vp);
if (VN_CACHED(vp)) {
filemap_fdatawrite(ip->i_mapping);
diff --git a/fs/xfs/linux-2.6/xfs_ioctl.c b/fs/xfs/linux-2.6/xfs_ioctl.c
index 4db47790415c..84478491609b 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl.c
@@ -138,7 +138,7 @@ xfs_find_handle(
}
/* we need the vnode */
- vp = LINVFS_GET_VP(inode);
+ vp = vn_from_inode(inode);
/* now we can grab the fsid */
memcpy(&handle.ha_fsid, vp->v_vfsp->vfs_altfsid, sizeof(xfs_fsid_t));
@@ -256,7 +256,7 @@ xfs_vget_fsop_handlereq(
}
vpp = XFS_ITOV(ip);
- inodep = LINVFS_GET_IP(vpp);
+ inodep = vn_to_inode(vpp);
xfs_iunlock(ip, XFS_ILOCK_SHARED);
*vp = vpp;
@@ -344,7 +344,7 @@ xfs_open_by_handle(
return -XFS_ERROR(-PTR_ERR(filp));
}
if (inode->i_mode & S_IFREG)
- filp->f_op = &linvfs_invis_file_operations;
+ filp->f_op = &xfs_invis_file_operations;
fd_install(new_fd, filp);
return new_fd;
@@ -715,7 +715,7 @@ xfs_ioctl(
xfs_inode_t *ip;
xfs_mount_t *mp;
- vp = LINVFS_GET_VP(inode);
+ vp = vn_from_inode(inode);
vn_trace_entry(vp, "xfs_ioctl", (inst_t *)__return_address);
@@ -1160,105 +1160,129 @@ xfs_ioc_xattr(
void __user *arg)
{
struct fsxattr fa;
- vattr_t va;
- int error;
+ struct vattr *vattr;
+ int error = 0;
int attr_flags;
unsigned int flags;
+ vattr = kmalloc(sizeof(*vattr), GFP_KERNEL);
+ if (unlikely(!vattr))
+ return -ENOMEM;
+
switch (cmd) {
case XFS_IOC_FSGETXATTR: {
- va.va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | \
- XFS_AT_NEXTENTS | XFS_AT_PROJID;
- VOP_GETATTR(vp, &va, 0, NULL, error);
- if (error)
- return -error;
+ vattr->va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | \
+ XFS_AT_NEXTENTS | XFS_AT_PROJID;
+ VOP_GETATTR(vp, vattr, 0, NULL, error);
+ if (unlikely(error)) {
+ error = -error;
+ break;
+ }
- fa.fsx_xflags = va.va_xflags;
- fa.fsx_extsize = va.va_extsize;
- fa.fsx_nextents = va.va_nextents;
- fa.fsx_projid = va.va_projid;
+ fa.fsx_xflags = vattr->va_xflags;
+ fa.fsx_extsize = vattr->va_extsize;
+ fa.fsx_nextents = vattr->va_nextents;
+ fa.fsx_projid = vattr->va_projid;
- if (copy_to_user(arg, &fa, sizeof(fa)))
- return -XFS_ERROR(EFAULT);
- return 0;
+ if (copy_to_user(arg, &fa, sizeof(fa))) {
+ error = -EFAULT;
+ break;
+ }
+ break;
}
case XFS_IOC_FSSETXATTR: {
- if (copy_from_user(&fa, arg, sizeof(fa)))
- return -XFS_ERROR(EFAULT);
+ if (copy_from_user(&fa, arg, sizeof(fa))) {
+ error = -EFAULT;
+ break;
+ }
attr_flags = 0;
if (filp->f_flags & (O_NDELAY|O_NONBLOCK))
attr_flags |= ATTR_NONBLOCK;
- va.va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | XFS_AT_PROJID;
- va.va_xflags = fa.fsx_xflags;
- va.va_extsize = fa.fsx_extsize;
- va.va_projid = fa.fsx_projid;
+ vattr->va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | XFS_AT_PROJID;
+ vattr->va_xflags = fa.fsx_xflags;
+ vattr->va_extsize = fa.fsx_extsize;
+ vattr->va_projid = fa.fsx_projid;
- VOP_SETATTR(vp, &va, attr_flags, NULL, error);
- if (!error)
- vn_revalidate(vp); /* update Linux inode flags */
- return -error;
+ VOP_SETATTR(vp, vattr, attr_flags, NULL, error);
+ if (likely(!error))
+ __vn_revalidate(vp, vattr); /* update flags */
+ error = -error;
+ break;
}
case XFS_IOC_FSGETXATTRA: {
- va.va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | \
- XFS_AT_ANEXTENTS | XFS_AT_PROJID;
- VOP_GETATTR(vp, &va, 0, NULL, error);
- if (error)
- return -error;
+ vattr->va_mask = XFS_AT_XFLAGS | XFS_AT_EXTSIZE | \
+ XFS_AT_ANEXTENTS | XFS_AT_PROJID;
+ VOP_GETATTR(vp, vattr, 0, NULL, error);
+ if (unlikely(error)) {
+ error = -error;
+ break;
+ }
- fa.fsx_xflags = va.va_xflags;
- fa.fsx_extsize = va.va_extsize;
- fa.fsx_nextents = va.va_anextents;
- fa.fsx_projid = va.va_projid;
+ fa.fsx_xflags = vattr->va_xflags;
+ fa.fsx_extsize = vattr->va_extsize;
+ fa.fsx_nextents = vattr->va_anextents;
+ fa.fsx_projid = vattr->va_projid;
- if (copy_to_user(arg, &fa, sizeof(fa)))
- return -XFS_ERROR(EFAULT);
- return 0;
+ if (copy_to_user(arg, &fa, sizeof(fa))) {
+ error = -EFAULT;
+ break;
+ }
+ break;
}
case XFS_IOC_GETXFLAGS: {
flags = xfs_di2lxflags(ip->i_d.di_flags);
if (copy_to_user(arg, &flags, sizeof(flags)))
- return -XFS_ERROR(EFAULT);
- return 0;
+ error = -EFAULT;
+ break;
}
case XFS_IOC_SETXFLAGS: {
- if (copy_from_user(&flags, arg, sizeof(flags)))
- return -XFS_ERROR(EFAULT);
+ if (copy_from_user(&flags, arg, sizeof(flags))) {
+ error = -EFAULT;
+ break;
+ }
if (flags & ~(LINUX_XFLAG_IMMUTABLE | LINUX_XFLAG_APPEND | \
LINUX_XFLAG_NOATIME | LINUX_XFLAG_NODUMP | \
- LINUX_XFLAG_SYNC))
- return -XFS_ERROR(EOPNOTSUPP);
+ LINUX_XFLAG_SYNC)) {
+ error = -EOPNOTSUPP;
+ break;
+ }
attr_flags = 0;
if (filp->f_flags & (O_NDELAY|O_NONBLOCK))
attr_flags |= ATTR_NONBLOCK;
- va.va_mask = XFS_AT_XFLAGS;
- va.va_xflags = xfs_merge_ioc_xflags(flags,
- xfs_ip2xflags(ip));
+ vattr->va_mask = XFS_AT_XFLAGS;
+ vattr->va_xflags = xfs_merge_ioc_xflags(flags,
+ xfs_ip2xflags(ip));
- VOP_SETATTR(vp, &va, attr_flags, NULL, error);
- if (!error)
- vn_revalidate(vp); /* update Linux inode flags */
- return -error;
+ VOP_SETATTR(vp, vattr, attr_flags, NULL, error);
+ if (likely(!error))
+ __vn_revalidate(vp, vattr); /* update flags */
+ error = -error;
+ break;
}
case XFS_IOC_GETVERSION: {
- flags = LINVFS_GET_IP(vp)->i_generation;
+ flags = vn_to_inode(vp)->i_generation;
if (copy_to_user(arg, &flags, sizeof(flags)))
- return -XFS_ERROR(EFAULT);
- return 0;
+ error = -EFAULT;
+ break;
}
default:
- return -ENOTTY;
+ error = -ENOTTY;
+ break;
}
+
+ kfree(vattr);
+ return error;
}
STATIC int
diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.c b/fs/xfs/linux-2.6/xfs_ioctl32.c
index a7c9ba1a9f7b..b6321abd9a81 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl32.c
+++ b/fs/xfs/linux-2.6/xfs_ioctl32.c
@@ -107,11 +107,11 @@ xfs_ioctl32_bulkstat(
#endif
STATIC long
-__linvfs_compat_ioctl(int mode, struct file *f, unsigned cmd, unsigned long arg)
+xfs_compat_ioctl(int mode, struct file *f, unsigned cmd, unsigned long arg)
{
int error;
struct inode *inode = f->f_dentry->d_inode;
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_to_inode(inode);
switch (cmd) {
case XFS_IOC_DIOINFO:
@@ -196,19 +196,19 @@ __linvfs_compat_ioctl(int mode, struct file *f, unsigned cmd, unsigned long arg)
}
long
-linvfs_compat_ioctl(
+xfs_file_compat_ioctl(
struct file *f,
unsigned cmd,
unsigned long arg)
{
- return __linvfs_compat_ioctl(0, f, cmd, arg);
+ return xfs_compat_ioctl(0, f, cmd, arg);
}
long
-linvfs_compat_invis_ioctl(
+xfs_file_compat_invis_ioctl(
struct file *f,
unsigned cmd,
unsigned long arg)
{
- return __linvfs_compat_ioctl(IO_INVIS, f, cmd, arg);
+ return xfs_compat_ioctl(IO_INVIS, f, cmd, arg);
}
diff --git a/fs/xfs/linux-2.6/xfs_ioctl32.h b/fs/xfs/linux-2.6/xfs_ioctl32.h
index 011c273bec50..02de6e62ee37 100644
--- a/fs/xfs/linux-2.6/xfs_ioctl32.h
+++ b/fs/xfs/linux-2.6/xfs_ioctl32.h
@@ -18,7 +18,7 @@
#ifndef __XFS_IOCTL32_H__
#define __XFS_IOCTL32_H__
-extern long linvfs_compat_ioctl(struct file *, unsigned, unsigned long);
-extern long linvfs_compat_invis_ioctl(struct file *f, unsigned, unsigned long);
+extern long xfs_file_compat_ioctl(struct file *, unsigned, unsigned long);
+extern long xfs_file_compat_invis_ioctl(struct file *, unsigned, unsigned long);
#endif /* __XFS_IOCTL32_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_iops.c b/fs/xfs/linux-2.6/xfs_iops.c
index d7f6f2d8ac8e..af487437bd7e 100644
--- a/fs/xfs/linux-2.6/xfs_iops.c
+++ b/fs/xfs/linux-2.6/xfs_iops.c
@@ -106,7 +106,7 @@ xfs_ichgtime(
xfs_inode_t *ip,
int flags)
{
- struct inode *inode = LINVFS_GET_IP(XFS_ITOV(ip));
+ struct inode *inode = vn_to_inode(XFS_ITOV(ip));
timespec_t tv;
nanotime(&tv);
@@ -198,22 +198,22 @@ xfs_ichgtime_fast(
* Pull the link count and size up from the xfs inode to the linux inode
*/
STATIC void
-validate_fields(
- struct inode *ip)
+xfs_validate_fields(
+ struct inode *ip,
+ struct vattr *vattr)
{
- vnode_t *vp = LINVFS_GET_VP(ip);
- vattr_t va;
+ vnode_t *vp = vn_from_inode(ip);
int error;
- va.va_mask = XFS_AT_NLINK|XFS_AT_SIZE|XFS_AT_NBLOCKS;
- VOP_GETATTR(vp, &va, ATTR_LAZY, NULL, error);
- if (likely(!error)) {
- ip->i_nlink = va.va_nlink;
- ip->i_blocks = va.va_nblocks;
+ vattr->va_mask = XFS_AT_NLINK|XFS_AT_SIZE|XFS_AT_NBLOCKS;
+ VOP_GETATTR(vp, vattr, ATTR_LAZY, NULL, error);
+ if (likely(!error)) {
+ ip->i_nlink = vattr->va_nlink;
+ ip->i_blocks = vattr->va_nblocks;
- /* we're under i_mutex so i_size can't change under us */
- if (i_size_read(ip) != va.va_size)
- i_size_write(ip, va.va_size);
+ /* we're under i_sem so i_size can't change under us */
+ if (i_size_read(ip) != vattr->va_size)
+ i_size_write(ip, vattr->va_size);
}
}
@@ -224,11 +224,11 @@ validate_fields(
* inode, of course, such that log replay can't cause these to be lost).
*/
STATIC int
-linvfs_init_security(
+xfs_init_security(
struct vnode *vp,
struct inode *dir)
{
- struct inode *ip = LINVFS_GET_IP(vp);
+ struct inode *ip = vn_to_inode(vp);
size_t length;
void *value;
char *name;
@@ -257,46 +257,46 @@ linvfs_init_security(
* XXX(hch): nfsd is broken, better fix it instead.
*/
STATIC inline int
-has_fs_struct(struct task_struct *task)
+xfs_has_fs_struct(struct task_struct *task)
{
return (task->fs != init_task.fs);
}
STATIC inline void
-cleanup_inode(
+xfs_cleanup_inode(
vnode_t *dvp,
vnode_t *vp,
- struct dentry *dentry,
+ struct dentry *dentry,
int mode)
{
struct dentry teardown = {};
- int err2;
+ int error;
/* Oh, the horror.
- * If we can't add the ACL or we fail in
- * linvfs_init_security we must back out.
+ * If we can't add the ACL or we fail in
+ * xfs_init_security we must back out.
* ENOSPC can hit here, among other things.
*/
- teardown.d_inode = LINVFS_GET_IP(vp);
+ teardown.d_inode = vn_to_inode(vp);
teardown.d_name = dentry->d_name;
if (S_ISDIR(mode))
- VOP_RMDIR(dvp, &teardown, NULL, err2);
+ VOP_RMDIR(dvp, &teardown, NULL, error);
else
- VOP_REMOVE(dvp, &teardown, NULL, err2);
+ VOP_REMOVE(dvp, &teardown, NULL, error);
VN_RELE(vp);
}
STATIC int
-linvfs_mknod(
+xfs_vn_mknod(
struct inode *dir,
struct dentry *dentry,
int mode,
dev_t rdev)
{
struct inode *ip;
- vattr_t va;
- vnode_t *vp = NULL, *dvp = LINVFS_GET_VP(dir);
+ vattr_t vattr = { 0 };
+ vnode_t *vp = NULL, *dvp = vn_from_inode(dir);
xfs_acl_t *default_acl = NULL;
attrexists_t test_default_acl = _ACL_DEFAULT_EXISTS;
int error;
@@ -305,99 +305,98 @@ linvfs_mknod(
* Irix uses Missed'em'V split, but doesn't want to see
* the upper 5 bits of (14bit) major.
*/
- if (!sysv_valid_dev(rdev) || MAJOR(rdev) & ~0x1ff)
+ if (unlikely(!sysv_valid_dev(rdev) || MAJOR(rdev) & ~0x1ff))
return -EINVAL;
- if (test_default_acl && test_default_acl(dvp)) {
- if (!_ACL_ALLOC(default_acl))
+ if (unlikely(test_default_acl && test_default_acl(dvp))) {
+ if (!_ACL_ALLOC(default_acl)) {
return -ENOMEM;
+ }
if (!_ACL_GET_DEFAULT(dvp, default_acl)) {
_ACL_FREE(default_acl);
default_acl = NULL;
}
}
- if (IS_POSIXACL(dir) && !default_acl && has_fs_struct(current))
+ if (IS_POSIXACL(dir) && !default_acl && xfs_has_fs_struct(current))
mode &= ~current->fs->umask;
- memset(&va, 0, sizeof(va));
- va.va_mask = XFS_AT_TYPE|XFS_AT_MODE;
- va.va_mode = mode;
+ vattr.va_mask = XFS_AT_TYPE|XFS_AT_MODE;
+ vattr.va_mode = mode;
switch (mode & S_IFMT) {
case S_IFCHR: case S_IFBLK: case S_IFIFO: case S_IFSOCK:
- va.va_rdev = sysv_encode_dev(rdev);
- va.va_mask |= XFS_AT_RDEV;
+ vattr.va_rdev = sysv_encode_dev(rdev);
+ vattr.va_mask |= XFS_AT_RDEV;
/*FALLTHROUGH*/
case S_IFREG:
- VOP_CREATE(dvp, dentry, &va, &vp, NULL, error);
+ VOP_CREATE(dvp, dentry, &vattr, &vp, NULL, error);
break;
case S_IFDIR:
- VOP_MKDIR(dvp, dentry, &va, &vp, NULL, error);
+ VOP_MKDIR(dvp, dentry, &vattr, &vp, NULL, error);
break;
default:
error = EINVAL;
break;
}
- if (!error)
- {
- error = linvfs_init_security(vp, dir);
+ if (unlikely(!error)) {
+ error = xfs_init_security(vp, dir);
if (error)
- cleanup_inode(dvp, vp, dentry, mode);
+ xfs_cleanup_inode(dvp, vp, dentry, mode);
}
- if (default_acl) {
+ if (unlikely(default_acl)) {
if (!error) {
- error = _ACL_INHERIT(vp, &va, default_acl);
- if (!error)
+ error = _ACL_INHERIT(vp, &vattr, default_acl);
+ if (!error)
VMODIFY(vp);
else
- cleanup_inode(dvp, vp, dentry, mode);
+ xfs_cleanup_inode(dvp, vp, dentry, mode);
}
_ACL_FREE(default_acl);
}
- if (!error) {
+ if (likely(!error)) {
ASSERT(vp);
- ip = LINVFS_GET_IP(vp);
+ ip = vn_to_inode(vp);
if (S_ISCHR(mode) || S_ISBLK(mode))
ip->i_rdev = rdev;
else if (S_ISDIR(mode))
- validate_fields(ip);
+ xfs_validate_fields(ip, &vattr);
d_instantiate(dentry, ip);
- validate_fields(dir);
+ xfs_validate_fields(dir, &vattr);
}
return -error;
}
STATIC int
-linvfs_create(
+xfs_vn_create(
struct inode *dir,
struct dentry *dentry,
int mode,
struct nameidata *nd)
{
- return linvfs_mknod(dir, dentry, mode, 0);
+ return xfs_vn_mknod(dir, dentry, mode, 0);
}
STATIC int
-linvfs_mkdir(
+xfs_vn_mkdir(
struct inode *dir,
struct dentry *dentry,
int mode)
{
- return linvfs_mknod(dir, dentry, mode|S_IFDIR, 0);
+ return xfs_vn_mknod(dir, dentry, mode|S_IFDIR, 0);
}
STATIC struct dentry *
-linvfs_lookup(
+xfs_vn_lookup(
struct inode *dir,
struct dentry *dentry,
struct nameidata *nd)
{
- struct vnode *vp = LINVFS_GET_VP(dir), *cvp;
+ struct vnode *vp = vn_from_inode(dir), *cvp;
int error;
if (dentry->d_name.len >= MAXNAMELEN)
@@ -411,11 +410,11 @@ linvfs_lookup(
return NULL;
}
- return d_splice_alias(LINVFS_GET_IP(cvp), dentry);
+ return d_splice_alias(vn_to_inode(cvp), dentry);
}
STATIC int
-linvfs_link(
+xfs_vn_link(
struct dentry *old_dentry,
struct inode *dir,
struct dentry *dentry)
@@ -423,99 +422,102 @@ linvfs_link(
struct inode *ip; /* inode of guy being linked to */
vnode_t *tdvp; /* target directory for new name/link */
vnode_t *vp; /* vp of name being linked */
+ vattr_t vattr;
int error;
ip = old_dentry->d_inode; /* inode being linked to */
if (S_ISDIR(ip->i_mode))
return -EPERM;
- tdvp = LINVFS_GET_VP(dir);
- vp = LINVFS_GET_VP(ip);
+ tdvp = vn_from_inode(dir);
+ vp = vn_from_inode(ip);
VOP_LINK(tdvp, vp, dentry, NULL, error);
- if (!error) {
+ if (likely(!error)) {
VMODIFY(tdvp);
VN_HOLD(vp);
- validate_fields(ip);
+ xfs_validate_fields(ip, &vattr);
d_instantiate(dentry, ip);
}
return -error;
}
STATIC int
-linvfs_unlink(
+xfs_vn_unlink(
struct inode *dir,
struct dentry *dentry)
{
struct inode *inode;
vnode_t *dvp; /* directory containing name to remove */
+ vattr_t vattr;
int error;
inode = dentry->d_inode;
- dvp = LINVFS_GET_VP(dir);
+ dvp = vn_from_inode(dir);
VOP_REMOVE(dvp, dentry, NULL, error);
- if (!error) {
- validate_fields(dir); /* For size only */
- validate_fields(inode);
+ if (likely(!error)) {
+ xfs_validate_fields(dir, &vattr); /* size needs update */
+ xfs_validate_fields(inode, &vattr);
}
-
return -error;
}
STATIC int
-linvfs_symlink(
+xfs_vn_symlink(
struct inode *dir,
struct dentry *dentry,
const char *symname)
{
struct inode *ip;
- vattr_t va;
+ vattr_t vattr = { 0 };
vnode_t *dvp; /* directory containing name of symlink */
vnode_t *cvp; /* used to lookup symlink to put in dentry */
int error;
- dvp = LINVFS_GET_VP(dir);
+ dvp = vn_from_inode(dir);
cvp = NULL;
- memset(&va, 0, sizeof(va));
- va.va_mode = S_IFLNK |
+ vattr.va_mode = S_IFLNK |
(irix_symlink_mode ? 0777 & ~current->fs->umask : S_IRWXUGO);
- va.va_mask = XFS_AT_TYPE|XFS_AT_MODE;
+ vattr.va_mask = XFS_AT_TYPE|XFS_AT_MODE;
error = 0;
- VOP_SYMLINK(dvp, dentry, &va, (char *)symname, &cvp, NULL, error);
+ VOP_SYMLINK(dvp, dentry, &vattr, (char *)symname, &cvp, NULL, error);
if (likely(!error && cvp)) {
- error = linvfs_init_security(cvp, dir);
+ error = xfs_init_security(cvp, dir);
if (likely(!error)) {
- ip = LINVFS_GET_IP(cvp);
+ ip = vn_to_inode(cvp);
d_instantiate(dentry, ip);
- validate_fields(dir);
- validate_fields(ip);
+ xfs_validate_fields(dir, &vattr);
+ xfs_validate_fields(ip, &vattr);
+ } else {
+ xfs_cleanup_inode(dvp, cvp, dentry, 0);
}
}
return -error;
}
STATIC int
-linvfs_rmdir(
+xfs_vn_rmdir(
struct inode *dir,
struct dentry *dentry)
{
struct inode *inode = dentry->d_inode;
- vnode_t *dvp = LINVFS_GET_VP(dir);
+ vnode_t *dvp = vn_from_inode(dir);
+ vattr_t vattr;
int error;
VOP_RMDIR(dvp, dentry, NULL, error);
- if (!error) {
- validate_fields(inode);
- validate_fields(dir);
+ if (likely(!error)) {
+ xfs_validate_fields(inode, &vattr);
+ xfs_validate_fields(dir, &vattr);
}
return -error;
}
STATIC int
-linvfs_rename(
+xfs_vn_rename(
struct inode *odir,
struct dentry *odentry,
struct inode *ndir,
@@ -524,22 +526,21 @@ linvfs_rename(
struct inode *new_inode = ndentry->d_inode;
vnode_t *fvp; /* from directory */
vnode_t *tvp; /* target directory */
+ vattr_t vattr;
int error;
- fvp = LINVFS_GET_VP(odir);
- tvp = LINVFS_GET_VP(ndir);
+ fvp = vn_from_inode(odir);
+ tvp = vn_from_inode(ndir);
VOP_RENAME(fvp, odentry, tvp, ndentry, NULL, error);
- if (error)
- return -error;
-
- if (new_inode)
- validate_fields(new_inode);
-
- validate_fields(odir);
- if (ndir != odir)
- validate_fields(ndir);
- return 0;
+ if (likely(!error)) {
+ if (new_inode)
+ xfs_validate_fields(new_inode, &vattr);
+ xfs_validate_fields(odir, &vattr);
+ if (ndir != odir)
+ xfs_validate_fields(ndir, &vattr);
+ }
+ return -error;
}
/*
@@ -548,7 +549,7 @@ linvfs_rename(
* uio is kmalloced for this reason...
*/
STATIC void *
-linvfs_follow_link(
+xfs_vn_follow_link(
struct dentry *dentry,
struct nameidata *nd)
{
@@ -574,7 +575,7 @@ linvfs_follow_link(
return NULL;
}
- vp = LINVFS_GET_VP(dentry->d_inode);
+ vp = vn_from_inode(dentry->d_inode);
iov.iov_base = link;
iov.iov_len = MAXPATHLEN;
@@ -599,7 +600,7 @@ linvfs_follow_link(
}
STATIC void
-linvfs_put_link(
+xfs_vn_put_link(
struct dentry *dentry,
struct nameidata *nd,
void *p)
@@ -612,12 +613,12 @@ linvfs_put_link(
#ifdef CONFIG_XFS_POSIX_ACL
STATIC int
-linvfs_permission(
+xfs_vn_permission(
struct inode *inode,
int mode,
struct nameidata *nd)
{
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
int error;
mode <<= 6; /* convert from linux to vnode access bits */
@@ -625,17 +626,17 @@ linvfs_permission(
return -error;
}
#else
-#define linvfs_permission NULL
+#define xfs_vn_permission NULL
#endif
STATIC int
-linvfs_getattr(
+xfs_vn_getattr(
struct vfsmount *mnt,
struct dentry *dentry,
struct kstat *stat)
{
struct inode *inode = dentry->d_inode;
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
int error = 0;
if (unlikely(vp->v_flag & VMODIFIED))
@@ -646,18 +647,17 @@ linvfs_getattr(
}
STATIC int
-linvfs_setattr(
+xfs_vn_setattr(
struct dentry *dentry,
struct iattr *attr)
{
struct inode *inode = dentry->d_inode;
unsigned int ia_valid = attr->ia_valid;
- vnode_t *vp = LINVFS_GET_VP(inode);
- vattr_t vattr;
+ vnode_t *vp = vn_from_inode(inode);
+ vattr_t vattr = { 0 };
int flags = 0;
int error;
- memset(&vattr, 0, sizeof(vattr_t));
if (ia_valid & ATTR_UID) {
vattr.va_mask |= XFS_AT_UID;
vattr.va_uid = attr->ia_uid;
@@ -699,28 +699,27 @@ linvfs_setattr(
#endif
VOP_SETATTR(vp, &vattr, flags, NULL, error);
- if (error)
- return -error;
- vn_revalidate(vp);
- return error;
+ if (likely(!error))
+ __vn_revalidate(vp, &vattr);
+ return -error;
}
STATIC void
-linvfs_truncate(
+xfs_vn_truncate(
struct inode *inode)
{
- block_truncate_page(inode->i_mapping, inode->i_size, linvfs_get_block);
+ block_truncate_page(inode->i_mapping, inode->i_size, xfs_get_block);
}
STATIC int
-linvfs_setxattr(
+xfs_vn_setxattr(
struct dentry *dentry,
const char *name,
const void *data,
size_t size,
int flags)
{
- vnode_t *vp = LINVFS_GET_VP(dentry->d_inode);
+ vnode_t *vp = vn_from_inode(dentry->d_inode);
char *attr = (char *)name;
attrnames_t *namesp;
int xflags = 0;
@@ -744,13 +743,13 @@ linvfs_setxattr(
}
STATIC ssize_t
-linvfs_getxattr(
+xfs_vn_getxattr(
struct dentry *dentry,
const char *name,
void *data,
size_t size)
{
- vnode_t *vp = LINVFS_GET_VP(dentry->d_inode);
+ vnode_t *vp = vn_from_inode(dentry->d_inode);
char *attr = (char *)name;
attrnames_t *namesp;
int xflags = 0;
@@ -774,12 +773,12 @@ linvfs_getxattr(
}
STATIC ssize_t
-linvfs_listxattr(
+xfs_vn_listxattr(
struct dentry *dentry,
char *data,
size_t size)
{
- vnode_t *vp = LINVFS_GET_VP(dentry->d_inode);
+ vnode_t *vp = vn_from_inode(dentry->d_inode);
int error, xflags = ATTR_KERNAMELS;
ssize_t result;
@@ -794,11 +793,11 @@ linvfs_listxattr(
}
STATIC int
-linvfs_removexattr(
+xfs_vn_removexattr(
struct dentry *dentry,
const char *name)
{
- vnode_t *vp = LINVFS_GET_VP(dentry->d_inode);
+ vnode_t *vp = vn_from_inode(dentry->d_inode);
char *attr = (char *)name;
attrnames_t *namesp;
int xflags = 0;
@@ -816,45 +815,45 @@ linvfs_removexattr(
}
-struct inode_operations linvfs_file_inode_operations = {
- .permission = linvfs_permission,
- .truncate = linvfs_truncate,
- .getattr = linvfs_getattr,
- .setattr = linvfs_setattr,
- .setxattr = linvfs_setxattr,
- .getxattr = linvfs_getxattr,
- .listxattr = linvfs_listxattr,
- .removexattr = linvfs_removexattr,
+struct inode_operations xfs_inode_operations = {
+ .permission = xfs_vn_permission,
+ .truncate = xfs_vn_truncate,
+ .getattr = xfs_vn_getattr,
+ .setattr = xfs_vn_setattr,
+ .setxattr = xfs_vn_setxattr,
+ .getxattr = xfs_vn_getxattr,
+ .listxattr = xfs_vn_listxattr,
+ .removexattr = xfs_vn_removexattr,
};
-struct inode_operations linvfs_dir_inode_operations = {
- .create = linvfs_create,
- .lookup = linvfs_lookup,
- .link = linvfs_link,
- .unlink = linvfs_unlink,
- .symlink = linvfs_symlink,
- .mkdir = linvfs_mkdir,
- .rmdir = linvfs_rmdir,
- .mknod = linvfs_mknod,
- .rename = linvfs_rename,
- .permission = linvfs_permission,
- .getattr = linvfs_getattr,
- .setattr = linvfs_setattr,
- .setxattr = linvfs_setxattr,
- .getxattr = linvfs_getxattr,
- .listxattr = linvfs_listxattr,
- .removexattr = linvfs_removexattr,
+struct inode_operations xfs_dir_inode_operations = {
+ .create = xfs_vn_create,
+ .lookup = xfs_vn_lookup,
+ .link = xfs_vn_link,
+ .unlink = xfs_vn_unlink,
+ .symlink = xfs_vn_symlink,
+ .mkdir = xfs_vn_mkdir,
+ .rmdir = xfs_vn_rmdir,
+ .mknod = xfs_vn_mknod,
+ .rename = xfs_vn_rename,
+ .permission = xfs_vn_permission,
+ .getattr = xfs_vn_getattr,
+ .setattr = xfs_vn_setattr,
+ .setxattr = xfs_vn_setxattr,
+ .getxattr = xfs_vn_getxattr,
+ .listxattr = xfs_vn_listxattr,
+ .removexattr = xfs_vn_removexattr,
};
-struct inode_operations linvfs_symlink_inode_operations = {
+struct inode_operations xfs_symlink_inode_operations = {
.readlink = generic_readlink,
- .follow_link = linvfs_follow_link,
- .put_link = linvfs_put_link,
- .permission = linvfs_permission,
- .getattr = linvfs_getattr,
- .setattr = linvfs_setattr,
- .setxattr = linvfs_setxattr,
- .getxattr = linvfs_getxattr,
- .listxattr = linvfs_listxattr,
- .removexattr = linvfs_removexattr,
+ .follow_link = xfs_vn_follow_link,
+ .put_link = xfs_vn_put_link,
+ .permission = xfs_vn_permission,
+ .getattr = xfs_vn_getattr,
+ .setattr = xfs_vn_setattr,
+ .setxattr = xfs_vn_setxattr,
+ .getxattr = xfs_vn_getxattr,
+ .listxattr = xfs_vn_listxattr,
+ .removexattr = xfs_vn_removexattr,
};
diff --git a/fs/xfs/linux-2.6/xfs_iops.h b/fs/xfs/linux-2.6/xfs_iops.h
index 6899a6b4a50a..a8417d7af5f9 100644
--- a/fs/xfs/linux-2.6/xfs_iops.h
+++ b/fs/xfs/linux-2.6/xfs_iops.h
@@ -18,13 +18,13 @@
#ifndef __XFS_IOPS_H__
#define __XFS_IOPS_H__
-extern struct inode_operations linvfs_file_inode_operations;
-extern struct inode_operations linvfs_dir_inode_operations;
-extern struct inode_operations linvfs_symlink_inode_operations;
+extern struct inode_operations xfs_inode_operations;
+extern struct inode_operations xfs_dir_inode_operations;
+extern struct inode_operations xfs_symlink_inode_operations;
-extern struct file_operations linvfs_file_operations;
-extern struct file_operations linvfs_invis_file_operations;
-extern struct file_operations linvfs_dir_operations;
+extern struct file_operations xfs_file_operations;
+extern struct file_operations xfs_dir_file_operations;
+extern struct file_operations xfs_invis_file_operations;
extern int xfs_ioctl(struct bhv_desc *, struct inode *, struct file *,
int, unsigned int, void __user *);
diff --git a/fs/xfs/linux-2.6/xfs_linux.h b/fs/xfs/linux-2.6/xfs_linux.h
index 67389b745526..1fe09f2d6519 100644
--- a/fs/xfs/linux-2.6/xfs_linux.h
+++ b/fs/xfs/linux-2.6/xfs_linux.h
@@ -73,6 +73,9 @@
#include <linux/list.h>
#include <linux/proc_fs.h>
#include <linux/sort.h>
+#include <linux/cpu.h>
+#include <linux/notifier.h>
+#include <linux/delay.h>
#include <asm/page.h>
#include <asm/div64.h>
@@ -100,6 +103,11 @@
*/
#undef HAVE_REFCACHE /* reference cache not needed for NFS in 2.6 */
#define HAVE_SENDFILE /* sendfile(2) exists in 2.6, but not in 2.4 */
+#ifdef CONFIG_SMP
+#define HAVE_PERCPU_SB /* per cpu superblock counters are a 2.6 feature */
+#else
+#undef HAVE_PERCPU_SB /* per cpu superblock counters are a 2.6 feature */
+#endif
/*
* State flag for unwritten extent buffers.
@@ -226,7 +234,7 @@ BUFFER_FNS(PrivateStart, unwritten);
#define xfs_sort(a,n,s,fn) sort(a,n,s,fn,NULL)
#define xfs_stack_trace() dump_stack()
#define xfs_itruncate_data(ip, off) \
- (-vmtruncate(LINVFS_GET_IP(XFS_ITOV(ip)), (off)))
+ (-vmtruncate(vn_to_inode(XFS_ITOV(ip)), (off)))
#define xfs_statvfs_fsid(statp, mp) \
({ u64 id = huge_encode_dev((mp)->m_ddev_targp->bt_dev); \
__kernel_fsid_t *fsid = &(statp)->f_fsid; \
diff --git a/fs/xfs/linux-2.6/xfs_lrw.c b/fs/xfs/linux-2.6/xfs_lrw.c
index e0ab45fbfebd..0169360475c4 100644
--- a/fs/xfs/linux-2.6/xfs_lrw.c
+++ b/fs/xfs/linux-2.6/xfs_lrw.c
@@ -83,7 +83,7 @@ xfs_rw_enter_trace(
(void *)((unsigned long)ioflags),
(void *)((unsigned long)((io->io_new_size >> 32) & 0xffffffff)),
(void *)((unsigned long)(io->io_new_size & 0xffffffff)),
- (void *)NULL,
+ (void *)((unsigned long)current_pid()),
(void *)NULL,
(void *)NULL,
(void *)NULL,
@@ -113,7 +113,7 @@ xfs_inval_cached_trace(
(void *)((unsigned long)(first & 0xffffffff)),
(void *)((unsigned long)((last >> 32) & 0xffffffff)),
(void *)((unsigned long)(last & 0xffffffff)),
- (void *)NULL,
+ (void *)((unsigned long)current_pid()),
(void *)NULL,
(void *)NULL,
(void *)NULL,
@@ -249,9 +249,8 @@ xfs_read(
if (n < size)
size = n;
- if (XFS_FORCED_SHUTDOWN(mp)) {
+ if (XFS_FORCED_SHUTDOWN(mp))
return -EIO;
- }
if (unlikely(ioflags & IO_ISDIRECT))
mutex_lock(&inode->i_mutex);
@@ -267,10 +266,14 @@ xfs_read(
dmflags, &locktype);
if (ret) {
xfs_iunlock(ip, XFS_IOLOCK_SHARED);
- goto unlock_isem;
+ goto unlock_mutex;
}
}
+ if (unlikely((ioflags & IO_ISDIRECT) && VN_CACHED(vp)))
+ VOP_FLUSHINVAL_PAGES(vp, ctooff(offtoct(*offset)),
+ -1, FI_REMAPF_LOCKED);
+
xfs_rw_enter_trace(XFS_READ_ENTER, &ip->i_iocore,
(void *)iovp, segs, *offset, ioflags);
ret = __generic_file_aio_read(iocb, iovp, segs, offset);
@@ -281,7 +284,7 @@ xfs_read(
xfs_iunlock(ip, XFS_IOLOCK_SHARED);
-unlock_isem:
+unlock_mutex:
if (unlikely(ioflags & IO_ISDIRECT))
mutex_unlock(&inode->i_mutex);
return ret;
@@ -432,7 +435,7 @@ xfs_zero_eof(
xfs_fsize_t isize, /* current inode size */
xfs_fsize_t end_size) /* terminal inode size */
{
- struct inode *ip = LINVFS_GET_IP(vp);
+ struct inode *ip = vn_to_inode(vp);
xfs_fileoff_t start_zero_fsb;
xfs_fileoff_t end_zero_fsb;
xfs_fileoff_t zero_count_fsb;
@@ -573,7 +576,7 @@ xfs_write(
vrwlock_t locktype;
size_t ocount = 0, count;
loff_t pos;
- int need_isem = 1, need_flush = 0;
+ int need_i_mutex = 1, need_flush = 0;
XFS_STATS_INC(xs_write_calls);
@@ -622,14 +625,14 @@ xfs_write(
return XFS_ERROR(-EINVAL);
if (!VN_CACHED(vp) && pos < i_size_read(inode))
- need_isem = 0;
+ need_i_mutex = 0;
if (VN_CACHED(vp))
need_flush = 1;
}
relock:
- if (need_isem) {
+ if (need_i_mutex) {
iolock = XFS_IOLOCK_EXCL;
locktype = VRWLOCK_WRITE;
@@ -651,7 +654,7 @@ start:
S_ISBLK(inode->i_mode));
if (error) {
xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
- goto out_unlock_isem;
+ goto out_unlock_mutex;
}
new_size = pos + count;
@@ -663,7 +666,7 @@ start:
loff_t savedsize = pos;
int dmflags = FILP_DELAY_FLAG(file);
- if (need_isem)
+ if (need_i_mutex)
dmflags |= DM_FLAGS_IMUX;
xfs_iunlock(xip, XFS_ILOCK_EXCL);
@@ -672,7 +675,7 @@ start:
dmflags, &locktype);
if (error) {
xfs_iunlock(xip, iolock);
- goto out_unlock_isem;
+ goto out_unlock_mutex;
}
xfs_ilock(xip, XFS_ILOCK_EXCL);
eventsent = 1;
@@ -710,7 +713,7 @@ start:
isize, pos + count);
if (error) {
xfs_iunlock(xip, XFS_ILOCK_EXCL|iolock);
- goto out_unlock_isem;
+ goto out_unlock_mutex;
}
}
xfs_iunlock(xip, XFS_ILOCK_EXCL);
@@ -731,7 +734,7 @@ start:
error = -remove_suid(file->f_dentry);
if (unlikely(error)) {
xfs_iunlock(xip, iolock);
- goto out_unlock_isem;
+ goto out_unlock_mutex;
}
}
@@ -747,14 +750,14 @@ retry:
-1, FI_REMAPF_LOCKED);
}
- if (need_isem) {
+ if (need_i_mutex) {
/* demote the lock now the cached pages are gone */
XFS_ILOCK_DEMOTE(mp, io, XFS_IOLOCK_EXCL);
mutex_unlock(&inode->i_mutex);
iolock = XFS_IOLOCK_SHARED;
locktype = VRWLOCK_WRITE_DIRECT;
- need_isem = 0;
+ need_i_mutex = 0;
}
xfs_rw_enter_trace(XFS_DIOWR_ENTER, io, (void *)iovp, segs,
@@ -772,7 +775,7 @@ retry:
pos += ret;
count -= ret;
- need_isem = 1;
+ need_i_mutex = 1;
ioflags &= ~IO_ISDIRECT;
xfs_iunlock(xip, iolock);
goto relock;
@@ -794,14 +797,14 @@ retry:
!(ioflags & IO_INVIS)) {
xfs_rwunlock(bdp, locktype);
- if (need_isem)
+ if (need_i_mutex)
mutex_unlock(&inode->i_mutex);
error = XFS_SEND_NAMESP(xip->i_mount, DM_EVENT_NOSPACE, vp,
DM_RIGHT_NULL, vp, DM_RIGHT_NULL, NULL, NULL,
0, 0, 0); /* Delay flag intentionally unused */
if (error)
goto out_nounlocks;
- if (need_isem)
+ if (need_i_mutex)
mutex_lock(&inode->i_mutex);
xfs_rwlock(bdp, locktype);
pos = xip->i_d.di_size;
@@ -905,9 +908,9 @@ retry:
if (error)
goto out_unlock_internal;
}
-
+
xfs_rwunlock(bdp, locktype);
- if (need_isem)
+ if (need_i_mutex)
mutex_unlock(&inode->i_mutex);
error = sync_page_range(inode, mapping, pos, ret);
@@ -918,8 +921,8 @@ retry:
out_unlock_internal:
xfs_rwunlock(bdp, locktype);
- out_unlock_isem:
- if (need_isem)
+ out_unlock_mutex:
+ if (need_i_mutex)
mutex_unlock(&inode->i_mutex);
out_nounlocks:
return -error;
diff --git a/fs/xfs/linux-2.6/xfs_stats.c b/fs/xfs/linux-2.6/xfs_stats.c
index 8955720a2c6b..713e6a7505d0 100644
--- a/fs/xfs/linux-2.6/xfs_stats.c
+++ b/fs/xfs/linux-2.6/xfs_stats.c
@@ -62,18 +62,15 @@ xfs_read_xfsstats(
while (j < xstats[i].endpoint) {
val = 0;
/* sum over all cpus */
- for (c = 0; c < NR_CPUS; c++) {
- if (!cpu_possible(c)) continue;
+ for_each_cpu(c)
val += *(((__u32*)&per_cpu(xfsstats, c) + j));
- }
len += sprintf(buffer + len, " %u", val);
j++;
}
buffer[len++] = '\n';
}
/* extra precision counters */
- for (i = 0; i < NR_CPUS; i++) {
- if (!cpu_possible(i)) continue;
+ for_each_cpu(i) {
xs_xstrat_bytes += per_cpu(xfsstats, i).xs_xstrat_bytes;
xs_write_bytes += per_cpu(xfsstats, i).xs_write_bytes;
xs_read_bytes += per_cpu(xfsstats, i).xs_read_bytes;
diff --git a/fs/xfs/linux-2.6/xfs_super.c b/fs/xfs/linux-2.6/xfs_super.c
index f22e426d9e42..8355faf8ffde 100644
--- a/fs/xfs/linux-2.6/xfs_super.c
+++ b/fs/xfs/linux-2.6/xfs_super.c
@@ -59,8 +59,8 @@
#include <linux/writeback.h>
#include <linux/kthread.h>
-STATIC struct quotactl_ops linvfs_qops;
-STATIC struct super_operations linvfs_sops;
+STATIC struct quotactl_ops xfs_quotactl_operations;
+STATIC struct super_operations xfs_super_operations;
STATIC kmem_zone_t *xfs_vnode_zone;
STATIC kmem_zone_t *xfs_ioend_zone;
mempool_t *xfs_ioend_pool;
@@ -76,8 +76,6 @@ xfs_args_allocate(
strncpy(args->fsname, sb->s_id, MAXNAMELEN);
/* Copy the already-parsed mount(2) flags we're interested in */
- if (sb->s_flags & MS_NOATIME)
- args->flags |= XFSMNT_NOATIME;
if (sb->s_flags & MS_DIRSYNC)
args->flags |= XFSMNT_DIRSYNC;
if (sb->s_flags & MS_SYNCHRONOUS)
@@ -129,21 +127,21 @@ xfs_set_inodeops(
{
switch (inode->i_mode & S_IFMT) {
case S_IFREG:
- inode->i_op = &linvfs_file_inode_operations;
- inode->i_fop = &linvfs_file_operations;
- inode->i_mapping->a_ops = &linvfs_aops;
+ inode->i_op = &xfs_inode_operations;
+ inode->i_fop = &xfs_file_operations;
+ inode->i_mapping->a_ops = &xfs_address_space_operations;
break;
case S_IFDIR:
- inode->i_op = &linvfs_dir_inode_operations;
- inode->i_fop = &linvfs_dir_operations;
+ inode->i_op = &xfs_dir_inode_operations;
+ inode->i_fop = &xfs_dir_file_operations;
break;
case S_IFLNK:
- inode->i_op = &linvfs_symlink_inode_operations;
+ inode->i_op = &xfs_symlink_inode_operations;
if (inode->i_blocks)
- inode->i_mapping->a_ops = &linvfs_aops;
+ inode->i_mapping->a_ops = &xfs_address_space_operations;
break;
default:
- inode->i_op = &linvfs_file_inode_operations;
+ inode->i_op = &xfs_inode_operations;
init_special_inode(inode, inode->i_mode, inode->i_rdev);
break;
}
@@ -155,7 +153,7 @@ xfs_revalidate_inode(
vnode_t *vp,
xfs_inode_t *ip)
{
- struct inode *inode = LINVFS_GET_IP(vp);
+ struct inode *inode = vn_to_inode(vp);
inode->i_mode = ip->i_d.di_mode;
inode->i_nlink = ip->i_d.di_nlink;
@@ -212,7 +210,7 @@ xfs_initialize_vnode(
int unlock)
{
xfs_inode_t *ip = XFS_BHVTOI(inode_bhv);
- struct inode *inode = LINVFS_GET_IP(vp);
+ struct inode *inode = vn_to_inode(vp);
if (!inode_bhv->bd_vobj) {
vp->v_vfsp = bhvtovfs(bdp);
@@ -230,7 +228,7 @@ xfs_initialize_vnode(
if (ip->i_d.di_mode != 0 && unlock && (inode->i_state & I_NEW)) {
xfs_revalidate_inode(XFS_BHVTOM(bdp), vp, ip);
xfs_set_inodeops(inode);
-
+
ip->i_flags &= ~XFS_INEW;
barrier();
@@ -334,43 +332,42 @@ xfs_blkdev_issue_flush(
}
STATIC struct inode *
-linvfs_alloc_inode(
+xfs_fs_alloc_inode(
struct super_block *sb)
{
vnode_t *vp;
- vp = kmem_cache_alloc(xfs_vnode_zone, kmem_flags_convert(KM_SLEEP));
- if (!vp)
+ vp = kmem_zone_alloc(xfs_vnode_zone, KM_SLEEP);
+ if (unlikely(!vp))
return NULL;
- return LINVFS_GET_IP(vp);
+ return vn_to_inode(vp);
}
STATIC void
-linvfs_destroy_inode(
+xfs_fs_destroy_inode(
struct inode *inode)
{
- kmem_zone_free(xfs_vnode_zone, LINVFS_GET_VP(inode));
+ kmem_zone_free(xfs_vnode_zone, vn_from_inode(inode));
}
STATIC void
-linvfs_inode_init_once(
- void *data,
- kmem_cache_t *cachep,
+xfs_fs_inode_init_once(
+ void *vnode,
+ kmem_zone_t *zonep,
unsigned long flags)
{
- vnode_t *vp = (vnode_t *)data;
-
if ((flags & (SLAB_CTOR_VERIFY|SLAB_CTOR_CONSTRUCTOR)) ==
- SLAB_CTOR_CONSTRUCTOR)
- inode_init_once(LINVFS_GET_IP(vp));
+ SLAB_CTOR_CONSTRUCTOR)
+ inode_init_once(vn_to_inode((vnode_t *)vnode));
}
STATIC int
-linvfs_init_zones(void)
+xfs_init_zones(void)
{
- xfs_vnode_zone = kmem_cache_create("xfs_vnode",
- sizeof(vnode_t), 0, SLAB_RECLAIM_ACCOUNT,
- linvfs_inode_init_once, NULL);
+ xfs_vnode_zone = kmem_zone_init_flags(sizeof(vnode_t), "xfs_vnode_t",
+ KM_ZONE_HWALIGN | KM_ZONE_RECLAIM |
+ KM_ZONE_SPREAD,
+ xfs_fs_inode_init_once);
if (!xfs_vnode_zone)
goto out;
@@ -379,14 +376,12 @@ linvfs_init_zones(void)
goto out_destroy_vnode_zone;
xfs_ioend_pool = mempool_create(4 * MAX_BUF_PER_PAGE,
- mempool_alloc_slab, mempool_free_slab,
- xfs_ioend_zone);
+ mempool_alloc_slab, mempool_free_slab,
+ xfs_ioend_zone);
if (!xfs_ioend_pool)
goto out_free_ioend_zone;
-
return 0;
-
out_free_ioend_zone:
kmem_zone_destroy(xfs_ioend_zone);
out_destroy_vnode_zone:
@@ -396,7 +391,7 @@ linvfs_init_zones(void)
}
STATIC void
-linvfs_destroy_zones(void)
+xfs_destroy_zones(void)
{
mempool_destroy(xfs_ioend_pool);
kmem_zone_destroy(xfs_vnode_zone);
@@ -407,14 +402,14 @@ linvfs_destroy_zones(void)
* Attempt to flush the inode, this will actually fail
* if the inode is pinned, but we dirty the inode again
* at the point when it is unpinned after a log write,
- * since this is when the inode itself becomes flushable.
+ * since this is when the inode itself becomes flushable.
*/
STATIC int
-linvfs_write_inode(
+xfs_fs_write_inode(
struct inode *inode,
int sync)
{
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
int error = 0, flags = FLUSH_INODE;
if (vp) {
@@ -434,13 +429,13 @@ linvfs_write_inode(
}
STATIC void
-linvfs_clear_inode(
+xfs_fs_clear_inode(
struct inode *inode)
{
- vnode_t *vp = LINVFS_GET_VP(inode);
+ vnode_t *vp = vn_from_inode(inode);
int error, cache;
- vn_trace_entry(vp, "clear_inode", (inst_t *)__return_address);
+ vn_trace_entry(vp, __FUNCTION__, (inst_t *)__return_address);
XFS_STATS_INC(vn_rele);
XFS_STATS_INC(vn_remove);
@@ -516,7 +511,7 @@ void
xfs_flush_inode(
xfs_inode_t *ip)
{
- struct inode *inode = LINVFS_GET_IP(XFS_ITOV(ip));
+ struct inode *inode = vn_to_inode(XFS_ITOV(ip));
struct vfs *vfs = XFS_MTOVFS(ip->i_mount);
igrab(inode);
@@ -541,7 +536,7 @@ void
xfs_flush_device(
xfs_inode_t *ip)
{
- struct inode *inode = LINVFS_GET_IP(XFS_ITOV(ip));
+ struct inode *inode = vn_to_inode(XFS_ITOV(ip));
struct vfs *vfs = XFS_MTOVFS(ip->i_mount);
igrab(inode);
@@ -550,7 +545,7 @@ xfs_flush_device(
xfs_log_force(ip->i_mount, (xfs_lsn_t)0, XFS_LOG_FORCE|XFS_LOG_SYNC);
}
-#define SYNCD_FLAGS (SYNC_FSDATA|SYNC_BDFLUSH|SYNC_ATTR)
+#define SYNCD_FLAGS (SYNC_FSDATA|SYNC_BDFLUSH|SYNC_ATTR|SYNC_REFCACHE)
STATIC void
vfs_sync_worker(
vfs_t *vfsp,
@@ -613,7 +608,7 @@ xfssyncd(
}
STATIC int
-linvfs_start_syncd(
+xfs_fs_start_syncd(
vfs_t *vfsp)
{
vfsp->vfs_sync_work.w_syncer = vfs_sync_worker;
@@ -625,20 +620,20 @@ linvfs_start_syncd(
}
STATIC void
-linvfs_stop_syncd(
+xfs_fs_stop_syncd(
vfs_t *vfsp)
{
kthread_stop(vfsp->vfs_sync_task);
}
STATIC void
-linvfs_put_super(
+xfs_fs_put_super(
struct super_block *sb)
{
- vfs_t *vfsp = LINVFS_GET_VFS(sb);
+ vfs_t *vfsp = vfs_from_sb(sb);
int error;
- linvfs_stop_syncd(vfsp);
+ xfs_fs_stop_syncd(vfsp);
VFS_SYNC(vfsp, SYNC_ATTR|SYNC_DELWRI, NULL, error);
if (!error)
VFS_UNMOUNT(vfsp, 0, NULL, error);
@@ -652,10 +647,10 @@ linvfs_put_super(
}
STATIC void
-linvfs_write_super(
+xfs_fs_write_super(
struct super_block *sb)
{
- vfs_t *vfsp = LINVFS_GET_VFS(sb);
+ vfs_t *vfsp = vfs_from_sb(sb);
int error;
if (sb->s_flags & MS_RDONLY) {
@@ -668,11 +663,11 @@ linvfs_write_super(
}
STATIC int
-linvfs_sync_super(
+xfs_fs_sync_super(
struct super_block *sb,
int wait)
{
- vfs_t *vfsp = LINVFS_GET_VFS(sb);
+ vfs_t *vfsp = vfs_from_sb(sb);
int error;
int flags = SYNC_FSDATA;
@@ -707,11 +702,11 @@ linvfs_sync_super(
}
STATIC int
-linvfs_statfs(
+xfs_fs_statfs(
struct super_block *sb,
struct kstatfs *statp)
{
- vfs_t *vfsp = LINVFS_GET_VFS(sb);
+ vfs_t *vfsp = vfs_from_sb(sb);
int error;
VFS_STATVFS(vfsp, statp, NULL, error);
@@ -719,12 +714,12 @@ linvfs_statfs(
}
STATIC int
-linvfs_remount(
+xfs_fs_remount(
struct super_block *sb,
int *flags,
char *options)
{
- vfs_t *vfsp = LINVFS_GET_VFS(sb);
+ vfs_t *vfsp = vfs_from_sb(sb);
struct xfs_mount_args *args = xfs_args_allocate(sb);
int error;
@@ -736,18 +731,18 @@ linvfs_remount(
}
STATIC void
-linvfs_freeze_fs(
+xfs_fs_lockfs(
struct super_block *sb)
{
- VFS_FREEZE(LINVFS_GET_VFS(sb));
+ VFS_FREEZE(vfs_from_sb(sb));
}
STATIC int
-linvfs_show_options(
+xfs_fs_show_options(
struct seq_file *m,
struct vfsmount *mnt)
{
- struct vfs *vfsp = LINVFS_GET_VFS(mnt->mnt_sb);
+ struct vfs *vfsp = vfs_from_sb(mnt->mnt_sb);
int error;
VFS_SHOWARGS(vfsp, m, error);
@@ -755,11 +750,11 @@ linvfs_show_options(
}
STATIC int
-linvfs_quotasync(
+xfs_fs_quotasync(
struct super_block *sb,
int type)
{
- struct vfs *vfsp = LINVFS_GET_VFS(sb);
+ struct vfs *vfsp = vfs_from_sb(sb);
int error;
VFS_QUOTACTL(vfsp, Q_XQUOTASYNC, 0, (caddr_t)NULL, error);
@@ -767,11 +762,11 @@ linvfs_quotasync(
}
STATIC int
-linvfs_getxstate(
+xfs_fs_getxstate(
struct super_block *sb,
struct fs_quota_stat *fqs)
{
- struct vfs *vfsp = LINVFS_GET_VFS(sb);
+ struct vfs *vfsp = vfs_from_sb(sb);
int error;
VFS_QUOTACTL(vfsp, Q_XGETQSTAT, 0, (caddr_t)fqs, error);
@@ -779,12 +774,12 @@ linvfs_getxstate(
}
STATIC int
-linvfs_setxstate(
+xfs_fs_setxstate(
struct super_block *sb,
unsigned int flags,
int op)
{
- struct vfs *vfsp = LINVFS_GET_VFS(sb);
+ struct vfs *vfsp = vfs_from_sb(sb);
int error;
VFS_QUOTACTL(vfsp, op, 0, (caddr_t)&flags, error);
@@ -792,13 +787,13 @@ linvfs_setxstate(
}
STATIC int
-linvfs_getxquota(
+xfs_fs_getxquota(
struct super_block *sb,
int type,
qid_t id,
struct fs_disk_quota *fdq)
{
- struct vfs *vfsp = LINVFS_GET_VFS(sb);
+ struct vfs *vfsp = vfs_from_sb(sb);
int error, getmode;
getmode = (type == USRQUOTA) ? Q_XGETQUOTA :
@@ -808,13 +803,13 @@ linvfs_getxquota(
}
STATIC int
-linvfs_setxquota(
+xfs_fs_setxquota(
struct super_block *sb,
int type,
qid_t id,
struct fs_disk_quota *fdq)
{
- struct vfs *vfsp = LINVFS_GET_VFS(sb);
+ struct vfs *vfsp = vfs_from_sb(sb);
int error, setmode;
setmode = (type == USRQUOTA) ? Q_XSETQLIM :
@@ -824,21 +819,17 @@ linvfs_setxquota(
}
STATIC int
-linvfs_fill_super(
+xfs_fs_fill_super(
struct super_block *sb,
void *data,
int silent)
{
vnode_t *rootvp;
- struct vfs *vfsp = vfs_allocate();
+ struct vfs *vfsp = vfs_allocate(sb);
struct xfs_mount_args *args = xfs_args_allocate(sb);
struct kstatfs statvfs;
int error, error2;
- vfsp->vfs_super = sb;
- LINVFS_SET_VFS(sb, vfsp);
- if (sb->s_flags & MS_RDONLY)
- vfsp->vfs_flag |= VFS_RDONLY;
bhv_insert_all_vfsops(vfsp);
VFS_PARSEARGS(vfsp, (char *)data, args, 0, error);
@@ -849,10 +840,10 @@ linvfs_fill_super(
sb_min_blocksize(sb, BBSIZE);
#ifdef CONFIG_XFS_EXPORT
- sb->s_export_op = &linvfs_export_ops;
+ sb->s_export_op = &xfs_export_operations;
#endif
- sb->s_qcop = &linvfs_qops;
- sb->s_op = &linvfs_sops;
+ sb->s_qcop = &xfs_quotactl_operations;
+ sb->s_op = &xfs_super_operations;
VFS_MOUNT(vfsp, args, NULL, error);
if (error) {
@@ -876,7 +867,7 @@ linvfs_fill_super(
if (error)
goto fail_unmount;
- sb->s_root = d_alloc_root(LINVFS_GET_IP(rootvp));
+ sb->s_root = d_alloc_root(vn_to_inode(rootvp));
if (!sb->s_root) {
error = ENOMEM;
goto fail_vnrele;
@@ -885,7 +876,7 @@ linvfs_fill_super(
error = EINVAL;
goto fail_vnrele;
}
- if ((error = linvfs_start_syncd(vfsp)))
+ if ((error = xfs_fs_start_syncd(vfsp)))
goto fail_vnrele;
vn_trace_exit(rootvp, __FUNCTION__, (inst_t *)__return_address);
@@ -910,41 +901,41 @@ fail_vfsop:
}
STATIC struct super_block *
-linvfs_get_sb(
+xfs_fs_get_sb(
struct file_system_type *fs_type,
int flags,
const char *dev_name,
void *data)
{
- return get_sb_bdev(fs_type, flags, dev_name, data, linvfs_fill_super);
-}
-
-STATIC struct super_operations linvfs_sops = {
- .alloc_inode = linvfs_alloc_inode,
- .destroy_inode = linvfs_destroy_inode,
- .write_inode = linvfs_write_inode,
- .clear_inode = linvfs_clear_inode,
- .put_super = linvfs_put_super,
- .write_super = linvfs_write_super,
- .sync_fs = linvfs_sync_super,
- .write_super_lockfs = linvfs_freeze_fs,
- .statfs = linvfs_statfs,
- .remount_fs = linvfs_remount,
- .show_options = linvfs_show_options,
+ return get_sb_bdev(fs_type, flags, dev_name, data, xfs_fs_fill_super);
+}
+
+STATIC struct super_operations xfs_super_operations = {
+ .alloc_inode = xfs_fs_alloc_inode,
+ .destroy_inode = xfs_fs_destroy_inode,
+ .write_inode = xfs_fs_write_inode,
+ .clear_inode = xfs_fs_clear_inode,
+ .put_super = xfs_fs_put_super,
+ .write_super = xfs_fs_write_super,
+ .sync_fs = xfs_fs_sync_super,
+ .write_super_lockfs = xfs_fs_lockfs,
+ .statfs = xfs_fs_statfs,
+ .remount_fs = xfs_fs_remount,
+ .show_options = xfs_fs_show_options,
};
-STATIC struct quotactl_ops linvfs_qops = {
- .quota_sync = linvfs_quotasync,
- .get_xstate = linvfs_getxstate,
- .set_xstate = linvfs_setxstate,
- .get_xquota = linvfs_getxquota,
- .set_xquota = linvfs_setxquota,
+STATIC struct quotactl_ops xfs_quotactl_operations = {
+ .quota_sync = xfs_fs_quotasync,
+ .get_xstate = xfs_fs_getxstate,
+ .set_xstate = xfs_fs_setxstate,
+ .get_xquota = xfs_fs_getxquota,
+ .set_xquota = xfs_fs_setxquota,
};
STATIC struct file_system_type xfs_fs_type = {
.owner = THIS_MODULE,
.name = "xfs",
- .get_sb = linvfs_get_sb,
+ .get_sb = xfs_fs_get_sb,
.kill_sb = kill_block_super,
.fs_flags = FS_REQUIRES_DEV,
};
@@ -965,7 +956,7 @@ init_xfs_fs( void )
ktrace_init(64);
- error = linvfs_init_zones();
+ error = xfs_init_zones();
if (error < 0)
goto undo_zones;
@@ -981,14 +972,13 @@ init_xfs_fs( void )
error = register_filesystem(&xfs_fs_type);
if (error)
goto undo_register;
- XFS_DM_INIT(&xfs_fs_type);
return 0;
undo_register:
xfs_buf_terminate();
undo_buffers:
- linvfs_destroy_zones();
+ xfs_destroy_zones();
undo_zones:
return error;
@@ -998,11 +988,10 @@ STATIC void __exit
exit_xfs_fs( void )
{
vfs_exitquota();
- XFS_DM_EXIT(&xfs_fs_type);
unregister_filesystem(&xfs_fs_type);
xfs_cleanup();
xfs_buf_terminate();
- linvfs_destroy_zones();
+ xfs_destroy_zones();
ktrace_uninit();
}
diff --git a/fs/xfs/linux-2.6/xfs_super.h b/fs/xfs/linux-2.6/xfs_super.h
index df59408dca06..376b96cb513a 100644
--- a/fs/xfs/linux-2.6/xfs_super.h
+++ b/fs/xfs/linux-2.6/xfs_super.h
@@ -98,11 +98,6 @@ extern void xfs_qm_exit(void);
XFS_DMAPI_STRING \
XFS_DBG_STRING /* DBG must be last */
-#define LINVFS_GET_VFS(s) \
- (vfs_t *)((s)->s_fs_info)
-#define LINVFS_SET_VFS(s, vfsp) \
- ((s)->s_fs_info = vfsp)
-
struct xfs_inode;
struct xfs_mount;
struct xfs_buftarg;
@@ -120,6 +115,6 @@ extern int xfs_blkdev_get(struct xfs_mount *, const char *,
extern void xfs_blkdev_put(struct block_device *);
extern void xfs_blkdev_issue_flush(struct xfs_buftarg *);
-extern struct export_operations linvfs_export_ops;
+extern struct export_operations xfs_export_operations;
#endif /* __XFS_SUPER_H__ */
diff --git a/fs/xfs/linux-2.6/xfs_sysctl.c b/fs/xfs/linux-2.6/xfs_sysctl.c
index a02564972420..7079cc837210 100644
--- a/fs/xfs/linux-2.6/xfs_sysctl.c
+++ b/fs/xfs/linux-2.6/xfs_sysctl.c
@@ -38,8 +38,7 @@ xfs_stats_clear_proc_handler(
if (!ret && write && *valp) {
printk("XFS Clearing xfsstats\n");
- for (c = 0; c < NR_CPUS; c++) {
- if (!cpu_possible(c)) continue;
+ for_each_cpu(c) {
preempt_disable();
/* save vn_active, it's a universal truth! */
vn_active = per_cpu(xfsstats, c).vn_active;
diff --git a/fs/xfs/linux-2.6/xfs_vfs.c b/fs/xfs/linux-2.6/xfs_vfs.c
index c855d62e5344..6f7c9f7a8624 100644
--- a/fs/xfs/linux-2.6/xfs_vfs.c
+++ b/fs/xfs/linux-2.6/xfs_vfs.c
@@ -227,7 +227,8 @@ vfs_freeze(
}
vfs_t *
-vfs_allocate( void )
+vfs_allocate(
+ struct super_block *sb)
{
struct vfs *vfsp;
@@ -236,9 +237,23 @@ vfs_allocate( void )
INIT_LIST_HEAD(&vfsp->vfs_sync_list);
spin_lock_init(&vfsp->vfs_sync_lock);
init_waitqueue_head(&vfsp->vfs_wait_single_sync_task);
+
+ vfsp->vfs_super = sb;
+ sb->s_fs_info = vfsp;
+
+ if (sb->s_flags & MS_RDONLY)
+ vfsp->vfs_flag |= VFS_RDONLY;
+
return vfsp;
}
+vfs_t *
+vfs_from_sb(
+ struct super_block *sb)
+{
+ return (vfs_t *)sb->s_fs_info;
+}
+
void
vfs_deallocate(
struct vfs *vfsp)
@@ -295,7 +310,7 @@ bhv_remove_all_vfsops(
bhv_remove_vfsops(vfsp, VFS_POSITION_DM);
if (!freebase)
return;
- mp = XFS_BHVTOM(bhv_lookup(VFS_BHVHEAD(vfsp), &xfs_vfsops));
+ mp = XFS_VFSTOM(vfsp);
VFS_REMOVEBHV(vfsp, &mp->m_bhv);
xfs_mount_free(mp, 0);
}
diff --git a/fs/xfs/linux-2.6/xfs_vfs.h b/fs/xfs/linux-2.6/xfs_vfs.h
index 57caf9eddee0..8fed356db055 100644
--- a/fs/xfs/linux-2.6/xfs_vfs.h
+++ b/fs/xfs/linux-2.6/xfs_vfs.h
@@ -193,7 +193,8 @@ typedef struct bhv_vfsops {
#define vfs_bhv_set_custom(b,o) ( (b)->bhv_custom = (void *)(o))
#define vfs_bhv_clr_custom(b) ( (b)->bhv_custom = NULL )
-extern vfs_t *vfs_allocate(void);
+extern vfs_t *vfs_allocate(struct super_block *);
+extern vfs_t *vfs_from_sb(struct super_block *);
extern void vfs_deallocate(vfs_t *);
extern void vfs_insertops(vfs_t *, bhv_vfsops_t *);
extern void vfs_insertbhv(vfs_t *, bhv_desc_t *, vfsops_t *, void *);
diff --git a/fs/xfs/linux-2.6/xfs_vnode.c b/fs/xfs/linux-2.6/xfs_vnode.c
index 260dd8415dd7..d27c25b27ccd 100644
--- a/fs/xfs/linux-2.6/xfs_vnode.c
+++ b/fs/xfs/linux-2.6/xfs_vnode.c
@@ -58,7 +58,7 @@ struct vnode *
vn_initialize(
struct inode *inode)
{
- struct vnode *vp = LINVFS_GET_VP(inode);
+ struct vnode *vp = vn_from_inode(inode);
XFS_STATS_INC(vn_active);
XFS_STATS_INC(vn_alloc);
@@ -83,7 +83,7 @@ vn_initialize(
vp->v_trace = ktrace_alloc(VNODE_TRACE_SIZE, KM_SLEEP);
#endif /* XFS_VNODE_TRACE */
- vn_trace_exit(vp, "vn_initialize", (inst_t *)__return_address);
+ vn_trace_exit(vp, __FUNCTION__, (inst_t *)__return_address);
return vp;
}
@@ -97,7 +97,7 @@ vn_revalidate_core(
struct vnode *vp,
vattr_t *vap)
{
- struct inode *inode = LINVFS_GET_IP(vp);
+ struct inode *inode = vn_to_inode(vp);
inode->i_mode = vap->va_mode;
inode->i_nlink = vap->va_nlink;
@@ -129,24 +129,31 @@ vn_revalidate_core(
* Revalidate the Linux inode from the vnode.
*/
int
-vn_revalidate(
- struct vnode *vp)
+__vn_revalidate(
+ struct vnode *vp,
+ struct vattr *vattr)
{
- vattr_t va;
int error;
- vn_trace_entry(vp, "vn_revalidate", (inst_t *)__return_address);
- ASSERT(vp->v_fbhv != NULL);
-
- va.va_mask = XFS_AT_STAT|XFS_AT_XFLAGS;
- VOP_GETATTR(vp, &va, 0, NULL, error);
- if (!error) {
- vn_revalidate_core(vp, &va);
+ vn_trace_entry(vp, __FUNCTION__, (inst_t *)__return_address);
+ vattr->va_mask = XFS_AT_STAT | XFS_AT_XFLAGS;
+ VOP_GETATTR(vp, vattr, 0, NULL, error);
+ if (likely(!error)) {
+ vn_revalidate_core(vp, vattr);
VUNMODIFY(vp);
}
return -error;
}
+int
+vn_revalidate(
+ struct vnode *vp)
+{
+ vattr_t vattr;
+
+ return __vn_revalidate(vp, &vattr);
+}
+
/*
* Add a reference to a referenced vnode.
*/
@@ -159,7 +166,7 @@ vn_hold(
XFS_STATS_INC(vn_hold);
VN_LOCK(vp);
- inode = igrab(LINVFS_GET_IP(vp));
+ inode = igrab(vn_to_inode(vp));
ASSERT(inode);
VN_UNLOCK(vp, 0);
diff --git a/fs/xfs/linux-2.6/xfs_vnode.h b/fs/xfs/linux-2.6/xfs_vnode.h
index 0fe2419461d6..06f5845e9568 100644
--- a/fs/xfs/linux-2.6/xfs_vnode.h
+++ b/fs/xfs/linux-2.6/xfs_vnode.h
@@ -116,8 +116,14 @@ typedef enum {
/*
* Vnode to Linux inode mapping.
*/
-#define LINVFS_GET_VP(inode) ((vnode_t *)list_entry(inode, vnode_t, v_inode))
-#define LINVFS_GET_IP(vp) (&(vp)->v_inode)
+static inline struct vnode *vn_from_inode(struct inode *inode)
+{
+ return (vnode_t *)list_entry(inode, vnode_t, v_inode);
+}
+static inline struct inode *vn_to_inode(struct vnode *vnode)
+{
+ return &vnode->v_inode;
+}
/*
* Vnode flags.
@@ -490,6 +496,7 @@ typedef struct vnode_map {
(vmap).v_ino = (vp)->v_inode.i_ino; }
extern int vn_revalidate(struct vnode *);
+extern int __vn_revalidate(struct vnode *, vattr_t *);
extern void vn_revalidate_core(struct vnode *, vattr_t *);
extern void vn_iowait(struct vnode *vp);
@@ -497,7 +504,7 @@ extern void vn_iowake(struct vnode *vp);
static inline int vn_count(struct vnode *vp)
{
- return atomic_read(&LINVFS_GET_IP(vp)->i_count);
+ return atomic_read(&vn_to_inode(vp)->i_count);
}
/*
@@ -511,16 +518,16 @@ extern vnode_t *vn_hold(struct vnode *);
vn_trace_hold(vp, __FILE__, __LINE__, (inst_t *)__return_address))
#define VN_RELE(vp) \
(vn_trace_rele(vp, __FILE__, __LINE__, (inst_t *)__return_address), \
- iput(LINVFS_GET_IP(vp)))
+ iput(vn_to_inode(vp)))
#else
#define VN_HOLD(vp) ((void)vn_hold(vp))
-#define VN_RELE(vp) (iput(LINVFS_GET_IP(vp)))
+#define VN_RELE(vp) (iput(vn_to_inode(vp)))
#endif
static inline struct vnode *vn_grab(struct vnode *vp)
{
- struct inode *inode = igrab(LINVFS_GET_IP(vp));
- return inode ? LINVFS_GET_VP(inode) : NULL;
+ struct inode *inode = igrab(vn_to_inode(vp));
+ return inode ? vn_from_inode(inode) : NULL;
}
/*
@@ -528,7 +535,7 @@ static inline struct vnode *vn_grab(struct vnode *vp)
*/
#define VNAME(dentry) ((char *) (dentry)->d_name.name)
#define VNAMELEN(dentry) ((dentry)->d_name.len)
-#define VNAME_TO_VNODE(dentry) (LINVFS_GET_VP((dentry)->d_inode))
+#define VNAME_TO_VNODE(dentry) (vn_from_inode((dentry)->d_inode))
/*
* Vnode spinlock manipulation.
@@ -557,12 +564,12 @@ static __inline__ void vn_flagclr(struct vnode *vp, uint flag)
*/
static inline void vn_mark_bad(struct vnode *vp)
{
- make_bad_inode(LINVFS_GET_IP(vp));
+ make_bad_inode(vn_to_inode(vp));
}
static inline int VN_BAD(struct vnode *vp)
{
- return is_bad_inode(LINVFS_GET_IP(vp));
+ return is_bad_inode(vn_to_inode(vp));
}
/*
@@ -587,9 +594,9 @@ static inline void vn_atime_to_time_t(struct vnode *vp, time_t *tt)
/*
* Some useful predicates.
*/
-#define VN_MAPPED(vp) mapping_mapped(LINVFS_GET_IP(vp)->i_mapping)
-#define VN_CACHED(vp) (LINVFS_GET_IP(vp)->i_mapping->nrpages)
-#define VN_DIRTY(vp) mapping_tagged(LINVFS_GET_IP(vp)->i_mapping, \
+#define VN_MAPPED(vp) mapping_mapped(vn_to_inode(vp)->i_mapping)
+#define VN_CACHED(vp) (vn_to_inode(vp)->i_mapping->nrpages)
+#define VN_DIRTY(vp) mapping_tagged(vn_to_inode(vp)->i_mapping, \
PAGECACHE_TAG_DIRTY)
#define VMODIFY(vp) VN_FLAGSET(vp, VMODIFIED)
#define VUNMODIFY(vp) VN_FLAGCLR(vp, VMODIFIED)
diff --git a/fs/xfs/quota/xfs_dquot_item.c b/fs/xfs/quota/xfs_dquot_item.c
index 2ec6b441849c..e4e5f05b841b 100644
--- a/fs/xfs/quota/xfs_dquot_item.c
+++ b/fs/xfs/quota/xfs_dquot_item.c
@@ -79,9 +79,11 @@ xfs_qm_dquot_logitem_format(
logvec->i_addr = (xfs_caddr_t)&logitem->qli_format;
logvec->i_len = sizeof(xfs_dq_logformat_t);
+ XLOG_VEC_SET_TYPE(logvec, XLOG_REG_TYPE_QFORMAT);
logvec++;
logvec->i_addr = (xfs_caddr_t)&logitem->qli_dquot->q_core;
logvec->i_len = sizeof(xfs_disk_dquot_t);
+ XLOG_VEC_SET_TYPE(logvec, XLOG_REG_TYPE_DQUOT);
ASSERT(2 == logitem->qli_item.li_desc->lid_size);
logitem->qli_format.qlf_size = 2;
diff --git a/fs/xfs/quota/xfs_qm.c b/fs/xfs/quota/xfs_qm.c
index 7c0e39dc6189..1fb757ef3f41 100644
--- a/fs/xfs/quota/xfs_qm.c
+++ b/fs/xfs/quota/xfs_qm.c
@@ -1704,9 +1704,9 @@ xfs_qm_get_rtblks(
xfs_qcnt_t *O_rtblks)
{
xfs_filblks_t rtblks; /* total rt blks */
+ xfs_extnum_t idx; /* extent record index */
xfs_ifork_t *ifp; /* inode fork pointer */
xfs_extnum_t nextents; /* number of extent entries */
- xfs_bmbt_rec_t *base; /* base of extent array */
xfs_bmbt_rec_t *ep; /* pointer to an extent entry */
int error;
@@ -1717,10 +1717,11 @@ xfs_qm_get_rtblks(
return error;
}
rtblks = 0;
- nextents = ifp->if_bytes / sizeof(xfs_bmbt_rec_t);
- base = &ifp->if_u1.if_extents[0];
- for (ep = base; ep < &base[nextents]; ep++)
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ for (idx = 0; idx < nextents; idx++) {
+ ep = xfs_iext_get_ext(ifp, idx);
rtblks += xfs_bmbt_get_blockcount(ep);
+ }
*O_rtblks = (xfs_qcnt_t)rtblks;
return 0;
}
@@ -2788,9 +2789,7 @@ xfs_qm_freelist_destroy(xfs_frlist_t *ql)
xfs_qm_dqdestroy(dqp);
dqp = nextdqp;
}
- /*
- * Don't bother about unlocking.
- */
+ mutex_unlock(&ql->qh_lock);
mutex_destroy(&ql->qh_lock);
ASSERT(ql->qh_nelems == 0);
diff --git a/fs/xfs/quota/xfs_qm_bhv.c b/fs/xfs/quota/xfs_qm_bhv.c
index 90402a1c3983..6838b36d95a9 100644
--- a/fs/xfs/quota/xfs_qm_bhv.c
+++ b/fs/xfs/quota/xfs_qm_bhv.c
@@ -374,7 +374,7 @@ xfs_qm_exit(void)
vfs_bhv_clr_custom(&xfs_qmops);
xfs_qm_cleanup_procfs();
if (qm_dqzone)
- kmem_cache_destroy(qm_dqzone);
+ kmem_zone_destroy(qm_dqzone);
if (qm_dqtrxzone)
- kmem_cache_destroy(qm_dqtrxzone);
+ kmem_zone_destroy(qm_dqtrxzone);
}
diff --git a/fs/xfs/support/ktrace.c b/fs/xfs/support/ktrace.c
index 841aa4c15b8a..addf5a7ea06c 100644
--- a/fs/xfs/support/ktrace.c
+++ b/fs/xfs/support/ktrace.c
@@ -39,8 +39,8 @@ ktrace_init(int zentries)
void
ktrace_uninit(void)
{
- kmem_cache_destroy(ktrace_hdr_zone);
- kmem_cache_destroy(ktrace_ent_zone);
+ kmem_zone_destroy(ktrace_hdr_zone);
+ kmem_zone_destroy(ktrace_ent_zone);
}
/*
diff --git a/fs/xfs/support/uuid.c b/fs/xfs/support/uuid.c
index a3d565a67734..e157015c70ff 100644
--- a/fs/xfs/support/uuid.c
+++ b/fs/xfs/support/uuid.c
@@ -21,13 +21,6 @@ static mutex_t uuid_monitor;
static int uuid_table_size;
static uuid_t *uuid_table;
-void
-uuid_init(void)
-{
- mutex_init(&uuid_monitor);
-}
-
-
/* IRIX interpretation of an uuid_t */
typedef struct {
__be32 uu_timelow;
@@ -50,7 +43,7 @@ uuid_getnodeuniq(uuid_t *uuid, int fsid [2])
fsid[0] = (be16_to_cpu(uup->uu_clockseq) << 16) |
be16_to_cpu(uup->uu_timemid);
- fsid[1] = be16_to_cpu(uup->uu_timelow);
+ fsid[1] = be32_to_cpu(uup->uu_timelow);
}
void
@@ -139,3 +132,9 @@ uuid_table_remove(uuid_t *uuid)
ASSERT(i < uuid_table_size);
mutex_unlock(&uuid_monitor);
}
+
+void
+uuid_init(void)
+{
+ mutex_init(&uuid_monitor);
+}
diff --git a/fs/xfs/xfs_acl.h b/fs/xfs/xfs_acl.h
index f9315bc960cb..538d0d65b04c 100644
--- a/fs/xfs/xfs_acl.h
+++ b/fs/xfs/xfs_acl.h
@@ -55,8 +55,8 @@ struct xfs_inode;
extern struct kmem_zone *xfs_acl_zone;
#define xfs_acl_zone_init(zone, name) \
- (zone) = kmem_zone_init(sizeof(xfs_acl_t), name)
-#define xfs_acl_zone_destroy(zone) kmem_cache_destroy(zone)
+ (zone) = kmem_zone_init(sizeof(xfs_acl_t), (name))
+#define xfs_acl_zone_destroy(zone) kmem_zone_destroy(zone)
extern int xfs_acl_inherit(struct vnode *, struct vattr *, xfs_acl_t *);
extern int xfs_acl_iaccess(struct xfs_inode *, mode_t, cred_t *);
diff --git a/fs/xfs/xfs_attr.c b/fs/xfs/xfs_attr.c
index e5e91e9c7e89..093fac476bda 100644
--- a/fs/xfs/xfs_attr.c
+++ b/fs/xfs/xfs_attr.c
@@ -1127,8 +1127,7 @@ xfs_attr_leaf_list(xfs_attr_list_context_t *context)
return(error);
ASSERT(bp != NULL);
leaf = bp->data;
- if (unlikely(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- != XFS_ATTR_LEAF_MAGIC)) {
+ if (unlikely(be16_to_cpu(leaf->hdr.info.magic) != XFS_ATTR_LEAF_MAGIC)) {
XFS_CORRUPTION_ERROR("xfs_attr_leaf_list", XFS_ERRLEVEL_LOW,
context->dp->i_mount, leaf);
xfs_da_brelse(NULL, bp);
@@ -1541,8 +1540,8 @@ xfs_attr_node_removename(xfs_da_args_t *args)
XFS_ATTR_FORK);
if (error)
goto out;
- ASSERT(INT_GET(((xfs_attr_leafblock_t *)
- bp->data)->hdr.info.magic, ARCH_CONVERT)
+ ASSERT(be16_to_cpu(((xfs_attr_leafblock_t *)
+ bp->data)->hdr.info.magic)
== XFS_ATTR_LEAF_MAGIC);
if ((forkoff = xfs_attr_shortform_allfit(bp, dp))) {
@@ -1763,7 +1762,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
return(error);
if (bp) {
node = bp->data;
- switch (INT_GET(node->hdr.info.magic, ARCH_CONVERT)) {
+ switch (be16_to_cpu(node->hdr.info.magic)) {
case XFS_DA_NODE_MAGIC:
xfs_attr_trace_l_cn("wrong blk", context, node);
xfs_da_brelse(NULL, bp);
@@ -1771,18 +1770,14 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
break;
case XFS_ATTR_LEAF_MAGIC:
leaf = bp->data;
- if (cursor->hashval >
- INT_GET(leaf->entries[
- INT_GET(leaf->hdr.count,
- ARCH_CONVERT)-1].hashval,
- ARCH_CONVERT)) {
+ if (cursor->hashval > be32_to_cpu(leaf->entries[
+ be16_to_cpu(leaf->hdr.count)-1].hashval)) {
xfs_attr_trace_l_cl("wrong blk",
context, leaf);
xfs_da_brelse(NULL, bp);
bp = NULL;
} else if (cursor->hashval <=
- INT_GET(leaf->entries[0].hashval,
- ARCH_CONVERT)) {
+ be32_to_cpu(leaf->entries[0].hashval)) {
xfs_attr_trace_l_cl("maybe wrong blk",
context, leaf);
xfs_da_brelse(NULL, bp);
@@ -1817,10 +1812,10 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
return(XFS_ERROR(EFSCORRUPTED));
}
node = bp->data;
- if (INT_GET(node->hdr.info.magic, ARCH_CONVERT)
+ if (be16_to_cpu(node->hdr.info.magic)
== XFS_ATTR_LEAF_MAGIC)
break;
- if (unlikely(INT_GET(node->hdr.info.magic, ARCH_CONVERT)
+ if (unlikely(be16_to_cpu(node->hdr.info.magic)
!= XFS_DA_NODE_MAGIC)) {
XFS_CORRUPTION_ERROR("xfs_attr_node_list(3)",
XFS_ERRLEVEL_LOW,
@@ -1830,19 +1825,17 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
return(XFS_ERROR(EFSCORRUPTED));
}
btree = node->btree;
- for (i = 0;
- i < INT_GET(node->hdr.count, ARCH_CONVERT);
+ for (i = 0; i < be16_to_cpu(node->hdr.count);
btree++, i++) {
if (cursor->hashval
- <= INT_GET(btree->hashval,
- ARCH_CONVERT)) {
- cursor->blkno = INT_GET(btree->before, ARCH_CONVERT);
+ <= be32_to_cpu(btree->hashval)) {
+ cursor->blkno = be32_to_cpu(btree->before);
xfs_attr_trace_l_cb("descending",
context, btree);
break;
}
}
- if (i == INT_GET(node->hdr.count, ARCH_CONVERT)) {
+ if (i == be16_to_cpu(node->hdr.count)) {
xfs_da_brelse(NULL, bp);
return(0);
}
@@ -1858,7 +1851,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
*/
for (;;) {
leaf = bp->data;
- if (unlikely(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
+ if (unlikely(be16_to_cpu(leaf->hdr.info.magic)
!= XFS_ATTR_LEAF_MAGIC)) {
XFS_CORRUPTION_ERROR("xfs_attr_node_list(4)",
XFS_ERRLEVEL_LOW,
@@ -1869,7 +1862,7 @@ xfs_attr_node_list(xfs_attr_list_context_t *context)
error = xfs_attr_leaf_list_int(bp, context);
if (error || !leaf->hdr.info.forw)
break; /* not really an error, buffer full or EOF */
- cursor->blkno = INT_GET(leaf->hdr.info.forw, ARCH_CONVERT);
+ cursor->blkno = be32_to_cpu(leaf->hdr.info.forw);
xfs_da_brelse(NULL, bp);
error = xfs_da_read_buf(NULL, context->dp, cursor->blkno, -1,
&bp, XFS_ATTR_FORK);
@@ -2232,9 +2225,10 @@ xfs_attr_trace_l_cn(char *where, struct xfs_attr_list_context *context,
: 0,
(__psunsigned_t)context->dupcnt,
(__psunsigned_t)context->flags,
- (__psunsigned_t)INT_GET(node->hdr.count, ARCH_CONVERT),
- (__psunsigned_t)INT_GET(node->btree[0].hashval, ARCH_CONVERT),
- (__psunsigned_t)INT_GET(node->btree[INT_GET(node->hdr.count, ARCH_CONVERT)-1].hashval, ARCH_CONVERT));
+ (__psunsigned_t)be16_to_cpu(node->hdr.count),
+ (__psunsigned_t)be32_to_cpu(node->btree[0].hashval),
+ (__psunsigned_t)be32_to_cpu(node->btree[
+ be16_to_cpu(node->hdr.count)-1].hashval));
}
/*
@@ -2261,8 +2255,8 @@ xfs_attr_trace_l_cb(char *where, struct xfs_attr_list_context *context,
: 0,
(__psunsigned_t)context->dupcnt,
(__psunsigned_t)context->flags,
- (__psunsigned_t)INT_GET(btree->hashval, ARCH_CONVERT),
- (__psunsigned_t)INT_GET(btree->before, ARCH_CONVERT),
+ (__psunsigned_t)be32_to_cpu(btree->hashval),
+ (__psunsigned_t)be32_to_cpu(btree->before),
(__psunsigned_t)NULL);
}
@@ -2290,9 +2284,10 @@ xfs_attr_trace_l_cl(char *where, struct xfs_attr_list_context *context,
: 0,
(__psunsigned_t)context->dupcnt,
(__psunsigned_t)context->flags,
- (__psunsigned_t)INT_GET(leaf->hdr.count, ARCH_CONVERT),
- (__psunsigned_t)INT_GET(leaf->entries[0].hashval, ARCH_CONVERT),
- (__psunsigned_t)INT_GET(leaf->entries[INT_GET(leaf->hdr.count, ARCH_CONVERT)-1].hashval, ARCH_CONVERT));
+ (__psunsigned_t)be16_to_cpu(leaf->hdr.count),
+ (__psunsigned_t)be32_to_cpu(leaf->entries[0].hashval),
+ (__psunsigned_t)be32_to_cpu(leaf->entries[
+ be16_to_cpu(leaf->hdr.count)-1].hashval));
}
/*
@@ -2522,7 +2517,7 @@ attr_user_capable(
struct vnode *vp,
cred_t *cred)
{
- struct inode *inode = LINVFS_GET_IP(vp);
+ struct inode *inode = vn_to_inode(vp);
if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
return -EPERM;
@@ -2540,7 +2535,7 @@ attr_trusted_capable(
struct vnode *vp,
cred_t *cred)
{
- struct inode *inode = LINVFS_GET_IP(vp);
+ struct inode *inode = vn_to_inode(vp);
if (IS_IMMUTABLE(inode) || IS_APPEND(inode))
return -EPERM;
diff --git a/fs/xfs/xfs_attr_leaf.c b/fs/xfs/xfs_attr_leaf.c
index fe91eac4e2a7..717682747bd2 100644
--- a/fs/xfs/xfs_attr_leaf.c
+++ b/fs/xfs/xfs_attr_leaf.c
@@ -194,7 +194,7 @@ xfs_attr_shortform_create(xfs_da_args_t *args)
xfs_idata_realloc(dp, sizeof(*hdr), XFS_ATTR_FORK);
hdr = (xfs_attr_sf_hdr_t *)ifp->if_u1.if_data;
hdr->count = 0;
- INT_SET(hdr->totsize, ARCH_CONVERT, sizeof(*hdr));
+ hdr->totsize = cpu_to_be16(sizeof(*hdr));
xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_ADATA);
}
@@ -224,8 +224,7 @@ xfs_attr_shortform_add(xfs_da_args_t *args, int forkoff)
ASSERT(ifp->if_flags & XFS_IFINLINE);
sf = (xfs_attr_shortform_t *)ifp->if_u1.if_data;
sfe = &sf->list[0];
- for (i = 0; i < INT_GET(sf->hdr.count, ARCH_CONVERT);
- sfe = XFS_ATTR_SF_NEXTENTRY(sfe), i++) {
+ for (i = 0; i < sf->hdr.count; sfe = XFS_ATTR_SF_NEXTENTRY(sfe), i++) {
#ifdef DEBUG
if (sfe->namelen != args->namelen)
continue;
@@ -248,13 +247,13 @@ xfs_attr_shortform_add(xfs_da_args_t *args, int forkoff)
sfe = (xfs_attr_sf_entry_t *)((char *)sf + offset);
sfe->namelen = args->namelen;
- INT_SET(sfe->valuelen, ARCH_CONVERT, args->valuelen);
+ sfe->valuelen = args->valuelen;
sfe->flags = (args->flags & ATTR_SECURE) ? XFS_ATTR_SECURE :
((args->flags & ATTR_ROOT) ? XFS_ATTR_ROOT : 0);
memcpy(sfe->nameval, args->name, args->namelen);
memcpy(&sfe->nameval[args->namelen], args->value, args->valuelen);
- INT_MOD(sf->hdr.count, ARCH_CONVERT, 1);
- INT_MOD(sf->hdr.totsize, ARCH_CONVERT, size);
+ sf->hdr.count++;
+ be16_add(&sf->hdr.totsize, size);
xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_ADATA);
xfs_sbversion_add_attr2(mp, args->trans);
@@ -277,7 +276,7 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
base = sizeof(xfs_attr_sf_hdr_t);
sf = (xfs_attr_shortform_t *)dp->i_afp->if_u1.if_data;
sfe = &sf->list[0];
- end = INT_GET(sf->hdr.count, ARCH_CONVERT);
+ end = sf->hdr.count;
for (i = 0; i < end; sfe = XFS_ATTR_SF_NEXTENTRY(sfe),
base += size, i++) {
size = XFS_ATTR_SF_ENTSIZE(sfe);
@@ -300,11 +299,11 @@ xfs_attr_shortform_remove(xfs_da_args_t *args)
* Fix up the attribute fork data, covering the hole
*/
end = base + size;
- totsize = INT_GET(sf->hdr.totsize, ARCH_CONVERT);
+ totsize = be16_to_cpu(sf->hdr.totsize);
if (end != totsize)
memmove(&((char *)sf)[base], &((char *)sf)[end], totsize - end);
- INT_MOD(sf->hdr.count, ARCH_CONVERT, -1);
- INT_MOD(sf->hdr.totsize, ARCH_CONVERT, -size);
+ sf->hdr.count--;
+ be16_add(&sf->hdr.totsize, -size);
/*
* Fix up the start offset of the attribute fork
@@ -360,7 +359,7 @@ xfs_attr_shortform_lookup(xfs_da_args_t *args)
ASSERT(ifp->if_flags & XFS_IFINLINE);
sf = (xfs_attr_shortform_t *)ifp->if_u1.if_data;
sfe = &sf->list[0];
- for (i = 0; i < INT_GET(sf->hdr.count, ARCH_CONVERT);
+ for (i = 0; i < sf->hdr.count;
sfe = XFS_ATTR_SF_NEXTENTRY(sfe), i++) {
if (sfe->namelen != args->namelen)
continue;
@@ -391,7 +390,7 @@ xfs_attr_shortform_getvalue(xfs_da_args_t *args)
ASSERT(args->dp->i_d.di_aformat == XFS_IFINLINE);
sf = (xfs_attr_shortform_t *)args->dp->i_afp->if_u1.if_data;
sfe = &sf->list[0];
- for (i = 0; i < INT_GET(sf->hdr.count, ARCH_CONVERT);
+ for (i = 0; i < sf->hdr.count;
sfe = XFS_ATTR_SF_NEXTENTRY(sfe), i++) {
if (sfe->namelen != args->namelen)
continue;
@@ -404,14 +403,14 @@ xfs_attr_shortform_getvalue(xfs_da_args_t *args)
((sfe->flags & XFS_ATTR_ROOT) != 0))
continue;
if (args->flags & ATTR_KERNOVAL) {
- args->valuelen = INT_GET(sfe->valuelen, ARCH_CONVERT);
+ args->valuelen = sfe->valuelen;
return(XFS_ERROR(EEXIST));
}
- if (args->valuelen < INT_GET(sfe->valuelen, ARCH_CONVERT)) {
- args->valuelen = INT_GET(sfe->valuelen, ARCH_CONVERT);
+ if (args->valuelen < sfe->valuelen) {
+ args->valuelen = sfe->valuelen;
return(XFS_ERROR(ERANGE));
}
- args->valuelen = INT_GET(sfe->valuelen, ARCH_CONVERT);
+ args->valuelen = sfe->valuelen;
memcpy(args->value, &sfe->nameval[args->namelen],
args->valuelen);
return(XFS_ERROR(EEXIST));
@@ -438,7 +437,7 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
dp = args->dp;
ifp = dp->i_afp;
sf = (xfs_attr_shortform_t *)ifp->if_u1.if_data;
- size = INT_GET(sf->hdr.totsize, ARCH_CONVERT);
+ size = be16_to_cpu(sf->hdr.totsize);
tmpbuffer = kmem_alloc(size, KM_SLEEP);
ASSERT(tmpbuffer != NULL);
memcpy(tmpbuffer, ifp->if_u1.if_data, size);
@@ -481,11 +480,11 @@ xfs_attr_shortform_to_leaf(xfs_da_args_t *args)
nargs.oknoent = 1;
sfe = &sf->list[0];
- for (i = 0; i < INT_GET(sf->hdr.count, ARCH_CONVERT); i++) {
+ for (i = 0; i < sf->hdr.count; i++) {
nargs.name = (char *)sfe->nameval;
nargs.namelen = sfe->namelen;
nargs.value = (char *)&sfe->nameval[nargs.namelen];
- nargs.valuelen = INT_GET(sfe->valuelen, ARCH_CONVERT);
+ nargs.valuelen = sfe->valuelen;
nargs.hashval = xfs_da_hashname((char *)sfe->nameval,
sfe->namelen);
nargs.flags = (sfe->flags & XFS_ATTR_SECURE) ? ATTR_SECURE :
@@ -514,11 +513,9 @@ xfs_attr_shortform_compare(const void *a, const void *b)
sa = (xfs_attr_sf_sort_t *)a;
sb = (xfs_attr_sf_sort_t *)b;
- if (INT_GET(sa->hash, ARCH_CONVERT)
- < INT_GET(sb->hash, ARCH_CONVERT)) {
+ if (sa->hash < sb->hash) {
return(-1);
- } else if (INT_GET(sa->hash, ARCH_CONVERT)
- > INT_GET(sb->hash, ARCH_CONVERT)) {
+ } else if (sa->hash > sb->hash) {
return(1);
} else {
return(sa->entno - sb->entno);
@@ -560,10 +557,8 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
* If the buffer is large enough, do not bother with sorting.
* Note the generous fudge factor of 16 overhead bytes per entry.
*/
- if ((dp->i_afp->if_bytes + INT_GET(sf->hdr.count, ARCH_CONVERT) * 16)
- < context->bufsize) {
- for (i = 0, sfe = &sf->list[0];
- i < INT_GET(sf->hdr.count, ARCH_CONVERT); i++) {
+ if ((dp->i_afp->if_bytes + sf->hdr.count * 16) < context->bufsize) {
+ for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) {
attrnames_t *namesp;
if (((context->flags & ATTR_SECURE) != 0) !=
@@ -584,14 +579,13 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
if (context->flags & ATTR_KERNOVAL) {
ASSERT(context->flags & ATTR_KERNAMELS);
context->count += namesp->attr_namelen +
- INT_GET(sfe->namelen, ARCH_CONVERT) + 1;
+ sfe->namelen + 1;
}
else {
if (xfs_attr_put_listent(context, namesp,
(char *)sfe->nameval,
(int)sfe->namelen,
- (int)INT_GET(sfe->valuelen,
- ARCH_CONVERT)))
+ (int)sfe->valuelen))
break;
}
sfe = XFS_ATTR_SF_NEXTENTRY(sfe);
@@ -603,7 +597,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
/*
* It didn't all fit, so we have to sort everything on hashval.
*/
- sbsize = INT_GET(sf->hdr.count, ARCH_CONVERT) * sizeof(*sbuf);
+ sbsize = sf->hdr.count * sizeof(*sbuf);
sbp = sbuf = kmem_alloc(sbsize, KM_SLEEP);
/*
@@ -611,8 +605,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
* the relevant info from only those that match into a buffer.
*/
nsbuf = 0;
- for (i = 0, sfe = &sf->list[0];
- i < INT_GET(sf->hdr.count, ARCH_CONVERT); i++) {
+ for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) {
if (unlikely(
((char *)sfe < (char *)sf) ||
((char *)sfe >= ((char *)sf + dp->i_afp->if_bytes)))) {
@@ -636,8 +629,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
continue;
}
sbp->entno = i;
- INT_SET(sbp->hash, ARCH_CONVERT,
- xfs_da_hashname((char *)sfe->nameval, sfe->namelen));
+ sbp->hash = xfs_da_hashname((char *)sfe->nameval, sfe->namelen);
sbp->name = (char *)sfe->nameval;
sbp->namelen = sfe->namelen;
/* These are bytes, and both on-disk, don't endian-flip */
@@ -660,12 +652,12 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
cursor->initted = 1;
cursor->blkno = 0;
for (sbp = sbuf, i = 0; i < nsbuf; i++, sbp++) {
- if (INT_GET(sbp->hash, ARCH_CONVERT) == cursor->hashval) {
+ if (sbp->hash == cursor->hashval) {
if (cursor->offset == count) {
break;
}
count++;
- } else if (INT_GET(sbp->hash, ARCH_CONVERT) > cursor->hashval) {
+ } else if (sbp->hash > cursor->hashval) {
break;
}
}
@@ -685,8 +677,8 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
((sbp->flags & XFS_ATTR_ROOT) ? &attr_trusted :
&attr_user);
- if (cursor->hashval != INT_GET(sbp->hash, ARCH_CONVERT)) {
- cursor->hashval = INT_GET(sbp->hash, ARCH_CONVERT);
+ if (cursor->hashval != sbp->hash) {
+ cursor->hashval = sbp->hash;
cursor->offset = 0;
}
if (context->flags & ATTR_KERNOVAL) {
@@ -696,7 +688,7 @@ xfs_attr_shortform_list(xfs_attr_list_context_t *context)
} else {
if (xfs_attr_put_listent(context, namesp,
sbp->name, sbp->namelen,
- INT_GET(sbp->valuelen, ARCH_CONVERT)))
+ sbp->valuelen))
break;
}
cursor->offset++;
@@ -720,12 +712,11 @@ xfs_attr_shortform_allfit(xfs_dabuf_t *bp, xfs_inode_t *dp)
int bytes, i;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
entry = &leaf->entries[0];
bytes = sizeof(struct xfs_attr_sf_hdr);
- for (i = 0; i < INT_GET(leaf->hdr.count, ARCH_CONVERT); entry++, i++) {
+ for (i = 0; i < be16_to_cpu(leaf->hdr.count); entry++, i++) {
if (entry->flags & XFS_ATTR_INCOMPLETE)
continue; /* don't copy partial entries */
if (!(entry->flags & XFS_ATTR_LOCAL))
@@ -733,11 +724,11 @@ xfs_attr_shortform_allfit(xfs_dabuf_t *bp, xfs_inode_t *dp)
name_loc = XFS_ATTR_LEAF_NAME_LOCAL(leaf, i);
if (name_loc->namelen >= XFS_ATTR_SF_ENTSIZE_MAX)
return(0);
- if (INT_GET(name_loc->valuelen, ARCH_CONVERT) >= XFS_ATTR_SF_ENTSIZE_MAX)
+ if (be16_to_cpu(name_loc->valuelen) >= XFS_ATTR_SF_ENTSIZE_MAX)
return(0);
bytes += sizeof(struct xfs_attr_sf_entry)-1
+ name_loc->namelen
- + INT_GET(name_loc->valuelen, ARCH_CONVERT);
+ + be16_to_cpu(name_loc->valuelen);
}
if ((dp->i_mount->m_flags & XFS_MOUNT_ATTR2) &&
(bytes == sizeof(struct xfs_attr_sf_hdr)))
@@ -766,8 +757,7 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff)
ASSERT(bp != NULL);
memcpy(tmpbuffer, bp->data, XFS_LBSIZE(dp->i_mount));
leaf = (xfs_attr_leafblock_t *)tmpbuffer;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
memset(bp->data, 0, XFS_LBSIZE(dp->i_mount));
/*
@@ -810,7 +800,7 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff)
nargs.trans = args->trans;
nargs.oknoent = 1;
entry = &leaf->entries[0];
- for (i = 0; i < INT_GET(leaf->hdr.count, ARCH_CONVERT); entry++, i++) {
+ for (i = 0; i < be16_to_cpu(leaf->hdr.count); entry++, i++) {
if (entry->flags & XFS_ATTR_INCOMPLETE)
continue; /* don't copy partial entries */
if (!entry->nameidx)
@@ -820,8 +810,8 @@ xfs_attr_leaf_to_shortform(xfs_dabuf_t *bp, xfs_da_args_t *args, int forkoff)
nargs.name = (char *)name_loc->nameval;
nargs.namelen = name_loc->namelen;
nargs.value = (char *)&name_loc->nameval[nargs.namelen];
- nargs.valuelen = INT_GET(name_loc->valuelen, ARCH_CONVERT);
- nargs.hashval = INT_GET(entry->hashval, ARCH_CONVERT);
+ nargs.valuelen = be16_to_cpu(name_loc->valuelen);
+ nargs.hashval = be32_to_cpu(entry->hashval);
nargs.flags = (entry->flags & XFS_ATTR_SECURE) ? ATTR_SECURE :
((entry->flags & XFS_ATTR_ROOT) ? ATTR_ROOT : 0);
xfs_attr_shortform_add(&nargs, forkoff);
@@ -875,13 +865,12 @@ xfs_attr_leaf_to_node(xfs_da_args_t *args)
goto out;
node = bp1->data;
leaf = bp2->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
/* both on-disk, don't endian-flip twice */
node->btree[0].hashval =
- leaf->entries[INT_GET(leaf->hdr.count, ARCH_CONVERT)-1 ].hashval;
- INT_SET(node->btree[0].before, ARCH_CONVERT, blkno);
- INT_SET(node->hdr.count, ARCH_CONVERT, 1);
+ leaf->entries[be16_to_cpu(leaf->hdr.count)-1 ].hashval;
+ node->btree[0].before = cpu_to_be32(blkno);
+ node->hdr.count = cpu_to_be16(1);
xfs_da_log_buf(args->trans, bp1, 0, XFS_LBSIZE(dp->i_mount) - 1);
error = 0;
out:
@@ -920,19 +909,16 @@ xfs_attr_leaf_create(xfs_da_args_t *args, xfs_dablk_t blkno, xfs_dabuf_t **bpp)
leaf = bp->data;
memset((char *)leaf, 0, XFS_LBSIZE(dp->i_mount));
hdr = &leaf->hdr;
- INT_SET(hdr->info.magic, ARCH_CONVERT, XFS_ATTR_LEAF_MAGIC);
- INT_SET(hdr->firstused, ARCH_CONVERT, XFS_LBSIZE(dp->i_mount));
+ hdr->info.magic = cpu_to_be16(XFS_ATTR_LEAF_MAGIC);
+ hdr->firstused = cpu_to_be16(XFS_LBSIZE(dp->i_mount));
if (!hdr->firstused) {
- INT_SET(hdr->firstused, ARCH_CONVERT,
+ hdr->firstused = cpu_to_be16(
XFS_LBSIZE(dp->i_mount) - XFS_ATTR_LEAF_NAME_ALIGN);
}
- INT_SET(hdr->freemap[0].base, ARCH_CONVERT,
- sizeof(xfs_attr_leaf_hdr_t));
- INT_SET(hdr->freemap[0].size, ARCH_CONVERT,
- INT_GET(hdr->firstused, ARCH_CONVERT)
- - INT_GET(hdr->freemap[0].base,
- ARCH_CONVERT));
+ hdr->freemap[0].base = cpu_to_be16(sizeof(xfs_attr_leaf_hdr_t));
+ hdr->freemap[0].size = cpu_to_be16(be16_to_cpu(hdr->firstused) -
+ sizeof(xfs_attr_leaf_hdr_t));
xfs_da_log_buf(args->trans, bp, 0, XFS_LBSIZE(dp->i_mount) - 1);
@@ -1004,10 +990,9 @@ xfs_attr_leaf_add(xfs_dabuf_t *bp, xfs_da_args_t *args)
int tablesize, entsize, sum, tmp, i;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
ASSERT((args->index >= 0)
- && (args->index <= INT_GET(leaf->hdr.count, ARCH_CONVERT)));
+ && (args->index <= be16_to_cpu(leaf->hdr.count)));
hdr = &leaf->hdr;
entsize = xfs_attr_leaf_newentsize(args->namelen, args->valuelen,
args->trans->t_mountp->m_sb.sb_blocksize, NULL);
@@ -1016,26 +1001,25 @@ xfs_attr_leaf_add(xfs_dabuf_t *bp, xfs_da_args_t *args)
* Search through freemap for first-fit on new name length.
* (may need to figure in size of entry struct too)
*/
- tablesize = (INT_GET(hdr->count, ARCH_CONVERT) + 1)
+ tablesize = (be16_to_cpu(hdr->count) + 1)
* sizeof(xfs_attr_leaf_entry_t)
+ sizeof(xfs_attr_leaf_hdr_t);
map = &hdr->freemap[XFS_ATTR_LEAF_MAPSIZE-1];
for (sum = 0, i = XFS_ATTR_LEAF_MAPSIZE-1; i >= 0; map--, i--) {
- if (tablesize > INT_GET(hdr->firstused, ARCH_CONVERT)) {
- sum += INT_GET(map->size, ARCH_CONVERT);
+ if (tablesize > be16_to_cpu(hdr->firstused)) {
+ sum += be16_to_cpu(map->size);
continue;
}
if (!map->size)
continue; /* no space in this map */
tmp = entsize;
- if (INT_GET(map->base, ARCH_CONVERT)
- < INT_GET(hdr->firstused, ARCH_CONVERT))
+ if (be16_to_cpu(map->base) < be16_to_cpu(hdr->firstused))
tmp += sizeof(xfs_attr_leaf_entry_t);
- if (INT_GET(map->size, ARCH_CONVERT) >= tmp) {
+ if (be16_to_cpu(map->size) >= tmp) {
tmp = xfs_attr_leaf_add_work(bp, args, i);
return(tmp);
}
- sum += INT_GET(map->size, ARCH_CONVERT);
+ sum += be16_to_cpu(map->size);
}
/*
@@ -1056,7 +1040,7 @@ xfs_attr_leaf_add(xfs_dabuf_t *bp, xfs_da_args_t *args)
* After compaction, the block is guaranteed to have only one
* free region, in freemap[0]. If it is not big enough, give up.
*/
- if (INT_GET(hdr->freemap[0].size, ARCH_CONVERT)
+ if (be16_to_cpu(hdr->freemap[0].size)
< (entsize + sizeof(xfs_attr_leaf_entry_t)))
return(XFS_ERROR(ENOSPC));
@@ -1079,45 +1063,42 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
int tmp, i;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
hdr = &leaf->hdr;
ASSERT((mapindex >= 0) && (mapindex < XFS_ATTR_LEAF_MAPSIZE));
- ASSERT((args->index >= 0)
- && (args->index <= INT_GET(hdr->count, ARCH_CONVERT)));
+ ASSERT((args->index >= 0) && (args->index <= be16_to_cpu(hdr->count)));
/*
* Force open some space in the entry array and fill it in.
*/
entry = &leaf->entries[args->index];
- if (args->index < INT_GET(hdr->count, ARCH_CONVERT)) {
- tmp = INT_GET(hdr->count, ARCH_CONVERT) - args->index;
+ if (args->index < be16_to_cpu(hdr->count)) {
+ tmp = be16_to_cpu(hdr->count) - args->index;
tmp *= sizeof(xfs_attr_leaf_entry_t);
memmove((char *)(entry+1), (char *)entry, tmp);
xfs_da_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, entry, tmp + sizeof(*entry)));
}
- INT_MOD(hdr->count, ARCH_CONVERT, 1);
+ be16_add(&hdr->count, 1);
/*
* Allocate space for the new string (at the end of the run).
*/
map = &hdr->freemap[mapindex];
mp = args->trans->t_mountp;
- ASSERT(INT_GET(map->base, ARCH_CONVERT) < XFS_LBSIZE(mp));
- ASSERT((INT_GET(map->base, ARCH_CONVERT) & 0x3) == 0);
- ASSERT(INT_GET(map->size, ARCH_CONVERT) >=
+ ASSERT(be16_to_cpu(map->base) < XFS_LBSIZE(mp));
+ ASSERT((be16_to_cpu(map->base) & 0x3) == 0);
+ ASSERT(be16_to_cpu(map->size) >=
xfs_attr_leaf_newentsize(args->namelen, args->valuelen,
mp->m_sb.sb_blocksize, NULL));
- ASSERT(INT_GET(map->size, ARCH_CONVERT) < XFS_LBSIZE(mp));
- ASSERT((INT_GET(map->size, ARCH_CONVERT) & 0x3) == 0);
- INT_MOD(map->size, ARCH_CONVERT,
+ ASSERT(be16_to_cpu(map->size) < XFS_LBSIZE(mp));
+ ASSERT((be16_to_cpu(map->size) & 0x3) == 0);
+ be16_add(&map->size,
-xfs_attr_leaf_newentsize(args->namelen, args->valuelen,
mp->m_sb.sb_blocksize, &tmp));
- INT_SET(entry->nameidx, ARCH_CONVERT,
- INT_GET(map->base, ARCH_CONVERT)
- + INT_GET(map->size, ARCH_CONVERT));
- INT_SET(entry->hashval, ARCH_CONVERT, args->hashval);
+ entry->nameidx = cpu_to_be16(be16_to_cpu(map->base) +
+ be16_to_cpu(map->size));
+ entry->hashval = cpu_to_be32(args->hashval);
entry->flags = tmp ? XFS_ATTR_LOCAL : 0;
entry->flags |= (args->flags & ATTR_SECURE) ? XFS_ATTR_SECURE :
((args->flags & ATTR_ROOT) ? XFS_ATTR_ROOT : 0);
@@ -1130,12 +1111,10 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
}
xfs_da_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, entry, sizeof(*entry)));
- ASSERT((args->index == 0) || (INT_GET(entry->hashval, ARCH_CONVERT)
- >= INT_GET((entry-1)->hashval,
- ARCH_CONVERT)));
- ASSERT((args->index == INT_GET(hdr->count, ARCH_CONVERT)-1) ||
- (INT_GET(entry->hashval, ARCH_CONVERT)
- <= (INT_GET((entry+1)->hashval, ARCH_CONVERT))));
+ ASSERT((args->index == 0) ||
+ (be32_to_cpu(entry->hashval) >= be32_to_cpu((entry-1)->hashval)));
+ ASSERT((args->index == be16_to_cpu(hdr->count)-1) ||
+ (be32_to_cpu(entry->hashval) <= be32_to_cpu((entry+1)->hashval)));
/*
* Copy the attribute name and value into the new space.
@@ -1149,10 +1128,10 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
if (entry->flags & XFS_ATTR_LOCAL) {
name_loc = XFS_ATTR_LEAF_NAME_LOCAL(leaf, args->index);
name_loc->namelen = args->namelen;
- INT_SET(name_loc->valuelen, ARCH_CONVERT, args->valuelen);
+ name_loc->valuelen = cpu_to_be16(args->valuelen);
memcpy((char *)name_loc->nameval, args->name, args->namelen);
memcpy((char *)&name_loc->nameval[args->namelen], args->value,
- INT_GET(name_loc->valuelen, ARCH_CONVERT));
+ be16_to_cpu(name_loc->valuelen));
} else {
name_rmt = XFS_ATTR_LEAF_NAME_REMOTE(leaf, args->index);
name_rmt->namelen = args->namelen;
@@ -1171,28 +1150,23 @@ xfs_attr_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int mapindex)
/*
* Update the control info for this leaf node
*/
- if (INT_GET(entry->nameidx, ARCH_CONVERT)
- < INT_GET(hdr->firstused, ARCH_CONVERT)) {
+ if (be16_to_cpu(entry->nameidx) < be16_to_cpu(hdr->firstused)) {
/* both on-disk, don't endian-flip twice */
hdr->firstused = entry->nameidx;
}
- ASSERT(INT_GET(hdr->firstused, ARCH_CONVERT)
- >= ((INT_GET(hdr->count, ARCH_CONVERT)
- * sizeof(*entry))+sizeof(*hdr)));
- tmp = (INT_GET(hdr->count, ARCH_CONVERT)-1)
- * sizeof(xfs_attr_leaf_entry_t)
+ ASSERT(be16_to_cpu(hdr->firstused) >=
+ ((be16_to_cpu(hdr->count) * sizeof(*entry)) + sizeof(*hdr)));
+ tmp = (be16_to_cpu(hdr->count)-1) * sizeof(xfs_attr_leaf_entry_t)
+ sizeof(xfs_attr_leaf_hdr_t);
map = &hdr->freemap[0];
for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; map++, i++) {
- if (INT_GET(map->base, ARCH_CONVERT) == tmp) {
- INT_MOD(map->base, ARCH_CONVERT,
- sizeof(xfs_attr_leaf_entry_t));
- INT_MOD(map->size, ARCH_CONVERT,
- -sizeof(xfs_attr_leaf_entry_t));
+ if (be16_to_cpu(map->base) == tmp) {
+ be16_add(&map->base, sizeof(xfs_attr_leaf_entry_t));
+ be16_add(&map->size,
+ -((int)sizeof(xfs_attr_leaf_entry_t)));
}
}
- INT_MOD(hdr->usedbytes, ARCH_CONVERT,
- xfs_attr_leaf_entsize(leaf, args->index));
+ be16_add(&hdr->usedbytes, xfs_attr_leaf_entsize(leaf, args->index));
xfs_da_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, hdr, sizeof(*hdr)));
return(0);
@@ -1223,28 +1197,25 @@ xfs_attr_leaf_compact(xfs_trans_t *trans, xfs_dabuf_t *bp)
hdr_s = &leaf_s->hdr;
hdr_d = &leaf_d->hdr;
hdr_d->info = hdr_s->info; /* struct copy */
- INT_SET(hdr_d->firstused, ARCH_CONVERT, XFS_LBSIZE(mp));
+ hdr_d->firstused = cpu_to_be16(XFS_LBSIZE(mp));
/* handle truncation gracefully */
if (!hdr_d->firstused) {
- INT_SET(hdr_d->firstused, ARCH_CONVERT,
+ hdr_d->firstused = cpu_to_be16(
XFS_LBSIZE(mp) - XFS_ATTR_LEAF_NAME_ALIGN);
}
hdr_d->usedbytes = 0;
hdr_d->count = 0;
hdr_d->holes = 0;
- INT_SET(hdr_d->freemap[0].base, ARCH_CONVERT,
- sizeof(xfs_attr_leaf_hdr_t));
- INT_SET(hdr_d->freemap[0].size, ARCH_CONVERT,
- INT_GET(hdr_d->firstused, ARCH_CONVERT)
- - INT_GET(hdr_d->freemap[0].base, ARCH_CONVERT));
+ hdr_d->freemap[0].base = cpu_to_be16(sizeof(xfs_attr_leaf_hdr_t));
+ hdr_d->freemap[0].size = cpu_to_be16(be16_to_cpu(hdr_d->firstused) -
+ sizeof(xfs_attr_leaf_hdr_t));
/*
* Copy all entry's in the same (sorted) order,
* but allocate name/value pairs packed and in sequence.
*/
xfs_attr_leaf_moveents(leaf_s, 0, leaf_d, 0,
- (int)INT_GET(hdr_s->count, ARCH_CONVERT), mp);
-
+ be16_to_cpu(hdr_s->count), mp);
xfs_da_log_buf(trans, bp, 0, XFS_LBSIZE(mp) - 1);
kmem_free(tmpbuffer, XFS_LBSIZE(mp));
@@ -1279,10 +1250,8 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
ASSERT(blk2->magic == XFS_ATTR_LEAF_MAGIC);
leaf1 = blk1->bp->data;
leaf2 = blk2->bp->data;
- ASSERT(INT_GET(leaf1->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
- ASSERT(INT_GET(leaf2->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf1->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf2->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
args = state->args;
/*
@@ -1319,22 +1288,21 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
/*
* Move any entries required from leaf to leaf:
*/
- if (count < INT_GET(hdr1->count, ARCH_CONVERT)) {
+ if (count < be16_to_cpu(hdr1->count)) {
/*
* Figure the total bytes to be added to the destination leaf.
*/
/* number entries being moved */
- count = INT_GET(hdr1->count, ARCH_CONVERT) - count;
- space = INT_GET(hdr1->usedbytes, ARCH_CONVERT) - totallen;
+ count = be16_to_cpu(hdr1->count) - count;
+ space = be16_to_cpu(hdr1->usedbytes) - totallen;
space += count * sizeof(xfs_attr_leaf_entry_t);
/*
* leaf2 is the destination, compact it if it looks tight.
*/
- max = INT_GET(hdr2->firstused, ARCH_CONVERT)
+ max = be16_to_cpu(hdr2->firstused)
- sizeof(xfs_attr_leaf_hdr_t);
- max -= INT_GET(hdr2->count, ARCH_CONVERT)
- * sizeof(xfs_attr_leaf_entry_t);
+ max -= be16_to_cpu(hdr2->count) * sizeof(xfs_attr_leaf_entry_t);
if (space > max) {
xfs_attr_leaf_compact(args->trans, blk2->bp);
}
@@ -1342,13 +1310,12 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
/*
* Move high entries from leaf1 to low end of leaf2.
*/
- xfs_attr_leaf_moveents(leaf1,
- INT_GET(hdr1->count, ARCH_CONVERT)-count,
+ xfs_attr_leaf_moveents(leaf1, be16_to_cpu(hdr1->count) - count,
leaf2, 0, count, state->mp);
xfs_da_log_buf(args->trans, blk1->bp, 0, state->blocksize-1);
xfs_da_log_buf(args->trans, blk2->bp, 0, state->blocksize-1);
- } else if (count > INT_GET(hdr1->count, ARCH_CONVERT)) {
+ } else if (count > be16_to_cpu(hdr1->count)) {
/*
* I assert that since all callers pass in an empty
* second buffer, this code should never execute.
@@ -1358,17 +1325,16 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* Figure the total bytes to be added to the destination leaf.
*/
/* number entries being moved */
- count -= INT_GET(hdr1->count, ARCH_CONVERT);
- space = totallen - INT_GET(hdr1->usedbytes, ARCH_CONVERT);
+ count -= be16_to_cpu(hdr1->count);
+ space = totallen - be16_to_cpu(hdr1->usedbytes);
space += count * sizeof(xfs_attr_leaf_entry_t);
/*
* leaf1 is the destination, compact it if it looks tight.
*/
- max = INT_GET(hdr1->firstused, ARCH_CONVERT)
+ max = be16_to_cpu(hdr1->firstused)
- sizeof(xfs_attr_leaf_hdr_t);
- max -= INT_GET(hdr1->count, ARCH_CONVERT)
- * sizeof(xfs_attr_leaf_entry_t);
+ max -= be16_to_cpu(hdr1->count) * sizeof(xfs_attr_leaf_entry_t);
if (space > max) {
xfs_attr_leaf_compact(args->trans, blk1->bp);
}
@@ -1377,8 +1343,7 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* Move low entries from leaf2 to high end of leaf1.
*/
xfs_attr_leaf_moveents(leaf2, 0, leaf1,
- (int)INT_GET(hdr1->count, ARCH_CONVERT), count,
- state->mp);
+ be16_to_cpu(hdr1->count), count, state->mp);
xfs_da_log_buf(args->trans, blk1->bp, 0, state->blocksize-1);
xfs_da_log_buf(args->trans, blk2->bp, 0, state->blocksize-1);
@@ -1387,12 +1352,10 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
/*
* Copy out last hashval in each block for B-tree code.
*/
- blk1->hashval =
- INT_GET(leaf1->entries[INT_GET(leaf1->hdr.count,
- ARCH_CONVERT)-1].hashval, ARCH_CONVERT);
- blk2->hashval =
- INT_GET(leaf2->entries[INT_GET(leaf2->hdr.count,
- ARCH_CONVERT)-1].hashval, ARCH_CONVERT);
+ blk1->hashval = be32_to_cpu(
+ leaf1->entries[be16_to_cpu(leaf1->hdr.count)-1].hashval);
+ blk2->hashval = be32_to_cpu(
+ leaf2->entries[be16_to_cpu(leaf2->hdr.count)-1].hashval);
/*
* Adjust the expected index for insertion.
@@ -1406,13 +1369,12 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* inserting. The index/blkno fields refer to the "old" entry,
* while the index2/blkno2 fields refer to the "new" entry.
*/
- if (blk1->index > INT_GET(leaf1->hdr.count, ARCH_CONVERT)) {
+ if (blk1->index > be16_to_cpu(leaf1->hdr.count)) {
ASSERT(state->inleaf == 0);
- blk2->index = blk1->index
- - INT_GET(leaf1->hdr.count, ARCH_CONVERT);
+ blk2->index = blk1->index - be16_to_cpu(leaf1->hdr.count);
args->index = args->index2 = blk2->index;
args->blkno = args->blkno2 = blk2->blkno;
- } else if (blk1->index == INT_GET(leaf1->hdr.count, ARCH_CONVERT)) {
+ } else if (blk1->index == be16_to_cpu(leaf1->hdr.count)) {
if (state->inleaf) {
args->index = blk1->index;
args->blkno = blk1->blkno;
@@ -1420,7 +1382,7 @@ xfs_attr_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
args->blkno2 = blk2->blkno;
} else {
blk2->index = blk1->index
- - INT_GET(leaf1->hdr.count, ARCH_CONVERT);
+ - be16_to_cpu(leaf1->hdr.count);
args->index = args->index2 = blk2->index;
args->blkno = args->blkno2 = blk2->blkno;
}
@@ -1464,15 +1426,14 @@ xfs_attr_leaf_figure_balance(xfs_da_state_t *state,
* Examine entries until we reduce the absolute difference in
* byte usage between the two blocks to a minimum.
*/
- max = INT_GET(hdr1->count, ARCH_CONVERT)
- + INT_GET(hdr2->count, ARCH_CONVERT);
+ max = be16_to_cpu(hdr1->count) + be16_to_cpu(hdr2->count);
half = (max+1) * sizeof(*entry);
- half += INT_GET(hdr1->usedbytes, ARCH_CONVERT)
- + INT_GET(hdr2->usedbytes, ARCH_CONVERT)
- + xfs_attr_leaf_newentsize(
- state->args->namelen,
- state->args->valuelen,
- state->blocksize, NULL);
+ half += be16_to_cpu(hdr1->usedbytes) +
+ be16_to_cpu(hdr2->usedbytes) +
+ xfs_attr_leaf_newentsize(
+ state->args->namelen,
+ state->args->valuelen,
+ state->blocksize, NULL);
half /= 2;
lastdelta = state->blocksize;
entry = &leaf1->entries[0];
@@ -1498,7 +1459,7 @@ xfs_attr_leaf_figure_balance(xfs_da_state_t *state,
/*
* Wrap around into the second block if necessary.
*/
- if (count == INT_GET(hdr1->count, ARCH_CONVERT)) {
+ if (count == be16_to_cpu(hdr1->count)) {
leaf1 = leaf2;
entry = &leaf1->entries[0];
index = 0;
@@ -1566,12 +1527,12 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
*/
blk = &state->path.blk[ state->path.active-1 ];
info = blk->bp->data;
- ASSERT(INT_GET(info->magic, ARCH_CONVERT) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(info->magic) == XFS_ATTR_LEAF_MAGIC);
leaf = (xfs_attr_leafblock_t *)info;
- count = INT_GET(leaf->hdr.count, ARCH_CONVERT);
+ count = be16_to_cpu(leaf->hdr.count);
bytes = sizeof(xfs_attr_leaf_hdr_t) +
count * sizeof(xfs_attr_leaf_entry_t) +
- INT_GET(leaf->hdr.usedbytes, ARCH_CONVERT);
+ be16_to_cpu(leaf->hdr.usedbytes);
if (bytes > (state->blocksize >> 1)) {
*action = 0; /* blk over 50%, don't try to join */
return(0);
@@ -1588,7 +1549,7 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
* Make altpath point to the block we want to keep and
* path point to the block we want to drop (this one).
*/
- forward = info->forw;
+ forward = (info->forw != 0);
memcpy(&state->altpath, &state->path, sizeof(state->path));
error = xfs_da_path_shift(state, &state->altpath, forward,
0, &retval);
@@ -1610,13 +1571,12 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
* to shrink an attribute list over time.
*/
/* start with smaller blk num */
- forward = (INT_GET(info->forw, ARCH_CONVERT)
- < INT_GET(info->back, ARCH_CONVERT));
+ forward = (be32_to_cpu(info->forw) < be32_to_cpu(info->back));
for (i = 0; i < 2; forward = !forward, i++) {
if (forward)
- blkno = INT_GET(info->forw, ARCH_CONVERT);
+ blkno = be32_to_cpu(info->forw);
else
- blkno = INT_GET(info->back, ARCH_CONVERT);
+ blkno = be32_to_cpu(info->back);
if (blkno == 0)
continue;
error = xfs_da_read_buf(state->args->trans, state->args->dp,
@@ -1626,14 +1586,13 @@ xfs_attr_leaf_toosmall(xfs_da_state_t *state, int *action)
ASSERT(bp != NULL);
leaf = (xfs_attr_leafblock_t *)info;
- count = INT_GET(leaf->hdr.count, ARCH_CONVERT);
+ count = be16_to_cpu(leaf->hdr.count);
bytes = state->blocksize - (state->blocksize>>2);
- bytes -= INT_GET(leaf->hdr.usedbytes, ARCH_CONVERT);
+ bytes -= be16_to_cpu(leaf->hdr.usedbytes);
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
- count += INT_GET(leaf->hdr.count, ARCH_CONVERT);
- bytes -= INT_GET(leaf->hdr.usedbytes, ARCH_CONVERT);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ count += be16_to_cpu(leaf->hdr.count);
+ bytes -= be16_to_cpu(leaf->hdr.usedbytes);
bytes -= count * sizeof(xfs_attr_leaf_entry_t);
bytes -= sizeof(xfs_attr_leaf_hdr_t);
xfs_da_brelse(state->args->trans, bp);
@@ -1685,21 +1644,18 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
xfs_mount_t *mp;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
hdr = &leaf->hdr;
mp = args->trans->t_mountp;
- ASSERT((INT_GET(hdr->count, ARCH_CONVERT) > 0)
- && (INT_GET(hdr->count, ARCH_CONVERT) < (XFS_LBSIZE(mp)/8)));
+ ASSERT((be16_to_cpu(hdr->count) > 0)
+ && (be16_to_cpu(hdr->count) < (XFS_LBSIZE(mp)/8)));
ASSERT((args->index >= 0)
- && (args->index < INT_GET(hdr->count, ARCH_CONVERT)));
- ASSERT(INT_GET(hdr->firstused, ARCH_CONVERT)
- >= ((INT_GET(hdr->count, ARCH_CONVERT)
- * sizeof(*entry))+sizeof(*hdr)));
+ && (args->index < be16_to_cpu(hdr->count)));
+ ASSERT(be16_to_cpu(hdr->firstused) >=
+ ((be16_to_cpu(hdr->count) * sizeof(*entry)) + sizeof(*hdr)));
entry = &leaf->entries[args->index];
- ASSERT(INT_GET(entry->nameidx, ARCH_CONVERT)
- >= INT_GET(hdr->firstused, ARCH_CONVERT));
- ASSERT(INT_GET(entry->nameidx, ARCH_CONVERT) < XFS_LBSIZE(mp));
+ ASSERT(be16_to_cpu(entry->nameidx) >= be16_to_cpu(hdr->firstused));
+ ASSERT(be16_to_cpu(entry->nameidx) < XFS_LBSIZE(mp));
/*
* Scan through free region table:
@@ -1707,33 +1663,30 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
* find smallest free region in case we need to replace it,
* adjust any map that borders the entry table,
*/
- tablesize = INT_GET(hdr->count, ARCH_CONVERT)
- * sizeof(xfs_attr_leaf_entry_t)
+ tablesize = be16_to_cpu(hdr->count) * sizeof(xfs_attr_leaf_entry_t)
+ sizeof(xfs_attr_leaf_hdr_t);
map = &hdr->freemap[0];
- tmp = INT_GET(map->size, ARCH_CONVERT);
+ tmp = be16_to_cpu(map->size);
before = after = -1;
smallest = XFS_ATTR_LEAF_MAPSIZE - 1;
entsize = xfs_attr_leaf_entsize(leaf, args->index);
for (i = 0; i < XFS_ATTR_LEAF_MAPSIZE; map++, i++) {
- ASSERT(INT_GET(map->base, ARCH_CONVERT) < XFS_LBSIZE(mp));
- ASSERT(INT_GET(map->size, ARCH_CONVERT) < XFS_LBSIZE(mp));
- if (INT_GET(map->base, ARCH_CONVERT) == tablesize) {
- INT_MOD(map->base, ARCH_CONVERT,
- -sizeof(xfs_attr_leaf_entry_t));
- INT_MOD(map->size, ARCH_CONVERT,
- sizeof(xfs_attr_leaf_entry_t));
+ ASSERT(be16_to_cpu(map->base) < XFS_LBSIZE(mp));
+ ASSERT(be16_to_cpu(map->size) < XFS_LBSIZE(mp));
+ if (be16_to_cpu(map->base) == tablesize) {
+ be16_add(&map->base,
+ -((int)sizeof(xfs_attr_leaf_entry_t)));
+ be16_add(&map->size, sizeof(xfs_attr_leaf_entry_t));
}
- if ((INT_GET(map->base, ARCH_CONVERT)
- + INT_GET(map->size, ARCH_CONVERT))
- == INT_GET(entry->nameidx, ARCH_CONVERT)) {
+ if ((be16_to_cpu(map->base) + be16_to_cpu(map->size))
+ == be16_to_cpu(entry->nameidx)) {
before = i;
- } else if (INT_GET(map->base, ARCH_CONVERT)
- == (INT_GET(entry->nameidx, ARCH_CONVERT) + entsize)) {
+ } else if (be16_to_cpu(map->base)
+ == (be16_to_cpu(entry->nameidx) + entsize)) {
after = i;
- } else if (INT_GET(map->size, ARCH_CONVERT) < tmp) {
- tmp = INT_GET(map->size, ARCH_CONVERT);
+ } else if (be16_to_cpu(map->size) < tmp) {
+ tmp = be16_to_cpu(map->size);
smallest = i;
}
}
@@ -1745,38 +1698,35 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
if ((before >= 0) || (after >= 0)) {
if ((before >= 0) && (after >= 0)) {
map = &hdr->freemap[before];
- INT_MOD(map->size, ARCH_CONVERT, entsize);
- INT_MOD(map->size, ARCH_CONVERT,
- INT_GET(hdr->freemap[after].size,
- ARCH_CONVERT));
+ be16_add(&map->size, entsize);
+ be16_add(&map->size,
+ be16_to_cpu(hdr->freemap[after].size));
hdr->freemap[after].base = 0;
hdr->freemap[after].size = 0;
} else if (before >= 0) {
map = &hdr->freemap[before];
- INT_MOD(map->size, ARCH_CONVERT, entsize);
+ be16_add(&map->size, entsize);
} else {
map = &hdr->freemap[after];
/* both on-disk, don't endian flip twice */
map->base = entry->nameidx;
- INT_MOD(map->size, ARCH_CONVERT, entsize);
+ be16_add(&map->size, entsize);
}
} else {
/*
* Replace smallest region (if it is smaller than free'd entry)
*/
map = &hdr->freemap[smallest];
- if (INT_GET(map->size, ARCH_CONVERT) < entsize) {
- INT_SET(map->base, ARCH_CONVERT,
- INT_GET(entry->nameidx, ARCH_CONVERT));
- INT_SET(map->size, ARCH_CONVERT, entsize);
+ if (be16_to_cpu(map->size) < entsize) {
+ map->base = cpu_to_be16(be16_to_cpu(entry->nameidx));
+ map->size = cpu_to_be16(entsize);
}
}
/*
* Did we remove the first entry?
*/
- if (INT_GET(entry->nameidx, ARCH_CONVERT)
- == INT_GET(hdr->firstused, ARCH_CONVERT))
+ if (be16_to_cpu(entry->nameidx) == be16_to_cpu(hdr->firstused))
smallest = 1;
else
smallest = 0;
@@ -1785,18 +1735,18 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
* Compress the remaining entries and zero out the removed stuff.
*/
memset(XFS_ATTR_LEAF_NAME(leaf, args->index), 0, entsize);
- INT_MOD(hdr->usedbytes, ARCH_CONVERT, -entsize);
+ be16_add(&hdr->usedbytes, -entsize);
xfs_da_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, XFS_ATTR_LEAF_NAME(leaf, args->index),
entsize));
- tmp = (INT_GET(hdr->count, ARCH_CONVERT) - args->index)
+ tmp = (be16_to_cpu(hdr->count) - args->index)
* sizeof(xfs_attr_leaf_entry_t);
memmove((char *)entry, (char *)(entry+1), tmp);
- INT_MOD(hdr->count, ARCH_CONVERT, -1);
+ be16_add(&hdr->count, -1);
xfs_da_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, entry, tmp + sizeof(*entry)));
- entry = &leaf->entries[INT_GET(hdr->count, ARCH_CONVERT)];
+ entry = &leaf->entries[be16_to_cpu(hdr->count)];
memset((char *)entry, 0, sizeof(xfs_attr_leaf_entry_t));
/*
@@ -1808,18 +1758,17 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
if (smallest) {
tmp = XFS_LBSIZE(mp);
entry = &leaf->entries[0];
- for (i = INT_GET(hdr->count, ARCH_CONVERT)-1;
- i >= 0; entry++, i--) {
- ASSERT(INT_GET(entry->nameidx, ARCH_CONVERT)
- >= INT_GET(hdr->firstused, ARCH_CONVERT));
- ASSERT(INT_GET(entry->nameidx, ARCH_CONVERT)
- < XFS_LBSIZE(mp));
- if (INT_GET(entry->nameidx, ARCH_CONVERT) < tmp)
- tmp = INT_GET(entry->nameidx, ARCH_CONVERT);
+ for (i = be16_to_cpu(hdr->count)-1; i >= 0; entry++, i--) {
+ ASSERT(be16_to_cpu(entry->nameidx) >=
+ be16_to_cpu(hdr->firstused));
+ ASSERT(be16_to_cpu(entry->nameidx) < XFS_LBSIZE(mp));
+
+ if (be16_to_cpu(entry->nameidx) < tmp)
+ tmp = be16_to_cpu(entry->nameidx);
}
- INT_SET(hdr->firstused, ARCH_CONVERT, tmp);
+ hdr->firstused = cpu_to_be16(tmp);
if (!hdr->firstused) {
- INT_SET(hdr->firstused, ARCH_CONVERT,
+ hdr->firstused = cpu_to_be16(
tmp - XFS_ATTR_LEAF_NAME_ALIGN);
}
} else {
@@ -1833,9 +1782,8 @@ xfs_attr_leaf_remove(xfs_dabuf_t *bp, xfs_da_args_t *args)
* "join" the leaf with a sibling if so.
*/
tmp = sizeof(xfs_attr_leaf_hdr_t);
- tmp += INT_GET(leaf->hdr.count, ARCH_CONVERT)
- * sizeof(xfs_attr_leaf_entry_t);
- tmp += INT_GET(leaf->hdr.usedbytes, ARCH_CONVERT);
+ tmp += be16_to_cpu(leaf->hdr.count) * sizeof(xfs_attr_leaf_entry_t);
+ tmp += be16_to_cpu(leaf->hdr.usedbytes);
return(tmp < mp->m_attr_magicpct); /* leaf is < 37% full */
}
@@ -1859,20 +1807,16 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
ASSERT(save_blk->magic == XFS_ATTR_LEAF_MAGIC);
drop_leaf = drop_blk->bp->data;
save_leaf = save_blk->bp->data;
- ASSERT(INT_GET(drop_leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
- ASSERT(INT_GET(save_leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(drop_leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(save_leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
drop_hdr = &drop_leaf->hdr;
save_hdr = &save_leaf->hdr;
/*
* Save last hashval from dying block for later Btree fixup.
*/
- drop_blk->hashval =
- INT_GET(drop_leaf->entries[INT_GET(drop_leaf->hdr.count,
- ARCH_CONVERT)-1].hashval,
- ARCH_CONVERT);
+ drop_blk->hashval = be32_to_cpu(
+ drop_leaf->entries[be16_to_cpu(drop_leaf->hdr.count)-1].hashval);
/*
* Check if we need a temp buffer, or can we do it in place.
@@ -1886,12 +1830,11 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
*/
if (xfs_attr_leaf_order(save_blk->bp, drop_blk->bp)) {
xfs_attr_leaf_moveents(drop_leaf, 0, save_leaf, 0,
- (int)INT_GET(drop_hdr->count, ARCH_CONVERT), mp);
+ be16_to_cpu(drop_hdr->count), mp);
} else {
xfs_attr_leaf_moveents(drop_leaf, 0, save_leaf,
- INT_GET(save_hdr->count, ARCH_CONVERT),
- (int)INT_GET(drop_hdr->count, ARCH_CONVERT),
- mp);
+ be16_to_cpu(save_hdr->count),
+ be16_to_cpu(drop_hdr->count), mp);
}
} else {
/*
@@ -1905,28 +1848,24 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
tmp_hdr = &tmp_leaf->hdr;
tmp_hdr->info = save_hdr->info; /* struct copy */
tmp_hdr->count = 0;
- INT_SET(tmp_hdr->firstused, ARCH_CONVERT, state->blocksize);
+ tmp_hdr->firstused = cpu_to_be16(state->blocksize);
if (!tmp_hdr->firstused) {
- INT_SET(tmp_hdr->firstused, ARCH_CONVERT,
+ tmp_hdr->firstused = cpu_to_be16(
state->blocksize - XFS_ATTR_LEAF_NAME_ALIGN);
}
tmp_hdr->usedbytes = 0;
if (xfs_attr_leaf_order(save_blk->bp, drop_blk->bp)) {
xfs_attr_leaf_moveents(drop_leaf, 0, tmp_leaf, 0,
- (int)INT_GET(drop_hdr->count, ARCH_CONVERT),
- mp);
+ be16_to_cpu(drop_hdr->count), mp);
xfs_attr_leaf_moveents(save_leaf, 0, tmp_leaf,
- INT_GET(tmp_leaf->hdr.count, ARCH_CONVERT),
- (int)INT_GET(save_hdr->count, ARCH_CONVERT),
- mp);
+ be16_to_cpu(tmp_leaf->hdr.count),
+ be16_to_cpu(save_hdr->count), mp);
} else {
xfs_attr_leaf_moveents(save_leaf, 0, tmp_leaf, 0,
- (int)INT_GET(save_hdr->count, ARCH_CONVERT),
- mp);
+ be16_to_cpu(save_hdr->count), mp);
xfs_attr_leaf_moveents(drop_leaf, 0, tmp_leaf,
- INT_GET(tmp_leaf->hdr.count, ARCH_CONVERT),
- (int)INT_GET(drop_hdr->count, ARCH_CONVERT),
- mp);
+ be16_to_cpu(tmp_leaf->hdr.count),
+ be16_to_cpu(drop_hdr->count), mp);
}
memcpy((char *)save_leaf, (char *)tmp_leaf, state->blocksize);
kmem_free(tmpbuffer, state->blocksize);
@@ -1938,10 +1877,8 @@ xfs_attr_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
/*
* Copy out last hashval in each block for B-tree code.
*/
- save_blk->hashval =
- INT_GET(save_leaf->entries[INT_GET(save_leaf->hdr.count,
- ARCH_CONVERT)-1].hashval,
- ARCH_CONVERT);
+ save_blk->hashval = be32_to_cpu(
+ save_leaf->entries[be16_to_cpu(save_leaf->hdr.count)-1].hashval);
}
/*========================================================================
@@ -1972,48 +1909,45 @@ xfs_attr_leaf_lookup_int(xfs_dabuf_t *bp, xfs_da_args_t *args)
xfs_dahash_t hashval;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
- ASSERT(INT_GET(leaf->hdr.count, ARCH_CONVERT)
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.count)
< (XFS_LBSIZE(args->dp->i_mount)/8));
/*
* Binary search. (note: small blocks will skip this loop)
*/
hashval = args->hashval;
- probe = span = INT_GET(leaf->hdr.count, ARCH_CONVERT) / 2;
+ probe = span = be16_to_cpu(leaf->hdr.count) / 2;
for (entry = &leaf->entries[probe]; span > 4;
entry = &leaf->entries[probe]) {
span /= 2;
- if (INT_GET(entry->hashval, ARCH_CONVERT) < hashval)
+ if (be32_to_cpu(entry->hashval) < hashval)
probe += span;
- else if (INT_GET(entry->hashval, ARCH_CONVERT) > hashval)
+ else if (be32_to_cpu(entry->hashval) > hashval)
probe -= span;
else
break;
}
ASSERT((probe >= 0) &&
(!leaf->hdr.count
- || (probe < INT_GET(leaf->hdr.count, ARCH_CONVERT))));
- ASSERT((span <= 4) || (INT_GET(entry->hashval, ARCH_CONVERT)
- == hashval));
+ || (probe < be16_to_cpu(leaf->hdr.count))));
+ ASSERT((span <= 4) || (be32_to_cpu(entry->hashval) == hashval));
/*
* Since we may have duplicate hashval's, find the first matching
* hashval in the leaf.
*/
- while ((probe > 0) && (INT_GET(entry->hashval, ARCH_CONVERT)
- >= hashval)) {
+ while ((probe > 0) && (be32_to_cpu(entry->hashval) >= hashval)) {
entry--;
probe--;
}
- while ((probe < INT_GET(leaf->hdr.count, ARCH_CONVERT))
- && (INT_GET(entry->hashval, ARCH_CONVERT) < hashval)) {
+ while ((probe < be16_to_cpu(leaf->hdr.count)) &&
+ (be32_to_cpu(entry->hashval) < hashval)) {
entry++;
probe++;
}
- if ((probe == INT_GET(leaf->hdr.count, ARCH_CONVERT))
- || (INT_GET(entry->hashval, ARCH_CONVERT) != hashval)) {
+ if ((probe == be16_to_cpu(leaf->hdr.count)) ||
+ (be32_to_cpu(entry->hashval) != hashval)) {
args->index = probe;
return(XFS_ERROR(ENOATTR));
}
@@ -2021,8 +1955,8 @@ xfs_attr_leaf_lookup_int(xfs_dabuf_t *bp, xfs_da_args_t *args)
/*
* Duplicate keys may be present, so search all of them for a match.
*/
- for ( ; (probe < INT_GET(leaf->hdr.count, ARCH_CONVERT))
- && (INT_GET(entry->hashval, ARCH_CONVERT) == hashval);
+ for ( ; (probe < be16_to_cpu(leaf->hdr.count)) &&
+ (be32_to_cpu(entry->hashval) == hashval);
entry++, probe++) {
/*
* GROT: Add code to remove incomplete entries.
@@ -2064,11 +1998,9 @@ xfs_attr_leaf_lookup_int(xfs_dabuf_t *bp, xfs_da_args_t *args)
((entry->flags & XFS_ATTR_ROOT) != 0))
continue;
args->index = probe;
- args->rmtblkno
- = INT_GET(name_rmt->valueblk, ARCH_CONVERT);
+ args->rmtblkno = be32_to_cpu(name_rmt->valueblk);
args->rmtblkcnt = XFS_B_TO_FSB(args->dp->i_mount,
- INT_GET(name_rmt->valuelen,
- ARCH_CONVERT));
+ be32_to_cpu(name_rmt->valuelen));
return(XFS_ERROR(EEXIST));
}
}
@@ -2090,18 +2022,17 @@ xfs_attr_leaf_getvalue(xfs_dabuf_t *bp, xfs_da_args_t *args)
xfs_attr_leaf_name_remote_t *name_rmt;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
- ASSERT(INT_GET(leaf->hdr.count, ARCH_CONVERT)
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.count)
< (XFS_LBSIZE(args->dp->i_mount)/8));
- ASSERT(args->index < ((int)INT_GET(leaf->hdr.count, ARCH_CONVERT)));
+ ASSERT(args->index < be16_to_cpu(leaf->hdr.count));
entry = &leaf->entries[args->index];
if (entry->flags & XFS_ATTR_LOCAL) {
name_loc = XFS_ATTR_LEAF_NAME_LOCAL(leaf, args->index);
ASSERT(name_loc->namelen == args->namelen);
ASSERT(memcmp(args->name, name_loc->nameval, args->namelen) == 0);
- valuelen = INT_GET(name_loc->valuelen, ARCH_CONVERT);
+ valuelen = be16_to_cpu(name_loc->valuelen);
if (args->flags & ATTR_KERNOVAL) {
args->valuelen = valuelen;
return(0);
@@ -2116,8 +2047,8 @@ xfs_attr_leaf_getvalue(xfs_dabuf_t *bp, xfs_da_args_t *args)
name_rmt = XFS_ATTR_LEAF_NAME_REMOTE(leaf, args->index);
ASSERT(name_rmt->namelen == args->namelen);
ASSERT(memcmp(args->name, name_rmt->name, args->namelen) == 0);
- valuelen = INT_GET(name_rmt->valuelen, ARCH_CONVERT);
- args->rmtblkno = INT_GET(name_rmt->valueblk, ARCH_CONVERT);
+ valuelen = be32_to_cpu(name_rmt->valuelen);
+ args->rmtblkno = be32_to_cpu(name_rmt->valueblk);
args->rmtblkcnt = XFS_B_TO_FSB(args->dp->i_mount, valuelen);
if (args->flags & ATTR_KERNOVAL) {
args->valuelen = valuelen;
@@ -2159,32 +2090,29 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
/*
* Set up environment.
*/
- ASSERT(INT_GET(leaf_s->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
- ASSERT(INT_GET(leaf_d->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf_s->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf_d->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
hdr_s = &leaf_s->hdr;
hdr_d = &leaf_d->hdr;
- ASSERT((INT_GET(hdr_s->count, ARCH_CONVERT) > 0)
- && (INT_GET(hdr_s->count, ARCH_CONVERT)
- < (XFS_LBSIZE(mp)/8)));
- ASSERT(INT_GET(hdr_s->firstused, ARCH_CONVERT) >=
- ((INT_GET(hdr_s->count, ARCH_CONVERT)
+ ASSERT((be16_to_cpu(hdr_s->count) > 0) &&
+ (be16_to_cpu(hdr_s->count) < (XFS_LBSIZE(mp)/8)));
+ ASSERT(be16_to_cpu(hdr_s->firstused) >=
+ ((be16_to_cpu(hdr_s->count)
* sizeof(*entry_s))+sizeof(*hdr_s)));
- ASSERT(INT_GET(hdr_d->count, ARCH_CONVERT) < (XFS_LBSIZE(mp)/8));
- ASSERT(INT_GET(hdr_d->firstused, ARCH_CONVERT) >=
- ((INT_GET(hdr_d->count, ARCH_CONVERT)
+ ASSERT(be16_to_cpu(hdr_d->count) < (XFS_LBSIZE(mp)/8));
+ ASSERT(be16_to_cpu(hdr_d->firstused) >=
+ ((be16_to_cpu(hdr_d->count)
* sizeof(*entry_d))+sizeof(*hdr_d)));
- ASSERT(start_s < INT_GET(hdr_s->count, ARCH_CONVERT));
- ASSERT(start_d <= INT_GET(hdr_d->count, ARCH_CONVERT));
- ASSERT(count <= INT_GET(hdr_s->count, ARCH_CONVERT));
+ ASSERT(start_s < be16_to_cpu(hdr_s->count));
+ ASSERT(start_d <= be16_to_cpu(hdr_d->count));
+ ASSERT(count <= be16_to_cpu(hdr_s->count));
/*
* Move the entries in the destination leaf up to make a hole?
*/
- if (start_d < INT_GET(hdr_d->count, ARCH_CONVERT)) {
- tmp = INT_GET(hdr_d->count, ARCH_CONVERT) - start_d;
+ if (start_d < be16_to_cpu(hdr_d->count)) {
+ tmp = be16_to_cpu(hdr_d->count) - start_d;
tmp *= sizeof(xfs_attr_leaf_entry_t);
entry_s = &leaf_d->entries[start_d];
entry_d = &leaf_d->entries[start_d + count];
@@ -2199,8 +2127,8 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
entry_d = &leaf_d->entries[start_d];
desti = start_d;
for (i = 0; i < count; entry_s++, entry_d++, desti++, i++) {
- ASSERT(INT_GET(entry_s->nameidx, ARCH_CONVERT)
- >= INT_GET(hdr_s->firstused, ARCH_CONVERT));
+ ASSERT(be16_to_cpu(entry_s->nameidx)
+ >= be16_to_cpu(hdr_s->firstused));
tmp = xfs_attr_leaf_entsize(leaf_s, start_s + i);
#ifdef GROT
/*
@@ -2210,35 +2138,35 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
*/
if (entry_s->flags & XFS_ATTR_INCOMPLETE) { /* skip partials? */
memset(XFS_ATTR_LEAF_NAME(leaf_s, start_s + i), 0, tmp);
- INT_MOD(hdr_s->usedbytes, ARCH_CONVERT, -tmp);
- INT_MOD(hdr_s->count, ARCH_CONVERT, -1);
+ be16_add(&hdr_s->usedbytes, -tmp);
+ be16_add(&hdr_s->count, -1);
entry_d--; /* to compensate for ++ in loop hdr */
desti--;
if ((start_s + i) < offset)
result++; /* insertion index adjustment */
} else {
#endif /* GROT */
- INT_MOD(hdr_d->firstused, ARCH_CONVERT, -tmp);
+ be16_add(&hdr_d->firstused, -tmp);
/* both on-disk, don't endian flip twice */
entry_d->hashval = entry_s->hashval;
/* both on-disk, don't endian flip twice */
entry_d->nameidx = hdr_d->firstused;
entry_d->flags = entry_s->flags;
- ASSERT(INT_GET(entry_d->nameidx, ARCH_CONVERT) + tmp
+ ASSERT(be16_to_cpu(entry_d->nameidx) + tmp
<= XFS_LBSIZE(mp));
memmove(XFS_ATTR_LEAF_NAME(leaf_d, desti),
XFS_ATTR_LEAF_NAME(leaf_s, start_s + i), tmp);
- ASSERT(INT_GET(entry_s->nameidx, ARCH_CONVERT) + tmp
+ ASSERT(be16_to_cpu(entry_s->nameidx) + tmp
<= XFS_LBSIZE(mp));
memset(XFS_ATTR_LEAF_NAME(leaf_s, start_s + i), 0, tmp);
- INT_MOD(hdr_s->usedbytes, ARCH_CONVERT, -tmp);
- INT_MOD(hdr_d->usedbytes, ARCH_CONVERT, tmp);
- INT_MOD(hdr_s->count, ARCH_CONVERT, -1);
- INT_MOD(hdr_d->count, ARCH_CONVERT, 1);
- tmp = INT_GET(hdr_d->count, ARCH_CONVERT)
+ be16_add(&hdr_s->usedbytes, -tmp);
+ be16_add(&hdr_d->usedbytes, tmp);
+ be16_add(&hdr_s->count, -1);
+ be16_add(&hdr_d->count, 1);
+ tmp = be16_to_cpu(hdr_d->count)
* sizeof(xfs_attr_leaf_entry_t)
+ sizeof(xfs_attr_leaf_hdr_t);
- ASSERT(INT_GET(hdr_d->firstused, ARCH_CONVERT) >= tmp);
+ ASSERT(be16_to_cpu(hdr_d->firstused) >= tmp);
#ifdef GROT
}
#endif /* GROT */
@@ -2247,7 +2175,7 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
/*
* Zero out the entries we just copied.
*/
- if (start_s == INT_GET(hdr_s->count, ARCH_CONVERT)) {
+ if (start_s == be16_to_cpu(hdr_s->count)) {
tmp = count * sizeof(xfs_attr_leaf_entry_t);
entry_s = &leaf_s->entries[start_s];
ASSERT(((char *)entry_s + tmp) <=
@@ -2258,15 +2186,14 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
* Move the remaining entries down to fill the hole,
* then zero the entries at the top.
*/
- tmp = INT_GET(hdr_s->count, ARCH_CONVERT) - count;
+ tmp = be16_to_cpu(hdr_s->count) - count;
tmp *= sizeof(xfs_attr_leaf_entry_t);
entry_s = &leaf_s->entries[start_s + count];
entry_d = &leaf_s->entries[start_s];
memmove((char *)entry_d, (char *)entry_s, tmp);
tmp = count * sizeof(xfs_attr_leaf_entry_t);
- entry_s = &leaf_s->entries[INT_GET(hdr_s->count,
- ARCH_CONVERT)];
+ entry_s = &leaf_s->entries[be16_to_cpu(hdr_s->count)];
ASSERT(((char *)entry_s + tmp) <=
((char *)leaf_s + XFS_LBSIZE(mp)));
memset((char *)entry_s, 0, tmp);
@@ -2275,14 +2202,11 @@ xfs_attr_leaf_moveents(xfs_attr_leafblock_t *leaf_s, int start_s,
/*
* Fill in the freemap information
*/
- INT_SET(hdr_d->freemap[0].base, ARCH_CONVERT,
- sizeof(xfs_attr_leaf_hdr_t));
- INT_MOD(hdr_d->freemap[0].base, ARCH_CONVERT,
- INT_GET(hdr_d->count, ARCH_CONVERT)
- * sizeof(xfs_attr_leaf_entry_t));
- INT_SET(hdr_d->freemap[0].size, ARCH_CONVERT,
- INT_GET(hdr_d->firstused, ARCH_CONVERT)
- - INT_GET(hdr_d->freemap[0].base, ARCH_CONVERT));
+ hdr_d->freemap[0].base = cpu_to_be16(sizeof(xfs_attr_leaf_hdr_t));
+ be16_add(&hdr_d->freemap[0].base, be16_to_cpu(hdr_d->count) *
+ sizeof(xfs_attr_leaf_entry_t));
+ hdr_d->freemap[0].size = cpu_to_be16(be16_to_cpu(hdr_d->firstused)
+ - be16_to_cpu(hdr_d->freemap[0].base));
hdr_d->freemap[1].base = 0;
hdr_d->freemap[2].base = 0;
hdr_d->freemap[1].size = 0;
@@ -2301,18 +2225,16 @@ xfs_attr_leaf_order(xfs_dabuf_t *leaf1_bp, xfs_dabuf_t *leaf2_bp)
leaf1 = leaf1_bp->data;
leaf2 = leaf2_bp->data;
- ASSERT((INT_GET(leaf1->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC) &&
- (INT_GET(leaf2->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC));
- if ( (INT_GET(leaf1->hdr.count, ARCH_CONVERT) > 0)
- && (INT_GET(leaf2->hdr.count, ARCH_CONVERT) > 0)
- && ( (INT_GET(leaf2->entries[ 0 ].hashval, ARCH_CONVERT) <
- INT_GET(leaf1->entries[ 0 ].hashval, ARCH_CONVERT))
- || (INT_GET(leaf2->entries[INT_GET(leaf2->hdr.count,
- ARCH_CONVERT)-1].hashval, ARCH_CONVERT) <
- INT_GET(leaf1->entries[INT_GET(leaf1->hdr.count,
- ARCH_CONVERT)-1].hashval, ARCH_CONVERT))) ) {
+ ASSERT((be16_to_cpu(leaf1->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC) &&
+ (be16_to_cpu(leaf2->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC));
+ if ((be16_to_cpu(leaf1->hdr.count) > 0) &&
+ (be16_to_cpu(leaf2->hdr.count) > 0) &&
+ ((be32_to_cpu(leaf2->entries[0].hashval) <
+ be32_to_cpu(leaf1->entries[0].hashval)) ||
+ (be32_to_cpu(leaf2->entries[
+ be16_to_cpu(leaf2->hdr.count)-1].hashval) <
+ be32_to_cpu(leaf1->entries[
+ be16_to_cpu(leaf1->hdr.count)-1].hashval)))) {
return(1);
}
return(0);
@@ -2327,14 +2249,12 @@ xfs_attr_leaf_lasthash(xfs_dabuf_t *bp, int *count)
xfs_attr_leafblock_t *leaf;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
if (count)
- *count = INT_GET(leaf->hdr.count, ARCH_CONVERT);
+ *count = be16_to_cpu(leaf->hdr.count);
if (!leaf->hdr.count)
return(0);
- return(INT_GET(leaf->entries[INT_GET(leaf->hdr.count,
- ARCH_CONVERT)-1].hashval, ARCH_CONVERT));
+ return be32_to_cpu(leaf->entries[be16_to_cpu(leaf->hdr.count)-1].hashval);
}
/*
@@ -2348,13 +2268,11 @@ xfs_attr_leaf_entsize(xfs_attr_leafblock_t *leaf, int index)
xfs_attr_leaf_name_remote_t *name_rmt;
int size;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
if (leaf->entries[index].flags & XFS_ATTR_LOCAL) {
name_loc = XFS_ATTR_LEAF_NAME_LOCAL(leaf, index);
size = XFS_ATTR_LEAF_ENTSIZE_LOCAL(name_loc->namelen,
- INT_GET(name_loc->valuelen,
- ARCH_CONVERT));
+ be16_to_cpu(name_loc->valuelen));
} else {
name_rmt = XFS_ATTR_LEAF_NAME_REMOTE(leaf, index);
size = XFS_ATTR_LEAF_ENTSIZE_REMOTE(name_rmt->namelen);
@@ -2412,22 +2330,20 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
*/
if (context->resynch) {
entry = &leaf->entries[0];
- for (i = 0; i < INT_GET(leaf->hdr.count, ARCH_CONVERT);
- entry++, i++) {
- if (INT_GET(entry->hashval, ARCH_CONVERT)
- == cursor->hashval) {
+ for (i = 0; i < be16_to_cpu(leaf->hdr.count); entry++, i++) {
+ if (be32_to_cpu(entry->hashval) == cursor->hashval) {
if (cursor->offset == context->dupcnt) {
context->dupcnt = 0;
break;
}
context->dupcnt++;
- } else if (INT_GET(entry->hashval, ARCH_CONVERT)
- > cursor->hashval) {
+ } else if (be32_to_cpu(entry->hashval) >
+ cursor->hashval) {
context->dupcnt = 0;
break;
}
}
- if (i == INT_GET(leaf->hdr.count, ARCH_CONVERT)) {
+ if (i == be16_to_cpu(leaf->hdr.count)) {
xfs_attr_trace_l_c("not found", context);
return(0);
}
@@ -2441,12 +2357,12 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
* We have found our place, start copying out the new attributes.
*/
retval = 0;
- for ( ; (i < INT_GET(leaf->hdr.count, ARCH_CONVERT))
+ for ( ; (i < be16_to_cpu(leaf->hdr.count))
&& (retval == 0); entry++, i++) {
attrnames_t *namesp;
- if (INT_GET(entry->hashval, ARCH_CONVERT) != cursor->hashval) {
- cursor->hashval = INT_GET(entry->hashval, ARCH_CONVERT);
+ if (be32_to_cpu(entry->hashval) != cursor->hashval) {
+ cursor->hashval = be32_to_cpu(entry->hashval);
cursor->offset = 0;
}
@@ -2475,8 +2391,7 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
retval = xfs_attr_put_listent(context, namesp,
(char *)name_loc->nameval,
(int)name_loc->namelen,
- (int)INT_GET(name_loc->valuelen,
- ARCH_CONVERT));
+ be16_to_cpu(name_loc->valuelen));
}
} else {
name_rmt = XFS_ATTR_LEAF_NAME_REMOTE(leaf, i);
@@ -2488,8 +2403,7 @@ xfs_attr_leaf_list_int(xfs_dabuf_t *bp, xfs_attr_list_context_t *context)
retval = xfs_attr_put_listent(context, namesp,
(char *)name_rmt->name,
(int)name_rmt->namelen,
- (int)INT_GET(name_rmt->valuelen,
- ARCH_CONVERT));
+ be32_to_cpu(name_rmt->valuelen));
}
}
if (retval == 0) {
@@ -2596,9 +2510,8 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
ASSERT(bp != NULL);
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
- ASSERT(args->index < INT_GET(leaf->hdr.count, ARCH_CONVERT));
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(args->index < be16_to_cpu(leaf->hdr.count));
ASSERT(args->index >= 0);
entry = &leaf->entries[ args->index ];
ASSERT(entry->flags & XFS_ATTR_INCOMPLETE);
@@ -2613,7 +2526,7 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
namelen = name_rmt->namelen;
name = (char *)name_rmt->name;
}
- ASSERT(INT_GET(entry->hashval, ARCH_CONVERT) == args->hashval);
+ ASSERT(be32_to_cpu(entry->hashval) == args->hashval);
ASSERT(namelen == args->namelen);
ASSERT(memcmp(name, args->name, namelen) == 0);
#endif /* DEBUG */
@@ -2625,8 +2538,8 @@ xfs_attr_leaf_clearflag(xfs_da_args_t *args)
if (args->rmtblkno) {
ASSERT((entry->flags & XFS_ATTR_LOCAL) == 0);
name_rmt = XFS_ATTR_LEAF_NAME_REMOTE(leaf, args->index);
- INT_SET(name_rmt->valueblk, ARCH_CONVERT, args->rmtblkno);
- INT_SET(name_rmt->valuelen, ARCH_CONVERT, args->valuelen);
+ name_rmt->valueblk = cpu_to_be32(args->rmtblkno);
+ name_rmt->valuelen = cpu_to_be32(args->valuelen);
xfs_da_log_buf(args->trans, bp,
XFS_DA_LOGRANGE(leaf, name_rmt, sizeof(*name_rmt)));
}
@@ -2663,9 +2576,8 @@ xfs_attr_leaf_setflag(xfs_da_args_t *args)
ASSERT(bp != NULL);
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
- ASSERT(args->index < INT_GET(leaf->hdr.count, ARCH_CONVERT));
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(args->index < be16_to_cpu(leaf->hdr.count));
ASSERT(args->index >= 0);
entry = &leaf->entries[ args->index ];
@@ -2736,16 +2648,14 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
}
leaf1 = bp1->data;
- ASSERT(INT_GET(leaf1->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
- ASSERT(args->index < INT_GET(leaf1->hdr.count, ARCH_CONVERT));
+ ASSERT(be16_to_cpu(leaf1->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(args->index < be16_to_cpu(leaf1->hdr.count));
ASSERT(args->index >= 0);
entry1 = &leaf1->entries[ args->index ];
leaf2 = bp2->data;
- ASSERT(INT_GET(leaf2->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
- ASSERT(args->index2 < INT_GET(leaf2->hdr.count, ARCH_CONVERT));
+ ASSERT(be16_to_cpu(leaf2->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(args->index2 < be16_to_cpu(leaf2->hdr.count));
ASSERT(args->index2 >= 0);
entry2 = &leaf2->entries[ args->index2 ];
@@ -2768,7 +2678,7 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
namelen2 = name_rmt->namelen;
name2 = (char *)name_rmt->name;
}
- ASSERT(INT_GET(entry1->hashval, ARCH_CONVERT) == INT_GET(entry2->hashval, ARCH_CONVERT));
+ ASSERT(be32_to_cpu(entry1->hashval) == be32_to_cpu(entry2->hashval));
ASSERT(namelen1 == namelen2);
ASSERT(memcmp(name1, name2, namelen1) == 0);
#endif /* DEBUG */
@@ -2782,8 +2692,8 @@ xfs_attr_leaf_flipflags(xfs_da_args_t *args)
if (args->rmtblkno) {
ASSERT((entry1->flags & XFS_ATTR_LOCAL) == 0);
name_rmt = XFS_ATTR_LEAF_NAME_REMOTE(leaf1, args->index);
- INT_SET(name_rmt->valueblk, ARCH_CONVERT, args->rmtblkno);
- INT_SET(name_rmt->valuelen, ARCH_CONVERT, args->valuelen);
+ name_rmt->valueblk = cpu_to_be32(args->rmtblkno);
+ name_rmt->valuelen = cpu_to_be32(args->valuelen);
xfs_da_log_buf(args->trans, bp1,
XFS_DA_LOGRANGE(leaf1, name_rmt, sizeof(*name_rmt)));
}
@@ -2842,9 +2752,9 @@ xfs_attr_root_inactive(xfs_trans_t **trans, xfs_inode_t *dp)
* This is a depth-first traversal!
*/
info = bp->data;
- if (INT_GET(info->magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC) {
+ if (be16_to_cpu(info->magic) == XFS_DA_NODE_MAGIC) {
error = xfs_attr_node_inactive(trans, dp, bp, 1);
- } else if (INT_GET(info->magic, ARCH_CONVERT) == XFS_ATTR_LEAF_MAGIC) {
+ } else if (be16_to_cpu(info->magic) == XFS_ATTR_LEAF_MAGIC) {
error = xfs_attr_leaf_inactive(trans, dp, bp);
} else {
error = XFS_ERROR(EIO);
@@ -2892,15 +2802,14 @@ xfs_attr_node_inactive(xfs_trans_t **trans, xfs_inode_t *dp, xfs_dabuf_t *bp,
}
node = bp->data;
- ASSERT(INT_GET(node->hdr.info.magic, ARCH_CONVERT)
- == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
parent_blkno = xfs_da_blkno(bp); /* save for re-read later */
- count = INT_GET(node->hdr.count, ARCH_CONVERT);
+ count = be16_to_cpu(node->hdr.count);
if (!count) {
xfs_da_brelse(*trans, bp);
return(0);
}
- child_fsb = INT_GET(node->btree[0].before, ARCH_CONVERT);
+ child_fsb = be32_to_cpu(node->btree[0].before);
xfs_da_brelse(*trans, bp); /* no locks for later trans */
/*
@@ -2927,12 +2836,10 @@ xfs_attr_node_inactive(xfs_trans_t **trans, xfs_inode_t *dp, xfs_dabuf_t *bp,
* Invalidate the subtree, however we have to.
*/
info = child_bp->data;
- if (INT_GET(info->magic, ARCH_CONVERT)
- == XFS_DA_NODE_MAGIC) {
+ if (be16_to_cpu(info->magic) == XFS_DA_NODE_MAGIC) {
error = xfs_attr_node_inactive(trans, dp,
child_bp, level+1);
- } else if (INT_GET(info->magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC) {
+ } else if (be16_to_cpu(info->magic) == XFS_ATTR_LEAF_MAGIC) {
error = xfs_attr_leaf_inactive(trans, dp,
child_bp);
} else {
@@ -2962,7 +2869,7 @@ xfs_attr_node_inactive(xfs_trans_t **trans, xfs_inode_t *dp, xfs_dabuf_t *bp,
&bp, XFS_ATTR_FORK);
if (error)
return(error);
- child_fsb = INT_GET(node->btree[i+1].before, ARCH_CONVERT);
+ child_fsb = be32_to_cpu(node->btree[i+1].before);
xfs_da_brelse(*trans, bp);
}
/*
@@ -2991,17 +2898,16 @@ xfs_attr_leaf_inactive(xfs_trans_t **trans, xfs_inode_t *dp, xfs_dabuf_t *bp)
int error, count, size, tmp, i;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT)
- == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_ATTR_LEAF_MAGIC);
/*
* Count the number of "remote" value extents.
*/
count = 0;
entry = &leaf->entries[0];
- for (i = 0; i < INT_GET(leaf->hdr.count, ARCH_CONVERT); entry++, i++) {
- if ( INT_GET(entry->nameidx, ARCH_CONVERT)
- && ((entry->flags & XFS_ATTR_LOCAL) == 0)) {
+ for (i = 0; i < be16_to_cpu(leaf->hdr.count); entry++, i++) {
+ if (be16_to_cpu(entry->nameidx) &&
+ ((entry->flags & XFS_ATTR_LOCAL) == 0)) {
name_rmt = XFS_ATTR_LEAF_NAME_REMOTE(leaf, i);
if (name_rmt->valueblk)
count++;
@@ -3027,17 +2933,14 @@ xfs_attr_leaf_inactive(xfs_trans_t **trans, xfs_inode_t *dp, xfs_dabuf_t *bp)
*/
lp = list;
entry = &leaf->entries[0];
- for (i = 0; i < INT_GET(leaf->hdr.count, ARCH_CONVERT); entry++, i++) {
- if ( INT_GET(entry->nameidx, ARCH_CONVERT)
- && ((entry->flags & XFS_ATTR_LOCAL) == 0)) {
+ for (i = 0; i < be16_to_cpu(leaf->hdr.count); entry++, i++) {
+ if (be16_to_cpu(entry->nameidx) &&
+ ((entry->flags & XFS_ATTR_LOCAL) == 0)) {
name_rmt = XFS_ATTR_LEAF_NAME_REMOTE(leaf, i);
if (name_rmt->valueblk) {
- /* both on-disk, don't endian flip twice */
- lp->valueblk = name_rmt->valueblk;
- INT_SET(lp->valuelen, ARCH_CONVERT,
- XFS_B_TO_FSB(dp->i_mount,
- INT_GET(name_rmt->valuelen,
- ARCH_CONVERT)));
+ lp->valueblk = be32_to_cpu(name_rmt->valueblk);
+ lp->valuelen = XFS_B_TO_FSB(dp->i_mount,
+ be32_to_cpu(name_rmt->valuelen));
lp++;
}
}
@@ -3050,10 +2953,8 @@ xfs_attr_leaf_inactive(xfs_trans_t **trans, xfs_inode_t *dp, xfs_dabuf_t *bp)
error = 0;
for (lp = list, i = 0; i < count; i++, lp++) {
tmp = xfs_attr_leaf_freextent(trans, dp,
- INT_GET(lp->valueblk,
- ARCH_CONVERT),
- INT_GET(lp->valuelen,
- ARCH_CONVERT));
+ lp->valueblk, lp->valuelen);
+
if (error == 0)
error = tmp; /* save only the 1st errno */
}
diff --git a/fs/xfs/xfs_attr_leaf.h b/fs/xfs/xfs_attr_leaf.h
index 541e34109bb9..51c3ee156b2f 100644
--- a/fs/xfs/xfs_attr_leaf.h
+++ b/fs/xfs/xfs_attr_leaf.h
@@ -73,39 +73,39 @@ struct xfs_trans;
#define XFS_ATTR_LEAF_MAPSIZE 3 /* how many freespace slots */
typedef struct xfs_attr_leaf_map { /* RLE map of free bytes */
- __uint16_t base; /* base of free region */
- __uint16_t size; /* length of free region */
+ __be16 base; /* base of free region */
+ __be16 size; /* length of free region */
} xfs_attr_leaf_map_t;
typedef struct xfs_attr_leaf_hdr { /* constant-structure header block */
xfs_da_blkinfo_t info; /* block type, links, etc. */
- __uint16_t count; /* count of active leaf_entry's */
- __uint16_t usedbytes; /* num bytes of names/values stored */
- __uint16_t firstused; /* first used byte in name area */
- __uint8_t holes; /* != 0 if blk needs compaction */
- __uint8_t pad1;
+ __be16 count; /* count of active leaf_entry's */
+ __be16 usedbytes; /* num bytes of names/values stored */
+ __be16 firstused; /* first used byte in name area */
+ __u8 holes; /* != 0 if blk needs compaction */
+ __u8 pad1;
xfs_attr_leaf_map_t freemap[XFS_ATTR_LEAF_MAPSIZE];
/* N largest free regions */
} xfs_attr_leaf_hdr_t;
typedef struct xfs_attr_leaf_entry { /* sorted on key, not name */
- xfs_dahash_t hashval; /* hash value of name */
- __uint16_t nameidx; /* index into buffer of name/value */
- __uint8_t flags; /* LOCAL/ROOT/SECURE/INCOMPLETE flag */
- __uint8_t pad2; /* unused pad byte */
+ __be32 hashval; /* hash value of name */
+ __be16 nameidx; /* index into buffer of name/value */
+ __u8 flags; /* LOCAL/ROOT/SECURE/INCOMPLETE flag */
+ __u8 pad2; /* unused pad byte */
} xfs_attr_leaf_entry_t;
typedef struct xfs_attr_leaf_name_local {
- __uint16_t valuelen; /* number of bytes in value */
- __uint8_t namelen; /* length of name bytes */
- __uint8_t nameval[1]; /* name/value bytes */
+ __be16 valuelen; /* number of bytes in value */
+ __u8 namelen; /* length of name bytes */
+ __u8 nameval[1]; /* name/value bytes */
} xfs_attr_leaf_name_local_t;
typedef struct xfs_attr_leaf_name_remote {
- xfs_dablk_t valueblk; /* block number of value bytes */
- __uint32_t valuelen; /* number of bytes in value */
- __uint8_t namelen; /* length of name bytes */
- __uint8_t name[1]; /* name bytes */
+ __be32 valueblk; /* block number of value bytes */
+ __be32 valuelen; /* number of bytes in value */
+ __u8 namelen; /* length of name bytes */
+ __u8 name[1]; /* name bytes */
} xfs_attr_leaf_name_remote_t;
typedef struct xfs_attr_leafblock {
@@ -143,8 +143,8 @@ typedef struct xfs_attr_leafblock {
static inline xfs_attr_leaf_name_remote_t *
xfs_attr_leaf_name_remote(xfs_attr_leafblock_t *leafp, int idx)
{
- return (xfs_attr_leaf_name_remote_t *) &((char *)
- (leafp))[INT_GET((leafp)->entries[idx].nameidx, ARCH_CONVERT)];
+ return (xfs_attr_leaf_name_remote_t *)
+ &((char *)leafp)[be16_to_cpu(leafp->entries[idx].nameidx)];
}
#define XFS_ATTR_LEAF_NAME_LOCAL(leafp,idx) \
@@ -152,16 +152,15 @@ xfs_attr_leaf_name_remote(xfs_attr_leafblock_t *leafp, int idx)
static inline xfs_attr_leaf_name_local_t *
xfs_attr_leaf_name_local(xfs_attr_leafblock_t *leafp, int idx)
{
- return (xfs_attr_leaf_name_local_t *) &((char *)
- (leafp))[INT_GET((leafp)->entries[idx].nameidx, ARCH_CONVERT)];
+ return (xfs_attr_leaf_name_local_t *)
+ &((char *)leafp)[be16_to_cpu(leafp->entries[idx].nameidx)];
}
#define XFS_ATTR_LEAF_NAME(leafp,idx) \
xfs_attr_leaf_name(leafp,idx)
static inline char *xfs_attr_leaf_name(xfs_attr_leafblock_t *leafp, int idx)
{
- return (&((char *)
- (leafp))[INT_GET((leafp)->entries[idx].nameidx, ARCH_CONVERT)]);
+ return &((char *)leafp)[be16_to_cpu(leafp->entries[idx].nameidx)];
}
/*
diff --git a/fs/xfs/xfs_attr_sf.h b/fs/xfs/xfs_attr_sf.h
index ffed6ca81a52..f67f917803b1 100644
--- a/fs/xfs/xfs_attr_sf.h
+++ b/fs/xfs/xfs_attr_sf.h
@@ -32,8 +32,8 @@ struct xfs_inode;
*/
typedef struct xfs_attr_shortform {
struct xfs_attr_sf_hdr { /* constant-structure header block */
- __uint16_t totsize; /* total bytes in shortform list */
- __uint8_t count; /* count of active entries */
+ __be16 totsize; /* total bytes in shortform list */
+ __u8 count; /* count of active entries */
} hdr;
struct xfs_attr_sf_entry {
__uint8_t namelen; /* actual length of name (no NULL) */
@@ -66,8 +66,8 @@ typedef struct xfs_attr_sf_sort {
#define XFS_ATTR_SF_NEXTENTRY(sfep) /* next entry in struct */ \
((xfs_attr_sf_entry_t *)((char *)(sfep) + XFS_ATTR_SF_ENTSIZE(sfep)))
#define XFS_ATTR_SF_TOTSIZE(dp) /* total space in use */ \
- (INT_GET(((xfs_attr_shortform_t *) \
- ((dp)->i_afp->if_u1.if_data))->hdr.totsize, ARCH_CONVERT))
+ (be16_to_cpu(((xfs_attr_shortform_t *) \
+ ((dp)->i_afp->if_u1.if_data))->hdr.totsize))
#if defined(XFS_ATTR_TRACE)
/*
diff --git a/fs/xfs/xfs_bmap.c b/fs/xfs/xfs_bmap.c
index 70625e577c70..2d702e4a74a3 100644
--- a/fs/xfs/xfs_bmap.c
+++ b/fs/xfs/xfs_bmap.c
@@ -89,7 +89,7 @@ xfs_bmap_add_attrfork_local(
int *flags); /* inode logging flags */
/*
- * Called by xfs_bmapi to update extent list structure and the btree
+ * Called by xfs_bmapi to update file extent records and the btree
* after allocating space (or doing a delayed allocation).
*/
STATIC int /* error */
@@ -97,7 +97,7 @@ xfs_bmap_add_extent(
xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* extent number to update/insert */
xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
xfs_fsblock_t *first, /* pointer to firstblock variable */
xfs_bmap_free_t *flist, /* list of extents to be freed */
int *logflagsp, /* inode logging flags */
@@ -113,7 +113,7 @@ xfs_bmap_add_extent_delay_real(
xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* extent number to update/insert */
xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
xfs_filblks_t *dnew, /* new delayed-alloc indirect blocks */
xfs_fsblock_t *first, /* pointer to firstblock variable */
xfs_bmap_free_t *flist, /* list of extents to be freed */
@@ -129,7 +129,7 @@ xfs_bmap_add_extent_hole_delay(
xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* extent number to update/insert */
xfs_btree_cur_t *cur, /* if null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
int *logflagsp,/* inode logging flags */
int rsvd); /* OK to allocate reserved blocks */
@@ -142,7 +142,7 @@ xfs_bmap_add_extent_hole_real(
xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* extent number to update/insert */
xfs_btree_cur_t *cur, /* if null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
int *logflagsp, /* inode logging flags */
int whichfork); /* data or attr fork */
@@ -155,7 +155,7 @@ xfs_bmap_add_extent_unwritten_real(
xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* extent number to update/insert */
xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
int *logflagsp); /* inode logging flags */
/*
@@ -169,7 +169,7 @@ xfs_bmap_alloc(
/*
* Transform a btree format file with only one leaf node, where the
* extents list will fit in the inode, into an extents format file.
- * Since the extent list is already in-core, all we have to do is
+ * Since the file extents are already in-core, all we have to do is
* give up the space for the btree root and pitch the leaf block.
*/
STATIC int /* error */
@@ -191,7 +191,7 @@ xfs_bmap_check_extents(
#endif
/*
- * Called by xfs_bmapi to update extent list structure and the btree
+ * Called by xfs_bmapi to update file extent records and the btree
* after removing space (or undoing a delayed allocation).
*/
STATIC int /* error */
@@ -201,7 +201,7 @@ xfs_bmap_del_extent(
xfs_extnum_t idx, /* extent number to update/insert */
xfs_bmap_free_t *flist, /* list of extents to be freed */
xfs_btree_cur_t *cur, /* if null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
int *logflagsp,/* inode logging flags */
int whichfork, /* data or attr fork */
int rsvd); /* OK to allocate reserved blocks */
@@ -217,18 +217,6 @@ xfs_bmap_del_free(
xfs_bmap_free_item_t *free); /* list item to be freed */
/*
- * Remove count entries from the extents array for inode "ip", starting
- * at index "idx". Copies the remaining items down over the deleted ones,
- * and gives back the excess memory.
- */
-STATIC void
-xfs_bmap_delete_exlist(
- xfs_inode_t *ip, /* incode inode pointer */
- xfs_extnum_t idx, /* starting delete index */
- xfs_extnum_t count, /* count of items to delete */
- int whichfork); /* data or attr fork */
-
-/*
* Convert an extents-format file into a btree-format file.
* The new file will have a root block (in the inode) and a single child block.
*/
@@ -244,18 +232,6 @@ xfs_bmap_extents_to_btree(
int whichfork); /* data or attr fork */
/*
- * Insert new item(s) in the extent list for inode "ip".
- * Count new items are inserted at offset idx.
- */
-STATIC void
-xfs_bmap_insert_exlist(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* starting index of new items */
- xfs_extnum_t count, /* number of inserted items */
- xfs_bmbt_irec_t *new, /* items to insert */
- int whichfork); /* data or attr fork */
-
-/*
* Convert a local file to an extents file.
* This code is sort of bogus, since the file data needs to get
* logged so it won't be lost. The bmap-level manipulations are ok, though.
@@ -316,7 +292,7 @@ xfs_bmap_trace_addentry(
int whichfork); /* data or attr fork */
/*
- * Add bmap trace entry prior to a call to xfs_bmap_delete_exlist.
+ * Add bmap trace entry prior to a call to xfs_iext_remove.
*/
STATIC void
xfs_bmap_trace_delete(
@@ -328,7 +304,7 @@ xfs_bmap_trace_delete(
int whichfork); /* data or attr fork */
/*
- * Add bmap trace entry prior to a call to xfs_bmap_insert_exlist, or
+ * Add bmap trace entry prior to a call to xfs_iext_insert, or
* reading in the extents list from the disk (in the btree).
*/
STATIC void
@@ -343,7 +319,7 @@ xfs_bmap_trace_insert(
int whichfork); /* data or attr fork */
/*
- * Add bmap trace entry after updating an extent list entry in place.
+ * Add bmap trace entry after updating an extent record in place.
*/
STATIC void
xfs_bmap_trace_post_update(
@@ -354,7 +330,7 @@ xfs_bmap_trace_post_update(
int whichfork); /* data or attr fork */
/*
- * Add bmap trace entry prior to updating an extent list entry in place.
+ * Add bmap trace entry prior to updating an extent record in place.
*/
STATIC void
xfs_bmap_trace_pre_update(
@@ -413,19 +389,24 @@ STATIC int
xfs_bmap_count_tree(
xfs_mount_t *mp,
xfs_trans_t *tp,
+ xfs_ifork_t *ifp,
xfs_fsblock_t blockno,
int levelin,
int *count);
STATIC int
xfs_bmap_count_leaves(
- xfs_bmbt_rec_t *frp,
+ xfs_ifork_t *ifp,
+ xfs_extnum_t idx,
int numrecs,
int *count);
STATIC int
xfs_bmap_disk_count_leaves(
- xfs_bmbt_rec_t *frp,
+ xfs_ifork_t *ifp,
+ xfs_mount_t *mp,
+ xfs_extnum_t idx,
+ xfs_bmbt_block_t *block,
int numrecs,
int *count);
@@ -537,7 +518,7 @@ xfs_bmap_add_attrfork_local(
}
/*
- * Called by xfs_bmapi to update extent list structure and the btree
+ * Called by xfs_bmapi to update file extent records and the btree
* after allocating space (or doing a delayed allocation).
*/
STATIC int /* error */
@@ -545,7 +526,7 @@ xfs_bmap_add_extent(
xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* extent number to update/insert */
xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
xfs_fsblock_t *first, /* pointer to firstblock variable */
xfs_bmap_free_t *flist, /* list of extents to be freed */
int *logflagsp, /* inode logging flags */
@@ -578,7 +559,7 @@ xfs_bmap_add_extent(
if (nextents == 0) {
xfs_bmap_trace_insert(fname, "insert empty", ip, 0, 1, new,
NULL, whichfork);
- xfs_bmap_insert_exlist(ip, 0, 1, new, whichfork);
+ xfs_iext_insert(ifp, 0, 1, new);
ASSERT(cur == NULL);
ifp->if_lastex = 0;
if (!ISNULLSTARTBLOCK(new->br_startblock)) {
@@ -614,7 +595,7 @@ xfs_bmap_add_extent(
/*
* Get the record referred to by idx.
*/
- xfs_bmbt_get_all(&ifp->if_u1.if_extents[idx], &prev);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx), &prev);
/*
* If it's a real allocation record, and the new allocation ends
* after the start of the referred to record, then we're filling
@@ -714,14 +695,13 @@ xfs_bmap_add_extent_delay_real(
xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* extent number to update/insert */
xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
xfs_filblks_t *dnew, /* new delayed-alloc indirect blocks */
xfs_fsblock_t *first, /* pointer to firstblock variable */
xfs_bmap_free_t *flist, /* list of extents to be freed */
int *logflagsp, /* inode logging flags */
int rsvd) /* OK to use reserved data block allocation */
{
- xfs_bmbt_rec_t *base; /* base of extent entry list */
xfs_btree_cur_t *cur; /* btree cursor */
int diff; /* temp value */
xfs_bmbt_rec_t *ep; /* extent entry for idx */
@@ -730,6 +710,7 @@ xfs_bmap_add_extent_delay_real(
static char fname[] = "xfs_bmap_add_extent_delay_real";
#endif
int i; /* temp state */
+ xfs_ifork_t *ifp; /* inode fork pointer */
xfs_fileoff_t new_endoff; /* end offset of new entry */
xfs_bmbt_irec_t r[3]; /* neighbor extent entries */
/* left is 0, right is 1, prev is 2 */
@@ -763,8 +744,8 @@ xfs_bmap_add_extent_delay_real(
* Set up a bunch of variables to make the tests simpler.
*/
cur = *curp;
- base = ip->i_df.if_u1.if_extents;
- ep = &base[idx];
+ ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+ ep = xfs_iext_get_ext(ifp, idx);
xfs_bmbt_get_all(ep, &PREV);
new_endoff = new->br_startoff + new->br_blockcount;
ASSERT(PREV.br_startoff <= new->br_startoff);
@@ -781,7 +762,7 @@ xfs_bmap_add_extent_delay_real(
* Don't set contiguous if the combined extent would be too large.
*/
if (STATE_SET_TEST(LEFT_VALID, idx > 0)) {
- xfs_bmbt_get_all(ep - 1, &LEFT);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &LEFT);
STATE_SET(LEFT_DELAY, ISNULLSTARTBLOCK(LEFT.br_startblock));
}
STATE_SET(LEFT_CONTIG,
@@ -798,7 +779,7 @@ xfs_bmap_add_extent_delay_real(
if (STATE_SET_TEST(RIGHT_VALID,
idx <
ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1)) {
- xfs_bmbt_get_all(ep + 1, &RIGHT);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx + 1), &RIGHT);
STATE_SET(RIGHT_DELAY, ISNULLSTARTBLOCK(RIGHT.br_startblock));
}
STATE_SET(RIGHT_CONTIG,
@@ -825,14 +806,14 @@ xfs_bmap_add_extent_delay_real(
*/
xfs_bmap_trace_pre_update(fname, "LF|RF|LC|RC", ip, idx - 1,
XFS_DATA_FORK);
- xfs_bmbt_set_blockcount(ep - 1,
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
LEFT.br_blockcount + PREV.br_blockcount +
RIGHT.br_blockcount);
xfs_bmap_trace_post_update(fname, "LF|RF|LC|RC", ip, idx - 1,
XFS_DATA_FORK);
xfs_bmap_trace_delete(fname, "LF|RF|LC|RC", ip, idx, 2,
XFS_DATA_FORK);
- xfs_bmap_delete_exlist(ip, idx, 2, XFS_DATA_FORK);
+ xfs_iext_remove(ifp, idx, 2);
ip->i_df.if_lastex = idx - 1;
ip->i_d.di_nextents--;
if (cur == NULL)
@@ -867,14 +848,14 @@ xfs_bmap_add_extent_delay_real(
*/
xfs_bmap_trace_pre_update(fname, "LF|RF|LC", ip, idx - 1,
XFS_DATA_FORK);
- xfs_bmbt_set_blockcount(ep - 1,
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
LEFT.br_blockcount + PREV.br_blockcount);
xfs_bmap_trace_post_update(fname, "LF|RF|LC", ip, idx - 1,
XFS_DATA_FORK);
ip->i_df.if_lastex = idx - 1;
xfs_bmap_trace_delete(fname, "LF|RF|LC", ip, idx, 1,
XFS_DATA_FORK);
- xfs_bmap_delete_exlist(ip, idx, 1, XFS_DATA_FORK);
+ xfs_iext_remove(ifp, idx, 1);
if (cur == NULL)
rval = XFS_ILOG_DEXT;
else {
@@ -908,7 +889,7 @@ xfs_bmap_add_extent_delay_real(
ip->i_df.if_lastex = idx;
xfs_bmap_trace_delete(fname, "LF|RF|RC", ip, idx + 1, 1,
XFS_DATA_FORK);
- xfs_bmap_delete_exlist(ip, idx + 1, 1, XFS_DATA_FORK);
+ xfs_iext_remove(ifp, idx + 1, 1);
if (cur == NULL)
rval = XFS_ILOG_DEXT;
else {
@@ -964,7 +945,7 @@ xfs_bmap_add_extent_delay_real(
*/
xfs_bmap_trace_pre_update(fname, "LF|LC", ip, idx - 1,
XFS_DATA_FORK);
- xfs_bmbt_set_blockcount(ep - 1,
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
LEFT.br_blockcount + new->br_blockcount);
xfs_bmbt_set_startoff(ep,
PREV.br_startoff + new->br_blockcount);
@@ -1010,7 +991,7 @@ xfs_bmap_add_extent_delay_real(
xfs_bmbt_set_blockcount(ep, temp);
xfs_bmap_trace_insert(fname, "LF", ip, idx, 1, new, NULL,
XFS_DATA_FORK);
- xfs_bmap_insert_exlist(ip, idx, 1, new, XFS_DATA_FORK);
+ xfs_iext_insert(ifp, idx, 1, new);
ip->i_df.if_lastex = idx;
ip->i_d.di_nextents++;
if (cur == NULL)
@@ -1039,8 +1020,7 @@ xfs_bmap_add_extent_delay_real(
temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
STARTBLOCKVAL(PREV.br_startblock) -
(cur ? cur->bc_private.b.allocated : 0));
- base = ip->i_df.if_u1.if_extents;
- ep = &base[idx + 1];
+ ep = xfs_iext_get_ext(ifp, idx + 1);
xfs_bmbt_set_startblock(ep, NULLSTARTBLOCK((int)temp));
xfs_bmap_trace_post_update(fname, "LF", ip, idx + 1,
XFS_DATA_FORK);
@@ -1058,7 +1038,8 @@ xfs_bmap_add_extent_delay_real(
xfs_bmap_trace_pre_update(fname, "RF|RC", ip, idx + 1,
XFS_DATA_FORK);
xfs_bmbt_set_blockcount(ep, temp);
- xfs_bmbt_set_allf(ep + 1, new->br_startoff, new->br_startblock,
+ xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, idx + 1),
+ new->br_startoff, new->br_startblock,
new->br_blockcount + RIGHT.br_blockcount,
RIGHT.br_state);
xfs_bmap_trace_post_update(fname, "RF|RC", ip, idx + 1,
@@ -1098,7 +1079,7 @@ xfs_bmap_add_extent_delay_real(
xfs_bmbt_set_blockcount(ep, temp);
xfs_bmap_trace_insert(fname, "RF", ip, idx + 1, 1,
new, NULL, XFS_DATA_FORK);
- xfs_bmap_insert_exlist(ip, idx + 1, 1, new, XFS_DATA_FORK);
+ xfs_iext_insert(ifp, idx + 1, 1, new);
ip->i_df.if_lastex = idx + 1;
ip->i_d.di_nextents++;
if (cur == NULL)
@@ -1127,8 +1108,7 @@ xfs_bmap_add_extent_delay_real(
temp = XFS_FILBLKS_MIN(xfs_bmap_worst_indlen(ip, temp),
STARTBLOCKVAL(PREV.br_startblock) -
(cur ? cur->bc_private.b.allocated : 0));
- base = ip->i_df.if_u1.if_extents;
- ep = &base[idx];
+ ep = xfs_iext_get_ext(ifp, idx);
xfs_bmbt_set_startblock(ep, NULLSTARTBLOCK((int)temp));
xfs_bmap_trace_post_update(fname, "RF", ip, idx, XFS_DATA_FORK);
*dnew = temp;
@@ -1149,7 +1129,7 @@ xfs_bmap_add_extent_delay_real(
r[1].br_blockcount = temp2;
xfs_bmap_trace_insert(fname, "0", ip, idx + 1, 2, &r[0], &r[1],
XFS_DATA_FORK);
- xfs_bmap_insert_exlist(ip, idx + 1, 2, &r[0], XFS_DATA_FORK);
+ xfs_iext_insert(ifp, idx + 1, 2, &r[0]);
ip->i_df.if_lastex = idx + 1;
ip->i_d.di_nextents++;
if (cur == NULL)
@@ -1204,13 +1184,13 @@ xfs_bmap_add_extent_delay_real(
}
}
}
- base = ip->i_df.if_u1.if_extents;
- ep = &base[idx];
+ ep = xfs_iext_get_ext(ifp, idx);
xfs_bmbt_set_startblock(ep, NULLSTARTBLOCK((int)temp));
xfs_bmap_trace_post_update(fname, "0", ip, idx, XFS_DATA_FORK);
xfs_bmap_trace_pre_update(fname, "0", ip, idx + 2,
XFS_DATA_FORK);
- xfs_bmbt_set_startblock(ep + 2, NULLSTARTBLOCK((int)temp2));
+ xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx + 2),
+ NULLSTARTBLOCK((int)temp2));
xfs_bmap_trace_post_update(fname, "0", ip, idx + 2,
XFS_DATA_FORK);
*dnew = temp + temp2;
@@ -1254,10 +1234,9 @@ xfs_bmap_add_extent_unwritten_real(
xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* extent number to update/insert */
xfs_btree_cur_t **curp, /* if *curp is null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
int *logflagsp) /* inode logging flags */
{
- xfs_bmbt_rec_t *base; /* base of extent entry list */
xfs_btree_cur_t *cur; /* btree cursor */
xfs_bmbt_rec_t *ep; /* extent entry for idx */
int error; /* error return value */
@@ -1265,6 +1244,7 @@ xfs_bmap_add_extent_unwritten_real(
static char fname[] = "xfs_bmap_add_extent_unwritten_real";
#endif
int i; /* temp state */
+ xfs_ifork_t *ifp; /* inode fork pointer */
xfs_fileoff_t new_endoff; /* end offset of new entry */
xfs_exntst_t newext; /* new extent state */
xfs_exntst_t oldext; /* old extent state */
@@ -1298,8 +1278,8 @@ xfs_bmap_add_extent_unwritten_real(
*/
error = 0;
cur = *curp;
- base = ip->i_df.if_u1.if_extents;
- ep = &base[idx];
+ ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+ ep = xfs_iext_get_ext(ifp, idx);
xfs_bmbt_get_all(ep, &PREV);
newext = new->br_state;
oldext = (newext == XFS_EXT_UNWRITTEN) ?
@@ -1320,7 +1300,7 @@ xfs_bmap_add_extent_unwritten_real(
* Don't set contiguous if the combined extent would be too large.
*/
if (STATE_SET_TEST(LEFT_VALID, idx > 0)) {
- xfs_bmbt_get_all(ep - 1, &LEFT);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &LEFT);
STATE_SET(LEFT_DELAY, ISNULLSTARTBLOCK(LEFT.br_startblock));
}
STATE_SET(LEFT_CONTIG,
@@ -1337,7 +1317,7 @@ xfs_bmap_add_extent_unwritten_real(
if (STATE_SET_TEST(RIGHT_VALID,
idx <
ip->i_df.if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - 1)) {
- xfs_bmbt_get_all(ep + 1, &RIGHT);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx + 1), &RIGHT);
STATE_SET(RIGHT_DELAY, ISNULLSTARTBLOCK(RIGHT.br_startblock));
}
STATE_SET(RIGHT_CONTIG,
@@ -1363,14 +1343,14 @@ xfs_bmap_add_extent_unwritten_real(
*/
xfs_bmap_trace_pre_update(fname, "LF|RF|LC|RC", ip, idx - 1,
XFS_DATA_FORK);
- xfs_bmbt_set_blockcount(ep - 1,
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
LEFT.br_blockcount + PREV.br_blockcount +
RIGHT.br_blockcount);
xfs_bmap_trace_post_update(fname, "LF|RF|LC|RC", ip, idx - 1,
XFS_DATA_FORK);
xfs_bmap_trace_delete(fname, "LF|RF|LC|RC", ip, idx, 2,
XFS_DATA_FORK);
- xfs_bmap_delete_exlist(ip, idx, 2, XFS_DATA_FORK);
+ xfs_iext_remove(ifp, idx, 2);
ip->i_df.if_lastex = idx - 1;
ip->i_d.di_nextents -= 2;
if (cur == NULL)
@@ -1409,14 +1389,14 @@ xfs_bmap_add_extent_unwritten_real(
*/
xfs_bmap_trace_pre_update(fname, "LF|RF|LC", ip, idx - 1,
XFS_DATA_FORK);
- xfs_bmbt_set_blockcount(ep - 1,
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
LEFT.br_blockcount + PREV.br_blockcount);
xfs_bmap_trace_post_update(fname, "LF|RF|LC", ip, idx - 1,
XFS_DATA_FORK);
ip->i_df.if_lastex = idx - 1;
xfs_bmap_trace_delete(fname, "LF|RF|LC", ip, idx, 1,
XFS_DATA_FORK);
- xfs_bmap_delete_exlist(ip, idx, 1, XFS_DATA_FORK);
+ xfs_iext_remove(ifp, idx, 1);
ip->i_d.di_nextents--;
if (cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
@@ -1456,7 +1436,7 @@ xfs_bmap_add_extent_unwritten_real(
ip->i_df.if_lastex = idx;
xfs_bmap_trace_delete(fname, "LF|RF|RC", ip, idx + 1, 1,
XFS_DATA_FORK);
- xfs_bmap_delete_exlist(ip, idx + 1, 1, XFS_DATA_FORK);
+ xfs_iext_remove(ifp, idx + 1, 1);
ip->i_d.di_nextents--;
if (cur == NULL)
rval = XFS_ILOG_CORE | XFS_ILOG_DEXT;
@@ -1516,7 +1496,7 @@ xfs_bmap_add_extent_unwritten_real(
*/
xfs_bmap_trace_pre_update(fname, "LF|LC", ip, idx - 1,
XFS_DATA_FORK);
- xfs_bmbt_set_blockcount(ep - 1,
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
LEFT.br_blockcount + new->br_blockcount);
xfs_bmbt_set_startoff(ep,
PREV.br_startoff + new->br_blockcount);
@@ -1571,7 +1551,7 @@ xfs_bmap_add_extent_unwritten_real(
xfs_bmap_trace_post_update(fname, "LF", ip, idx, XFS_DATA_FORK);
xfs_bmap_trace_insert(fname, "LF", ip, idx, 1, new, NULL,
XFS_DATA_FORK);
- xfs_bmap_insert_exlist(ip, idx, 1, new, XFS_DATA_FORK);
+ xfs_iext_insert(ifp, idx, 1, new);
ip->i_df.if_lastex = idx;
ip->i_d.di_nextents++;
if (cur == NULL)
@@ -1609,7 +1589,8 @@ xfs_bmap_add_extent_unwritten_real(
PREV.br_blockcount - new->br_blockcount);
xfs_bmap_trace_post_update(fname, "RF|RC", ip, idx,
XFS_DATA_FORK);
- xfs_bmbt_set_allf(ep + 1, new->br_startoff, new->br_startblock,
+ xfs_bmbt_set_allf(xfs_iext_get_ext(ifp, idx + 1),
+ new->br_startoff, new->br_startblock,
new->br_blockcount + RIGHT.br_blockcount, newext);
xfs_bmap_trace_post_update(fname, "RF|RC", ip, idx + 1,
XFS_DATA_FORK);
@@ -1649,7 +1630,7 @@ xfs_bmap_add_extent_unwritten_real(
xfs_bmap_trace_post_update(fname, "RF", ip, idx, XFS_DATA_FORK);
xfs_bmap_trace_insert(fname, "RF", ip, idx + 1, 1,
new, NULL, XFS_DATA_FORK);
- xfs_bmap_insert_exlist(ip, idx + 1, 1, new, XFS_DATA_FORK);
+ xfs_iext_insert(ifp, idx + 1, 1, new);
ip->i_df.if_lastex = idx + 1;
ip->i_d.di_nextents++;
if (cur == NULL)
@@ -1696,7 +1677,7 @@ xfs_bmap_add_extent_unwritten_real(
r[1].br_state = oldext;
xfs_bmap_trace_insert(fname, "0", ip, idx + 1, 2, &r[0], &r[1],
XFS_DATA_FORK);
- xfs_bmap_insert_exlist(ip, idx + 1, 2, &r[0], XFS_DATA_FORK);
+ xfs_iext_insert(ifp, idx + 1, 2, &r[0]);
ip->i_df.if_lastex = idx + 1;
ip->i_d.di_nextents += 2;
if (cur == NULL)
@@ -1770,15 +1751,15 @@ xfs_bmap_add_extent_hole_delay(
xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* extent number to update/insert */
xfs_btree_cur_t *cur, /* if null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
int *logflagsp, /* inode logging flags */
int rsvd) /* OK to allocate reserved blocks */
{
- xfs_bmbt_rec_t *base; /* base of extent entry list */
- xfs_bmbt_rec_t *ep; /* extent list entry for idx */
+ xfs_bmbt_rec_t *ep; /* extent record for idx */
#ifdef XFS_BMAP_TRACE
static char fname[] = "xfs_bmap_add_extent_hole_delay";
#endif
+ xfs_ifork_t *ifp; /* inode fork pointer */
xfs_bmbt_irec_t left; /* left neighbor extent entry */
xfs_filblks_t newlen=0; /* new indirect size */
xfs_filblks_t oldlen=0; /* old indirect size */
@@ -1799,15 +1780,15 @@ xfs_bmap_add_extent_hole_delay(
((state &= ~MASK(b)), 0))
#define SWITCH_STATE (state & MASK2(LEFT_CONTIG, RIGHT_CONTIG))
- base = ip->i_df.if_u1.if_extents;
- ep = &base[idx];
+ ifp = XFS_IFORK_PTR(ip, XFS_DATA_FORK);
+ ep = xfs_iext_get_ext(ifp, idx);
state = 0;
ASSERT(ISNULLSTARTBLOCK(new->br_startblock));
/*
* Check and set flags if this segment has a left neighbor
*/
if (STATE_SET_TEST(LEFT_VALID, idx > 0)) {
- xfs_bmbt_get_all(ep - 1, &left);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &left);
STATE_SET(LEFT_DELAY, ISNULLSTARTBLOCK(left.br_startblock));
}
/*
@@ -1844,23 +1825,24 @@ xfs_bmap_add_extent_hole_delay(
/*
* New allocation is contiguous with delayed allocations
* on the left and on the right.
- * Merge all three into a single extent list entry.
+ * Merge all three into a single extent record.
*/
temp = left.br_blockcount + new->br_blockcount +
right.br_blockcount;
xfs_bmap_trace_pre_update(fname, "LC|RC", ip, idx - 1,
XFS_DATA_FORK);
- xfs_bmbt_set_blockcount(ep - 1, temp);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), temp);
oldlen = STARTBLOCKVAL(left.br_startblock) +
STARTBLOCKVAL(new->br_startblock) +
STARTBLOCKVAL(right.br_startblock);
newlen = xfs_bmap_worst_indlen(ip, temp);
- xfs_bmbt_set_startblock(ep - 1, NULLSTARTBLOCK((int)newlen));
+ xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx - 1),
+ NULLSTARTBLOCK((int)newlen));
xfs_bmap_trace_post_update(fname, "LC|RC", ip, idx - 1,
XFS_DATA_FORK);
xfs_bmap_trace_delete(fname, "LC|RC", ip, idx, 1,
XFS_DATA_FORK);
- xfs_bmap_delete_exlist(ip, idx, 1, XFS_DATA_FORK);
+ xfs_iext_remove(ifp, idx, 1);
ip->i_df.if_lastex = idx - 1;
break;
@@ -1873,11 +1855,12 @@ xfs_bmap_add_extent_hole_delay(
temp = left.br_blockcount + new->br_blockcount;
xfs_bmap_trace_pre_update(fname, "LC", ip, idx - 1,
XFS_DATA_FORK);
- xfs_bmbt_set_blockcount(ep - 1, temp);
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1), temp);
oldlen = STARTBLOCKVAL(left.br_startblock) +
STARTBLOCKVAL(new->br_startblock);
newlen = xfs_bmap_worst_indlen(ip, temp);
- xfs_bmbt_set_startblock(ep - 1, NULLSTARTBLOCK((int)newlen));
+ xfs_bmbt_set_startblock(xfs_iext_get_ext(ifp, idx - 1),
+ NULLSTARTBLOCK((int)newlen));
xfs_bmap_trace_post_update(fname, "LC", ip, idx - 1,
XFS_DATA_FORK);
ip->i_df.if_lastex = idx - 1;
@@ -1909,7 +1892,7 @@ xfs_bmap_add_extent_hole_delay(
oldlen = newlen = 0;
xfs_bmap_trace_insert(fname, "0", ip, idx, 1, new, NULL,
XFS_DATA_FORK);
- xfs_bmap_insert_exlist(ip, idx, 1, new, XFS_DATA_FORK);
+ xfs_iext_insert(ifp, idx, 1, new);
ip->i_df.if_lastex = idx;
break;
}
@@ -1940,7 +1923,7 @@ xfs_bmap_add_extent_hole_real(
xfs_inode_t *ip, /* incore inode pointer */
xfs_extnum_t idx, /* extent number to update/insert */
xfs_btree_cur_t *cur, /* if null, not a btree */
- xfs_bmbt_irec_t *new, /* new data to put in extent list */
+ xfs_bmbt_irec_t *new, /* new data to add to file extents */
int *logflagsp, /* inode logging flags */
int whichfork) /* data or attr fork */
{
@@ -1970,13 +1953,13 @@ xfs_bmap_add_extent_hole_real(
ifp = XFS_IFORK_PTR(ip, whichfork);
ASSERT(idx <= ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t));
- ep = &ifp->if_u1.if_extents[idx];
+ ep = xfs_iext_get_ext(ifp, idx);
state = 0;
/*
* Check and set flags if this segment has a left neighbor.
*/
if (STATE_SET_TEST(LEFT_VALID, idx > 0)) {
- xfs_bmbt_get_all(ep - 1, &left);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, idx - 1), &left);
STATE_SET(LEFT_DELAY, ISNULLSTARTBLOCK(left.br_startblock));
}
/*
@@ -2019,18 +2002,18 @@ xfs_bmap_add_extent_hole_real(
/*
* New allocation is contiguous with real allocations on the
* left and on the right.
- * Merge all three into a single extent list entry.
+ * Merge all three into a single extent record.
*/
xfs_bmap_trace_pre_update(fname, "LC|RC", ip, idx - 1,
whichfork);
- xfs_bmbt_set_blockcount(ep - 1,
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
left.br_blockcount + new->br_blockcount +
right.br_blockcount);
xfs_bmap_trace_post_update(fname, "LC|RC", ip, idx - 1,
whichfork);
xfs_bmap_trace_delete(fname, "LC|RC", ip,
idx, 1, whichfork);
- xfs_bmap_delete_exlist(ip, idx, 1, whichfork);
+ xfs_iext_remove(ifp, idx, 1);
ifp->if_lastex = idx - 1;
XFS_IFORK_NEXT_SET(ip, whichfork,
XFS_IFORK_NEXTENTS(ip, whichfork) - 1);
@@ -2062,7 +2045,7 @@ xfs_bmap_add_extent_hole_real(
* Merge the new allocation with the left neighbor.
*/
xfs_bmap_trace_pre_update(fname, "LC", ip, idx - 1, whichfork);
- xfs_bmbt_set_blockcount(ep - 1,
+ xfs_bmbt_set_blockcount(xfs_iext_get_ext(ifp, idx - 1),
left.br_blockcount + new->br_blockcount);
xfs_bmap_trace_post_update(fname, "LC", ip, idx - 1, whichfork);
ifp->if_lastex = idx - 1;
@@ -2116,7 +2099,7 @@ xfs_bmap_add_extent_hole_real(
*/
xfs_bmap_trace_insert(fname, "0", ip, idx, 1, new, NULL,
whichfork);
- xfs_bmap_insert_exlist(ip, idx, 1, new, whichfork);
+ xfs_iext_insert(ifp, idx, 1, new);
ifp->if_lastex = idx;
XFS_IFORK_NEXT_SET(ip, whichfork,
XFS_IFORK_NEXTENTS(ip, whichfork) + 1);
@@ -2311,25 +2294,15 @@ xfs_bmap_extsize_align(
#define XFS_ALLOC_GAP_UNITS 4
-/*
- * xfs_bmap_alloc is called by xfs_bmapi to allocate an extent for a file.
- * It figures out where to ask the underlying allocator to put the new extent.
- */
STATIC int
-xfs_bmap_alloc(
+xfs_bmap_adjacent(
xfs_bmalloca_t *ap) /* bmap alloc argument struct */
{
xfs_fsblock_t adjust; /* adjustment to block numbers */
- xfs_alloctype_t atype=0; /* type for allocation routines */
- int error; /* error return value */
xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */
xfs_mount_t *mp; /* mount point structure */
int nullfb; /* true if ap->firstblock isn't set */
int rt; /* true if inode is realtime */
- xfs_extlen_t prod = 0; /* product factor for allocators */
- xfs_extlen_t ralen = 0; /* realtime allocation length */
- xfs_extlen_t align; /* minimum allocation alignment */
- xfs_rtblock_t rtx;
#define ISVALID(x,y) \
(rt ? \
@@ -2338,75 +2311,10 @@ xfs_bmap_alloc(
XFS_FSB_TO_AGNO(mp, x) < mp->m_sb.sb_agcount && \
XFS_FSB_TO_AGBNO(mp, x) < mp->m_sb.sb_agblocks)
- /*
- * Set up variables.
- */
mp = ap->ip->i_mount;
nullfb = ap->firstblock == NULLFSBLOCK;
rt = XFS_IS_REALTIME_INODE(ap->ip) && ap->userdata;
fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, ap->firstblock);
- if (rt) {
- align = ap->ip->i_d.di_extsize ?
- ap->ip->i_d.di_extsize : mp->m_sb.sb_rextsize;
- /* Set prod to match the extent size */
- prod = align / mp->m_sb.sb_rextsize;
-
- error = xfs_bmap_extsize_align(mp, ap->gotp, ap->prevp,
- align, rt, ap->eof, 0,
- ap->conv, &ap->off, &ap->alen);
- if (error)
- return error;
- ASSERT(ap->alen);
- ASSERT(ap->alen % mp->m_sb.sb_rextsize == 0);
-
- /*
- * If the offset & length are not perfectly aligned
- * then kill prod, it will just get us in trouble.
- */
- if (do_mod(ap->off, align) || ap->alen % align)
- prod = 1;
- /*
- * Set ralen to be the actual requested length in rtextents.
- */
- ralen = ap->alen / mp->m_sb.sb_rextsize;
- /*
- * If the old value was close enough to MAXEXTLEN that
- * we rounded up to it, cut it back so it's valid again.
- * Note that if it's a really large request (bigger than
- * MAXEXTLEN), we don't hear about that number, and can't
- * adjust the starting point to match it.
- */
- if (ralen * mp->m_sb.sb_rextsize >= MAXEXTLEN)
- ralen = MAXEXTLEN / mp->m_sb.sb_rextsize;
- /*
- * If it's an allocation to an empty file at offset 0,
- * pick an extent that will space things out in the rt area.
- */
- if (ap->eof && ap->off == 0) {
- error = xfs_rtpick_extent(mp, ap->tp, ralen, &rtx);
- if (error)
- return error;
- ap->rval = rtx * mp->m_sb.sb_rextsize;
- } else
- ap->rval = 0;
- } else {
- align = (ap->userdata && ap->ip->i_d.di_extsize &&
- (ap->ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)) ?
- ap->ip->i_d.di_extsize : 0;
- if (unlikely(align)) {
- error = xfs_bmap_extsize_align(mp, ap->gotp, ap->prevp,
- align, rt,
- ap->eof, 0, ap->conv,
- &ap->off, &ap->alen);
- ASSERT(!error);
- ASSERT(ap->alen);
- }
- if (nullfb)
- ap->rval = XFS_INO_TO_FSB(mp, ap->ip->i_ino);
- else
- ap->rval = ap->firstblock;
- }
-
/*
* If allocating at eof, and there's a previous real block,
* try to use it's last block as our starting point.
@@ -2531,287 +2439,384 @@ xfs_bmap_alloc(
else if (gotbno != NULLFSBLOCK)
ap->rval = gotbno;
}
+#undef ISVALID
+ return 0;
+}
+
+STATIC int
+xfs_bmap_rtalloc(
+ xfs_bmalloca_t *ap) /* bmap alloc argument struct */
+{
+ xfs_alloctype_t atype = 0; /* type for allocation routines */
+ int error; /* error return value */
+ xfs_mount_t *mp; /* mount point structure */
+ xfs_extlen_t prod = 0; /* product factor for allocators */
+ xfs_extlen_t ralen = 0; /* realtime allocation length */
+ xfs_extlen_t align; /* minimum allocation alignment */
+ xfs_rtblock_t rtx; /* realtime extent number */
+ xfs_rtblock_t rtb;
+
+ mp = ap->ip->i_mount;
+ align = ap->ip->i_d.di_extsize ?
+ ap->ip->i_d.di_extsize : mp->m_sb.sb_rextsize;
+ prod = align / mp->m_sb.sb_rextsize;
+ error = xfs_bmap_extsize_align(mp, ap->gotp, ap->prevp,
+ align, 1, ap->eof, 0,
+ ap->conv, &ap->off, &ap->alen);
+ if (error)
+ return error;
+ ASSERT(ap->alen);
+ ASSERT(ap->alen % mp->m_sb.sb_rextsize == 0);
+
+ /*
+ * If the offset & length are not perfectly aligned
+ * then kill prod, it will just get us in trouble.
+ */
+ if (do_mod(ap->off, align) || ap->alen % align)
+ prod = 1;
+ /*
+ * Set ralen to be the actual requested length in rtextents.
+ */
+ ralen = ap->alen / mp->m_sb.sb_rextsize;
+ /*
+ * If the old value was close enough to MAXEXTLEN that
+ * we rounded up to it, cut it back so it's valid again.
+ * Note that if it's a really large request (bigger than
+ * MAXEXTLEN), we don't hear about that number, and can't
+ * adjust the starting point to match it.
+ */
+ if (ralen * mp->m_sb.sb_rextsize >= MAXEXTLEN)
+ ralen = MAXEXTLEN / mp->m_sb.sb_rextsize;
+ /*
+ * If it's an allocation to an empty file at offset 0,
+ * pick an extent that will space things out in the rt area.
+ */
+ if (ap->eof && ap->off == 0) {
+ error = xfs_rtpick_extent(mp, ap->tp, ralen, &rtx);
+ if (error)
+ return error;
+ ap->rval = rtx * mp->m_sb.sb_rextsize;
+ } else {
+ ap->rval = 0;
+ }
+
+ xfs_bmap_adjacent(ap);
+
+ /*
+ * Realtime allocation, done through xfs_rtallocate_extent.
+ */
+ atype = ap->rval == 0 ? XFS_ALLOCTYPE_ANY_AG : XFS_ALLOCTYPE_NEAR_BNO;
+ do_div(ap->rval, mp->m_sb.sb_rextsize);
+ rtb = ap->rval;
+ ap->alen = ralen;
+ if ((error = xfs_rtallocate_extent(ap->tp, ap->rval, 1, ap->alen,
+ &ralen, atype, ap->wasdel, prod, &rtb)))
+ return error;
+ if (rtb == NULLFSBLOCK && prod > 1 &&
+ (error = xfs_rtallocate_extent(ap->tp, ap->rval, 1,
+ ap->alen, &ralen, atype,
+ ap->wasdel, 1, &rtb)))
+ return error;
+ ap->rval = rtb;
+ if (ap->rval != NULLFSBLOCK) {
+ ap->rval *= mp->m_sb.sb_rextsize;
+ ralen *= mp->m_sb.sb_rextsize;
+ ap->alen = ralen;
+ ap->ip->i_d.di_nblocks += ralen;
+ xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE);
+ if (ap->wasdel)
+ ap->ip->i_delayed_blks -= ralen;
+ /*
+ * Adjust the disk quota also. This was reserved
+ * earlier.
+ */
+ XFS_TRANS_MOD_DQUOT_BYINO(mp, ap->tp, ap->ip,
+ ap->wasdel ? XFS_TRANS_DQ_DELRTBCOUNT :
+ XFS_TRANS_DQ_RTBCOUNT, (long) ralen);
+ } else {
+ ap->alen = 0;
+ }
+ return 0;
+}
+
+STATIC int
+xfs_bmap_btalloc(
+ xfs_bmalloca_t *ap) /* bmap alloc argument struct */
+{
+ xfs_mount_t *mp; /* mount point structure */
+ xfs_alloctype_t atype = 0; /* type for allocation routines */
+ xfs_extlen_t align; /* minimum allocation alignment */
+ xfs_agnumber_t ag;
+ xfs_agnumber_t fb_agno; /* ag number of ap->firstblock */
+ xfs_agnumber_t startag;
+ xfs_alloc_arg_t args;
+ xfs_extlen_t blen;
+ xfs_extlen_t delta;
+ xfs_extlen_t longest;
+ xfs_extlen_t need;
+ xfs_extlen_t nextminlen = 0;
+ xfs_perag_t *pag;
+ int nullfb; /* true if ap->firstblock isn't set */
+ int isaligned;
+ int notinit;
+ int tryagain;
+ int error;
+
+ mp = ap->ip->i_mount;
+ align = (ap->userdata && ap->ip->i_d.di_extsize &&
+ (ap->ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE)) ?
+ ap->ip->i_d.di_extsize : 0;
+ if (unlikely(align)) {
+ error = xfs_bmap_extsize_align(mp, ap->gotp, ap->prevp,
+ align, 0, ap->eof, 0, ap->conv,
+ &ap->off, &ap->alen);
+ ASSERT(!error);
+ ASSERT(ap->alen);
+ }
+ nullfb = ap->firstblock == NULLFSBLOCK;
+ fb_agno = nullfb ? NULLAGNUMBER : XFS_FSB_TO_AGNO(mp, ap->firstblock);
+ if (nullfb)
+ ap->rval = XFS_INO_TO_FSB(mp, ap->ip->i_ino);
+ else
+ ap->rval = ap->firstblock;
+
+ xfs_bmap_adjacent(ap);
+
/*
* If allowed, use ap->rval; otherwise must use firstblock since
* it's in the right allocation group.
*/
- if (nullfb || rt || XFS_FSB_TO_AGNO(mp, ap->rval) == fb_agno)
+ if (nullfb || XFS_FSB_TO_AGNO(mp, ap->rval) == fb_agno)
;
else
ap->rval = ap->firstblock;
/*
- * Realtime allocation, done through xfs_rtallocate_extent.
+ * Normal allocation, done through xfs_alloc_vextent.
*/
- if (rt) {
-#ifndef __KERNEL__
- ASSERT(0);
-#else
- xfs_rtblock_t rtb;
-
- atype = ap->rval == 0 ?
- XFS_ALLOCTYPE_ANY_AG : XFS_ALLOCTYPE_NEAR_BNO;
- do_div(ap->rval, mp->m_sb.sb_rextsize);
- rtb = ap->rval;
- ap->alen = ralen;
- if ((error = xfs_rtallocate_extent(ap->tp, ap->rval, 1, ap->alen,
- &ralen, atype, ap->wasdel, prod, &rtb)))
- return error;
- if (rtb == NULLFSBLOCK && prod > 1 &&
- (error = xfs_rtallocate_extent(ap->tp, ap->rval, 1,
- ap->alen, &ralen, atype,
- ap->wasdel, 1, &rtb)))
- return error;
- ap->rval = rtb;
- if (ap->rval != NULLFSBLOCK) {
- ap->rval *= mp->m_sb.sb_rextsize;
- ralen *= mp->m_sb.sb_rextsize;
- ap->alen = ralen;
- ap->ip->i_d.di_nblocks += ralen;
- xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE);
- if (ap->wasdel)
- ap->ip->i_delayed_blks -= ralen;
+ tryagain = isaligned = 0;
+ args.tp = ap->tp;
+ args.mp = mp;
+ args.fsbno = ap->rval;
+ args.maxlen = MIN(ap->alen, mp->m_sb.sb_agblocks);
+ blen = 0;
+ if (nullfb) {
+ args.type = XFS_ALLOCTYPE_START_BNO;
+ args.total = ap->total;
+ /*
+ * Find the longest available space.
+ * We're going to try for the whole allocation at once.
+ */
+ startag = ag = XFS_FSB_TO_AGNO(mp, args.fsbno);
+ notinit = 0;
+ down_read(&mp->m_peraglock);
+ while (blen < ap->alen) {
+ pag = &mp->m_perag[ag];
+ if (!pag->pagf_init &&
+ (error = xfs_alloc_pagf_init(mp, args.tp,
+ ag, XFS_ALLOC_FLAG_TRYLOCK))) {
+ up_read(&mp->m_peraglock);
+ return error;
+ }
/*
- * Adjust the disk quota also. This was reserved
- * earlier.
+ * See xfs_alloc_fix_freelist...
*/
- XFS_TRANS_MOD_DQUOT_BYINO(mp, ap->tp, ap->ip,
- ap->wasdel ? XFS_TRANS_DQ_DELRTBCOUNT :
- XFS_TRANS_DQ_RTBCOUNT,
- (long) ralen);
- } else
- ap->alen = 0;
-#endif /* __KERNEL__ */
+ if (pag->pagf_init) {
+ need = XFS_MIN_FREELIST_PAG(pag, mp);
+ delta = need > pag->pagf_flcount ?
+ need - pag->pagf_flcount : 0;
+ longest = (pag->pagf_longest > delta) ?
+ (pag->pagf_longest - delta) :
+ (pag->pagf_flcount > 0 ||
+ pag->pagf_longest > 0);
+ if (blen < longest)
+ blen = longest;
+ } else
+ notinit = 1;
+ if (++ag == mp->m_sb.sb_agcount)
+ ag = 0;
+ if (ag == startag)
+ break;
+ }
+ up_read(&mp->m_peraglock);
+ /*
+ * Since the above loop did a BUF_TRYLOCK, it is
+ * possible that there is space for this request.
+ */
+ if (notinit || blen < ap->minlen)
+ args.minlen = ap->minlen;
+ /*
+ * If the best seen length is less than the request
+ * length, use the best as the minimum.
+ */
+ else if (blen < ap->alen)
+ args.minlen = blen;
+ /*
+ * Otherwise we've seen an extent as big as alen,
+ * use that as the minimum.
+ */
+ else
+ args.minlen = ap->alen;
+ } else if (ap->low) {
+ args.type = XFS_ALLOCTYPE_FIRST_AG;
+ args.total = args.minlen = ap->minlen;
+ } else {
+ args.type = XFS_ALLOCTYPE_NEAR_BNO;
+ args.total = ap->total;
+ args.minlen = ap->minlen;
+ }
+ if (unlikely(ap->userdata && ap->ip->i_d.di_extsize &&
+ (ap->ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE))) {
+ args.prod = ap->ip->i_d.di_extsize;
+ if ((args.mod = (xfs_extlen_t)do_mod(ap->off, args.prod)))
+ args.mod = (xfs_extlen_t)(args.prod - args.mod);
+ } else if (unlikely(mp->m_sb.sb_blocksize >= NBPP)) {
+ args.prod = 1;
+ args.mod = 0;
+ } else {
+ args.prod = NBPP >> mp->m_sb.sb_blocklog;
+ if ((args.mod = (xfs_extlen_t)(do_mod(ap->off, args.prod))))
+ args.mod = (xfs_extlen_t)(args.prod - args.mod);
}
/*
- * Normal allocation, done through xfs_alloc_vextent.
+ * If we are not low on available data blocks, and the
+ * underlying logical volume manager is a stripe, and
+ * the file offset is zero then try to allocate data
+ * blocks on stripe unit boundary.
+ * NOTE: ap->aeof is only set if the allocation length
+ * is >= the stripe unit and the allocation offset is
+ * at the end of file.
*/
- else {
- xfs_agnumber_t ag;
- xfs_alloc_arg_t args;
- xfs_extlen_t blen;
- xfs_extlen_t delta;
- int isaligned;
- xfs_extlen_t longest;
- xfs_extlen_t need;
- xfs_extlen_t nextminlen=0;
- int notinit;
- xfs_perag_t *pag;
- xfs_agnumber_t startag;
- int tryagain;
-
- tryagain = isaligned = 0;
- args.tp = ap->tp;
- args.mp = mp;
- args.fsbno = ap->rval;
- args.maxlen = MIN(ap->alen, mp->m_sb.sb_agblocks);
- blen = 0;
- if (nullfb) {
- args.type = XFS_ALLOCTYPE_START_BNO;
- args.total = ap->total;
- /*
- * Find the longest available space.
- * We're going to try for the whole allocation at once.
- */
- startag = ag = XFS_FSB_TO_AGNO(mp, args.fsbno);
- notinit = 0;
- down_read(&mp->m_peraglock);
- while (blen < ap->alen) {
- pag = &mp->m_perag[ag];
- if (!pag->pagf_init &&
- (error = xfs_alloc_pagf_init(mp, args.tp,
- ag, XFS_ALLOC_FLAG_TRYLOCK))) {
- up_read(&mp->m_peraglock);
- return error;
- }
- /*
- * See xfs_alloc_fix_freelist...
- */
- if (pag->pagf_init) {
- need = XFS_MIN_FREELIST_PAG(pag, mp);
- delta = need > pag->pagf_flcount ?
- need - pag->pagf_flcount : 0;
- longest = (pag->pagf_longest > delta) ?
- (pag->pagf_longest - delta) :
- (pag->pagf_flcount > 0 ||
- pag->pagf_longest > 0);
- if (blen < longest)
- blen = longest;
- } else
- notinit = 1;
- if (++ag == mp->m_sb.sb_agcount)
- ag = 0;
- if (ag == startag)
- break;
- }
- up_read(&mp->m_peraglock);
+ if (!ap->low && ap->aeof) {
+ if (!ap->off) {
+ args.alignment = mp->m_dalign;
+ atype = args.type;
+ isaligned = 1;
/*
- * Since the above loop did a BUF_TRYLOCK, it is
- * possible that there is space for this request.
+ * Adjust for alignment
*/
- if (notinit || blen < ap->minlen)
- args.minlen = ap->minlen;
+ if (blen > args.alignment && blen <= ap->alen)
+ args.minlen = blen - args.alignment;
+ args.minalignslop = 0;
+ } else {
/*
- * If the best seen length is less than the request
- * length, use the best as the minimum.
+ * First try an exact bno allocation.
+ * If it fails then do a near or start bno
+ * allocation with alignment turned on.
*/
- else if (blen < ap->alen)
- args.minlen = blen;
+ atype = args.type;
+ tryagain = 1;
+ args.type = XFS_ALLOCTYPE_THIS_BNO;
+ args.alignment = 1;
/*
- * Otherwise we've seen an extent as big as alen,
- * use that as the minimum.
+ * Compute the minlen+alignment for the
+ * next case. Set slop so that the value
+ * of minlen+alignment+slop doesn't go up
+ * between the calls.
*/
+ if (blen > mp->m_dalign && blen <= ap->alen)
+ nextminlen = blen - mp->m_dalign;
else
- args.minlen = ap->alen;
- } else if (ap->low) {
- args.type = XFS_ALLOCTYPE_FIRST_AG;
- args.total = args.minlen = ap->minlen;
- } else {
- args.type = XFS_ALLOCTYPE_NEAR_BNO;
- args.total = ap->total;
- args.minlen = ap->minlen;
- }
- if (unlikely(ap->userdata && ap->ip->i_d.di_extsize &&
- (ap->ip->i_d.di_flags & XFS_DIFLAG_EXTSIZE))) {
- args.prod = ap->ip->i_d.di_extsize;
- if ((args.mod = (xfs_extlen_t)do_mod(ap->off, args.prod)))
- args.mod = (xfs_extlen_t)(args.prod - args.mod);
- } else if (unlikely(mp->m_sb.sb_blocksize >= NBPP)) {
- args.prod = 1;
- args.mod = 0;
- } else {
- args.prod = NBPP >> mp->m_sb.sb_blocklog;
- if ((args.mod = (xfs_extlen_t)(do_mod(ap->off, args.prod))))
- args.mod = (xfs_extlen_t)(args.prod - args.mod);
+ nextminlen = args.minlen;
+ if (nextminlen + mp->m_dalign > args.minlen + 1)
+ args.minalignslop =
+ nextminlen + mp->m_dalign -
+ args.minlen - 1;
+ else
+ args.minalignslop = 0;
}
+ } else {
+ args.alignment = 1;
+ args.minalignslop = 0;
+ }
+ args.minleft = ap->minleft;
+ args.wasdel = ap->wasdel;
+ args.isfl = 0;
+ args.userdata = ap->userdata;
+ if ((error = xfs_alloc_vextent(&args)))
+ return error;
+ if (tryagain && args.fsbno == NULLFSBLOCK) {
/*
- * If we are not low on available data blocks, and the
- * underlying logical volume manager is a stripe, and
- * the file offset is zero then try to allocate data
- * blocks on stripe unit boundary.
- * NOTE: ap->aeof is only set if the allocation length
- * is >= the stripe unit and the allocation offset is
- * at the end of file.
+ * Exact allocation failed. Now try with alignment
+ * turned on.
*/
- if (!ap->low && ap->aeof) {
- if (!ap->off) {
- args.alignment = mp->m_dalign;
- atype = args.type;
- isaligned = 1;
- /*
- * Adjust for alignment
- */
- if (blen > args.alignment && blen <= ap->alen)
- args.minlen = blen - args.alignment;
- args.minalignslop = 0;
- } else {
- /*
- * First try an exact bno allocation.
- * If it fails then do a near or start bno
- * allocation with alignment turned on.
- */
- atype = args.type;
- tryagain = 1;
- args.type = XFS_ALLOCTYPE_THIS_BNO;
- args.alignment = 1;
- /*
- * Compute the minlen+alignment for the
- * next case. Set slop so that the value
- * of minlen+alignment+slop doesn't go up
- * between the calls.
- */
- if (blen > mp->m_dalign && blen <= ap->alen)
- nextminlen = blen - mp->m_dalign;
- else
- nextminlen = args.minlen;
- if (nextminlen + mp->m_dalign > args.minlen + 1)
- args.minalignslop =
- nextminlen + mp->m_dalign -
- args.minlen - 1;
- else
- args.minalignslop = 0;
- }
- } else {
- args.alignment = 1;
- args.minalignslop = 0;
- }
- args.minleft = ap->minleft;
- args.wasdel = ap->wasdel;
- args.isfl = 0;
- args.userdata = ap->userdata;
+ args.type = atype;
+ args.fsbno = ap->rval;
+ args.alignment = mp->m_dalign;
+ args.minlen = nextminlen;
+ args.minalignslop = 0;
+ isaligned = 1;
+ if ((error = xfs_alloc_vextent(&args)))
+ return error;
+ }
+ if (isaligned && args.fsbno == NULLFSBLOCK) {
+ /*
+ * allocation failed, so turn off alignment and
+ * try again.
+ */
+ args.type = atype;
+ args.fsbno = ap->rval;
+ args.alignment = 0;
+ if ((error = xfs_alloc_vextent(&args)))
+ return error;
+ }
+ if (args.fsbno == NULLFSBLOCK && nullfb &&
+ args.minlen > ap->minlen) {
+ args.minlen = ap->minlen;
+ args.type = XFS_ALLOCTYPE_START_BNO;
+ args.fsbno = ap->rval;
if ((error = xfs_alloc_vextent(&args)))
return error;
- if (tryagain && args.fsbno == NULLFSBLOCK) {
- /*
- * Exact allocation failed. Now try with alignment
- * turned on.
- */
- args.type = atype;
- args.fsbno = ap->rval;
- args.alignment = mp->m_dalign;
- args.minlen = nextminlen;
- args.minalignslop = 0;
- isaligned = 1;
- if ((error = xfs_alloc_vextent(&args)))
- return error;
- }
- if (isaligned && args.fsbno == NULLFSBLOCK) {
- /*
- * allocation failed, so turn off alignment and
- * try again.
- */
- args.type = atype;
- args.fsbno = ap->rval;
- args.alignment = 0;
- if ((error = xfs_alloc_vextent(&args)))
- return error;
- }
- if (args.fsbno == NULLFSBLOCK && nullfb &&
- args.minlen > ap->minlen) {
- args.minlen = ap->minlen;
- args.type = XFS_ALLOCTYPE_START_BNO;
- args.fsbno = ap->rval;
- if ((error = xfs_alloc_vextent(&args)))
- return error;
- }
- if (args.fsbno == NULLFSBLOCK && nullfb) {
- args.fsbno = 0;
- args.type = XFS_ALLOCTYPE_FIRST_AG;
- args.total = ap->minlen;
- args.minleft = 0;
- if ((error = xfs_alloc_vextent(&args)))
- return error;
- ap->low = 1;
- }
- if (args.fsbno != NULLFSBLOCK) {
- ap->firstblock = ap->rval = args.fsbno;
- ASSERT(nullfb || fb_agno == args.agno ||
- (ap->low && fb_agno < args.agno));
- ap->alen = args.len;
- ap->ip->i_d.di_nblocks += args.len;
- xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE);
- if (ap->wasdel)
- ap->ip->i_delayed_blks -= args.len;
- /*
- * Adjust the disk quota also. This was reserved
- * earlier.
- */
- XFS_TRANS_MOD_DQUOT_BYINO(mp, ap->tp, ap->ip,
- ap->wasdel ? XFS_TRANS_DQ_DELBCOUNT :
- XFS_TRANS_DQ_BCOUNT,
- (long) args.len);
- } else {
- ap->rval = NULLFSBLOCK;
- ap->alen = 0;
- }
+ }
+ if (args.fsbno == NULLFSBLOCK && nullfb) {
+ args.fsbno = 0;
+ args.type = XFS_ALLOCTYPE_FIRST_AG;
+ args.total = ap->minlen;
+ args.minleft = 0;
+ if ((error = xfs_alloc_vextent(&args)))
+ return error;
+ ap->low = 1;
+ }
+ if (args.fsbno != NULLFSBLOCK) {
+ ap->firstblock = ap->rval = args.fsbno;
+ ASSERT(nullfb || fb_agno == args.agno ||
+ (ap->low && fb_agno < args.agno));
+ ap->alen = args.len;
+ ap->ip->i_d.di_nblocks += args.len;
+ xfs_trans_log_inode(ap->tp, ap->ip, XFS_ILOG_CORE);
+ if (ap->wasdel)
+ ap->ip->i_delayed_blks -= args.len;
+ /*
+ * Adjust the disk quota also. This was reserved
+ * earlier.
+ */
+ XFS_TRANS_MOD_DQUOT_BYINO(mp, ap->tp, ap->ip,
+ ap->wasdel ? XFS_TRANS_DQ_DELBCOUNT :
+ XFS_TRANS_DQ_BCOUNT,
+ (long) args.len);
+ } else {
+ ap->rval = NULLFSBLOCK;
+ ap->alen = 0;
}
return 0;
-#undef ISVALID
+}
+
+/*
+ * xfs_bmap_alloc is called by xfs_bmapi to allocate an extent for a file.
+ * It figures out where to ask the underlying allocator to put the new extent.
+ */
+STATIC int
+xfs_bmap_alloc(
+ xfs_bmalloca_t *ap) /* bmap alloc argument struct */
+{
+ if ((ap->ip->i_d.di_flags & XFS_DIFLAG_REALTIME) && ap->userdata)
+ return xfs_bmap_rtalloc(ap);
+ return xfs_bmap_btalloc(ap);
}
/*
* Transform a btree format file with only one leaf node, where the
* extents list will fit in the inode, into an extents format file.
- * Since the extent list is already in-core, all we have to do is
+ * Since the file extents are already in-core, all we have to do is
* give up the space for the btree root and pitch the leaf block.
*/
STATIC int /* error */
@@ -2868,7 +2873,7 @@ xfs_bmap_btree_to_extents(
}
/*
- * Called by xfs_bmapi to update extent list structure and the btree
+ * Called by xfs_bmapi to update file extent records and the btree
* after removing space (or undoing a delayed allocation).
*/
STATIC int /* error */
@@ -2878,7 +2883,7 @@ xfs_bmap_del_extent(
xfs_extnum_t idx, /* extent number to update/delete */
xfs_bmap_free_t *flist, /* list of extents to be freed */
xfs_btree_cur_t *cur, /* if null, not a btree */
- xfs_bmbt_irec_t *del, /* data to remove from extent list */
+ xfs_bmbt_irec_t *del, /* data to remove from extents */
int *logflagsp, /* inode logging flags */
int whichfork, /* data or attr fork */
int rsvd) /* OK to allocate reserved blocks */
@@ -2903,7 +2908,6 @@ xfs_bmap_del_extent(
xfs_filblks_t nblks; /* quota/sb block count */
xfs_bmbt_irec_t new; /* new record to be inserted */
/* REFERENCED */
- xfs_extnum_t nextents; /* number of extents in list */
uint qfield; /* quota field to update */
xfs_filblks_t temp; /* for indirect length calculations */
xfs_filblks_t temp2; /* for indirect length calculations */
@@ -2911,10 +2915,10 @@ xfs_bmap_del_extent(
XFS_STATS_INC(xs_del_exlist);
mp = ip->i_mount;
ifp = XFS_IFORK_PTR(ip, whichfork);
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- ASSERT(idx >= 0 && idx < nextents);
+ ASSERT((idx >= 0) && (idx < ifp->if_bytes /
+ (uint)sizeof(xfs_bmbt_rec_t)));
ASSERT(del->br_blockcount > 0);
- ep = &ifp->if_u1.if_extents[idx];
+ ep = xfs_iext_get_ext(ifp, idx);
xfs_bmbt_get_all(ep, &got);
ASSERT(got.br_startoff <= del->br_startoff);
del_endoff = del->br_startoff + del->br_blockcount;
@@ -2990,7 +2994,7 @@ xfs_bmap_del_extent(
* Matches the whole extent. Delete the entry.
*/
xfs_bmap_trace_delete(fname, "3", ip, idx, 1, whichfork);
- xfs_bmap_delete_exlist(ip, idx, 1, whichfork);
+ xfs_iext_remove(ifp, idx, 1);
ifp->if_lastex = idx;
if (delay)
break;
@@ -3160,7 +3164,7 @@ xfs_bmap_del_extent(
xfs_bmap_trace_post_update(fname, "0", ip, idx, whichfork);
xfs_bmap_trace_insert(fname, "0", ip, idx + 1, 1, &new, NULL,
whichfork);
- xfs_bmap_insert_exlist(ip, idx + 1, 1, &new, whichfork);
+ xfs_iext_insert(ifp, idx + 1, 1, &new);
ifp->if_lastex = idx + 1;
break;
}
@@ -3213,31 +3217,6 @@ xfs_bmap_del_free(
}
/*
- * Remove count entries from the extents array for inode "ip", starting
- * at index "idx". Copies the remaining items down over the deleted ones,
- * and gives back the excess memory.
- */
-STATIC void
-xfs_bmap_delete_exlist(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* starting delete index */
- xfs_extnum_t count, /* count of items to delete */
- int whichfork) /* data or attr fork */
-{
- xfs_bmbt_rec_t *base; /* base of extent list */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_extnum_t nextents; /* number of extents in list after */
-
- ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT(ifp->if_flags & XFS_IFEXTENTS);
- base = ifp->if_u1.if_extents;
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t) - count;
- memmove(&base[idx], &base[idx + count],
- (nextents - idx) * sizeof(*base));
- xfs_iext_realloc(ip, -count, whichfork);
-}
-
-/*
* Convert an extents-format file into a btree-format file.
* The new file will have a root block (in the inode) and a single child block.
*/
@@ -3258,13 +3237,13 @@ xfs_bmap_extents_to_btree(
xfs_bmbt_rec_t *arp; /* child record pointer */
xfs_bmbt_block_t *block; /* btree root block */
xfs_btree_cur_t *cur; /* bmap btree cursor */
- xfs_bmbt_rec_t *ep; /* extent list pointer */
+ xfs_bmbt_rec_t *ep; /* extent record pointer */
int error; /* error return value */
- xfs_extnum_t i, cnt; /* extent list index */
+ xfs_extnum_t i, cnt; /* extent record index */
xfs_ifork_t *ifp; /* inode fork pointer */
xfs_bmbt_key_t *kp; /* root block key pointer */
xfs_mount_t *mp; /* mount structure */
- xfs_extnum_t nextents; /* extent list size */
+ xfs_extnum_t nextents; /* number of file extents */
xfs_bmbt_ptr_t *pp; /* root block address pointer */
ifp = XFS_IFORK_PTR(ip, whichfork);
@@ -3343,7 +3322,8 @@ xfs_bmap_extents_to_btree(
ablock->bb_rightsib = cpu_to_be64(NULLDFSBNO);
arp = XFS_BMAP_REC_IADDR(ablock, 1, cur);
nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- for (ep = ifp->if_u1.if_extents, cnt = i = 0; i < nextents; i++, ep++) {
+ for (cnt = i = 0; i < nextents; i++) {
+ ep = xfs_iext_get_ext(ifp, i);
if (!ISNULLSTARTBLOCK(xfs_bmbt_get_startblock(ep))) {
arp->l0 = INT_GET(ep->l0, ARCH_CONVERT);
arp->l1 = INT_GET(ep->l1, ARCH_CONVERT);
@@ -3373,34 +3353,6 @@ xfs_bmap_extents_to_btree(
}
/*
- * Insert new item(s) in the extent list for inode "ip".
- * Count new items are inserted at offset idx.
- */
-STATIC void
-xfs_bmap_insert_exlist(
- xfs_inode_t *ip, /* incore inode pointer */
- xfs_extnum_t idx, /* starting index of new items */
- xfs_extnum_t count, /* number of inserted items */
- xfs_bmbt_irec_t *new, /* items to insert */
- int whichfork) /* data or attr fork */
-{
- xfs_bmbt_rec_t *base; /* extent list base */
- xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_extnum_t nextents; /* extent list size */
- xfs_extnum_t to; /* extent list index */
-
- ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT(ifp->if_flags & XFS_IFEXTENTS);
- xfs_iext_realloc(ip, count, whichfork);
- base = ifp->if_u1.if_extents;
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- memmove(&base[idx + count], &base[idx],
- (nextents - (idx + count)) * sizeof(*base));
- for (to = idx; to < idx + count; to++, new++)
- xfs_bmbt_set_all(&base[to], new);
-}
-
-/*
* Helper routine to reset inode di_forkoff field when switching
* attribute fork from local to extent format - we reset it where
* possible to make space available for inline data fork extents.
@@ -3457,12 +3409,13 @@ xfs_bmap_local_to_extents(
error = 0;
if (ifp->if_bytes) {
xfs_alloc_arg_t args; /* allocation arguments */
- xfs_buf_t *bp; /* buffer for extent list block */
- xfs_bmbt_rec_t *ep; /* extent list pointer */
+ xfs_buf_t *bp; /* buffer for extent block */
+ xfs_bmbt_rec_t *ep; /* extent record pointer */
args.tp = tp;
args.mp = ip->i_mount;
- ASSERT(ifp->if_flags & XFS_IFINLINE);
+ ASSERT((ifp->if_flags &
+ (XFS_IFINLINE|XFS_IFEXTENTS|XFS_IFEXTIREC)) == XFS_IFINLINE);
/*
* Allocate a block. We know we need only one, since the
* file currently fits in an inode.
@@ -3492,8 +3445,8 @@ xfs_bmap_local_to_extents(
xfs_trans_log_buf(tp, bp, 0, ifp->if_bytes - 1);
xfs_bmap_forkoff_reset(args.mp, ip, whichfork);
xfs_idata_realloc(ip, -ifp->if_bytes, whichfork);
- xfs_iext_realloc(ip, 1, whichfork);
- ep = ifp->if_u1.if_extents;
+ xfs_iext_add(ifp, 0, 1);
+ ep = xfs_iext_get_ext(ifp, 0);
xfs_bmbt_set_allf(ep, 0, args.fsbno, 1, XFS_EXT_NORM);
xfs_bmap_trace_post_update(fname, "new", ip, 0, whichfork);
XFS_IFORK_NEXT_SET(ip, whichfork, 1);
@@ -3518,7 +3471,7 @@ xfs_bmbt_rec_t * /* pointer to found extent entry */
xfs_bmap_do_search_extents(
xfs_bmbt_rec_t *base, /* base of extent list */
xfs_extnum_t lastx, /* last extent index used */
- xfs_extnum_t nextents, /* extent list size */
+ xfs_extnum_t nextents, /* number of file extents */
xfs_fileoff_t bno, /* block number searched for */
int *eofp, /* out: end of file found */
xfs_extnum_t *lastxp, /* out: last extent index */
@@ -3569,9 +3522,9 @@ xfs_bmap_do_search_extents(
got.br_blockcount = xfs_bmbt_get_blockcount(ep);
*eofp = 0;
} else {
- /* binary search the extents array */
low = 0;
high = nextents - 1;
+ /* binary search the extents array */
while (low <= high) {
XFS_STATS_INC(xs_cmp_exlist);
lastx = (low + high) >> 1;
@@ -3622,6 +3575,57 @@ xfs_bmap_do_search_extents(
}
/*
+ * Search the extent records for the entry containing block bno.
+ * If bno lies in a hole, point to the next entry. If bno lies
+ * past eof, *eofp will be set, and *prevp will contain the last
+ * entry (null if none). Else, *lastxp will be set to the index
+ * of the found entry; *gotp will contain the entry.
+ */
+xfs_bmbt_rec_t * /* pointer to found extent entry */
+xfs_bmap_search_multi_extents(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_fileoff_t bno, /* block number searched for */
+ int *eofp, /* out: end of file found */
+ xfs_extnum_t *lastxp, /* out: last extent index */
+ xfs_bmbt_irec_t *gotp, /* out: extent entry found */
+ xfs_bmbt_irec_t *prevp) /* out: previous extent entry found */
+{
+ xfs_bmbt_rec_t *ep; /* extent record pointer */
+ xfs_extnum_t lastx; /* last extent index */
+
+ /*
+ * Initialize the extent entry structure to catch access to
+ * uninitialized br_startblock field.
+ */
+ gotp->br_startoff = 0xffa5a5a5a5a5a5a5LL;
+ gotp->br_blockcount = 0xa55a5a5a5a5a5a5aLL;
+ gotp->br_state = XFS_EXT_INVALID;
+#if XFS_BIG_BLKNOS
+ gotp->br_startblock = 0xffffa5a5a5a5a5a5LL;
+#else
+ gotp->br_startblock = 0xffffa5a5;
+#endif
+ prevp->br_startoff = NULLFILEOFF;
+
+ ep = xfs_iext_bno_to_ext(ifp, bno, &lastx);
+ if (lastx > 0) {
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp, lastx - 1), prevp);
+ }
+ if (lastx < (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t))) {
+ xfs_bmbt_get_all(ep, gotp);
+ *eofp = 0;
+ } else {
+ if (lastx > 0) {
+ *gotp = *prevp;
+ }
+ *eofp = 1;
+ ep = NULL;
+ }
+ *lastxp = lastx;
+ return ep;
+}
+
+/*
* Search the extents list for the inode, for the extent containing bno.
* If bno lies in a hole, point to the next entry. If bno lies past eof,
* *eofp will be set, and *prevp will contain the last entry (null if none).
@@ -3639,20 +3643,14 @@ xfs_bmap_search_extents(
xfs_bmbt_irec_t *prevp) /* out: previous extent entry found */
{
xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_bmbt_rec_t *base; /* base of extent list */
- xfs_extnum_t lastx; /* last extent index used */
- xfs_extnum_t nextents; /* extent list size */
- xfs_bmbt_rec_t *ep; /* extent list entry pointer */
+ xfs_bmbt_rec_t *ep; /* extent record pointer */
int rt; /* realtime flag */
XFS_STATS_INC(xs_look_exlist);
ifp = XFS_IFORK_PTR(ip, whichfork);
- lastx = ifp->if_lastex;
- nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- base = &ifp->if_u1.if_extents[0];
- ep = xfs_bmap_do_search_extents(base, lastx, nextents, bno, eofp,
- lastxp, gotp, prevp);
+ ep = xfs_bmap_search_multi_extents(ifp, bno, eofp, lastxp, gotp, prevp);
+
rt = (whichfork == XFS_DATA_FORK) && XFS_IS_REALTIME_INODE(ip);
if (unlikely(!rt && !gotp->br_startblock && (*lastxp != NULLEXTNUM))) {
cmn_err(CE_PANIC,"Access to block zero: fs: <%s> inode: %lld "
@@ -3732,7 +3730,7 @@ xfs_bmap_trace_addentry(
}
/*
- * Add bmap trace entry prior to a call to xfs_bmap_delete_exlist.
+ * Add bmap trace entry prior to a call to xfs_iext_remove.
*/
STATIC void
xfs_bmap_trace_delete(
@@ -3747,13 +3745,13 @@ xfs_bmap_trace_delete(
ifp = XFS_IFORK_PTR(ip, whichfork);
xfs_bmap_trace_addentry(XFS_BMAP_KTRACE_DELETE, fname, desc, ip, idx,
- cnt, &ifp->if_u1.if_extents[idx],
- cnt == 2 ? &ifp->if_u1.if_extents[idx + 1] : NULL,
+ cnt, xfs_iext_get_ext(ifp, idx),
+ cnt == 2 ? xfs_iext_get_ext(ifp, idx + 1) : NULL,
whichfork);
}
/*
- * Add bmap trace entry prior to a call to xfs_bmap_insert_exlist, or
+ * Add bmap trace entry prior to a call to xfs_iext_insert, or
* reading in the extents list from the disk (in the btree).
*/
STATIC void
@@ -3783,7 +3781,7 @@ xfs_bmap_trace_insert(
}
/*
- * Add bmap trace entry after updating an extent list entry in place.
+ * Add bmap trace entry after updating an extent record in place.
*/
STATIC void
xfs_bmap_trace_post_update(
@@ -3797,11 +3795,11 @@ xfs_bmap_trace_post_update(
ifp = XFS_IFORK_PTR(ip, whichfork);
xfs_bmap_trace_addentry(XFS_BMAP_KTRACE_POST_UP, fname, desc, ip, idx,
- 1, &ifp->if_u1.if_extents[idx], NULL, whichfork);
+ 1, xfs_iext_get_ext(ifp, idx), NULL, whichfork);
}
/*
- * Add bmap trace entry prior to updating an extent list entry in place.
+ * Add bmap trace entry prior to updating an extent record in place.
*/
STATIC void
xfs_bmap_trace_pre_update(
@@ -3815,7 +3813,7 @@ xfs_bmap_trace_pre_update(
ifp = XFS_IFORK_PTR(ip, whichfork);
xfs_bmap_trace_addentry(XFS_BMAP_KTRACE_PRE_UP, fname, desc, ip, idx, 1,
- &ifp->if_u1.if_extents[idx], NULL, whichfork);
+ xfs_iext_get_ext(ifp, idx), NULL, whichfork);
}
#endif /* XFS_BMAP_TRACE */
@@ -3892,7 +3890,7 @@ xfs_bmap_add_attrfork(
int rsvd) /* xact may use reserved blks */
{
xfs_fsblock_t firstblock; /* 1st block/ag allocated */
- xfs_bmap_free_t flist; /* freed extent list */
+ xfs_bmap_free_t flist; /* freed extent records */
xfs_mount_t *mp; /* mount structure */
xfs_trans_t *tp; /* transaction pointer */
unsigned long s; /* spinlock spl value */
@@ -4146,7 +4144,7 @@ xfs_bmap_finish(
xfs_efd_log_item_t *efd; /* extent free data */
xfs_efi_log_item_t *efi; /* extent free intention */
int error; /* error return value */
- xfs_bmap_free_item_t *free; /* free extent list item */
+ xfs_bmap_free_item_t *free; /* free extent item */
unsigned int logres; /* new log reservation */
unsigned int logcount; /* new log count */
xfs_mount_t *mp; /* filesystem mount structure */
@@ -4242,9 +4240,9 @@ xfs_bmap_first_unused(
xfs_fileoff_t *first_unused, /* unused block */
int whichfork) /* data or attr fork */
{
- xfs_bmbt_rec_t *base; /* base of extent array */
xfs_bmbt_rec_t *ep; /* pointer to an extent entry */
int error; /* error return value */
+ int idx; /* extent record index */
xfs_ifork_t *ifp; /* inode fork pointer */
xfs_fileoff_t lastaddr; /* last block number seen */
xfs_fileoff_t lowest; /* lowest useful block */
@@ -4265,10 +4263,8 @@ xfs_bmap_first_unused(
return error;
lowest = *first_unused;
nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- base = &ifp->if_u1.if_extents[0];
- for (lastaddr = 0, max = lowest, ep = base;
- ep < &base[nextents];
- ep++) {
+ for (idx = 0, lastaddr = 0, max = lowest; idx < nextents; idx++) {
+ ep = xfs_iext_get_ext(ifp, idx);
off = xfs_bmbt_get_startoff(ep);
/*
* See if the hole before this extent will work.
@@ -4287,8 +4283,8 @@ xfs_bmap_first_unused(
/*
* Returns the file-relative block number of the last block + 1 before
* last_block (input value) in the file.
- * This is not based on i_size, it is based on the extent list.
- * Returns 0 for local files, as they do not have an extent list.
+ * This is not based on i_size, it is based on the extent records.
+ * Returns 0 for local files, as they do not have extent records.
*/
int /* error */
xfs_bmap_last_before(
@@ -4335,8 +4331,8 @@ xfs_bmap_last_before(
/*
* Returns the file-relative block number of the first block past eof in
- * the file. This is not based on i_size, it is based on the extent list.
- * Returns 0 for local files, as they do not have an extent list.
+ * the file. This is not based on i_size, it is based on the extent records.
+ * Returns 0 for local files, as they do not have extent records.
*/
int /* error */
xfs_bmap_last_offset(
@@ -4345,7 +4341,6 @@ xfs_bmap_last_offset(
xfs_fileoff_t *last_block, /* last block */
int whichfork) /* data or attr fork */
{
- xfs_bmbt_rec_t *base; /* base of extent array */
xfs_bmbt_rec_t *ep; /* pointer to last extent */
int error; /* error return value */
xfs_ifork_t *ifp; /* inode fork pointer */
@@ -4368,9 +4363,7 @@ xfs_bmap_last_offset(
*last_block = 0;
return 0;
}
- base = &ifp->if_u1.if_extents[0];
- ASSERT(base != NULL);
- ep = &base[nextents - 1];
+ ep = xfs_iext_get_ext(ifp, nextents - 1);
*last_block = xfs_bmbt_get_startoff(ep) + xfs_bmbt_get_blockcount(ep);
return 0;
}
@@ -4400,7 +4393,7 @@ xfs_bmap_one_block(
return 0;
ifp = XFS_IFORK_PTR(ip, whichfork);
ASSERT(ifp->if_flags & XFS_IFEXTENTS);
- ep = ifp->if_u1.if_extents;
+ ep = xfs_iext_get_ext(ifp, 0);
xfs_bmbt_get_all(ep, &s);
rval = s.br_startoff == 0 && s.br_blockcount == 1;
if (rval && whichfork == XFS_DATA_FORK)
@@ -4435,7 +4428,6 @@ xfs_bmap_read_extents(
xfs_bmbt_ptr_t *pp; /* pointer to block address */
/* REFERENCED */
xfs_extnum_t room; /* number of entries there's room for */
- xfs_bmbt_rec_t *trp; /* target record pointer */
bno = NULLFSBLOCK;
mp = ip->i_mount;
@@ -4478,16 +4470,16 @@ xfs_bmap_read_extents(
/*
* Here with bp and block set to the leftmost leaf node in the tree.
*/
- room = ifp->if_bytes / (uint)sizeof(*trp);
- trp = ifp->if_u1.if_extents;
+ room = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
i = 0;
/*
- * Loop over all leaf nodes. Copy information to the extent list.
+ * Loop over all leaf nodes. Copy information to the extent records.
*/
for (;;) {
- xfs_bmbt_rec_t *frp, *temp;
+ xfs_bmbt_rec_t *frp, *trp;
xfs_fsblock_t nextbno;
xfs_extnum_t num_recs;
+ xfs_extnum_t start;
num_recs = be16_to_cpu(block->bb_numrecs);
@@ -4511,12 +4503,13 @@ xfs_bmap_read_extents(
if (nextbno != NULLFSBLOCK)
xfs_btree_reada_bufl(mp, nextbno, 1);
/*
- * Copy records into the extent list.
+ * Copy records into the extent records.
*/
frp = XFS_BTREE_REC_ADDR(mp->m_sb.sb_blocksize, xfs_bmbt,
block, 1, mp->m_bmap_dmxr[0]);
- temp = trp;
- for (j = 0; j < num_recs; j++, frp++, trp++) {
+ start = i;
+ for (j = 0; j < num_recs; j++, i++, frp++) {
+ trp = xfs_iext_get_ext(ifp, i);
trp->l0 = INT_GET(frp->l0, ARCH_CONVERT);
trp->l1 = INT_GET(frp->l1, ARCH_CONVERT);
}
@@ -4526,14 +4519,14 @@ xfs_bmap_read_extents(
* any "older" data bmap btree records for a
* set bit in the "extent flag" position.
*/
- if (unlikely(xfs_check_nostate_extents(temp, num_recs))) {
+ if (unlikely(xfs_check_nostate_extents(ifp,
+ start, num_recs))) {
XFS_ERROR_REPORT("xfs_bmap_read_extents(2)",
XFS_ERRLEVEL_LOW,
ip->i_mount);
goto error0;
}
}
- i += num_recs;
xfs_trans_brelse(tp, bp);
bno = nextbno;
/*
@@ -4546,7 +4539,7 @@ xfs_bmap_read_extents(
return error;
block = XFS_BUF_TO_BMBT_BLOCK(bp);
}
- ASSERT(i == ifp->if_bytes / (uint)sizeof(*trp));
+ ASSERT(i == (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)));
ASSERT(i == XFS_IFORK_NEXTENTS(ip, whichfork));
xfs_bmap_trace_exlist(fname, ip, i, whichfork);
return 0;
@@ -4557,7 +4550,7 @@ error0:
#ifdef XFS_BMAP_TRACE
/*
- * Add bmap trace insert entries for all the contents of the extent list.
+ * Add bmap trace insert entries for all the contents of the extent records.
*/
void
xfs_bmap_trace_exlist(
@@ -4566,16 +4559,15 @@ xfs_bmap_trace_exlist(
xfs_extnum_t cnt, /* count of entries in the list */
int whichfork) /* data or attr fork */
{
- xfs_bmbt_rec_t *base; /* base of extent list */
- xfs_bmbt_rec_t *ep; /* current entry in extent list */
- xfs_extnum_t idx; /* extent list entry number */
+ xfs_bmbt_rec_t *ep; /* current extent record */
+ xfs_extnum_t idx; /* extent record index */
xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_bmbt_irec_t s; /* extent list record */
+ xfs_bmbt_irec_t s; /* file extent record */
ifp = XFS_IFORK_PTR(ip, whichfork);
- ASSERT(cnt == ifp->if_bytes / (uint)sizeof(*base));
- base = ifp->if_u1.if_extents;
- for (idx = 0, ep = base; idx < cnt; idx++, ep++) {
+ ASSERT(cnt == (ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t)));
+ for (idx = 0; idx < cnt; idx++) {
+ ep = xfs_iext_get_ext(ifp, idx);
xfs_bmbt_get_all(ep, &s);
xfs_bmap_trace_insert(fname, "exlist", ip, idx, 1, &s, NULL,
whichfork);
@@ -4661,14 +4653,10 @@ xfs_bmapi(
xfs_bmalloca_t bma; /* args for xfs_bmap_alloc */
xfs_btree_cur_t *cur; /* bmap btree cursor */
xfs_fileoff_t end; /* end of mapped file region */
- int eof; /* we've hit the end of extent list */
- char contig; /* allocation must be one extent */
- char delay; /* this request is for delayed alloc */
- char exact; /* don't do all of wasdelayed extent */
- char convert; /* unwritten extent I/O completion */
- xfs_bmbt_rec_t *ep; /* extent list entry pointer */
+ int eof; /* we've hit the end of extents */
+ xfs_bmbt_rec_t *ep; /* extent record pointer */
int error; /* error return */
- xfs_bmbt_irec_t got; /* current extent list record */
+ xfs_bmbt_irec_t got; /* current file extent record */
xfs_ifork_t *ifp; /* inode fork pointer */
xfs_extlen_t indlen; /* indirect blocks length */
xfs_extnum_t lastx; /* last useful extent number */
@@ -4680,17 +4668,13 @@ xfs_bmapi(
int nallocs; /* number of extents alloc\'d */
xfs_extnum_t nextents; /* number of extents in file */
xfs_fileoff_t obno; /* old block number (offset) */
- xfs_bmbt_irec_t prev; /* previous extent list record */
+ xfs_bmbt_irec_t prev; /* previous file extent record */
int tmp_logflags; /* temp flags holder */
int whichfork; /* data or attr fork */
char inhole; /* current location is hole in file */
- char stateless; /* ignore state flag set */
- char trim; /* output trimmed to match range */
- char userdata; /* allocating non-metadata */
char wasdelay; /* old extent was delayed */
char wr; /* this is a write request */
char rt; /* this is a realtime file */
- char rsvd; /* OK to allocate reserved blocks */
#ifdef DEBUG
xfs_fileoff_t orig_bno; /* original block number value */
int orig_flags; /* original flags arg value */
@@ -4727,15 +4711,8 @@ xfs_bmapi(
XFS_STATS_INC(xs_blk_mapw);
else
XFS_STATS_INC(xs_blk_mapr);
- delay = (flags & XFS_BMAPI_DELAY) != 0;
- trim = (flags & XFS_BMAPI_ENTIRE) == 0;
- userdata = (flags & XFS_BMAPI_METADATA) == 0;
- convert = (flags & XFS_BMAPI_CONVERT) != 0;
- exact = (flags & XFS_BMAPI_EXACT) != 0;
- rsvd = (flags & XFS_BMAPI_RSVBLOCKS) != 0;
- contig = (flags & XFS_BMAPI_CONTIG) != 0;
/*
- * stateless is used to combine extents which
+ * IGSTATE flag is used to combine extents which
* differ only due to the state of the extents.
* This technique is used from xfs_getbmap()
* when the caller does not wish to see the
@@ -4751,10 +4728,9 @@ xfs_bmapi(
* xfs_strat_comp(), where the xfs_bmapi() call
* is transactioned, and the extents combined.
*/
- stateless = (flags & XFS_BMAPI_IGSTATE) != 0;
- if (stateless && wr) /* if writing unwritten space, no */
- wr = 0; /* allocations are allowed */
- ASSERT(wr || !delay);
+ if ((flags & XFS_BMAPI_IGSTATE) && wr) /* if writing unwritten space */
+ wr = 0; /* no allocations are allowed */
+ ASSERT(wr || !(flags & XFS_BMAPI_DELAY));
logflags = 0;
nallocs = 0;
cur = NULL;
@@ -4789,7 +4765,7 @@ xfs_bmapi(
if (eof && !wr)
got.br_startoff = end;
inhole = eof || got.br_startoff > bno;
- wasdelay = wr && !inhole && !delay &&
+ wasdelay = wr && !inhole && !(flags & XFS_BMAPI_DELAY) &&
ISNULLSTARTBLOCK(got.br_startblock);
/*
* First, deal with the hole before the allocated space
@@ -4801,11 +4777,11 @@ xfs_bmapi(
* allocate the stuff asked for in this bmap call
* but that wouldn't be as good.
*/
- if (wasdelay && !exact) {
+ if (wasdelay && !(flags & XFS_BMAPI_EXACT)) {
alen = (xfs_extlen_t)got.br_blockcount;
aoff = got.br_startoff;
if (lastx != NULLEXTNUM && lastx) {
- ep = &ifp->if_u1.if_extents[lastx - 1];
+ ep = xfs_iext_get_ext(ifp, lastx - 1);
xfs_bmbt_get_all(ep, &prev);
}
} else if (wasdelay) {
@@ -4823,8 +4799,8 @@ xfs_bmapi(
got.br_startoff - bno);
aoff = bno;
}
- minlen = contig ? alen : 1;
- if (delay) {
+ minlen = (flags & XFS_BMAPI_CONTIG) ? alen : 1;
+ if (flags & XFS_BMAPI_DELAY) {
xfs_extlen_t extsz;
/* Figure out the extent size, adjust alen */
@@ -4837,7 +4813,9 @@ xfs_bmapi(
if (extsz) {
error = xfs_bmap_extsize_align(mp,
&got, &prev, extsz,
- rt, eof, delay, convert,
+ rt, eof,
+ flags&XFS_BMAPI_DELAY,
+ flags&XFS_BMAPI_CONVERT,
&aoff, &alen);
ASSERT(!error);
}
@@ -4875,24 +4853,29 @@ xfs_bmapi(
if (rt) {
error = xfs_mod_incore_sb(mp,
XFS_SBS_FREXTENTS,
- -(extsz), rsvd);
+ -(extsz), (flags &
+ XFS_BMAPI_RSVBLOCKS));
} else {
error = xfs_mod_incore_sb(mp,
XFS_SBS_FDBLOCKS,
- -(alen), rsvd);
+ -(alen), (flags &
+ XFS_BMAPI_RSVBLOCKS));
}
if (!error) {
error = xfs_mod_incore_sb(mp,
XFS_SBS_FDBLOCKS,
- -(indlen), rsvd);
+ -(indlen), (flags &
+ XFS_BMAPI_RSVBLOCKS));
if (error && rt)
xfs_mod_incore_sb(mp,
XFS_SBS_FREXTENTS,
- extsz, rsvd);
+ extsz, (flags &
+ XFS_BMAPI_RSVBLOCKS));
else if (error)
xfs_mod_incore_sb(mp,
XFS_SBS_FDBLOCKS,
- alen, rsvd);
+ alen, (flags &
+ XFS_BMAPI_RSVBLOCKS));
}
if (error) {
@@ -4925,7 +4908,7 @@ xfs_bmapi(
/* Indicate if this is the first user data
* in the file, or just any user data.
*/
- if (userdata) {
+ if (!(flags & XFS_BMAPI_METADATA)) {
bma.userdata = (aoff == 0) ?
XFS_ALLOC_INITIAL_USER_DATA :
XFS_ALLOC_USERDATA;
@@ -4937,7 +4920,7 @@ xfs_bmapi(
bma.firstblock = *firstblock;
bma.alen = alen;
bma.off = aoff;
- bma.conv = convert;
+ bma.conv = (flags & XFS_BMAPI_CONVERT);
bma.wasdel = wasdelay;
bma.minlen = minlen;
bma.low = flist->xbf_low;
@@ -4948,7 +4931,8 @@ xfs_bmapi(
* is larger than a stripe unit.
*/
if (mp->m_dalign && alen >= mp->m_dalign &&
- userdata && whichfork == XFS_DATA_FORK) {
+ (!(flags & XFS_BMAPI_METADATA)) &&
+ (whichfork == XFS_DATA_FORK)) {
if ((error = xfs_bmap_isaeof(ip, aoff,
whichfork, &bma.aeof)))
goto error0;
@@ -5011,19 +4995,19 @@ xfs_bmapi(
}
error = xfs_bmap_add_extent(ip, lastx, &cur, &got,
firstblock, flist, &tmp_logflags, whichfork,
- rsvd);
+ (flags & XFS_BMAPI_RSVBLOCKS));
logflags |= tmp_logflags;
if (error)
goto error0;
lastx = ifp->if_lastex;
- ep = &ifp->if_u1.if_extents[lastx];
+ ep = xfs_iext_get_ext(ifp, lastx);
nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
xfs_bmbt_get_all(ep, &got);
ASSERT(got.br_startoff <= aoff);
ASSERT(got.br_startoff + got.br_blockcount >=
aoff + alen);
#ifdef DEBUG
- if (delay) {
+ if (flags & XFS_BMAPI_DELAY) {
ASSERT(ISNULLSTARTBLOCK(got.br_startblock));
ASSERT(STARTBLOCKVAL(got.br_startblock) > 0);
}
@@ -5052,14 +5036,15 @@ xfs_bmapi(
* Then deal with the allocated space we found.
*/
ASSERT(ep != NULL);
- if (trim && (got.br_startoff + got.br_blockcount > obno)) {
+ if (!(flags & XFS_BMAPI_ENTIRE) &&
+ (got.br_startoff + got.br_blockcount > obno)) {
if (obno > bno)
bno = obno;
ASSERT((bno >= obno) || (n == 0));
ASSERT(bno < end);
mval->br_startoff = bno;
if (ISNULLSTARTBLOCK(got.br_startblock)) {
- ASSERT(!wr || delay);
+ ASSERT(!wr || (flags & XFS_BMAPI_DELAY));
mval->br_startblock = DELAYSTARTBLOCK;
} else
mval->br_startblock =
@@ -5081,7 +5066,7 @@ xfs_bmapi(
} else {
*mval = got;
if (ISNULLSTARTBLOCK(mval->br_startblock)) {
- ASSERT(!wr || delay);
+ ASSERT(!wr || (flags & XFS_BMAPI_DELAY));
mval->br_startblock = DELAYSTARTBLOCK;
}
}
@@ -5107,12 +5092,12 @@ xfs_bmapi(
mval->br_state = XFS_EXT_NORM;
error = xfs_bmap_add_extent(ip, lastx, &cur, mval,
firstblock, flist, &tmp_logflags, whichfork,
- rsvd);
+ (flags & XFS_BMAPI_RSVBLOCKS));
logflags |= tmp_logflags;
if (error)
goto error0;
lastx = ifp->if_lastex;
- ep = &ifp->if_u1.if_extents[lastx];
+ ep = xfs_iext_get_ext(ifp, lastx);
nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
xfs_bmbt_get_all(ep, &got);
/*
@@ -5124,9 +5109,10 @@ xfs_bmapi(
continue;
}
- ASSERT(!trim ||
+ ASSERT((flags & XFS_BMAPI_ENTIRE) ||
((mval->br_startoff + mval->br_blockcount) <= end));
- ASSERT(!trim || (mval->br_blockcount <= len) ||
+ ASSERT((flags & XFS_BMAPI_ENTIRE) ||
+ (mval->br_blockcount <= len) ||
(mval->br_startoff < obno));
bno = mval->br_startoff + mval->br_blockcount;
len = end - bno;
@@ -5141,7 +5127,8 @@ xfs_bmapi(
mval[-1].br_startblock != HOLESTARTBLOCK &&
mval->br_startblock ==
mval[-1].br_startblock + mval[-1].br_blockcount &&
- (stateless || mval[-1].br_state == mval->br_state)) {
+ ((flags & XFS_BMAPI_IGSTATE) ||
+ mval[-1].br_state == mval->br_state)) {
ASSERT(mval->br_startoff ==
mval[-1].br_startoff + mval[-1].br_blockcount);
mval[-1].br_blockcount += mval->br_blockcount;
@@ -5168,8 +5155,7 @@ xfs_bmapi(
/*
* Else go on to the next record.
*/
- ep++;
- lastx++;
+ ep = xfs_iext_get_ext(ifp, ++lastx);
if (lastx >= nextents) {
eof = 1;
prev = got;
@@ -5199,7 +5185,7 @@ xfs_bmapi(
error0:
/*
* Log everything. Do this after conversion, there's no point in
- * logging the extent list if we've converted to btree format.
+ * logging the extent records if we've converted to btree format.
*/
if ((logflags & XFS_ILOG_FEXT(whichfork)) &&
XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
@@ -5252,12 +5238,12 @@ xfs_bmapi_single(
xfs_fsblock_t *fsb, /* output: mapped block */
xfs_fileoff_t bno) /* starting file offs. mapped */
{
- int eof; /* we've hit the end of extent list */
+ int eof; /* we've hit the end of extents */
int error; /* error return */
- xfs_bmbt_irec_t got; /* current extent list record */
+ xfs_bmbt_irec_t got; /* current file extent record */
xfs_ifork_t *ifp; /* inode fork pointer */
xfs_extnum_t lastx; /* last useful extent number */
- xfs_bmbt_irec_t prev; /* previous extent list record */
+ xfs_bmbt_irec_t prev; /* previous file extent record */
ifp = XFS_IFORK_PTR(ip, whichfork);
if (unlikely(
@@ -5312,18 +5298,18 @@ xfs_bunmapi(
xfs_btree_cur_t *cur; /* bmap btree cursor */
xfs_bmbt_irec_t del; /* extent being deleted */
int eof; /* is deleting at eof */
- xfs_bmbt_rec_t *ep; /* extent list entry pointer */
+ xfs_bmbt_rec_t *ep; /* extent record pointer */
int error; /* error return value */
xfs_extnum_t extno; /* extent number in list */
- xfs_bmbt_irec_t got; /* current extent list entry */
+ xfs_bmbt_irec_t got; /* current extent record */
xfs_ifork_t *ifp; /* inode fork pointer */
int isrt; /* freeing in rt area */
xfs_extnum_t lastx; /* last extent index used */
int logflags; /* transaction logging flags */
xfs_extlen_t mod; /* rt extent offset */
xfs_mount_t *mp; /* mount structure */
- xfs_extnum_t nextents; /* size of extent list */
- xfs_bmbt_irec_t prev; /* previous extent list entry */
+ xfs_extnum_t nextents; /* number of file extents */
+ xfs_bmbt_irec_t prev; /* previous extent record */
xfs_fileoff_t start; /* first file offset deleted */
int tmp_logflags; /* partial logging flags */
int wasdel; /* was a delayed alloc extent */
@@ -5369,7 +5355,7 @@ xfs_bunmapi(
* file, back up to the last block if so...
*/
if (eof) {
- ep = &ifp->if_u1.if_extents[--lastx];
+ ep = xfs_iext_get_ext(ifp, --lastx);
xfs_bmbt_get_all(ep, &got);
bno = got.br_startoff + got.br_blockcount - 1;
}
@@ -5393,7 +5379,7 @@ xfs_bunmapi(
if (got.br_startoff > bno) {
if (--lastx < 0)
break;
- ep--;
+ ep = xfs_iext_get_ext(ifp, lastx);
xfs_bmbt_get_all(ep, &got);
}
/*
@@ -5440,7 +5426,8 @@ xfs_bunmapi(
del.br_blockcount : mod;
if (bno < got.br_startoff) {
if (--lastx >= 0)
- xfs_bmbt_get_all(--ep, &got);
+ xfs_bmbt_get_all(xfs_iext_get_ext(
+ ifp, lastx), &got);
}
continue;
}
@@ -5500,7 +5487,8 @@ xfs_bunmapi(
* try again.
*/
ASSERT(lastx > 0);
- xfs_bmbt_get_all(ep - 1, &prev);
+ xfs_bmbt_get_all(xfs_iext_get_ext(ifp,
+ lastx - 1), &prev);
ASSERT(prev.br_state == XFS_EXT_NORM);
ASSERT(!ISNULLSTARTBLOCK(prev.br_startblock));
ASSERT(del.br_startblock ==
@@ -5587,12 +5575,12 @@ nodelete:
* If not done go on to the next (previous) record.
* Reset ep in case the extents array was re-alloced.
*/
- ep = &ifp->if_u1.if_extents[lastx];
+ ep = xfs_iext_get_ext(ifp, lastx);
if (bno != (xfs_fileoff_t)-1 && bno >= start) {
if (lastx >= XFS_IFORK_NEXTENTS(ip, whichfork) ||
xfs_bmbt_get_startoff(ep) > bno) {
- lastx--;
- ep--;
+ if (--lastx >= 0)
+ ep = xfs_iext_get_ext(ifp, lastx);
}
if (lastx >= 0)
xfs_bmbt_get_all(ep, &got);
@@ -5636,7 +5624,7 @@ nodelete:
error0:
/*
* Log everything. Do this after conversion, there's no point in
- * logging the extent list if we've converted to btree format.
+ * logging the extent records if we've converted to btree format.
*/
if ((logflags & XFS_ILOG_FEXT(whichfork)) &&
XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_EXTENTS)
@@ -5892,9 +5880,9 @@ xfs_bmap_isaeof(
{
int error; /* error return value */
xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_bmbt_rec_t *lastrec; /* extent list entry pointer */
- xfs_extnum_t nextents; /* size of extent list */
- xfs_bmbt_irec_t s; /* expanded extent list entry */
+ xfs_bmbt_rec_t *lastrec; /* extent record pointer */
+ xfs_extnum_t nextents; /* number of file extents */
+ xfs_bmbt_irec_t s; /* expanded extent record */
ASSERT(whichfork == XFS_DATA_FORK);
ifp = XFS_IFORK_PTR(ip, whichfork);
@@ -5909,7 +5897,7 @@ xfs_bmap_isaeof(
/*
* Go to the last extent
*/
- lastrec = &ifp->if_u1.if_extents[nextents - 1];
+ lastrec = xfs_iext_get_ext(ifp, nextents - 1);
xfs_bmbt_get_all(lastrec, &s);
/*
* Check we are allocating in the last extent (for delayed allocations)
@@ -5936,8 +5924,8 @@ xfs_bmap_eof(
xfs_fsblock_t blockcount; /* extent block count */
int error; /* error return value */
xfs_ifork_t *ifp; /* inode fork pointer */
- xfs_bmbt_rec_t *lastrec; /* extent list entry pointer */
- xfs_extnum_t nextents; /* size of extent list */
+ xfs_bmbt_rec_t *lastrec; /* extent record pointer */
+ xfs_extnum_t nextents; /* number of file extents */
xfs_fileoff_t startoff; /* extent starting file offset */
ASSERT(whichfork == XFS_DATA_FORK);
@@ -5953,7 +5941,7 @@ xfs_bmap_eof(
/*
* Go to the last extent
*/
- lastrec = &ifp->if_u1.if_extents[nextents - 1];
+ lastrec = xfs_iext_get_ext(ifp, nextents - 1);
startoff = xfs_bmbt_get_startoff(lastrec);
blockcount = xfs_bmbt_get_blockcount(lastrec);
*eof = endoff >= startoff + blockcount;
@@ -5969,18 +5957,21 @@ xfs_bmap_check_extents(
xfs_inode_t *ip, /* incore inode pointer */
int whichfork) /* data or attr fork */
{
- xfs_bmbt_rec_t *base; /* base of extents list */
xfs_bmbt_rec_t *ep; /* current extent entry */
+ xfs_extnum_t idx; /* extent record index */
xfs_ifork_t *ifp; /* inode fork pointer */
xfs_extnum_t nextents; /* number of extents in list */
+ xfs_bmbt_rec_t *nextp; /* next extent entry */
ifp = XFS_IFORK_PTR(ip, whichfork);
ASSERT(ifp->if_flags & XFS_IFEXTENTS);
- base = ifp->if_u1.if_extents;
nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
- for (ep = base; ep < &base[nextents - 1]; ep++) {
+ ep = xfs_iext_get_ext(ifp, 0);
+ for (idx = 0; idx < nextents - 1; idx++) {
+ nextp = xfs_iext_get_ext(ifp, idx + 1);
xfs_btree_check_rec(XFS_BTNUM_BMAP, (void *)ep,
- (void *)(ep + 1));
+ (void *)(nextp));
+ ep = nextp;
}
}
@@ -6119,12 +6110,14 @@ xfs_bmap_check_leaf_extents(
xfs_fsblock_t bno; /* block # of "block" */
xfs_buf_t *bp; /* buffer for "block" */
int error; /* error return value */
- xfs_extnum_t i=0; /* index into the extents list */
+ xfs_extnum_t i=0, j; /* index into the extents list */
xfs_ifork_t *ifp; /* fork structure */
int level; /* btree level, for checking */
xfs_mount_t *mp; /* file system mount structure */
xfs_bmbt_ptr_t *pp; /* pointer to block address */
- xfs_bmbt_rec_t *ep, *lastp; /* extent pointers in block entry */
+ xfs_bmbt_rec_t *ep; /* pointer to current extent */
+ xfs_bmbt_rec_t *lastp; /* pointer to previous extent */
+ xfs_bmbt_rec_t *nextp; /* pointer to next extent */
int bp_release = 0;
if (XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE) {
@@ -6194,7 +6187,6 @@ xfs_bmap_check_leaf_extents(
*/
lastp = NULL;
for (;;) {
- xfs_bmbt_rec_t *frp;
xfs_fsblock_t nextbno;
xfs_extnum_t num_recs;
@@ -6213,18 +6205,20 @@ xfs_bmap_check_leaf_extents(
* conform with the first entry in this one.
*/
- frp = XFS_BTREE_REC_ADDR(mp->m_sb.sb_blocksize, xfs_bmbt,
+ ep = XFS_BTREE_REC_ADDR(mp->m_sb.sb_blocksize, xfs_bmbt,
block, 1, mp->m_bmap_dmxr[0]);
-
- for (ep = frp;ep < frp + (num_recs - 1); ep++) {
+ for (j = 1; j < num_recs; j++) {
+ nextp = XFS_BTREE_REC_ADDR(mp->m_sb.sb_blocksize, xfs_bmbt,
+ block, j + 1, mp->m_bmap_dmxr[0]);
if (lastp) {
xfs_btree_check_rec(XFS_BTNUM_BMAP,
(void *)lastp, (void *)ep);
}
xfs_btree_check_rec(XFS_BTNUM_BMAP, (void *)ep,
- (void *)(ep + 1));
+ (void *)(nextp));
+ lastp = ep;
+ ep = nextp;
}
- lastp = frp + num_recs - 1; /* For the next iteration */
i += num_recs;
if (bp_release) {
@@ -6288,7 +6282,7 @@ xfs_bmap_count_blocks(
mp = ip->i_mount;
ifp = XFS_IFORK_PTR(ip, whichfork);
if ( XFS_IFORK_FORMAT(ip, whichfork) == XFS_DINODE_FMT_EXTENTS ) {
- if (unlikely(xfs_bmap_count_leaves(ifp->if_u1.if_extents,
+ if (unlikely(xfs_bmap_count_leaves(ifp, 0,
ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t),
count) < 0)) {
XFS_ERROR_REPORT("xfs_bmap_count_blocks(1)",
@@ -6310,7 +6304,7 @@ xfs_bmap_count_blocks(
ASSERT(XFS_FSB_TO_AGBNO(mp, INT_GET(*pp, ARCH_CONVERT)) < mp->m_sb.sb_agblocks);
bno = INT_GET(*pp, ARCH_CONVERT);
- if (unlikely(xfs_bmap_count_tree(mp, tp, bno, level, count) < 0)) {
+ if (unlikely(xfs_bmap_count_tree(mp, tp, ifp, bno, level, count) < 0)) {
XFS_ERROR_REPORT("xfs_bmap_count_blocks(2)", XFS_ERRLEVEL_LOW,
mp);
return XFS_ERROR(EFSCORRUPTED);
@@ -6327,6 +6321,7 @@ int /* error */
xfs_bmap_count_tree(
xfs_mount_t *mp, /* file system mount point */
xfs_trans_t *tp, /* transaction pointer */
+ xfs_ifork_t *ifp, /* inode fork pointer */
xfs_fsblock_t blockno, /* file system block number */
int levelin, /* level in btree */
int *count) /* Count of blocks */
@@ -6339,7 +6334,6 @@ xfs_bmap_count_tree(
xfs_fsblock_t nextbno;
xfs_bmbt_block_t *block, *nextblock;
int numrecs;
- xfs_bmbt_rec_t *frp;
if ((error = xfs_btree_read_bufl(mp, tp, bno, 0, &bp, XFS_BMAP_BTREE_REF)))
return error;
@@ -6364,7 +6358,7 @@ xfs_bmap_count_tree(
xfs_bmbt, block, 1, mp->m_bmap_dmxr[1]);
bno = INT_GET(*pp, ARCH_CONVERT);
if (unlikely((error =
- xfs_bmap_count_tree(mp, tp, bno, level, count)) < 0)) {
+ xfs_bmap_count_tree(mp, tp, ifp, bno, level, count)) < 0)) {
xfs_trans_brelse(tp, bp);
XFS_ERROR_REPORT("xfs_bmap_count_tree(1)",
XFS_ERRLEVEL_LOW, mp);
@@ -6376,9 +6370,8 @@ xfs_bmap_count_tree(
for (;;) {
nextbno = be64_to_cpu(block->bb_rightsib);
numrecs = be16_to_cpu(block->bb_numrecs);
- frp = XFS_BTREE_REC_ADDR(mp->m_sb.sb_blocksize,
- xfs_bmbt, block, 1, mp->m_bmap_dmxr[0]);
- if (unlikely(xfs_bmap_disk_count_leaves(frp, numrecs, count) < 0)) {
+ if (unlikely(xfs_bmap_disk_count_leaves(ifp, mp,
+ 0, block, numrecs, count) < 0)) {
xfs_trans_brelse(tp, bp);
XFS_ERROR_REPORT("xfs_bmap_count_tree(2)",
XFS_ERRLEVEL_LOW, mp);
@@ -6399,33 +6392,45 @@ xfs_bmap_count_tree(
}
/*
- * Count leaf blocks given a pointer to an extent list.
+ * Count leaf blocks given a range of extent records.
*/
int
xfs_bmap_count_leaves(
- xfs_bmbt_rec_t *frp,
+ xfs_ifork_t *ifp,
+ xfs_extnum_t idx,
int numrecs,
int *count)
{
int b;
+ xfs_bmbt_rec_t *frp;
- for ( b = 1; b <= numrecs; b++, frp++)
+ for (b = 0; b < numrecs; b++) {
+ frp = xfs_iext_get_ext(ifp, idx + b);
*count += xfs_bmbt_get_blockcount(frp);
+ }
return 0;
}
/*
- * Count leaf blocks given a pointer to an extent list originally in btree format.
+ * Count leaf blocks given a range of extent records originally
+ * in btree format.
*/
int
xfs_bmap_disk_count_leaves(
- xfs_bmbt_rec_t *frp,
+ xfs_ifork_t *ifp,
+ xfs_mount_t *mp,
+ xfs_extnum_t idx,
+ xfs_bmbt_block_t *block,
int numrecs,
int *count)
{
int b;
+ xfs_bmbt_rec_t *frp;
- for ( b = 1; b <= numrecs; b++, frp++)
+ for (b = 1; b <= numrecs; b++) {
+ frp = XFS_BTREE_REC_ADDR(mp->m_sb.sb_blocksize,
+ xfs_bmbt, block, idx + b, mp->m_bmap_dmxr[0]);
*count += xfs_bmbt_disk_get_blockcount(frp);
+ }
return 0;
}
diff --git a/fs/xfs/xfs_bmap.h b/fs/xfs/xfs_bmap.h
index 12cc63dfc2c4..011ccaa9a1c0 100644
--- a/fs/xfs/xfs_bmap.h
+++ b/fs/xfs/xfs_bmap.h
@@ -20,6 +20,7 @@
struct getbmap;
struct xfs_bmbt_irec;
+struct xfs_ifork;
struct xfs_inode;
struct xfs_mount;
struct xfs_trans;
@@ -347,9 +348,28 @@ xfs_bmap_count_blocks(
*/
int
xfs_check_nostate_extents(
- xfs_bmbt_rec_t *ep,
+ struct xfs_ifork *ifp,
+ xfs_extnum_t idx,
xfs_extnum_t num);
+/*
+ * Call xfs_bmap_do_search_extents() to search for the extent
+ * record containing block bno. If in multi-level in-core extent
+ * allocation mode, find and extract the target extent buffer,
+ * otherwise just use the direct extent list.
+ */
+xfs_bmbt_rec_t *
+xfs_bmap_search_multi_extents(struct xfs_ifork *, xfs_fileoff_t, int *,
+ xfs_extnum_t *, xfs_bmbt_irec_t *, xfs_bmbt_irec_t *);
+
+/*
+ * Search an extent list for the extent which includes block
+ * bno.
+ */
+xfs_bmbt_rec_t *xfs_bmap_do_search_extents(xfs_bmbt_rec_t *,
+ xfs_extnum_t, xfs_extnum_t, xfs_fileoff_t, int *,
+ xfs_extnum_t *, xfs_bmbt_irec_t *, xfs_bmbt_irec_t *);
+
#endif /* __KERNEL__ */
#endif /* __XFS_BMAP_H__ */
diff --git a/fs/xfs/xfs_bmap_btree.c b/fs/xfs/xfs_bmap_btree.c
index 3f1383d160e8..bea44709afbe 100644
--- a/fs/xfs/xfs_bmap_btree.c
+++ b/fs/xfs/xfs_bmap_btree.c
@@ -2754,7 +2754,7 @@ xfs_bmbt_update(
}
/*
- * Check an extent list, which has just been read, for
+ * Check extent records, which have just been read, for
* any bit in the extent flag field. ASSERT on debug
* kernels, as this condition should not occur.
* Return an error condition (1) if any flags found,
@@ -2763,10 +2763,14 @@ xfs_bmbt_update(
int
xfs_check_nostate_extents(
- xfs_bmbt_rec_t *ep,
+ xfs_ifork_t *ifp,
+ xfs_extnum_t idx,
xfs_extnum_t num)
{
- for (; num > 0; num--, ep++) {
+ xfs_bmbt_rec_t *ep;
+
+ for (; num > 0; num--, idx++) {
+ ep = xfs_iext_get_ext(ifp, idx);
if ((ep->l0 >>
(64 - BMBT_EXNTFLAG_BITLEN)) != 0) {
ASSERT(0);
diff --git a/fs/xfs/xfs_bmap_btree.h b/fs/xfs/xfs_bmap_btree.h
index e095a2d344ae..6478cfa0e539 100644
--- a/fs/xfs/xfs_bmap_btree.h
+++ b/fs/xfs/xfs_bmap_btree.h
@@ -372,14 +372,6 @@ extern int xfs_bmbt_get_rec(struct xfs_btree_cur *, xfs_fileoff_t *,
xfs_exntst_t *, int *);
#endif
-/*
- * Search an extent list for the extent which includes block
- * bno.
- */
-xfs_bmbt_rec_t *xfs_bmap_do_search_extents(xfs_bmbt_rec_t *,
- xfs_extnum_t, xfs_extnum_t, xfs_fileoff_t, int *,
- xfs_extnum_t *, xfs_bmbt_irec_t *, xfs_bmbt_irec_t *);
-
#endif /* __KERNEL__ */
#endif /* __XFS_BMAP_BTREE_H__ */
diff --git a/fs/xfs/xfs_clnt.h b/fs/xfs/xfs_clnt.h
index f57cc9ac875e..022fff62085b 100644
--- a/fs/xfs/xfs_clnt.h
+++ b/fs/xfs/xfs_clnt.h
@@ -68,8 +68,6 @@ struct xfs_mount_args {
* enforcement */
#define XFSMNT_PQUOTAENF 0x00000040 /* IRIX project quota limit
* enforcement */
-#define XFSMNT_NOATIME 0x00000100 /* don't modify access
- * times on reads */
#define XFSMNT_NOALIGN 0x00000200 /* don't allocate at
* stripe boundaries*/
#define XFSMNT_RETERR 0x00000400 /* return error to user */
diff --git a/fs/xfs/xfs_da_btree.c b/fs/xfs/xfs_da_btree.c
index 473671fa5c13..4bae3a76c678 100644
--- a/fs/xfs/xfs_da_btree.c
+++ b/fs/xfs/xfs_da_btree.c
@@ -126,10 +126,10 @@ xfs_da_node_create(xfs_da_args_t *args, xfs_dablk_t blkno, int level,
node = bp->data;
node->hdr.info.forw = 0;
node->hdr.info.back = 0;
- INT_SET(node->hdr.info.magic, ARCH_CONVERT, XFS_DA_NODE_MAGIC);
+ node->hdr.info.magic = cpu_to_be16(XFS_DA_NODE_MAGIC);
node->hdr.info.pad = 0;
node->hdr.count = 0;
- INT_SET(node->hdr.level, ARCH_CONVERT, level);
+ node->hdr.level = cpu_to_be16(level);
xfs_da_log_buf(tp, bp,
XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr)));
@@ -290,28 +290,28 @@ xfs_da_split(xfs_da_state_t *state)
node = oldblk->bp->data;
if (node->hdr.info.forw) {
- if (INT_GET(node->hdr.info.forw, ARCH_CONVERT) == addblk->blkno) {
+ if (be32_to_cpu(node->hdr.info.forw) == addblk->blkno) {
bp = addblk->bp;
} else {
ASSERT(state->extravalid);
bp = state->extrablk.bp;
}
node = bp->data;
- INT_SET(node->hdr.info.back, ARCH_CONVERT, oldblk->blkno);
+ node->hdr.info.back = cpu_to_be32(oldblk->blkno);
xfs_da_log_buf(state->args->trans, bp,
XFS_DA_LOGRANGE(node, &node->hdr.info,
sizeof(node->hdr.info)));
}
node = oldblk->bp->data;
- if (INT_GET(node->hdr.info.back, ARCH_CONVERT)) {
- if (INT_GET(node->hdr.info.back, ARCH_CONVERT) == addblk->blkno) {
+ if (node->hdr.info.back) {
+ if (be32_to_cpu(node->hdr.info.back) == addblk->blkno) {
bp = addblk->bp;
} else {
ASSERT(state->extravalid);
bp = state->extrablk.bp;
}
node = bp->data;
- INT_SET(node->hdr.info.forw, ARCH_CONVERT, oldblk->blkno);
+ node->hdr.info.forw = cpu_to_be32(oldblk->blkno);
xfs_da_log_buf(state->args->trans, bp,
XFS_DA_LOGRANGE(node, &node->hdr.info,
sizeof(node->hdr.info)));
@@ -359,14 +359,14 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
ASSERT(bp != NULL);
node = bp->data;
oldroot = blk1->bp->data;
- if (INT_GET(oldroot->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC) {
- size = (int)((char *)&oldroot->btree[INT_GET(oldroot->hdr.count, ARCH_CONVERT)] -
+ if (be16_to_cpu(oldroot->hdr.info.magic) == XFS_DA_NODE_MAGIC) {
+ size = (int)((char *)&oldroot->btree[be16_to_cpu(oldroot->hdr.count)] -
(char *)oldroot);
} else {
ASSERT(XFS_DIR_IS_V2(mp));
- ASSERT(INT_GET(oldroot->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(oldroot->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
leaf = (xfs_dir2_leaf_t *)oldroot;
- size = (int)((char *)&leaf->ents[INT_GET(leaf->hdr.count, ARCH_CONVERT)] -
+ size = (int)((char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] -
(char *)leaf);
}
memcpy(node, oldroot, size);
@@ -381,18 +381,18 @@ xfs_da_root_split(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
error = xfs_da_node_create(args,
args->whichfork == XFS_DATA_FORK &&
XFS_DIR_IS_V2(mp) ? mp->m_dirleafblk : 0,
- INT_GET(node->hdr.level, ARCH_CONVERT) + 1, &bp, args->whichfork);
+ be16_to_cpu(node->hdr.level) + 1, &bp, args->whichfork);
if (error)
return(error);
node = bp->data;
- INT_SET(node->btree[0].hashval, ARCH_CONVERT, blk1->hashval);
- INT_SET(node->btree[0].before, ARCH_CONVERT, blk1->blkno);
- INT_SET(node->btree[1].hashval, ARCH_CONVERT, blk2->hashval);
- INT_SET(node->btree[1].before, ARCH_CONVERT, blk2->blkno);
- INT_SET(node->hdr.count, ARCH_CONVERT, 2);
+ node->btree[0].hashval = cpu_to_be32(blk1->hashval);
+ node->btree[0].before = cpu_to_be32(blk1->blkno);
+ node->btree[1].hashval = cpu_to_be32(blk2->hashval);
+ node->btree[1].before = cpu_to_be32(blk2->blkno);
+ node->hdr.count = cpu_to_be16(2);
#ifdef DEBUG
- if (INT_GET(oldroot->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC) {
+ if (be16_to_cpu(oldroot->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC) {
ASSERT(blk1->blkno >= mp->m_dirleafblk &&
blk1->blkno < mp->m_dirfreeblk);
ASSERT(blk2->blkno >= mp->m_dirleafblk &&
@@ -424,7 +424,7 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
int useextra;
node = oldblk->bp->data;
- ASSERT(INT_GET(node->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
/*
* With V2 the extra block is data or freespace.
@@ -435,7 +435,7 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
/*
* Do we have to split the node?
*/
- if ((INT_GET(node->hdr.count, ARCH_CONVERT) + newcount) > state->node_ents) {
+ if ((be16_to_cpu(node->hdr.count) + newcount) > state->node_ents) {
/*
* Allocate a new node, add to the doubly linked chain of
* nodes, then move some of our excess entries into it.
@@ -472,7 +472,7 @@ xfs_da_node_split(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
* If we had double-split op below us, then add the extra block too.
*/
node = oldblk->bp->data;
- if (oldblk->index <= INT_GET(node->hdr.count, ARCH_CONVERT)) {
+ if (oldblk->index <= be16_to_cpu(node->hdr.count)) {
oldblk->index++;
xfs_da_node_add(state, oldblk, addblk);
if (useextra) {
@@ -516,17 +516,17 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* Figure out how many entries need to move, and in which direction.
* Swap the nodes around if that makes it simpler.
*/
- if ((INT_GET(node1->hdr.count, ARCH_CONVERT) > 0) && (INT_GET(node2->hdr.count, ARCH_CONVERT) > 0) &&
- ((INT_GET(node2->btree[ 0 ].hashval, ARCH_CONVERT) < INT_GET(node1->btree[ 0 ].hashval, ARCH_CONVERT)) ||
- (INT_GET(node2->btree[ INT_GET(node2->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT) <
- INT_GET(node1->btree[ INT_GET(node1->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT)))) {
+ if ((be16_to_cpu(node1->hdr.count) > 0) && (be16_to_cpu(node2->hdr.count) > 0) &&
+ ((be32_to_cpu(node2->btree[0].hashval) < be32_to_cpu(node1->btree[0].hashval)) ||
+ (be32_to_cpu(node2->btree[be16_to_cpu(node2->hdr.count)-1].hashval) <
+ be32_to_cpu(node1->btree[be16_to_cpu(node1->hdr.count)-1].hashval)))) {
tmpnode = node1;
node1 = node2;
node2 = tmpnode;
}
- ASSERT(INT_GET(node1->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
- ASSERT(INT_GET(node2->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
- count = (INT_GET(node1->hdr.count, ARCH_CONVERT) - INT_GET(node2->hdr.count, ARCH_CONVERT)) / 2;
+ ASSERT(be16_to_cpu(node1->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(node2->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ count = (be16_to_cpu(node1->hdr.count) - be16_to_cpu(node2->hdr.count)) / 2;
if (count == 0)
return;
tp = state->args->trans;
@@ -537,7 +537,7 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
/*
* Move elements in node2 up to make a hole.
*/
- if ((tmp = INT_GET(node2->hdr.count, ARCH_CONVERT)) > 0) {
+ if ((tmp = be16_to_cpu(node2->hdr.count)) > 0) {
tmp *= (uint)sizeof(xfs_da_node_entry_t);
btree_s = &node2->btree[0];
btree_d = &node2->btree[count];
@@ -548,13 +548,12 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
* Move the req'd B-tree elements from high in node1 to
* low in node2.
*/
- INT_MOD(node2->hdr.count, ARCH_CONVERT, count);
+ be16_add(&node2->hdr.count, count);
tmp = count * (uint)sizeof(xfs_da_node_entry_t);
- btree_s = &node1->btree[INT_GET(node1->hdr.count, ARCH_CONVERT) - count];
+ btree_s = &node1->btree[be16_to_cpu(node1->hdr.count) - count];
btree_d = &node2->btree[0];
memcpy(btree_d, btree_s, tmp);
- INT_MOD(node1->hdr.count, ARCH_CONVERT, -(count));
-
+ be16_add(&node1->hdr.count, -count);
} else {
/*
* Move the req'd B-tree elements from low in node2 to
@@ -563,21 +562,21 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
count = -count;
tmp = count * (uint)sizeof(xfs_da_node_entry_t);
btree_s = &node2->btree[0];
- btree_d = &node1->btree[INT_GET(node1->hdr.count, ARCH_CONVERT)];
+ btree_d = &node1->btree[be16_to_cpu(node1->hdr.count)];
memcpy(btree_d, btree_s, tmp);
- INT_MOD(node1->hdr.count, ARCH_CONVERT, count);
+ be16_add(&node1->hdr.count, count);
xfs_da_log_buf(tp, blk1->bp,
XFS_DA_LOGRANGE(node1, btree_d, tmp));
/*
* Move elements in node2 down to fill the hole.
*/
- tmp = INT_GET(node2->hdr.count, ARCH_CONVERT) - count;
+ tmp = be16_to_cpu(node2->hdr.count) - count;
tmp *= (uint)sizeof(xfs_da_node_entry_t);
btree_s = &node2->btree[count];
btree_d = &node2->btree[0];
memmove(btree_d, btree_s, tmp);
- INT_MOD(node2->hdr.count, ARCH_CONVERT, -(count));
+ be16_add(&node2->hdr.count, -count);
}
/*
@@ -588,7 +587,7 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
xfs_da_log_buf(tp, blk2->bp,
XFS_DA_LOGRANGE(node2, &node2->hdr,
sizeof(node2->hdr) +
- sizeof(node2->btree[0]) * INT_GET(node2->hdr.count, ARCH_CONVERT)));
+ sizeof(node2->btree[0]) * be16_to_cpu(node2->hdr.count)));
/*
* Record the last hashval from each block for upward propagation.
@@ -596,15 +595,15 @@ xfs_da_node_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
*/
node1 = blk1->bp->data;
node2 = blk2->bp->data;
- blk1->hashval = INT_GET(node1->btree[ INT_GET(node1->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT);
- blk2->hashval = INT_GET(node2->btree[ INT_GET(node2->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT);
+ blk1->hashval = be32_to_cpu(node1->btree[be16_to_cpu(node1->hdr.count)-1].hashval);
+ blk2->hashval = be32_to_cpu(node2->btree[be16_to_cpu(node2->hdr.count)-1].hashval);
/*
* Adjust the expected index for insertion.
*/
- if (blk1->index >= INT_GET(node1->hdr.count, ARCH_CONVERT)) {
- blk2->index = blk1->index - INT_GET(node1->hdr.count, ARCH_CONVERT);
- blk1->index = INT_GET(node1->hdr.count, ARCH_CONVERT) + 1; /* make it invalid */
+ if (blk1->index >= be16_to_cpu(node1->hdr.count)) {
+ blk2->index = blk1->index - be16_to_cpu(node1->hdr.count);
+ blk1->index = be16_to_cpu(node1->hdr.count) + 1; /* make it invalid */
}
}
@@ -622,8 +621,8 @@ xfs_da_node_add(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
node = oldblk->bp->data;
mp = state->mp;
- ASSERT(INT_GET(node->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
- ASSERT((oldblk->index >= 0) && (oldblk->index <= INT_GET(node->hdr.count, ARCH_CONVERT)));
+ ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ ASSERT((oldblk->index >= 0) && (oldblk->index <= be16_to_cpu(node->hdr.count)));
ASSERT(newblk->blkno != 0);
if (state->args->whichfork == XFS_DATA_FORK && XFS_DIR_IS_V2(mp))
ASSERT(newblk->blkno >= mp->m_dirleafblk &&
@@ -634,22 +633,22 @@ xfs_da_node_add(xfs_da_state_t *state, xfs_da_state_blk_t *oldblk,
*/
tmp = 0;
btree = &node->btree[ oldblk->index ];
- if (oldblk->index < INT_GET(node->hdr.count, ARCH_CONVERT)) {
- tmp = (INT_GET(node->hdr.count, ARCH_CONVERT) - oldblk->index) * (uint)sizeof(*btree);
+ if (oldblk->index < be16_to_cpu(node->hdr.count)) {
+ tmp = (be16_to_cpu(node->hdr.count) - oldblk->index) * (uint)sizeof(*btree);
memmove(btree + 1, btree, tmp);
}
- INT_SET(btree->hashval, ARCH_CONVERT, newblk->hashval);
- INT_SET(btree->before, ARCH_CONVERT, newblk->blkno);
+ btree->hashval = cpu_to_be32(newblk->hashval);
+ btree->before = cpu_to_be32(newblk->blkno);
xfs_da_log_buf(state->args->trans, oldblk->bp,
XFS_DA_LOGRANGE(node, btree, tmp + sizeof(*btree)));
- INT_MOD(node->hdr.count, ARCH_CONVERT, +1);
+ be16_add(&node->hdr.count, 1);
xfs_da_log_buf(state->args->trans, oldblk->bp,
XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr)));
/*
* Copy the last hash value from the oldblk to propagate upwards.
*/
- oldblk->hashval = INT_GET(node->btree[ INT_GET(node->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT);
+ oldblk->hashval = be32_to_cpu(node->btree[be16_to_cpu(node->hdr.count)-1 ].hashval);
}
/*========================================================================
@@ -768,21 +767,21 @@ xfs_da_root_join(xfs_da_state_t *state, xfs_da_state_blk_t *root_blk)
ASSERT(args != NULL);
ASSERT(root_blk->magic == XFS_DA_NODE_MAGIC);
oldroot = root_blk->bp->data;
- ASSERT(INT_GET(oldroot->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(oldroot->hdr.info.magic) == XFS_DA_NODE_MAGIC);
ASSERT(!oldroot->hdr.info.forw);
ASSERT(!oldroot->hdr.info.back);
/*
* If the root has more than one child, then don't do anything.
*/
- if (INT_GET(oldroot->hdr.count, ARCH_CONVERT) > 1)
+ if (be16_to_cpu(oldroot->hdr.count) > 1)
return(0);
/*
* Read in the (only) child block, then copy those bytes into
* the root block's buffer and free the original child block.
*/
- child = INT_GET(oldroot->btree[ 0 ].before, ARCH_CONVERT);
+ child = be32_to_cpu(oldroot->btree[0].before);
ASSERT(child != 0);
error = xfs_da_read_buf(args->trans, args->dp, child, -1, &bp,
args->whichfork);
@@ -790,11 +789,11 @@ xfs_da_root_join(xfs_da_state_t *state, xfs_da_state_blk_t *root_blk)
return(error);
ASSERT(bp != NULL);
blkinfo = bp->data;
- if (INT_GET(oldroot->hdr.level, ARCH_CONVERT) == 1) {
- ASSERT(INT_GET(blkinfo->magic, ARCH_CONVERT) == XFS_DIRX_LEAF_MAGIC(state->mp) ||
- INT_GET(blkinfo->magic, ARCH_CONVERT) == XFS_ATTR_LEAF_MAGIC);
+ if (be16_to_cpu(oldroot->hdr.level) == 1) {
+ ASSERT(be16_to_cpu(blkinfo->magic) == XFS_DIRX_LEAF_MAGIC(state->mp) ||
+ be16_to_cpu(blkinfo->magic) == XFS_ATTR_LEAF_MAGIC);
} else {
- ASSERT(INT_GET(blkinfo->magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(blkinfo->magic) == XFS_DA_NODE_MAGIC);
}
ASSERT(!blkinfo->forw);
ASSERT(!blkinfo->back);
@@ -830,9 +829,9 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
*/
blk = &state->path.blk[ state->path.active-1 ];
info = blk->bp->data;
- ASSERT(INT_GET(info->magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(info->magic) == XFS_DA_NODE_MAGIC);
node = (xfs_da_intnode_t *)info;
- count = INT_GET(node->hdr.count, ARCH_CONVERT);
+ count = be16_to_cpu(node->hdr.count);
if (count > (state->node_ents >> 1)) {
*action = 0; /* blk over 50%, don't try to join */
return(0); /* blk over 50%, don't try to join */
@@ -849,7 +848,7 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
* Make altpath point to the block we want to keep and
* path point to the block we want to drop (this one).
*/
- forward = info->forw;
+ forward = (info->forw != 0);
memcpy(&state->altpath, &state->path, sizeof(state->path));
error = xfs_da_path_shift(state, &state->altpath, forward,
0, &retval);
@@ -871,13 +870,12 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
* to shrink a directory over time.
*/
/* start with smaller blk num */
- forward = (INT_GET(info->forw, ARCH_CONVERT)
- < INT_GET(info->back, ARCH_CONVERT));
+ forward = (be32_to_cpu(info->forw) < be32_to_cpu(info->back));
for (i = 0; i < 2; forward = !forward, i++) {
if (forward)
- blkno = INT_GET(info->forw, ARCH_CONVERT);
+ blkno = be32_to_cpu(info->forw);
else
- blkno = INT_GET(info->back, ARCH_CONVERT);
+ blkno = be32_to_cpu(info->back);
if (blkno == 0)
continue;
error = xfs_da_read_buf(state->args->trans, state->args->dp,
@@ -889,10 +887,10 @@ xfs_da_node_toosmall(xfs_da_state_t *state, int *action)
node = (xfs_da_intnode_t *)info;
count = state->node_ents;
count -= state->node_ents >> 2;
- count -= INT_GET(node->hdr.count, ARCH_CONVERT);
+ count -= be16_to_cpu(node->hdr.count);
node = bp->data;
- ASSERT(INT_GET(node->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
- count -= INT_GET(node->hdr.count, ARCH_CONVERT);
+ ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ count -= be16_to_cpu(node->hdr.count);
xfs_da_brelse(state->args->trans, bp);
if (count >= 0)
break; /* fits with at least 25% to spare */
@@ -973,16 +971,16 @@ xfs_da_fixhashpath(xfs_da_state_t *state, xfs_da_state_path_t *path)
}
for (blk--, level--; level >= 0; blk--, level--) {
node = blk->bp->data;
- ASSERT(INT_GET(node->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
btree = &node->btree[ blk->index ];
- if (INT_GET(btree->hashval, ARCH_CONVERT) == lasthash)
+ if (be32_to_cpu(btree->hashval) == lasthash)
break;
blk->hashval = lasthash;
- INT_SET(btree->hashval, ARCH_CONVERT, lasthash);
+ btree->hashval = cpu_to_be32(lasthash);
xfs_da_log_buf(state->args->trans, blk->bp,
XFS_DA_LOGRANGE(node, btree, sizeof(*btree)));
- lasthash = INT_GET(node->btree[ INT_GET(node->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT);
+ lasthash = be32_to_cpu(node->btree[be16_to_cpu(node->hdr.count)-1].hashval);
}
}
@@ -997,25 +995,25 @@ xfs_da_node_remove(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk)
int tmp;
node = drop_blk->bp->data;
- ASSERT(drop_blk->index < INT_GET(node->hdr.count, ARCH_CONVERT));
+ ASSERT(drop_blk->index < be16_to_cpu(node->hdr.count));
ASSERT(drop_blk->index >= 0);
/*
* Copy over the offending entry, or just zero it out.
*/
btree = &node->btree[drop_blk->index];
- if (drop_blk->index < (INT_GET(node->hdr.count, ARCH_CONVERT)-1)) {
- tmp = INT_GET(node->hdr.count, ARCH_CONVERT) - drop_blk->index - 1;
+ if (drop_blk->index < (be16_to_cpu(node->hdr.count)-1)) {
+ tmp = be16_to_cpu(node->hdr.count) - drop_blk->index - 1;
tmp *= (uint)sizeof(xfs_da_node_entry_t);
memmove(btree, btree + 1, tmp);
xfs_da_log_buf(state->args->trans, drop_blk->bp,
XFS_DA_LOGRANGE(node, btree, tmp));
- btree = &node->btree[ INT_GET(node->hdr.count, ARCH_CONVERT)-1 ];
+ btree = &node->btree[be16_to_cpu(node->hdr.count)-1];
}
memset((char *)btree, 0, sizeof(xfs_da_node_entry_t));
xfs_da_log_buf(state->args->trans, drop_blk->bp,
XFS_DA_LOGRANGE(node, btree, sizeof(*btree)));
- INT_MOD(node->hdr.count, ARCH_CONVERT, -1);
+ be16_add(&node->hdr.count, -1);
xfs_da_log_buf(state->args->trans, drop_blk->bp,
XFS_DA_LOGRANGE(node, &node->hdr, sizeof(node->hdr)));
@@ -1023,7 +1021,7 @@ xfs_da_node_remove(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk)
* Copy the last hash value from the block to propagate upwards.
*/
btree--;
- drop_blk->hashval = INT_GET(btree->hashval, ARCH_CONVERT);
+ drop_blk->hashval = be32_to_cpu(btree->hashval);
}
/*
@@ -1041,40 +1039,40 @@ xfs_da_node_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
drop_node = drop_blk->bp->data;
save_node = save_blk->bp->data;
- ASSERT(INT_GET(drop_node->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
- ASSERT(INT_GET(save_node->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(drop_node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(save_node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
tp = state->args->trans;
/*
* If the dying block has lower hashvals, then move all the
* elements in the remaining block up to make a hole.
*/
- if ((INT_GET(drop_node->btree[ 0 ].hashval, ARCH_CONVERT) < INT_GET(save_node->btree[ 0 ].hashval, ARCH_CONVERT)) ||
- (INT_GET(drop_node->btree[ INT_GET(drop_node->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT) <
- INT_GET(save_node->btree[ INT_GET(save_node->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT)))
+ if ((be32_to_cpu(drop_node->btree[0].hashval) < be32_to_cpu(save_node->btree[ 0 ].hashval)) ||
+ (be32_to_cpu(drop_node->btree[be16_to_cpu(drop_node->hdr.count)-1].hashval) <
+ be32_to_cpu(save_node->btree[be16_to_cpu(save_node->hdr.count)-1].hashval)))
{
- btree = &save_node->btree[ INT_GET(drop_node->hdr.count, ARCH_CONVERT) ];
- tmp = INT_GET(save_node->hdr.count, ARCH_CONVERT) * (uint)sizeof(xfs_da_node_entry_t);
+ btree = &save_node->btree[be16_to_cpu(drop_node->hdr.count)];
+ tmp = be16_to_cpu(save_node->hdr.count) * (uint)sizeof(xfs_da_node_entry_t);
memmove(btree, &save_node->btree[0], tmp);
btree = &save_node->btree[0];
xfs_da_log_buf(tp, save_blk->bp,
XFS_DA_LOGRANGE(save_node, btree,
- (INT_GET(save_node->hdr.count, ARCH_CONVERT) + INT_GET(drop_node->hdr.count, ARCH_CONVERT)) *
+ (be16_to_cpu(save_node->hdr.count) + be16_to_cpu(drop_node->hdr.count)) *
sizeof(xfs_da_node_entry_t)));
} else {
- btree = &save_node->btree[ INT_GET(save_node->hdr.count, ARCH_CONVERT) ];
+ btree = &save_node->btree[be16_to_cpu(save_node->hdr.count)];
xfs_da_log_buf(tp, save_blk->bp,
XFS_DA_LOGRANGE(save_node, btree,
- INT_GET(drop_node->hdr.count, ARCH_CONVERT) *
+ be16_to_cpu(drop_node->hdr.count) *
sizeof(xfs_da_node_entry_t)));
}
/*
* Move all the B-tree elements from drop_blk to save_blk.
*/
- tmp = INT_GET(drop_node->hdr.count, ARCH_CONVERT) * (uint)sizeof(xfs_da_node_entry_t);
+ tmp = be16_to_cpu(drop_node->hdr.count) * (uint)sizeof(xfs_da_node_entry_t);
memcpy(btree, &drop_node->btree[0], tmp);
- INT_MOD(save_node->hdr.count, ARCH_CONVERT, INT_GET(drop_node->hdr.count, ARCH_CONVERT));
+ be16_add(&save_node->hdr.count, be16_to_cpu(drop_node->hdr.count));
xfs_da_log_buf(tp, save_blk->bp,
XFS_DA_LOGRANGE(save_node, &save_node->hdr,
@@ -1083,7 +1081,7 @@ xfs_da_node_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
/*
* Save the last hashval in the remaining block for upward propagation.
*/
- save_blk->hashval = INT_GET(save_node->btree[ INT_GET(save_node->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT);
+ save_blk->hashval = be32_to_cpu(save_node->btree[be16_to_cpu(save_node->hdr.count)-1].hashval);
}
/*========================================================================
@@ -1138,46 +1136,46 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result)
return(error);
}
curr = blk->bp->data;
- ASSERT(INT_GET(curr->magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC ||
- INT_GET(curr->magic, ARCH_CONVERT) == XFS_DIRX_LEAF_MAGIC(state->mp) ||
- INT_GET(curr->magic, ARCH_CONVERT) == XFS_ATTR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(curr->magic) == XFS_DA_NODE_MAGIC ||
+ be16_to_cpu(curr->magic) == XFS_DIRX_LEAF_MAGIC(state->mp) ||
+ be16_to_cpu(curr->magic) == XFS_ATTR_LEAF_MAGIC);
/*
* Search an intermediate node for a match.
*/
- blk->magic = INT_GET(curr->magic, ARCH_CONVERT);
- if (INT_GET(curr->magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC) {
+ blk->magic = be16_to_cpu(curr->magic);
+ if (blk->magic == XFS_DA_NODE_MAGIC) {
node = blk->bp->data;
- blk->hashval = INT_GET(node->btree[ INT_GET(node->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT);
+ blk->hashval = be32_to_cpu(node->btree[be16_to_cpu(node->hdr.count)-1].hashval);
/*
* Binary search. (note: small blocks will skip loop)
*/
- max = INT_GET(node->hdr.count, ARCH_CONVERT);
+ max = be16_to_cpu(node->hdr.count);
probe = span = max / 2;
hashval = args->hashval;
for (btree = &node->btree[probe]; span > 4;
btree = &node->btree[probe]) {
span /= 2;
- if (INT_GET(btree->hashval, ARCH_CONVERT) < hashval)
+ if (be32_to_cpu(btree->hashval) < hashval)
probe += span;
- else if (INT_GET(btree->hashval, ARCH_CONVERT) > hashval)
+ else if (be32_to_cpu(btree->hashval) > hashval)
probe -= span;
else
break;
}
ASSERT((probe >= 0) && (probe < max));
- ASSERT((span <= 4) || (INT_GET(btree->hashval, ARCH_CONVERT) == hashval));
+ ASSERT((span <= 4) || (be32_to_cpu(btree->hashval) == hashval));
/*
* Since we may have duplicate hashval's, find the first
* matching hashval in the node.
*/
- while ((probe > 0) && (INT_GET(btree->hashval, ARCH_CONVERT) >= hashval)) {
+ while ((probe > 0) && (be32_to_cpu(btree->hashval) >= hashval)) {
btree--;
probe--;
}
- while ((probe < max) && (INT_GET(btree->hashval, ARCH_CONVERT) < hashval)) {
+ while ((probe < max) && (be32_to_cpu(btree->hashval) < hashval)) {
btree++;
probe++;
}
@@ -1187,21 +1185,21 @@ xfs_da_node_lookup_int(xfs_da_state_t *state, int *result)
*/
if (probe == max) {
blk->index = max-1;
- blkno = INT_GET(node->btree[ max-1 ].before, ARCH_CONVERT);
+ blkno = be32_to_cpu(node->btree[max-1].before);
} else {
blk->index = probe;
- blkno = INT_GET(btree->before, ARCH_CONVERT);
+ blkno = be32_to_cpu(btree->before);
}
}
- else if (INT_GET(curr->magic, ARCH_CONVERT) == XFS_ATTR_LEAF_MAGIC) {
+ else if (be16_to_cpu(curr->magic) == XFS_ATTR_LEAF_MAGIC) {
blk->hashval = xfs_attr_leaf_lasthash(blk->bp, NULL);
break;
}
- else if (INT_GET(curr->magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC) {
+ else if (be16_to_cpu(curr->magic) == XFS_DIR_LEAF_MAGIC) {
blk->hashval = xfs_dir_leaf_lasthash(blk->bp, NULL);
break;
}
- else if (INT_GET(curr->magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC) {
+ else if (be16_to_cpu(curr->magic) == XFS_DIR2_LEAFN_MAGIC) {
blk->hashval = xfs_dir2_leafn_lasthash(blk->bp, NULL);
break;
}
@@ -1274,8 +1272,8 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
ASSERT(old_blk->magic == XFS_DA_NODE_MAGIC ||
old_blk->magic == XFS_DIRX_LEAF_MAGIC(state->mp) ||
old_blk->magic == XFS_ATTR_LEAF_MAGIC);
- ASSERT(old_blk->magic == INT_GET(old_info->magic, ARCH_CONVERT));
- ASSERT(new_blk->magic == INT_GET(new_info->magic, ARCH_CONVERT));
+ ASSERT(old_blk->magic == be16_to_cpu(old_info->magic));
+ ASSERT(new_blk->magic == be16_to_cpu(new_info->magic));
ASSERT(old_blk->magic == new_blk->magic);
switch (old_blk->magic) {
@@ -1302,47 +1300,44 @@ xfs_da_blk_link(xfs_da_state_t *state, xfs_da_state_blk_t *old_blk,
/*
* Link new block in before existing block.
*/
- INT_SET(new_info->forw, ARCH_CONVERT, old_blk->blkno);
- new_info->back = old_info->back; /* INT_: direct copy */
- if (INT_GET(old_info->back, ARCH_CONVERT)) {
+ new_info->forw = cpu_to_be32(old_blk->blkno);
+ new_info->back = old_info->back;
+ if (old_info->back) {
error = xfs_da_read_buf(args->trans, args->dp,
- INT_GET(old_info->back,
- ARCH_CONVERT), -1, &bp,
- args->whichfork);
+ be32_to_cpu(old_info->back),
+ -1, &bp, args->whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
tmp_info = bp->data;
- ASSERT(INT_GET(tmp_info->magic, ARCH_CONVERT) == INT_GET(old_info->magic, ARCH_CONVERT));
- ASSERT(INT_GET(tmp_info->forw, ARCH_CONVERT) == old_blk->blkno);
- INT_SET(tmp_info->forw, ARCH_CONVERT, new_blk->blkno);
+ ASSERT(be16_to_cpu(tmp_info->magic) == be16_to_cpu(old_info->magic));
+ ASSERT(be32_to_cpu(tmp_info->forw) == old_blk->blkno);
+ tmp_info->forw = cpu_to_be32(new_blk->blkno);
xfs_da_log_buf(args->trans, bp, 0, sizeof(*tmp_info)-1);
xfs_da_buf_done(bp);
}
- INT_SET(old_info->back, ARCH_CONVERT, new_blk->blkno);
+ old_info->back = cpu_to_be32(new_blk->blkno);
} else {
/*
* Link new block in after existing block.
*/
- new_info->forw = old_info->forw; /* INT_: direct copy */
- INT_SET(new_info->back, ARCH_CONVERT, old_blk->blkno);
- if (INT_GET(old_info->forw, ARCH_CONVERT)) {
+ new_info->forw = old_info->forw;
+ new_info->back = cpu_to_be32(old_blk->blkno);
+ if (old_info->forw) {
error = xfs_da_read_buf(args->trans, args->dp,
- INT_GET(old_info->forw, ARCH_CONVERT), -1, &bp,
- args->whichfork);
+ be32_to_cpu(old_info->forw),
+ -1, &bp, args->whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
tmp_info = bp->data;
- ASSERT(INT_GET(tmp_info->magic, ARCH_CONVERT)
- == INT_GET(old_info->magic, ARCH_CONVERT));
- ASSERT(INT_GET(tmp_info->back, ARCH_CONVERT)
- == old_blk->blkno);
- INT_SET(tmp_info->back, ARCH_CONVERT, new_blk->blkno);
+ ASSERT(tmp_info->magic == old_info->magic);
+ ASSERT(be32_to_cpu(tmp_info->back) == old_blk->blkno);
+ tmp_info->back = cpu_to_be32(new_blk->blkno);
xfs_da_log_buf(args->trans, bp, 0, sizeof(*tmp_info)-1);
xfs_da_buf_done(bp);
}
- INT_SET(old_info->forw, ARCH_CONVERT, new_blk->blkno);
+ old_info->forw = cpu_to_be32(new_blk->blkno);
}
xfs_da_log_buf(args->trans, old_blk->bp, 0, sizeof(*tmp_info) - 1);
@@ -1360,13 +1355,13 @@ xfs_da_node_order(xfs_dabuf_t *node1_bp, xfs_dabuf_t *node2_bp)
node1 = node1_bp->data;
node2 = node2_bp->data;
- ASSERT((INT_GET(node1->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC) &&
- (INT_GET(node2->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC));
- if ((INT_GET(node1->hdr.count, ARCH_CONVERT) > 0) && (INT_GET(node2->hdr.count, ARCH_CONVERT) > 0) &&
- ((INT_GET(node2->btree[ 0 ].hashval, ARCH_CONVERT) <
- INT_GET(node1->btree[ 0 ].hashval, ARCH_CONVERT)) ||
- (INT_GET(node2->btree[ INT_GET(node2->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT) <
- INT_GET(node1->btree[ INT_GET(node1->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT)))) {
+ ASSERT((be16_to_cpu(node1->hdr.info.magic) == XFS_DA_NODE_MAGIC) &&
+ (be16_to_cpu(node2->hdr.info.magic) == XFS_DA_NODE_MAGIC));
+ if ((be16_to_cpu(node1->hdr.count) > 0) && (be16_to_cpu(node2->hdr.count) > 0) &&
+ ((be32_to_cpu(node2->btree[0].hashval) <
+ be32_to_cpu(node1->btree[0].hashval)) ||
+ (be32_to_cpu(node2->btree[be16_to_cpu(node2->hdr.count)-1].hashval) <
+ be32_to_cpu(node1->btree[be16_to_cpu(node1->hdr.count)-1].hashval)))) {
return(1);
}
return(0);
@@ -1381,12 +1376,12 @@ xfs_da_node_lasthash(xfs_dabuf_t *bp, int *count)
xfs_da_intnode_t *node;
node = bp->data;
- ASSERT(INT_GET(node->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
if (count)
- *count = INT_GET(node->hdr.count, ARCH_CONVERT);
+ *count = be16_to_cpu(node->hdr.count);
if (!node->hdr.count)
return(0);
- return(INT_GET(node->btree[ INT_GET(node->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT));
+ return be32_to_cpu(node->btree[be16_to_cpu(node->hdr.count)-1].hashval);
}
/*
@@ -1411,50 +1406,47 @@ xfs_da_blk_unlink(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
ASSERT(save_blk->magic == XFS_DA_NODE_MAGIC ||
save_blk->magic == XFS_DIRX_LEAF_MAGIC(state->mp) ||
save_blk->magic == XFS_ATTR_LEAF_MAGIC);
- ASSERT(save_blk->magic == INT_GET(save_info->magic, ARCH_CONVERT));
- ASSERT(drop_blk->magic == INT_GET(drop_info->magic, ARCH_CONVERT));
+ ASSERT(save_blk->magic == be16_to_cpu(save_info->magic));
+ ASSERT(drop_blk->magic == be16_to_cpu(drop_info->magic));
ASSERT(save_blk->magic == drop_blk->magic);
- ASSERT((INT_GET(save_info->forw, ARCH_CONVERT) == drop_blk->blkno) ||
- (INT_GET(save_info->back, ARCH_CONVERT) == drop_blk->blkno));
- ASSERT((INT_GET(drop_info->forw, ARCH_CONVERT) == save_blk->blkno) ||
- (INT_GET(drop_info->back, ARCH_CONVERT) == save_blk->blkno));
+ ASSERT((be32_to_cpu(save_info->forw) == drop_blk->blkno) ||
+ (be32_to_cpu(save_info->back) == drop_blk->blkno));
+ ASSERT((be32_to_cpu(drop_info->forw) == save_blk->blkno) ||
+ (be32_to_cpu(drop_info->back) == save_blk->blkno));
/*
* Unlink the leaf block from the doubly linked chain of leaves.
*/
- if (INT_GET(save_info->back, ARCH_CONVERT) == drop_blk->blkno) {
- save_info->back = drop_info->back; /* INT_: direct copy */
- if (INT_GET(drop_info->back, ARCH_CONVERT)) {
+ if (be32_to_cpu(save_info->back) == drop_blk->blkno) {
+ save_info->back = drop_info->back;
+ if (drop_info->back) {
error = xfs_da_read_buf(args->trans, args->dp,
- INT_GET(drop_info->back,
- ARCH_CONVERT), -1, &bp,
- args->whichfork);
+ be32_to_cpu(drop_info->back),
+ -1, &bp, args->whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
tmp_info = bp->data;
- ASSERT(INT_GET(tmp_info->magic, ARCH_CONVERT) == INT_GET(save_info->magic, ARCH_CONVERT));
- ASSERT(INT_GET(tmp_info->forw, ARCH_CONVERT) == drop_blk->blkno);
- INT_SET(tmp_info->forw, ARCH_CONVERT, save_blk->blkno);
+ ASSERT(tmp_info->magic == save_info->magic);
+ ASSERT(be32_to_cpu(tmp_info->forw) == drop_blk->blkno);
+ tmp_info->forw = cpu_to_be32(save_blk->blkno);
xfs_da_log_buf(args->trans, bp, 0,
sizeof(*tmp_info) - 1);
xfs_da_buf_done(bp);
}
} else {
- save_info->forw = drop_info->forw; /* INT_: direct copy */
- if (INT_GET(drop_info->forw, ARCH_CONVERT)) {
+ save_info->forw = drop_info->forw;
+ if (drop_info->forw) {
error = xfs_da_read_buf(args->trans, args->dp,
- INT_GET(drop_info->forw, ARCH_CONVERT), -1, &bp,
- args->whichfork);
+ be32_to_cpu(drop_info->forw),
+ -1, &bp, args->whichfork);
if (error)
return(error);
ASSERT(bp != NULL);
tmp_info = bp->data;
- ASSERT(INT_GET(tmp_info->magic, ARCH_CONVERT)
- == INT_GET(save_info->magic, ARCH_CONVERT));
- ASSERT(INT_GET(tmp_info->back, ARCH_CONVERT)
- == drop_blk->blkno);
- INT_SET(tmp_info->back, ARCH_CONVERT, save_blk->blkno);
+ ASSERT(tmp_info->magic == save_info->magic);
+ ASSERT(be32_to_cpu(tmp_info->back) == drop_blk->blkno);
+ tmp_info->back = cpu_to_be32(save_blk->blkno);
xfs_da_log_buf(args->trans, bp, 0,
sizeof(*tmp_info) - 1);
xfs_da_buf_done(bp);
@@ -1497,14 +1489,14 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
for (blk = &path->blk[level]; level >= 0; blk--, level--) {
ASSERT(blk->bp != NULL);
node = blk->bp->data;
- ASSERT(INT_GET(node->hdr.info.magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
- if (forward && (blk->index < INT_GET(node->hdr.count, ARCH_CONVERT)-1)) {
+ ASSERT(be16_to_cpu(node->hdr.info.magic) == XFS_DA_NODE_MAGIC);
+ if (forward && (blk->index < be16_to_cpu(node->hdr.count)-1)) {
blk->index++;
- blkno = INT_GET(node->btree[ blk->index ].before, ARCH_CONVERT);
+ blkno = be32_to_cpu(node->btree[blk->index].before);
break;
} else if (!forward && (blk->index > 0)) {
blk->index--;
- blkno = INT_GET(node->btree[ blk->index ].before, ARCH_CONVERT);
+ blkno = be32_to_cpu(node->btree[blk->index].before);
break;
}
}
@@ -1536,18 +1528,18 @@ xfs_da_path_shift(xfs_da_state_t *state, xfs_da_state_path_t *path,
return(error);
ASSERT(blk->bp != NULL);
info = blk->bp->data;
- ASSERT(INT_GET(info->magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC ||
- INT_GET(info->magic, ARCH_CONVERT) == XFS_DIRX_LEAF_MAGIC(state->mp) ||
- INT_GET(info->magic, ARCH_CONVERT) == XFS_ATTR_LEAF_MAGIC);
- blk->magic = INT_GET(info->magic, ARCH_CONVERT);
- if (INT_GET(info->magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC) {
+ ASSERT(be16_to_cpu(info->magic) == XFS_DA_NODE_MAGIC ||
+ be16_to_cpu(info->magic) == XFS_DIRX_LEAF_MAGIC(state->mp) ||
+ be16_to_cpu(info->magic) == XFS_ATTR_LEAF_MAGIC);
+ blk->magic = be16_to_cpu(info->magic);
+ if (blk->magic == XFS_DA_NODE_MAGIC) {
node = (xfs_da_intnode_t *)info;
- blk->hashval = INT_GET(node->btree[ INT_GET(node->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT);
+ blk->hashval = be32_to_cpu(node->btree[be16_to_cpu(node->hdr.count)-1].hashval);
if (forward)
blk->index = 0;
else
- blk->index = INT_GET(node->hdr.count, ARCH_CONVERT)-1;
- blkno = INT_GET(node->btree[ blk->index ].before, ARCH_CONVERT);
+ blk->index = be16_to_cpu(node->hdr.count)-1;
+ blkno = be32_to_cpu(node->btree[blk->index].before);
} else {
ASSERT(level == path->active-1);
blk->index = 0;
@@ -1788,40 +1780,40 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
/*
* Get values from the moved block.
*/
- if (INT_GET(dead_info->magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC) {
+ if (be16_to_cpu(dead_info->magic) == XFS_DIR_LEAF_MAGIC) {
ASSERT(XFS_DIR_IS_V1(mp));
dead_leaf = (xfs_dir_leafblock_t *)dead_info;
dead_level = 0;
dead_hash =
INT_GET(dead_leaf->entries[INT_GET(dead_leaf->hdr.count, ARCH_CONVERT) - 1].hashval, ARCH_CONVERT);
- } else if (INT_GET(dead_info->magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC) {
+ } else if (be16_to_cpu(dead_info->magic) == XFS_DIR2_LEAFN_MAGIC) {
ASSERT(XFS_DIR_IS_V2(mp));
dead_leaf2 = (xfs_dir2_leaf_t *)dead_info;
dead_level = 0;
- dead_hash = INT_GET(dead_leaf2->ents[INT_GET(dead_leaf2->hdr.count, ARCH_CONVERT) - 1].hashval, ARCH_CONVERT);
+ dead_hash = be32_to_cpu(dead_leaf2->ents[be16_to_cpu(dead_leaf2->hdr.count) - 1].hashval);
} else {
- ASSERT(INT_GET(dead_info->magic, ARCH_CONVERT) == XFS_DA_NODE_MAGIC);
+ ASSERT(be16_to_cpu(dead_info->magic) == XFS_DA_NODE_MAGIC);
dead_node = (xfs_da_intnode_t *)dead_info;
- dead_level = INT_GET(dead_node->hdr.level, ARCH_CONVERT);
- dead_hash = INT_GET(dead_node->btree[INT_GET(dead_node->hdr.count, ARCH_CONVERT) - 1].hashval, ARCH_CONVERT);
+ dead_level = be16_to_cpu(dead_node->hdr.level);
+ dead_hash = be32_to_cpu(dead_node->btree[be16_to_cpu(dead_node->hdr.count) - 1].hashval);
}
sib_buf = par_buf = NULL;
/*
* If the moved block has a left sibling, fix up the pointers.
*/
- if ((sib_blkno = INT_GET(dead_info->back, ARCH_CONVERT))) {
+ if ((sib_blkno = be32_to_cpu(dead_info->back))) {
if ((error = xfs_da_read_buf(tp, ip, sib_blkno, -1, &sib_buf, w)))
goto done;
sib_info = sib_buf->data;
if (unlikely(
- INT_GET(sib_info->forw, ARCH_CONVERT) != last_blkno ||
- INT_GET(sib_info->magic, ARCH_CONVERT) != INT_GET(dead_info->magic, ARCH_CONVERT))) {
+ be32_to_cpu(sib_info->forw) != last_blkno ||
+ sib_info->magic != dead_info->magic)) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(2)",
XFS_ERRLEVEL_LOW, mp);
error = XFS_ERROR(EFSCORRUPTED);
goto done;
}
- INT_SET(sib_info->forw, ARCH_CONVERT, dead_blkno);
+ sib_info->forw = cpu_to_be32(dead_blkno);
xfs_da_log_buf(tp, sib_buf,
XFS_DA_LOGRANGE(sib_info, &sib_info->forw,
sizeof(sib_info->forw)));
@@ -1831,20 +1823,19 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
/*
* If the moved block has a right sibling, fix up the pointers.
*/
- if ((sib_blkno = INT_GET(dead_info->forw, ARCH_CONVERT))) {
+ if ((sib_blkno = be32_to_cpu(dead_info->forw))) {
if ((error = xfs_da_read_buf(tp, ip, sib_blkno, -1, &sib_buf, w)))
goto done;
sib_info = sib_buf->data;
if (unlikely(
- INT_GET(sib_info->back, ARCH_CONVERT) != last_blkno
- || INT_GET(sib_info->magic, ARCH_CONVERT)
- != INT_GET(dead_info->magic, ARCH_CONVERT))) {
+ be32_to_cpu(sib_info->back) != last_blkno ||
+ sib_info->magic != dead_info->magic)) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(3)",
XFS_ERRLEVEL_LOW, mp);
error = XFS_ERROR(EFSCORRUPTED);
goto done;
}
- INT_SET(sib_info->back, ARCH_CONVERT, dead_blkno);
+ sib_info->back = cpu_to_be32(dead_blkno);
xfs_da_log_buf(tp, sib_buf,
XFS_DA_LOGRANGE(sib_info, &sib_info->back,
sizeof(sib_info->back)));
@@ -1861,26 +1852,26 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
goto done;
par_node = par_buf->data;
if (unlikely(
- INT_GET(par_node->hdr.info.magic, ARCH_CONVERT) != XFS_DA_NODE_MAGIC ||
- (level >= 0 && level != INT_GET(par_node->hdr.level, ARCH_CONVERT) + 1))) {
+ be16_to_cpu(par_node->hdr.info.magic) != XFS_DA_NODE_MAGIC ||
+ (level >= 0 && level != be16_to_cpu(par_node->hdr.level) + 1))) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(4)",
XFS_ERRLEVEL_LOW, mp);
error = XFS_ERROR(EFSCORRUPTED);
goto done;
}
- level = INT_GET(par_node->hdr.level, ARCH_CONVERT);
+ level = be16_to_cpu(par_node->hdr.level);
for (entno = 0;
- entno < INT_GET(par_node->hdr.count, ARCH_CONVERT) &&
- INT_GET(par_node->btree[entno].hashval, ARCH_CONVERT) < dead_hash;
+ entno < be16_to_cpu(par_node->hdr.count) &&
+ be32_to_cpu(par_node->btree[entno].hashval) < dead_hash;
entno++)
continue;
- if (unlikely(entno == INT_GET(par_node->hdr.count, ARCH_CONVERT))) {
+ if (unlikely(entno == be16_to_cpu(par_node->hdr.count))) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(5)",
XFS_ERRLEVEL_LOW, mp);
error = XFS_ERROR(EFSCORRUPTED);
goto done;
}
- par_blkno = INT_GET(par_node->btree[entno].before, ARCH_CONVERT);
+ par_blkno = be32_to_cpu(par_node->btree[entno].before);
if (level == dead_level + 1)
break;
xfs_da_brelse(tp, par_buf);
@@ -1892,13 +1883,13 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
*/
for (;;) {
for (;
- entno < INT_GET(par_node->hdr.count, ARCH_CONVERT) &&
- INT_GET(par_node->btree[entno].before, ARCH_CONVERT) != last_blkno;
+ entno < be16_to_cpu(par_node->hdr.count) &&
+ be32_to_cpu(par_node->btree[entno].before) != last_blkno;
entno++)
continue;
- if (entno < INT_GET(par_node->hdr.count, ARCH_CONVERT))
+ if (entno < be16_to_cpu(par_node->hdr.count))
break;
- par_blkno = INT_GET(par_node->hdr.info.forw, ARCH_CONVERT);
+ par_blkno = be32_to_cpu(par_node->hdr.info.forw);
xfs_da_brelse(tp, par_buf);
par_buf = NULL;
if (unlikely(par_blkno == 0)) {
@@ -1911,8 +1902,8 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
goto done;
par_node = par_buf->data;
if (unlikely(
- INT_GET(par_node->hdr.level, ARCH_CONVERT) != level ||
- INT_GET(par_node->hdr.info.magic, ARCH_CONVERT) != XFS_DA_NODE_MAGIC)) {
+ be16_to_cpu(par_node->hdr.level) != level ||
+ be16_to_cpu(par_node->hdr.info.magic) != XFS_DA_NODE_MAGIC)) {
XFS_ERROR_REPORT("xfs_da_swap_lastblock(7)",
XFS_ERRLEVEL_LOW, mp);
error = XFS_ERROR(EFSCORRUPTED);
@@ -1923,7 +1914,7 @@ xfs_da_swap_lastblock(xfs_da_args_t *args, xfs_dablk_t *dead_blknop,
/*
* Update the parent entry pointing to the moved block.
*/
- INT_SET(par_node->btree[entno].before, ARCH_CONVERT, dead_blkno);
+ par_node->btree[entno].before = cpu_to_be32(dead_blkno);
xfs_da_log_buf(tp, par_buf,
XFS_DA_LOGRANGE(par_node, &par_node->btree[entno].before,
sizeof(par_node->btree[entno].before)));
@@ -2203,8 +2194,8 @@ xfs_da_do_buf(
info = rbp->data;
data = rbp->data;
free = rbp->data;
- magic = INT_GET(info->magic, ARCH_CONVERT);
- magic1 = INT_GET(data->hdr.magic, ARCH_CONVERT);
+ magic = be16_to_cpu(info->magic);
+ magic1 = be32_to_cpu(data->hdr.magic);
if (unlikely(
XFS_TEST_ERROR((magic != XFS_DA_NODE_MAGIC) &&
(magic != XFS_DIR_LEAF_MAGIC) &&
@@ -2213,7 +2204,7 @@ xfs_da_do_buf(
(magic != XFS_DIR2_LEAFN_MAGIC) &&
(magic1 != XFS_DIR2_BLOCK_MAGIC) &&
(magic1 != XFS_DIR2_DATA_MAGIC) &&
- (INT_GET(free->hdr.magic, ARCH_CONVERT) != XFS_DIR2_FREE_MAGIC),
+ (be32_to_cpu(free->hdr.magic) != XFS_DIR2_FREE_MAGIC),
mp, XFS_ERRTAG_DA_READ_BUF,
XFS_RANDOM_DA_READ_BUF))) {
xfs_buftrace("DA READ ERROR", rbp->bps[0]);
diff --git a/fs/xfs/xfs_da_btree.h b/fs/xfs/xfs_da_btree.h
index 41352113721a..243a730d5ec8 100644
--- a/fs/xfs/xfs_da_btree.h
+++ b/fs/xfs/xfs_da_btree.h
@@ -45,10 +45,10 @@ struct zone;
(XFS_DIR_IS_V1(mp) ? XFS_DIR_LEAF_MAGIC : XFS_DIR2_LEAFN_MAGIC)
typedef struct xfs_da_blkinfo {
- xfs_dablk_t forw; /* previous block in list */
- xfs_dablk_t back; /* following block in list */
- __uint16_t magic; /* validity check on block */
- __uint16_t pad; /* unused */
+ __be32 forw; /* previous block in list */
+ __be32 back; /* following block in list */
+ __be16 magic; /* validity check on block */
+ __be16 pad; /* unused */
} xfs_da_blkinfo_t;
/*
@@ -65,12 +65,12 @@ typedef struct xfs_da_blkinfo {
typedef struct xfs_da_intnode {
struct xfs_da_node_hdr { /* constant-structure header block */
xfs_da_blkinfo_t info; /* block type, links, etc. */
- __uint16_t count; /* count of active entries */
- __uint16_t level; /* level above leaves (leaf == 0) */
+ __be16 count; /* count of active entries */
+ __be16 level; /* level above leaves (leaf == 0) */
} hdr;
struct xfs_da_node_entry {
- xfs_dahash_t hashval; /* hash value for this descendant */
- xfs_dablk_t before; /* Btree block before this key */
+ __be32 hashval; /* hash value for this descendant */
+ __be32 before; /* Btree block before this key */
} btree[1]; /* variable sized array of keys */
} xfs_da_intnode_t;
typedef struct xfs_da_node_hdr xfs_da_node_hdr_t;
diff --git a/fs/xfs/xfs_dfrag.c b/fs/xfs/xfs_dfrag.c
index c6191d00ad27..4968a6358e61 100644
--- a/fs/xfs/xfs_dfrag.c
+++ b/fs/xfs/xfs_dfrag.c
@@ -83,7 +83,7 @@ xfs_swapext(
/* Pull information for the target fd */
if (((fp = fget((int)sxp->sx_fdtarget)) == NULL) ||
- ((vp = LINVFS_GET_VP(fp->f_dentry->d_inode)) == NULL)) {
+ ((vp = vn_from_inode(fp->f_dentry->d_inode)) == NULL)) {
error = XFS_ERROR(EINVAL);
goto error0;
}
@@ -95,7 +95,7 @@ xfs_swapext(
}
if (((tfp = fget((int)sxp->sx_fdtmp)) == NULL) ||
- ((tvp = LINVFS_GET_VP(tfp->f_dentry->d_inode)) == NULL)) {
+ ((tvp = vn_from_inode(tfp->f_dentry->d_inode)) == NULL)) {
error = XFS_ERROR(EINVAL);
goto error0;
}
diff --git a/fs/xfs/xfs_dir.c b/fs/xfs/xfs_dir.c
index bb87d2a700a9..9cc702a839a3 100644
--- a/fs/xfs/xfs_dir.c
+++ b/fs/xfs/xfs_dir.c
@@ -634,7 +634,7 @@ xfs_dir_leaf_removename(xfs_da_args_t *args, int *count, int *totallen)
return(retval);
ASSERT(bp != NULL);
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
retval = xfs_dir_leaf_lookup_int(bp, args, &index);
if (retval == EEXIST) {
(void)xfs_dir_leaf_remove(args->trans, bp, index);
@@ -912,7 +912,7 @@ xfs_dir_node_getdents(xfs_trans_t *trans, xfs_inode_t *dp, uio_t *uio,
return(error);
if (bp)
leaf = bp->data;
- if (bp && INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) != XFS_DIR_LEAF_MAGIC) {
+ if (bp && be16_to_cpu(leaf->hdr.info.magic) != XFS_DIR_LEAF_MAGIC) {
xfs_dir_trace_g_dub("node: block not a leaf",
dp, uio, bno);
xfs_da_brelse(trans, bp);
@@ -949,17 +949,17 @@ xfs_dir_node_getdents(xfs_trans_t *trans, xfs_inode_t *dp, uio_t *uio,
if (bp == NULL)
return(XFS_ERROR(EFSCORRUPTED));
node = bp->data;
- if (INT_GET(node->hdr.info.magic, ARCH_CONVERT) != XFS_DA_NODE_MAGIC)
+ if (be16_to_cpu(node->hdr.info.magic) != XFS_DA_NODE_MAGIC)
break;
btree = &node->btree[0];
xfs_dir_trace_g_dun("node: node detail", dp, uio, node);
- for (i = 0; i < INT_GET(node->hdr.count, ARCH_CONVERT); btree++, i++) {
- if (INT_GET(btree->hashval, ARCH_CONVERT) >= cookhash) {
- bno = INT_GET(btree->before, ARCH_CONVERT);
+ for (i = 0; i < be16_to_cpu(node->hdr.count); btree++, i++) {
+ if (be32_to_cpu(btree->hashval) >= cookhash) {
+ bno = be32_to_cpu(btree->before);
break;
}
}
- if (i == INT_GET(node->hdr.count, ARCH_CONVERT)) {
+ if (i == be16_to_cpu(node->hdr.count)) {
xfs_da_brelse(trans, bp);
xfs_dir_trace_g_du("node: hash beyond EOF",
dp, uio);
@@ -982,7 +982,7 @@ xfs_dir_node_getdents(xfs_trans_t *trans, xfs_inode_t *dp, uio_t *uio,
*/
for (;;) {
leaf = bp->data;
- if (unlikely(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) != XFS_DIR_LEAF_MAGIC)) {
+ if (unlikely(be16_to_cpu(leaf->hdr.info.magic) != XFS_DIR_LEAF_MAGIC)) {
xfs_dir_trace_g_dul("node: not a leaf", dp, uio, leaf);
xfs_da_brelse(trans, bp);
XFS_CORRUPTION_ERROR("xfs_dir_node_getdents(1)",
@@ -990,7 +990,7 @@ xfs_dir_node_getdents(xfs_trans_t *trans, xfs_inode_t *dp, uio_t *uio,
return XFS_ERROR(EFSCORRUPTED);
}
xfs_dir_trace_g_dul("node: leaf detail", dp, uio, leaf);
- if ((nextbno = INT_GET(leaf->hdr.info.forw, ARCH_CONVERT))) {
+ if ((nextbno = be32_to_cpu(leaf->hdr.info.forw))) {
nextda = xfs_da_reada_buf(trans, dp, nextbno,
XFS_DATA_FORK);
} else
@@ -1118,21 +1118,20 @@ void
xfs_dir_trace_g_dun(char *where, xfs_inode_t *dp, uio_t *uio,
xfs_da_intnode_t *node)
{
- int last = INT_GET(node->hdr.count, ARCH_CONVERT) - 1;
+ int last = be16_to_cpu(node->hdr.count) - 1;
xfs_dir_trace_enter(XFS_DIR_KTRACE_G_DUN, where,
(void *)dp, (void *)dp->i_mount,
(void *)((unsigned long)(uio->uio_offset >> 32)),
(void *)((unsigned long)(uio->uio_offset & 0xFFFFFFFF)),
(void *)(unsigned long)uio->uio_resid,
+ (void *)(unsigned long)be32_to_cpu(node->hdr.info.forw),
(void *)(unsigned long)
- INT_GET(node->hdr.info.forw, ARCH_CONVERT),
+ be16_to_cpu(node->hdr.count),
(void *)(unsigned long)
- INT_GET(node->hdr.count, ARCH_CONVERT),
+ be32_to_cpu(node->btree[0].hashval),
(void *)(unsigned long)
- INT_GET(node->btree[0].hashval, ARCH_CONVERT),
- (void *)(unsigned long)
- INT_GET(node->btree[last].hashval, ARCH_CONVERT),
+ be32_to_cpu(node->btree[last].hashval),
NULL, NULL, NULL);
}
@@ -1150,8 +1149,7 @@ xfs_dir_trace_g_dul(char *where, xfs_inode_t *dp, uio_t *uio,
(void *)((unsigned long)(uio->uio_offset >> 32)),
(void *)((unsigned long)(uio->uio_offset & 0xFFFFFFFF)),
(void *)(unsigned long)uio->uio_resid,
- (void *)(unsigned long)
- INT_GET(leaf->hdr.info.forw, ARCH_CONVERT),
+ (void *)(unsigned long)be32_to_cpu(leaf->hdr.info.forw),
(void *)(unsigned long)
INT_GET(leaf->hdr.count, ARCH_CONVERT),
(void *)(unsigned long)
diff --git a/fs/xfs/xfs_dir2.h b/fs/xfs/xfs_dir2.h
index 3158f5dc431f..7dd364b1e038 100644
--- a/fs/xfs/xfs_dir2.h
+++ b/fs/xfs/xfs_dir2.h
@@ -55,16 +55,16 @@ typedef __uint32_t xfs_dir2_db_t;
/*
* Byte offset in a directory.
*/
-typedef xfs_off_t xfs_dir2_off_t;
+typedef xfs_off_t xfs_dir2_off_t;
/*
* For getdents, argument struct for put routines.
*/
typedef int (*xfs_dir2_put_t)(struct xfs_dir2_put_args *pa);
typedef struct xfs_dir2_put_args {
- xfs_off_t cook; /* cookie of (next) entry */
+ xfs_off_t cook; /* cookie of (next) entry */
xfs_intino_t ino; /* inode number */
- struct xfs_dirent *dbp; /* buffer pointer */
+ xfs_dirent_t *dbp; /* buffer pointer */
char *name; /* directory entry name */
int namelen; /* length of name */
int done; /* output: set if value was stored */
@@ -75,18 +75,13 @@ typedef struct xfs_dir2_put_args {
/*
* Other interfaces used by the rest of the dir v2 code.
*/
-extern int
- xfs_dir2_grow_inode(struct xfs_da_args *args, int space,
- xfs_dir2_db_t *dbp);
-
-extern int
- xfs_dir2_isblock(struct xfs_trans *tp, struct xfs_inode *dp, int *vp);
-
-extern int
- xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp, int *vp);
-
-extern int
- xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db,
- struct xfs_dabuf *bp);
+extern int xfs_dir2_grow_inode(struct xfs_da_args *args, int space,
+ xfs_dir2_db_t *dbp);
+extern int xfs_dir2_isblock(struct xfs_trans *tp, struct xfs_inode *dp,
+ int *vp);
+extern int xfs_dir2_isleaf(struct xfs_trans *tp, struct xfs_inode *dp,
+ int *vp);
+extern int xfs_dir2_shrink_inode(struct xfs_da_args *args, xfs_dir2_db_t db,
+ struct xfs_dabuf *bp);
#endif /* __XFS_DIR2_H__ */
diff --git a/fs/xfs/xfs_dir2_block.c b/fs/xfs/xfs_dir2_block.c
index 31bc99faa704..bd5cee6aa51a 100644
--- a/fs/xfs/xfs_dir2_block.c
+++ b/fs/xfs/xfs_dir2_block.c
@@ -81,7 +81,7 @@ xfs_dir2_block_addname(
xfs_mount_t *mp; /* filesystem mount point */
int needlog; /* need to log header */
int needscan; /* need to rescan freespace */
- xfs_dir2_data_off_t *tagp; /* pointer to tag value */
+ __be16 *tagp; /* pointer to tag value */
xfs_trans_t *tp; /* transaction structure */
xfs_dir2_trace_args("block_addname", args);
@@ -100,8 +100,7 @@ xfs_dir2_block_addname(
/*
* Check the magic number, corrupted if wrong.
*/
- if (unlikely(INT_GET(block->hdr.magic, ARCH_CONVERT)
- != XFS_DIR2_BLOCK_MAGIC)) {
+ if (unlikely(be32_to_cpu(block->hdr.magic) != XFS_DIR2_BLOCK_MAGIC)) {
XFS_CORRUPTION_ERROR("xfs_dir2_block_addname",
XFS_ERRLEVEL_LOW, mp, block);
xfs_da_brelse(tp, bp);
@@ -121,38 +120,38 @@ xfs_dir2_block_addname(
/*
* Tag just before the first leaf entry.
*/
- tagp = (xfs_dir2_data_off_t *)blp - 1;
+ tagp = (__be16 *)blp - 1;
/*
* Data object just before the first leaf entry.
*/
- enddup = (xfs_dir2_data_unused_t *)((char *)block + INT_GET(*tagp, ARCH_CONVERT));
+ enddup = (xfs_dir2_data_unused_t *)((char *)block + be16_to_cpu(*tagp));
/*
* If it's not free then can't do this add without cleaning up:
* the space before the first leaf entry needs to be free so it
* can be expanded to hold the pointer to the new entry.
*/
- if (INT_GET(enddup->freetag, ARCH_CONVERT) != XFS_DIR2_DATA_FREE_TAG)
+ if (be16_to_cpu(enddup->freetag) != XFS_DIR2_DATA_FREE_TAG)
dup = enddup = NULL;
/*
* Check out the biggest freespace and see if it's the same one.
*/
else {
dup = (xfs_dir2_data_unused_t *)
- ((char *)block + INT_GET(bf[0].offset, ARCH_CONVERT));
+ ((char *)block + be16_to_cpu(bf[0].offset));
if (dup == enddup) {
/*
* It is the biggest freespace, is it too small
* to hold the new leaf too?
*/
- if (INT_GET(dup->length, ARCH_CONVERT) < len + (uint)sizeof(*blp)) {
+ if (be16_to_cpu(dup->length) < len + (uint)sizeof(*blp)) {
/*
* Yes, we use the second-largest
* entry instead if it works.
*/
- if (INT_GET(bf[1].length, ARCH_CONVERT) >= len)
+ if (be16_to_cpu(bf[1].length) >= len)
dup = (xfs_dir2_data_unused_t *)
((char *)block +
- INT_GET(bf[1].offset, ARCH_CONVERT));
+ be16_to_cpu(bf[1].offset));
else
dup = NULL;
}
@@ -161,7 +160,7 @@ xfs_dir2_block_addname(
* Not the same free entry,
* just check its length.
*/
- if (INT_GET(dup->length, ARCH_CONVERT) < len) {
+ if (be16_to_cpu(dup->length) < len) {
dup = NULL;
}
}
@@ -172,9 +171,9 @@ xfs_dir2_block_addname(
* If there are stale entries we'll use one for the leaf.
* Is the biggest entry enough to avoid compaction?
*/
- else if (INT_GET(bf[0].length, ARCH_CONVERT) >= len) {
+ else if (be16_to_cpu(bf[0].length) >= len) {
dup = (xfs_dir2_data_unused_t *)
- ((char *)block + INT_GET(bf[0].offset, ARCH_CONVERT));
+ ((char *)block + be16_to_cpu(bf[0].offset));
compact = 0;
}
/*
@@ -184,20 +183,20 @@ xfs_dir2_block_addname(
/*
* Tag just before the first leaf entry.
*/
- tagp = (xfs_dir2_data_off_t *)blp - 1;
+ tagp = (__be16 *)blp - 1;
/*
* Data object just before the first leaf entry.
*/
- dup = (xfs_dir2_data_unused_t *)((char *)block + INT_GET(*tagp, ARCH_CONVERT));
+ dup = (xfs_dir2_data_unused_t *)((char *)block + be16_to_cpu(*tagp));
/*
* If it's not free then the data will go where the
* leaf data starts now, if it works at all.
*/
- if (INT_GET(dup->freetag, ARCH_CONVERT) == XFS_DIR2_DATA_FREE_TAG) {
- if (INT_GET(dup->length, ARCH_CONVERT) + (INT_GET(btp->stale, ARCH_CONVERT) - 1) *
+ if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
+ if (be16_to_cpu(dup->length) + (be32_to_cpu(btp->stale) - 1) *
(uint)sizeof(*blp) < len)
dup = NULL;
- } else if ((INT_GET(btp->stale, ARCH_CONVERT) - 1) * (uint)sizeof(*blp) < len)
+ } else if ((be32_to_cpu(btp->stale) - 1) * (uint)sizeof(*blp) < len)
dup = NULL;
else
dup = (xfs_dir2_data_unused_t *)blp;
@@ -243,11 +242,11 @@ xfs_dir2_block_addname(
int fromidx; /* source leaf index */
int toidx; /* target leaf index */
- for (fromidx = toidx = INT_GET(btp->count, ARCH_CONVERT) - 1,
+ for (fromidx = toidx = be32_to_cpu(btp->count) - 1,
highstale = lfloghigh = -1;
fromidx >= 0;
fromidx--) {
- if (INT_GET(blp[fromidx].address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR) {
+ if (be32_to_cpu(blp[fromidx].address) == XFS_DIR2_NULL_DATAPTR) {
if (highstale == -1)
highstale = toidx;
else {
@@ -260,15 +259,15 @@ xfs_dir2_block_addname(
blp[toidx] = blp[fromidx];
toidx--;
}
- lfloglow = toidx + 1 - (INT_GET(btp->stale, ARCH_CONVERT) - 1);
- lfloghigh -= INT_GET(btp->stale, ARCH_CONVERT) - 1;
- INT_MOD(btp->count, ARCH_CONVERT, -(INT_GET(btp->stale, ARCH_CONVERT) - 1));
+ lfloglow = toidx + 1 - (be32_to_cpu(btp->stale) - 1);
+ lfloghigh -= be32_to_cpu(btp->stale) - 1;
+ be32_add(&btp->count, -(be32_to_cpu(btp->stale) - 1));
xfs_dir2_data_make_free(tp, bp,
(xfs_dir2_data_aoff_t)((char *)blp - (char *)block),
- (xfs_dir2_data_aoff_t)((INT_GET(btp->stale, ARCH_CONVERT) - 1) * sizeof(*blp)),
+ (xfs_dir2_data_aoff_t)((be32_to_cpu(btp->stale) - 1) * sizeof(*blp)),
&needlog, &needscan);
- blp += INT_GET(btp->stale, ARCH_CONVERT) - 1;
- INT_SET(btp->stale, ARCH_CONVERT, 1);
+ blp += be32_to_cpu(btp->stale) - 1;
+ btp->stale = cpu_to_be32(1);
/*
* If we now need to rebuild the bestfree map, do so.
* This needs to happen before the next call to use_free.
@@ -283,23 +282,23 @@ xfs_dir2_block_addname(
* Set leaf logging boundaries to impossible state.
* For the no-stale case they're set explicitly.
*/
- else if (INT_GET(btp->stale, ARCH_CONVERT)) {
- lfloglow = INT_GET(btp->count, ARCH_CONVERT);
+ else if (btp->stale) {
+ lfloglow = be32_to_cpu(btp->count);
lfloghigh = -1;
}
/*
* Find the slot that's first lower than our hash value, -1 if none.
*/
- for (low = 0, high = INT_GET(btp->count, ARCH_CONVERT) - 1; low <= high; ) {
+ for (low = 0, high = be32_to_cpu(btp->count) - 1; low <= high; ) {
mid = (low + high) >> 1;
- if ((hash = INT_GET(blp[mid].hashval, ARCH_CONVERT)) == args->hashval)
+ if ((hash = be32_to_cpu(blp[mid].hashval)) == args->hashval)
break;
if (hash < args->hashval)
low = mid + 1;
else
high = mid - 1;
}
- while (mid >= 0 && INT_GET(blp[mid].hashval, ARCH_CONVERT) >= args->hashval) {
+ while (mid >= 0 && be32_to_cpu(blp[mid].hashval) >= args->hashval) {
mid--;
}
/*
@@ -311,14 +310,14 @@ xfs_dir2_block_addname(
*/
xfs_dir2_data_use_free(tp, bp, enddup,
(xfs_dir2_data_aoff_t)
- ((char *)enddup - (char *)block + INT_GET(enddup->length, ARCH_CONVERT) -
+ ((char *)enddup - (char *)block + be16_to_cpu(enddup->length) -
sizeof(*blp)),
(xfs_dir2_data_aoff_t)sizeof(*blp),
&needlog, &needscan);
/*
* Update the tail (entry count).
*/
- INT_MOD(btp->count, ARCH_CONVERT, +1);
+ be32_add(&btp->count, 1);
/*
* If we now need to rebuild the bestfree map, do so.
* This needs to happen before the next call to use_free.
@@ -346,12 +345,12 @@ xfs_dir2_block_addname(
else {
for (lowstale = mid;
lowstale >= 0 &&
- INT_GET(blp[lowstale].address, ARCH_CONVERT) != XFS_DIR2_NULL_DATAPTR;
+ be32_to_cpu(blp[lowstale].address) != XFS_DIR2_NULL_DATAPTR;
lowstale--)
continue;
for (highstale = mid + 1;
- highstale < INT_GET(btp->count, ARCH_CONVERT) &&
- INT_GET(blp[highstale].address, ARCH_CONVERT) != XFS_DIR2_NULL_DATAPTR &&
+ highstale < be32_to_cpu(btp->count) &&
+ be32_to_cpu(blp[highstale].address) != XFS_DIR2_NULL_DATAPTR &&
(lowstale < 0 || mid - lowstale > highstale - mid);
highstale++)
continue;
@@ -359,7 +358,7 @@ xfs_dir2_block_addname(
* Move entries toward the low-numbered stale entry.
*/
if (lowstale >= 0 &&
- (highstale == INT_GET(btp->count, ARCH_CONVERT) ||
+ (highstale == be32_to_cpu(btp->count) ||
mid - lowstale <= highstale - mid)) {
if (mid - lowstale)
memmove(&blp[lowstale], &blp[lowstale + 1],
@@ -371,7 +370,7 @@ xfs_dir2_block_addname(
* Move entries toward the high-numbered stale entry.
*/
else {
- ASSERT(highstale < INT_GET(btp->count, ARCH_CONVERT));
+ ASSERT(highstale < be32_to_cpu(btp->count));
mid++;
if (highstale - mid)
memmove(&blp[mid + 1], &blp[mid],
@@ -379,7 +378,7 @@ xfs_dir2_block_addname(
lfloglow = MIN(mid, lfloglow);
lfloghigh = MAX(highstale, lfloghigh);
}
- INT_MOD(btp->stale, ARCH_CONVERT, -1);
+ be32_add(&btp->stale, -1);
}
/*
* Point to the new data entry.
@@ -388,8 +387,9 @@ xfs_dir2_block_addname(
/*
* Fill in the leaf entry.
*/
- INT_SET(blp[mid].hashval, ARCH_CONVERT, args->hashval);
- INT_SET(blp[mid].address, ARCH_CONVERT, XFS_DIR2_BYTE_TO_DATAPTR(mp, (char *)dep - (char *)block));
+ blp[mid].hashval = cpu_to_be32(args->hashval);
+ blp[mid].address = cpu_to_be32(XFS_DIR2_BYTE_TO_DATAPTR(mp,
+ (char *)dep - (char *)block));
xfs_dir2_block_log_leaf(tp, bp, lfloglow, lfloghigh);
/*
* Mark space for the data entry used.
@@ -404,7 +404,7 @@ xfs_dir2_block_addname(
dep->namelen = args->namelen;
memcpy(dep->name, args->name, args->namelen);
tagp = XFS_DIR2_DATA_ENTRY_TAG_P(dep);
- INT_SET(*tagp, ARCH_CONVERT, (xfs_dir2_data_off_t)((char *)dep - (char *)block));
+ *tagp = cpu_to_be16((char *)dep - (char *)block);
/*
* Clean up the bestfree array and log the header, tail, and entry.
*/
@@ -485,8 +485,8 @@ xfs_dir2_block_getdents(
/*
* Unused, skip it.
*/
- if (INT_GET(dup->freetag, ARCH_CONVERT) == XFS_DIR2_DATA_FREE_TAG) {
- ptr += INT_GET(dup->length, ARCH_CONVERT);
+ if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
+ ptr += be16_to_cpu(dup->length);
continue;
}
@@ -622,7 +622,7 @@ xfs_dir2_block_lookup(
* Get the offset from the leaf entry, to point to the data.
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)block + XFS_DIR2_DATAPTR_TO_OFF(mp, INT_GET(blp[ent].address, ARCH_CONVERT)));
+ ((char *)block + XFS_DIR2_DATAPTR_TO_OFF(mp, be32_to_cpu(blp[ent].address)));
/*
* Fill in inode number, release the block.
*/
@@ -674,10 +674,10 @@ xfs_dir2_block_lookup_int(
* Loop doing a binary search for our hash value.
* Find our entry, ENOENT if it's not there.
*/
- for (low = 0, high = INT_GET(btp->count, ARCH_CONVERT) - 1; ; ) {
+ for (low = 0, high = be32_to_cpu(btp->count) - 1; ; ) {
ASSERT(low <= high);
mid = (low + high) >> 1;
- if ((hash = INT_GET(blp[mid].hashval, ARCH_CONVERT)) == args->hashval)
+ if ((hash = be32_to_cpu(blp[mid].hashval)) == args->hashval)
break;
if (hash < args->hashval)
low = mid + 1;
@@ -692,7 +692,7 @@ xfs_dir2_block_lookup_int(
/*
* Back up to the first one with the right hash value.
*/
- while (mid > 0 && INT_GET(blp[mid - 1].hashval, ARCH_CONVERT) == args->hashval) {
+ while (mid > 0 && be32_to_cpu(blp[mid - 1].hashval) == args->hashval) {
mid--;
}
/*
@@ -700,7 +700,7 @@ xfs_dir2_block_lookup_int(
* right hash value looking for our name.
*/
do {
- if ((addr = INT_GET(blp[mid].address, ARCH_CONVERT)) == XFS_DIR2_NULL_DATAPTR)
+ if ((addr = be32_to_cpu(blp[mid].address)) == XFS_DIR2_NULL_DATAPTR)
continue;
/*
* Get pointer to the entry from the leaf.
@@ -717,7 +717,7 @@ xfs_dir2_block_lookup_int(
*entno = mid;
return 0;
}
- } while (++mid < INT_GET(btp->count, ARCH_CONVERT) && INT_GET(blp[mid].hashval, ARCH_CONVERT) == hash);
+ } while (++mid < be32_to_cpu(btp->count) && be32_to_cpu(blp[mid].hashval) == hash);
/*
* No match, release the buffer and return ENOENT.
*/
@@ -767,7 +767,7 @@ xfs_dir2_block_removename(
* Point to the data entry using the leaf entry.
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)block + XFS_DIR2_DATAPTR_TO_OFF(mp, INT_GET(blp[ent].address, ARCH_CONVERT)));
+ ((char *)block + XFS_DIR2_DATAPTR_TO_OFF(mp, be32_to_cpu(blp[ent].address)));
/*
* Mark the data entry's space free.
*/
@@ -778,12 +778,12 @@ xfs_dir2_block_removename(
/*
* Fix up the block tail.
*/
- INT_MOD(btp->stale, ARCH_CONVERT, +1);
+ be32_add(&btp->stale, 1);
xfs_dir2_block_log_tail(tp, bp);
/*
* Remove the leaf entry by marking it stale.
*/
- INT_SET(blp[ent].address, ARCH_CONVERT, XFS_DIR2_NULL_DATAPTR);
+ blp[ent].address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
xfs_dir2_block_log_leaf(tp, bp, ent, ent);
/*
* Fix up bestfree, log the header if necessary.
@@ -843,7 +843,7 @@ xfs_dir2_block_replace(
* Point to the data entry we need to change.
*/
dep = (xfs_dir2_data_entry_t *)
- ((char *)block + XFS_DIR2_DATAPTR_TO_OFF(mp, INT_GET(blp[ent].address, ARCH_CONVERT)));
+ ((char *)block + XFS_DIR2_DATAPTR_TO_OFF(mp, be32_to_cpu(blp[ent].address)));
ASSERT(INT_GET(dep->inumber, ARCH_CONVERT) != args->inumber);
/*
* Change the inode number to the new value.
@@ -868,8 +868,8 @@ xfs_dir2_block_sort(
la = a;
lb = b;
- return INT_GET(la->hashval, ARCH_CONVERT) < INT_GET(lb->hashval, ARCH_CONVERT) ? -1 :
- (INT_GET(la->hashval, ARCH_CONVERT) > INT_GET(lb->hashval, ARCH_CONVERT) ? 1 : 0);
+ return be32_to_cpu(la->hashval) < be32_to_cpu(lb->hashval) ? -1 :
+ (be32_to_cpu(la->hashval) > be32_to_cpu(lb->hashval) ? 1 : 0);
}
/*
@@ -881,7 +881,7 @@ xfs_dir2_leaf_to_block(
xfs_dabuf_t *lbp, /* leaf buffer */
xfs_dabuf_t *dbp) /* data buffer */
{
- xfs_dir2_data_off_t *bestsp; /* leaf bests table */
+ __be16 *bestsp; /* leaf bests table */
xfs_dir2_block_t *block; /* block structure */
xfs_dir2_block_tail_t *btp; /* block tail */
xfs_inode_t *dp; /* incore directory inode */
@@ -896,7 +896,7 @@ xfs_dir2_leaf_to_block(
int needscan; /* need to scan for bestfree */
xfs_dir2_sf_hdr_t sfh; /* shortform header */
int size; /* bytes used */
- xfs_dir2_data_off_t *tagp; /* end of entry (tag) */
+ __be16 *tagp; /* end of entry (tag) */
int to; /* block/leaf to index */
xfs_trans_t *tp; /* transaction pointer */
@@ -905,7 +905,7 @@ xfs_dir2_leaf_to_block(
tp = args->trans;
mp = dp->i_mount;
leaf = lbp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAF1_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC);
ltp = XFS_DIR2_LEAF_TAIL_P(mp, leaf);
/*
* If there are data blocks other than the first one, take this
@@ -915,11 +915,11 @@ xfs_dir2_leaf_to_block(
*/
while (dp->i_d.di_size > mp->m_dirblksize) {
bestsp = XFS_DIR2_LEAF_BESTS_P(ltp);
- if (INT_GET(bestsp[INT_GET(ltp->bestcount, ARCH_CONVERT) - 1], ARCH_CONVERT) ==
+ if (be16_to_cpu(bestsp[be32_to_cpu(ltp->bestcount) - 1]) ==
mp->m_dirblksize - (uint)sizeof(block->hdr)) {
if ((error =
xfs_dir2_leaf_trim_data(args, lbp,
- (xfs_dir2_db_t)(INT_GET(ltp->bestcount, ARCH_CONVERT) - 1))))
+ (xfs_dir2_db_t)(be32_to_cpu(ltp->bestcount) - 1))))
goto out;
} else {
error = 0;
@@ -935,28 +935,29 @@ xfs_dir2_leaf_to_block(
goto out;
}
block = dbp->data;
- ASSERT(INT_GET(block->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC);
+ ASSERT(be32_to_cpu(block->hdr.magic) == XFS_DIR2_DATA_MAGIC);
/*
* Size of the "leaf" area in the block.
*/
size = (uint)sizeof(block->tail) +
- (uint)sizeof(*lep) * (INT_GET(leaf->hdr.count, ARCH_CONVERT) - INT_GET(leaf->hdr.stale, ARCH_CONVERT));
+ (uint)sizeof(*lep) * (be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale));
/*
* Look at the last data entry.
*/
- tagp = (xfs_dir2_data_off_t *)((char *)block + mp->m_dirblksize) - 1;
- dup = (xfs_dir2_data_unused_t *)((char *)block + INT_GET(*tagp, ARCH_CONVERT));
+ tagp = (__be16 *)((char *)block + mp->m_dirblksize) - 1;
+ dup = (xfs_dir2_data_unused_t *)((char *)block + be16_to_cpu(*tagp));
/*
* If it's not free or is too short we can't do it.
*/
- if (INT_GET(dup->freetag, ARCH_CONVERT) != XFS_DIR2_DATA_FREE_TAG || INT_GET(dup->length, ARCH_CONVERT) < size) {
+ if (be16_to_cpu(dup->freetag) != XFS_DIR2_DATA_FREE_TAG ||
+ be16_to_cpu(dup->length) < size) {
error = 0;
goto out;
}
/*
* Start converting it to block form.
*/
- INT_SET(block->hdr.magic, ARCH_CONVERT, XFS_DIR2_BLOCK_MAGIC);
+ block->hdr.magic = cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
needlog = 1;
needscan = 0;
/*
@@ -968,20 +969,20 @@ xfs_dir2_leaf_to_block(
* Initialize the block tail.
*/
btp = XFS_DIR2_BLOCK_TAIL_P(mp, block);
- INT_SET(btp->count, ARCH_CONVERT, INT_GET(leaf->hdr.count, ARCH_CONVERT) - INT_GET(leaf->hdr.stale, ARCH_CONVERT));
+ btp->count = cpu_to_be32(be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale));
btp->stale = 0;
xfs_dir2_block_log_tail(tp, dbp);
/*
* Initialize the block leaf area. We compact out stale entries.
*/
lep = XFS_DIR2_BLOCK_LEAF_P(btp);
- for (from = to = 0; from < INT_GET(leaf->hdr.count, ARCH_CONVERT); from++) {
- if (INT_GET(leaf->ents[from].address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR)
+ for (from = to = 0; from < be16_to_cpu(leaf->hdr.count); from++) {
+ if (be32_to_cpu(leaf->ents[from].address) == XFS_DIR2_NULL_DATAPTR)
continue;
lep[to++] = leaf->ents[from];
}
- ASSERT(to == INT_GET(btp->count, ARCH_CONVERT));
- xfs_dir2_block_log_leaf(tp, dbp, 0, INT_GET(btp->count, ARCH_CONVERT) - 1);
+ ASSERT(to == be32_to_cpu(btp->count));
+ xfs_dir2_block_log_leaf(tp, dbp, 0, be32_to_cpu(btp->count) - 1);
/*
* Scan the bestfree if we need it and log the data block header.
*/
@@ -1043,7 +1044,7 @@ xfs_dir2_sf_to_block(
int offset; /* target block offset */
xfs_dir2_sf_entry_t *sfep; /* sf entry pointer */
xfs_dir2_sf_t *sfp; /* shortform structure */
- xfs_dir2_data_off_t *tagp; /* end of data entry */
+ __be16 *tagp; /* end of data entry */
xfs_trans_t *tp; /* transaction pointer */
xfs_dir2_trace_args("sf_to_block", args);
@@ -1095,12 +1096,12 @@ xfs_dir2_sf_to_block(
return error;
}
block = bp->data;
- INT_SET(block->hdr.magic, ARCH_CONVERT, XFS_DIR2_BLOCK_MAGIC);
+ block->hdr.magic = cpu_to_be32(XFS_DIR2_BLOCK_MAGIC);
/*
* Compute size of block "tail" area.
*/
i = (uint)sizeof(*btp) +
- (INT_GET(sfp->hdr.count, ARCH_CONVERT) + 2) * (uint)sizeof(xfs_dir2_leaf_entry_t);
+ (sfp->hdr.count + 2) * (uint)sizeof(xfs_dir2_leaf_entry_t);
/*
* The whole thing is initialized to free by the init routine.
* Say we're using the leaf and tail area.
@@ -1114,7 +1115,7 @@ xfs_dir2_sf_to_block(
* Fill in the tail.
*/
btp = XFS_DIR2_BLOCK_TAIL_P(mp, block);
- INT_SET(btp->count, ARCH_CONVERT, INT_GET(sfp->hdr.count, ARCH_CONVERT) + 2); /* ., .. */
+ btp->count = cpu_to_be32(sfp->hdr.count + 2); /* ., .. */
btp->stale = 0;
blp = XFS_DIR2_BLOCK_LEAF_P(btp);
endoffset = (uint)((char *)blp - (char *)block);
@@ -1123,7 +1124,7 @@ xfs_dir2_sf_to_block(
*/
xfs_dir2_data_use_free(tp, bp, dup,
(xfs_dir2_data_aoff_t)((char *)dup - (char *)block),
- INT_GET(dup->length, ARCH_CONVERT), &needlog, &needscan);
+ be16_to_cpu(dup->length), &needlog, &needscan);
/*
* Create entry for .
*/
@@ -1133,10 +1134,11 @@ xfs_dir2_sf_to_block(
dep->namelen = 1;
dep->name[0] = '.';
tagp = XFS_DIR2_DATA_ENTRY_TAG_P(dep);
- INT_SET(*tagp, ARCH_CONVERT, (xfs_dir2_data_off_t)((char *)dep - (char *)block));
+ *tagp = cpu_to_be16((char *)dep - (char *)block);
xfs_dir2_data_log_entry(tp, bp, dep);
- INT_SET(blp[0].hashval, ARCH_CONVERT, xfs_dir_hash_dot);
- INT_SET(blp[0].address, ARCH_CONVERT, XFS_DIR2_BYTE_TO_DATAPTR(mp, (char *)dep - (char *)block));
+ blp[0].hashval = cpu_to_be32(xfs_dir_hash_dot);
+ blp[0].address = cpu_to_be32(XFS_DIR2_BYTE_TO_DATAPTR(mp,
+ (char *)dep - (char *)block));
/*
* Create entry for ..
*/
@@ -1146,15 +1148,16 @@ xfs_dir2_sf_to_block(
dep->namelen = 2;
dep->name[0] = dep->name[1] = '.';
tagp = XFS_DIR2_DATA_ENTRY_TAG_P(dep);
- INT_SET(*tagp, ARCH_CONVERT, (xfs_dir2_data_off_t)((char *)dep - (char *)block));
+ *tagp = cpu_to_be16((char *)dep - (char *)block);
xfs_dir2_data_log_entry(tp, bp, dep);
- INT_SET(blp[1].hashval, ARCH_CONVERT, xfs_dir_hash_dotdot);
- INT_SET(blp[1].address, ARCH_CONVERT, XFS_DIR2_BYTE_TO_DATAPTR(mp, (char *)dep - (char *)block));
+ blp[1].hashval = cpu_to_be32(xfs_dir_hash_dotdot);
+ blp[1].address = cpu_to_be32(XFS_DIR2_BYTE_TO_DATAPTR(mp,
+ (char *)dep - (char *)block));
offset = XFS_DIR2_DATA_FIRST_OFFSET;
/*
* Loop over existing entries, stuff them in.
*/
- if ((i = 0) == INT_GET(sfp->hdr.count, ARCH_CONVERT))
+ if ((i = 0) == sfp->hdr.count)
sfep = NULL;
else
sfep = XFS_DIR2_SF_FIRSTENTRY(sfp);
@@ -1176,15 +1179,14 @@ xfs_dir2_sf_to_block(
if (offset < newoffset) {
dup = (xfs_dir2_data_unused_t *)
((char *)block + offset);
- INT_SET(dup->freetag, ARCH_CONVERT, XFS_DIR2_DATA_FREE_TAG);
- INT_SET(dup->length, ARCH_CONVERT, newoffset - offset);
- INT_SET(*XFS_DIR2_DATA_UNUSED_TAG_P(dup), ARCH_CONVERT,
- (xfs_dir2_data_off_t)
+ dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
+ dup->length = cpu_to_be16(newoffset - offset);
+ *XFS_DIR2_DATA_UNUSED_TAG_P(dup) = cpu_to_be16(
((char *)dup - (char *)block));
xfs_dir2_data_log_unused(tp, bp, dup);
(void)xfs_dir2_data_freeinsert((xfs_dir2_data_t *)block,
dup, &dummy);
- offset += INT_GET(dup->length, ARCH_CONVERT);
+ offset += be16_to_cpu(dup->length);
continue;
}
/*
@@ -1196,13 +1198,14 @@ xfs_dir2_sf_to_block(
dep->namelen = sfep->namelen;
memcpy(dep->name, sfep->name, dep->namelen);
tagp = XFS_DIR2_DATA_ENTRY_TAG_P(dep);
- INT_SET(*tagp, ARCH_CONVERT, (xfs_dir2_data_off_t)((char *)dep - (char *)block));
+ *tagp = cpu_to_be16((char *)dep - (char *)block);
xfs_dir2_data_log_entry(tp, bp, dep);
- INT_SET(blp[2 + i].hashval, ARCH_CONVERT, xfs_da_hashname((char *)sfep->name, sfep->namelen));
- INT_SET(blp[2 + i].address, ARCH_CONVERT, XFS_DIR2_BYTE_TO_DATAPTR(mp,
+ blp[2 + i].hashval = cpu_to_be32(xfs_da_hashname(
+ (char *)sfep->name, sfep->namelen));
+ blp[2 + i].address = cpu_to_be32(XFS_DIR2_BYTE_TO_DATAPTR(mp,
(char *)dep - (char *)block));
offset = (int)((char *)(tagp + 1) - (char *)block);
- if (++i == INT_GET(sfp->hdr.count, ARCH_CONVERT))
+ if (++i == sfp->hdr.count)
sfep = NULL;
else
sfep = XFS_DIR2_SF_NEXTENTRY(sfp, sfep);
@@ -1212,13 +1215,13 @@ xfs_dir2_sf_to_block(
/*
* Sort the leaf entries by hash value.
*/
- xfs_sort(blp, INT_GET(btp->count, ARCH_CONVERT), sizeof(*blp), xfs_dir2_block_sort);
+ xfs_sort(blp, be32_to_cpu(btp->count), sizeof(*blp), xfs_dir2_block_sort);
/*
* Log the leaf entry area and tail.
* Already logged the header in data_init, ignore needlog.
*/
ASSERT(needscan == 0);
- xfs_dir2_block_log_leaf(tp, bp, 0, INT_GET(btp->count, ARCH_CONVERT) - 1);
+ xfs_dir2_block_log_leaf(tp, bp, 0, be32_to_cpu(btp->count) - 1);
xfs_dir2_block_log_tail(tp, bp);
xfs_dir2_data_check(dp, bp);
xfs_da_buf_done(bp);
diff --git a/fs/xfs/xfs_dir2_block.h b/fs/xfs/xfs_dir2_block.h
index a2e5cb98a838..6722effd0b20 100644
--- a/fs/xfs/xfs_dir2_block.h
+++ b/fs/xfs/xfs_dir2_block.h
@@ -43,8 +43,8 @@ struct xfs_trans;
#define XFS_DIR2_BLOCK_MAGIC 0x58443242 /* XD2B: for one block dirs */
typedef struct xfs_dir2_block_tail {
- __uint32_t count; /* count of leaf entries */
- __uint32_t stale; /* count of stale lf entries */
+ __be32 count; /* count of leaf entries */
+ __be32 stale; /* count of stale lf entries */
} xfs_dir2_block_tail_t;
/*
@@ -75,8 +75,7 @@ xfs_dir2_block_tail_p(struct xfs_mount *mp, xfs_dir2_block_t *block)
static inline struct xfs_dir2_leaf_entry *
xfs_dir2_block_leaf_p(xfs_dir2_block_tail_t *btp)
{
- return (((struct xfs_dir2_leaf_entry *)
- (btp)) - INT_GET((btp)->count, ARCH_CONVERT));
+ return ((struct xfs_dir2_leaf_entry *)btp) - be32_to_cpu(btp->count);
}
/*
diff --git a/fs/xfs/xfs_dir2_data.c b/fs/xfs/xfs_dir2_data.c
index 5b7c47e2f14a..bb3d03ff002b 100644
--- a/fs/xfs/xfs_dir2_data.c
+++ b/fs/xfs/xfs_dir2_data.c
@@ -70,11 +70,11 @@ xfs_dir2_data_check(
mp = dp->i_mount;
d = bp->data;
- ASSERT(INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC ||
- INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
+ be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
bf = d->hdr.bestfree;
p = (char *)d->u;
- if (INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC) {
+ if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
btp = XFS_DIR2_BLOCK_TAIL_P(mp, (xfs_dir2_block_t *)d);
lep = XFS_DIR2_BLOCK_LEAF_P(btp);
endp = (char *)lep;
@@ -96,8 +96,8 @@ xfs_dir2_data_check(
ASSERT(!bf[2].offset);
freeseen |= 1 << 2;
}
- ASSERT(INT_GET(bf[0].length, ARCH_CONVERT) >= INT_GET(bf[1].length, ARCH_CONVERT));
- ASSERT(INT_GET(bf[1].length, ARCH_CONVERT) >= INT_GET(bf[2].length, ARCH_CONVERT));
+ ASSERT(be16_to_cpu(bf[0].length) >= be16_to_cpu(bf[1].length));
+ ASSERT(be16_to_cpu(bf[1].length) >= be16_to_cpu(bf[2].length));
/*
* Loop over the data/unused entries.
*/
@@ -108,18 +108,20 @@ xfs_dir2_data_check(
* If we find it, account for that, else make sure it
* doesn't need to be there.
*/
- if (INT_GET(dup->freetag, ARCH_CONVERT) == XFS_DIR2_DATA_FREE_TAG) {
+ if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
ASSERT(lastfree == 0);
- ASSERT(INT_GET(*XFS_DIR2_DATA_UNUSED_TAG_P(dup), ARCH_CONVERT) ==
+ ASSERT(be16_to_cpu(*XFS_DIR2_DATA_UNUSED_TAG_P(dup)) ==
(char *)dup - (char *)d);
dfp = xfs_dir2_data_freefind(d, dup);
if (dfp) {
i = (int)(dfp - bf);
ASSERT((freeseen & (1 << i)) == 0);
freeseen |= 1 << i;
- } else
- ASSERT(INT_GET(dup->length, ARCH_CONVERT) <= INT_GET(bf[2].length, ARCH_CONVERT));
- p += INT_GET(dup->length, ARCH_CONVERT);
+ } else {
+ ASSERT(be16_to_cpu(dup->length) <=
+ be16_to_cpu(bf[2].length));
+ }
+ p += be16_to_cpu(dup->length);
lastfree = 1;
continue;
}
@@ -132,21 +134,21 @@ xfs_dir2_data_check(
dep = (xfs_dir2_data_entry_t *)p;
ASSERT(dep->namelen != 0);
ASSERT(xfs_dir_ino_validate(mp, INT_GET(dep->inumber, ARCH_CONVERT)) == 0);
- ASSERT(INT_GET(*XFS_DIR2_DATA_ENTRY_TAG_P(dep), ARCH_CONVERT) ==
+ ASSERT(be16_to_cpu(*XFS_DIR2_DATA_ENTRY_TAG_P(dep)) ==
(char *)dep - (char *)d);
count++;
lastfree = 0;
- if (INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC) {
+ if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
addr = XFS_DIR2_DB_OFF_TO_DATAPTR(mp, mp->m_dirdatablk,
(xfs_dir2_data_aoff_t)
((char *)dep - (char *)d));
hash = xfs_da_hashname((char *)dep->name, dep->namelen);
- for (i = 0; i < INT_GET(btp->count, ARCH_CONVERT); i++) {
- if (INT_GET(lep[i].address, ARCH_CONVERT) == addr &&
- INT_GET(lep[i].hashval, ARCH_CONVERT) == hash)
+ for (i = 0; i < be32_to_cpu(btp->count); i++) {
+ if (be32_to_cpu(lep[i].address) == addr &&
+ be32_to_cpu(lep[i].hashval) == hash)
break;
}
- ASSERT(i < INT_GET(btp->count, ARCH_CONVERT));
+ ASSERT(i < be32_to_cpu(btp->count));
}
p += XFS_DIR2_DATA_ENTSIZE(dep->namelen);
}
@@ -154,15 +156,15 @@ xfs_dir2_data_check(
* Need to have seen all the entries and all the bestfree slots.
*/
ASSERT(freeseen == 7);
- if (INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC) {
- for (i = stale = 0; i < INT_GET(btp->count, ARCH_CONVERT); i++) {
- if (INT_GET(lep[i].address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR)
+ if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
+ for (i = stale = 0; i < be32_to_cpu(btp->count); i++) {
+ if (be32_to_cpu(lep[i].address) == XFS_DIR2_NULL_DATAPTR)
stale++;
if (i > 0)
- ASSERT(INT_GET(lep[i].hashval, ARCH_CONVERT) >= INT_GET(lep[i - 1].hashval, ARCH_CONVERT));
+ ASSERT(be32_to_cpu(lep[i].hashval) >= be32_to_cpu(lep[i - 1].hashval));
}
- ASSERT(count == INT_GET(btp->count, ARCH_CONVERT) - INT_GET(btp->stale, ARCH_CONVERT));
- ASSERT(stale == INT_GET(btp->stale, ARCH_CONVERT));
+ ASSERT(count == be32_to_cpu(btp->count) - be32_to_cpu(btp->stale));
+ ASSERT(stale == be32_to_cpu(btp->stale));
}
}
#endif
@@ -190,8 +192,8 @@ xfs_dir2_data_freefind(
* Check order, non-overlapping entries, and if we find the
* one we're looking for it has to be exact.
*/
- ASSERT(INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC ||
- INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
+ be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
for (dfp = &d->hdr.bestfree[0], seenzero = matched = 0;
dfp < &d->hdr.bestfree[XFS_DIR2_DATA_FD_COUNT];
dfp++) {
@@ -201,23 +203,24 @@ xfs_dir2_data_freefind(
continue;
}
ASSERT(seenzero == 0);
- if (INT_GET(dfp->offset, ARCH_CONVERT) == off) {
+ if (be16_to_cpu(dfp->offset) == off) {
matched = 1;
- ASSERT(INT_GET(dfp->length, ARCH_CONVERT) == INT_GET(dup->length, ARCH_CONVERT));
- } else if (off < INT_GET(dfp->offset, ARCH_CONVERT))
- ASSERT(off + INT_GET(dup->length, ARCH_CONVERT) <= INT_GET(dfp->offset, ARCH_CONVERT));
+ ASSERT(dfp->length == dup->length);
+ } else if (off < be16_to_cpu(dfp->offset))
+ ASSERT(off + be16_to_cpu(dup->length) <= be16_to_cpu(dfp->offset));
else
- ASSERT(INT_GET(dfp->offset, ARCH_CONVERT) + INT_GET(dfp->length, ARCH_CONVERT) <= off);
- ASSERT(matched || INT_GET(dfp->length, ARCH_CONVERT) >= INT_GET(dup->length, ARCH_CONVERT));
+ ASSERT(be16_to_cpu(dfp->offset) + be16_to_cpu(dfp->length) <= off);
+ ASSERT(matched || be16_to_cpu(dfp->length) >= be16_to_cpu(dup->length));
if (dfp > &d->hdr.bestfree[0])
- ASSERT(INT_GET(dfp[-1].length, ARCH_CONVERT) >= INT_GET(dfp[0].length, ARCH_CONVERT));
+ ASSERT(be16_to_cpu(dfp[-1].length) >= be16_to_cpu(dfp[0].length));
}
#endif
/*
* If this is smaller than the smallest bestfree entry,
* it can't be there since they're sorted.
*/
- if (INT_GET(dup->length, ARCH_CONVERT) < INT_GET(d->hdr.bestfree[XFS_DIR2_DATA_FD_COUNT - 1].length, ARCH_CONVERT))
+ if (be16_to_cpu(dup->length) <
+ be16_to_cpu(d->hdr.bestfree[XFS_DIR2_DATA_FD_COUNT - 1].length))
return NULL;
/*
* Look at the three bestfree entries for our guy.
@@ -227,7 +230,7 @@ xfs_dir2_data_freefind(
dfp++) {
if (!dfp->offset)
return NULL;
- if (INT_GET(dfp->offset, ARCH_CONVERT) == off)
+ if (be16_to_cpu(dfp->offset) == off)
return dfp;
}
/*
@@ -249,29 +252,29 @@ xfs_dir2_data_freeinsert(
xfs_dir2_data_free_t new; /* new bestfree entry */
#ifdef __KERNEL__
- ASSERT(INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC ||
- INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
+ be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
#endif
dfp = d->hdr.bestfree;
- INT_COPY(new.length, dup->length, ARCH_CONVERT);
- INT_SET(new.offset, ARCH_CONVERT, (xfs_dir2_data_off_t)((char *)dup - (char *)d));
+ new.length = dup->length;
+ new.offset = cpu_to_be16((char *)dup - (char *)d);
/*
* Insert at position 0, 1, or 2; or not at all.
*/
- if (INT_GET(new.length, ARCH_CONVERT) > INT_GET(dfp[0].length, ARCH_CONVERT)) {
+ if (be16_to_cpu(new.length) > be16_to_cpu(dfp[0].length)) {
dfp[2] = dfp[1];
dfp[1] = dfp[0];
dfp[0] = new;
*loghead = 1;
return &dfp[0];
}
- if (INT_GET(new.length, ARCH_CONVERT) > INT_GET(dfp[1].length, ARCH_CONVERT)) {
+ if (be16_to_cpu(new.length) > be16_to_cpu(dfp[1].length)) {
dfp[2] = dfp[1];
dfp[1] = new;
*loghead = 1;
return &dfp[1];
}
- if (INT_GET(new.length, ARCH_CONVERT) > INT_GET(dfp[2].length, ARCH_CONVERT)) {
+ if (be16_to_cpu(new.length) > be16_to_cpu(dfp[2].length)) {
dfp[2] = new;
*loghead = 1;
return &dfp[2];
@@ -289,8 +292,8 @@ xfs_dir2_data_freeremove(
int *loghead) /* out: log data header */
{
#ifdef __KERNEL__
- ASSERT(INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC ||
- INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
+ be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
#endif
/*
* It's the first entry, slide the next 2 up.
@@ -334,8 +337,8 @@ xfs_dir2_data_freescan(
char *p; /* current entry pointer */
#ifdef __KERNEL__
- ASSERT(INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC ||
- INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
+ be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
#endif
/*
* Start by clearing the table.
@@ -348,7 +351,7 @@ xfs_dir2_data_freescan(
p = (char *)d->u;
if (aendp)
endp = aendp;
- else if (INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC) {
+ else if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC) {
btp = XFS_DIR2_BLOCK_TAIL_P(mp, (xfs_dir2_block_t *)d);
endp = (char *)XFS_DIR2_BLOCK_LEAF_P(btp);
} else
@@ -361,11 +364,11 @@ xfs_dir2_data_freescan(
/*
* If it's a free entry, insert it.
*/
- if (INT_GET(dup->freetag, ARCH_CONVERT) == XFS_DIR2_DATA_FREE_TAG) {
+ if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
ASSERT((char *)dup - (char *)d ==
- INT_GET(*XFS_DIR2_DATA_UNUSED_TAG_P(dup), ARCH_CONVERT));
+ be16_to_cpu(*XFS_DIR2_DATA_UNUSED_TAG_P(dup)));
xfs_dir2_data_freeinsert(d, dup, loghead);
- p += INT_GET(dup->length, ARCH_CONVERT);
+ p += be16_to_cpu(dup->length);
}
/*
* For active entries, check their tags and skip them.
@@ -373,7 +376,7 @@ xfs_dir2_data_freescan(
else {
dep = (xfs_dir2_data_entry_t *)p;
ASSERT((char *)dep - (char *)d ==
- INT_GET(*XFS_DIR2_DATA_ENTRY_TAG_P(dep), ARCH_CONVERT));
+ be16_to_cpu(*XFS_DIR2_DATA_ENTRY_TAG_P(dep)));
p += XFS_DIR2_DATA_ENTSIZE(dep->namelen);
}
}
@@ -415,8 +418,8 @@ xfs_dir2_data_init(
* Initialize the header.
*/
d = bp->data;
- INT_SET(d->hdr.magic, ARCH_CONVERT, XFS_DIR2_DATA_MAGIC);
- INT_SET(d->hdr.bestfree[0].offset, ARCH_CONVERT, (xfs_dir2_data_off_t)sizeof(d->hdr));
+ d->hdr.magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
+ d->hdr.bestfree[0].offset = cpu_to_be16(sizeof(d->hdr));
for (i = 1; i < XFS_DIR2_DATA_FD_COUNT; i++) {
d->hdr.bestfree[i].length = 0;
d->hdr.bestfree[i].offset = 0;
@@ -425,13 +428,12 @@ xfs_dir2_data_init(
* Set up an unused entry for the block's body.
*/
dup = &d->u[0].unused;
- INT_SET(dup->freetag, ARCH_CONVERT, XFS_DIR2_DATA_FREE_TAG);
+ dup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
t=mp->m_dirblksize - (uint)sizeof(d->hdr);
- INT_SET(d->hdr.bestfree[0].length, ARCH_CONVERT, t);
- INT_SET(dup->length, ARCH_CONVERT, t);
- INT_SET(*XFS_DIR2_DATA_UNUSED_TAG_P(dup), ARCH_CONVERT,
- (xfs_dir2_data_off_t)((char *)dup - (char *)d));
+ d->hdr.bestfree[0].length = cpu_to_be16(t);
+ dup->length = cpu_to_be16(t);
+ *XFS_DIR2_DATA_UNUSED_TAG_P(dup) = cpu_to_be16((char *)dup - (char *)d);
/*
* Log it and return it.
*/
@@ -453,8 +455,8 @@ xfs_dir2_data_log_entry(
xfs_dir2_data_t *d; /* data block pointer */
d = bp->data;
- ASSERT(INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC ||
- INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
+ be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
xfs_da_log_buf(tp, bp, (uint)((char *)dep - (char *)d),
(uint)((char *)(XFS_DIR2_DATA_ENTRY_TAG_P(dep) + 1) -
(char *)d - 1));
@@ -471,8 +473,8 @@ xfs_dir2_data_log_header(
xfs_dir2_data_t *d; /* data block pointer */
d = bp->data;
- ASSERT(INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC ||
- INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
+ be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
xfs_da_log_buf(tp, bp, (uint)((char *)&d->hdr - (char *)d),
(uint)(sizeof(d->hdr) - 1));
}
@@ -489,8 +491,8 @@ xfs_dir2_data_log_unused(
xfs_dir2_data_t *d; /* data block pointer */
d = bp->data;
- ASSERT(INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC ||
- INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
+ be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
/*
* Log the first part of the unused entry.
*/
@@ -533,12 +535,12 @@ xfs_dir2_data_make_free(
/*
* Figure out where the end of the data area is.
*/
- if (INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC)
+ if (be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC)
endptr = (char *)d + mp->m_dirblksize;
else {
xfs_dir2_block_tail_t *btp; /* block tail */
- ASSERT(INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
btp = XFS_DIR2_BLOCK_TAIL_P(mp, (xfs_dir2_block_t *)d);
endptr = (char *)XFS_DIR2_BLOCK_LEAF_P(btp);
}
@@ -547,11 +549,11 @@ xfs_dir2_data_make_free(
* the previous entry and see if it's free.
*/
if (offset > sizeof(d->hdr)) {
- xfs_dir2_data_off_t *tagp; /* tag just before us */
+ __be16 *tagp; /* tag just before us */
- tagp = (xfs_dir2_data_off_t *)((char *)d + offset) - 1;
- prevdup = (xfs_dir2_data_unused_t *)((char *)d + INT_GET(*tagp, ARCH_CONVERT));
- if (INT_GET(prevdup->freetag, ARCH_CONVERT) != XFS_DIR2_DATA_FREE_TAG)
+ tagp = (__be16 *)((char *)d + offset) - 1;
+ prevdup = (xfs_dir2_data_unused_t *)((char *)d + be16_to_cpu(*tagp));
+ if (be16_to_cpu(prevdup->freetag) != XFS_DIR2_DATA_FREE_TAG)
prevdup = NULL;
} else
prevdup = NULL;
@@ -562,7 +564,7 @@ xfs_dir2_data_make_free(
if ((char *)d + offset + len < endptr) {
postdup =
(xfs_dir2_data_unused_t *)((char *)d + offset + len);
- if (INT_GET(postdup->freetag, ARCH_CONVERT) != XFS_DIR2_DATA_FREE_TAG)
+ if (be16_to_cpu(postdup->freetag) != XFS_DIR2_DATA_FREE_TAG)
postdup = NULL;
} else
postdup = NULL;
@@ -586,13 +588,13 @@ xfs_dir2_data_make_free(
* since the third bestfree is there, there might be more
* entries.
*/
- needscan = d->hdr.bestfree[2].length;
+ needscan = (d->hdr.bestfree[2].length != 0);
/*
* Fix up the new big freespace.
*/
- INT_MOD(prevdup->length, ARCH_CONVERT, len + INT_GET(postdup->length, ARCH_CONVERT));
- INT_SET(*XFS_DIR2_DATA_UNUSED_TAG_P(prevdup), ARCH_CONVERT,
- (xfs_dir2_data_off_t)((char *)prevdup - (char *)d));
+ be16_add(&prevdup->length, len + be16_to_cpu(postdup->length));
+ *XFS_DIR2_DATA_UNUSED_TAG_P(prevdup) =
+ cpu_to_be16((char *)prevdup - (char *)d);
xfs_dir2_data_log_unused(tp, bp, prevdup);
if (!needscan) {
/*
@@ -614,7 +616,7 @@ xfs_dir2_data_make_free(
*/
dfp = xfs_dir2_data_freeinsert(d, prevdup, needlogp);
ASSERT(dfp == &d->hdr.bestfree[0]);
- ASSERT(INT_GET(dfp->length, ARCH_CONVERT) == INT_GET(prevdup->length, ARCH_CONVERT));
+ ASSERT(dfp->length == prevdup->length);
ASSERT(!dfp[1].length);
ASSERT(!dfp[2].length);
}
@@ -624,9 +626,9 @@ xfs_dir2_data_make_free(
*/
else if (prevdup) {
dfp = xfs_dir2_data_freefind(d, prevdup);
- INT_MOD(prevdup->length, ARCH_CONVERT, len);
- INT_SET(*XFS_DIR2_DATA_UNUSED_TAG_P(prevdup), ARCH_CONVERT,
- (xfs_dir2_data_off_t)((char *)prevdup - (char *)d));
+ be16_add(&prevdup->length, len);
+ *XFS_DIR2_DATA_UNUSED_TAG_P(prevdup) =
+ cpu_to_be16((char *)prevdup - (char *)d);
xfs_dir2_data_log_unused(tp, bp, prevdup);
/*
* If the previous entry was in the table, the new entry
@@ -640,8 +642,10 @@ xfs_dir2_data_make_free(
/*
* Otherwise we need a scan if the new entry is big enough.
*/
- else
- needscan = INT_GET(prevdup->length, ARCH_CONVERT) > INT_GET(d->hdr.bestfree[2].length, ARCH_CONVERT);
+ else {
+ needscan = be16_to_cpu(prevdup->length) >
+ be16_to_cpu(d->hdr.bestfree[2].length);
+ }
}
/*
* The following entry is free, merge with it.
@@ -649,10 +653,10 @@ xfs_dir2_data_make_free(
else if (postdup) {
dfp = xfs_dir2_data_freefind(d, postdup);
newdup = (xfs_dir2_data_unused_t *)((char *)d + offset);
- INT_SET(newdup->freetag, ARCH_CONVERT, XFS_DIR2_DATA_FREE_TAG);
- INT_SET(newdup->length, ARCH_CONVERT, len + INT_GET(postdup->length, ARCH_CONVERT));
- INT_SET(*XFS_DIR2_DATA_UNUSED_TAG_P(newdup), ARCH_CONVERT,
- (xfs_dir2_data_off_t)((char *)newdup - (char *)d));
+ newdup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
+ newdup->length = cpu_to_be16(len + be16_to_cpu(postdup->length));
+ *XFS_DIR2_DATA_UNUSED_TAG_P(newdup) =
+ cpu_to_be16((char *)newdup - (char *)d);
xfs_dir2_data_log_unused(tp, bp, newdup);
/*
* If the following entry was in the table, the new entry
@@ -666,18 +670,20 @@ xfs_dir2_data_make_free(
/*
* Otherwise we need a scan if the new entry is big enough.
*/
- else
- needscan = INT_GET(newdup->length, ARCH_CONVERT) > INT_GET(d->hdr.bestfree[2].length, ARCH_CONVERT);
+ else {
+ needscan = be16_to_cpu(newdup->length) >
+ be16_to_cpu(d->hdr.bestfree[2].length);
+ }
}
/*
* Neither neighbor is free. Make a new entry.
*/
else {
newdup = (xfs_dir2_data_unused_t *)((char *)d + offset);
- INT_SET(newdup->freetag, ARCH_CONVERT, XFS_DIR2_DATA_FREE_TAG);
- INT_SET(newdup->length, ARCH_CONVERT, len);
- INT_SET(*XFS_DIR2_DATA_UNUSED_TAG_P(newdup), ARCH_CONVERT,
- (xfs_dir2_data_off_t)((char *)newdup - (char *)d));
+ newdup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
+ newdup->length = cpu_to_be16(len);
+ *XFS_DIR2_DATA_UNUSED_TAG_P(newdup) =
+ cpu_to_be16((char *)newdup - (char *)d);
xfs_dir2_data_log_unused(tp, bp, newdup);
(void)xfs_dir2_data_freeinsert(d, newdup, needlogp);
}
@@ -707,18 +713,18 @@ xfs_dir2_data_use_free(
int oldlen; /* old unused entry's length */
d = bp->data;
- ASSERT(INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC ||
- INT_GET(d->hdr.magic, ARCH_CONVERT) == XFS_DIR2_BLOCK_MAGIC);
- ASSERT(INT_GET(dup->freetag, ARCH_CONVERT) == XFS_DIR2_DATA_FREE_TAG);
+ ASSERT(be32_to_cpu(d->hdr.magic) == XFS_DIR2_DATA_MAGIC ||
+ be32_to_cpu(d->hdr.magic) == XFS_DIR2_BLOCK_MAGIC);
+ ASSERT(be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG);
ASSERT(offset >= (char *)dup - (char *)d);
- ASSERT(offset + len <= (char *)dup + INT_GET(dup->length, ARCH_CONVERT) - (char *)d);
- ASSERT((char *)dup - (char *)d == INT_GET(*XFS_DIR2_DATA_UNUSED_TAG_P(dup), ARCH_CONVERT));
+ ASSERT(offset + len <= (char *)dup + be16_to_cpu(dup->length) - (char *)d);
+ ASSERT((char *)dup - (char *)d == be16_to_cpu(*XFS_DIR2_DATA_UNUSED_TAG_P(dup)));
/*
* Look up the entry in the bestfree table.
*/
dfp = xfs_dir2_data_freefind(d, dup);
- oldlen = INT_GET(dup->length, ARCH_CONVERT);
- ASSERT(dfp || oldlen <= INT_GET(d->hdr.bestfree[2].length, ARCH_CONVERT));
+ oldlen = be16_to_cpu(dup->length);
+ ASSERT(dfp || oldlen <= be16_to_cpu(d->hdr.bestfree[2].length));
/*
* Check for alignment with front and back of the entry.
*/
@@ -732,7 +738,7 @@ xfs_dir2_data_use_free(
*/
if (matchfront && matchback) {
if (dfp) {
- needscan = d->hdr.bestfree[2].offset;
+ needscan = (d->hdr.bestfree[2].offset != 0);
if (!needscan)
xfs_dir2_data_freeremove(d, dfp, needlogp);
}
@@ -743,10 +749,10 @@ xfs_dir2_data_use_free(
*/
else if (matchfront) {
newdup = (xfs_dir2_data_unused_t *)((char *)d + offset + len);
- INT_SET(newdup->freetag, ARCH_CONVERT, XFS_DIR2_DATA_FREE_TAG);
- INT_SET(newdup->length, ARCH_CONVERT, oldlen - len);
- INT_SET(*XFS_DIR2_DATA_UNUSED_TAG_P(newdup), ARCH_CONVERT,
- (xfs_dir2_data_off_t)((char *)newdup - (char *)d));
+ newdup->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
+ newdup->length = cpu_to_be16(oldlen - len);
+ *XFS_DIR2_DATA_UNUSED_TAG_P(newdup) =
+ cpu_to_be16((char *)newdup - (char *)d);
xfs_dir2_data_log_unused(tp, bp, newdup);
/*
* If it was in the table, remove it and add the new one.
@@ -755,8 +761,8 @@ xfs_dir2_data_use_free(
xfs_dir2_data_freeremove(d, dfp, needlogp);
dfp = xfs_dir2_data_freeinsert(d, newdup, needlogp);
ASSERT(dfp != NULL);
- ASSERT(INT_GET(dfp->length, ARCH_CONVERT) == INT_GET(newdup->length, ARCH_CONVERT));
- ASSERT(INT_GET(dfp->offset, ARCH_CONVERT) == (char *)newdup - (char *)d);
+ ASSERT(dfp->length == newdup->length);
+ ASSERT(be16_to_cpu(dfp->offset) == (char *)newdup - (char *)d);
/*
* If we got inserted at the last slot,
* that means we don't know if there was a better
@@ -771,10 +777,9 @@ xfs_dir2_data_use_free(
*/
else if (matchback) {
newdup = dup;
- INT_SET(newdup->length, ARCH_CONVERT, (xfs_dir2_data_off_t)
- (((char *)d + offset) - (char *)newdup));
- INT_SET(*XFS_DIR2_DATA_UNUSED_TAG_P(newdup), ARCH_CONVERT,
- (xfs_dir2_data_off_t)((char *)newdup - (char *)d));
+ newdup->length = cpu_to_be16(((char *)d + offset) - (char *)newdup);
+ *XFS_DIR2_DATA_UNUSED_TAG_P(newdup) =
+ cpu_to_be16((char *)newdup - (char *)d);
xfs_dir2_data_log_unused(tp, bp, newdup);
/*
* If it was in the table, remove it and add the new one.
@@ -783,8 +788,8 @@ xfs_dir2_data_use_free(
xfs_dir2_data_freeremove(d, dfp, needlogp);
dfp = xfs_dir2_data_freeinsert(d, newdup, needlogp);
ASSERT(dfp != NULL);
- ASSERT(INT_GET(dfp->length, ARCH_CONVERT) == INT_GET(newdup->length, ARCH_CONVERT));
- ASSERT(INT_GET(dfp->offset, ARCH_CONVERT) == (char *)newdup - (char *)d);
+ ASSERT(dfp->length == newdup->length);
+ ASSERT(be16_to_cpu(dfp->offset) == (char *)newdup - (char *)d);
/*
* If we got inserted at the last slot,
* that means we don't know if there was a better
@@ -799,16 +804,15 @@ xfs_dir2_data_use_free(
*/
else {
newdup = dup;
- INT_SET(newdup->length, ARCH_CONVERT, (xfs_dir2_data_off_t)
- (((char *)d + offset) - (char *)newdup));
- INT_SET(*XFS_DIR2_DATA_UNUSED_TAG_P(newdup), ARCH_CONVERT,
- (xfs_dir2_data_off_t)((char *)newdup - (char *)d));
+ newdup->length = cpu_to_be16(((char *)d + offset) - (char *)newdup);
+ *XFS_DIR2_DATA_UNUSED_TAG_P(newdup) =
+ cpu_to_be16((char *)newdup - (char *)d);
xfs_dir2_data_log_unused(tp, bp, newdup);
newdup2 = (xfs_dir2_data_unused_t *)((char *)d + offset + len);
- INT_SET(newdup2->freetag, ARCH_CONVERT, XFS_DIR2_DATA_FREE_TAG);
- INT_SET(newdup2->length, ARCH_CONVERT, oldlen - len - INT_GET(newdup->length, ARCH_CONVERT));
- INT_SET(*XFS_DIR2_DATA_UNUSED_TAG_P(newdup2), ARCH_CONVERT,
- (xfs_dir2_data_off_t)((char *)newdup2 - (char *)d));
+ newdup2->freetag = cpu_to_be16(XFS_DIR2_DATA_FREE_TAG);
+ newdup2->length = cpu_to_be16(oldlen - len - be16_to_cpu(newdup->length));
+ *XFS_DIR2_DATA_UNUSED_TAG_P(newdup2) =
+ cpu_to_be16((char *)newdup2 - (char *)d);
xfs_dir2_data_log_unused(tp, bp, newdup2);
/*
* If the old entry was in the table, we need to scan
@@ -819,7 +823,7 @@ xfs_dir2_data_use_free(
* the 2 new will work.
*/
if (dfp) {
- needscan = d->hdr.bestfree[2].length;
+ needscan = (d->hdr.bestfree[2].length != 0);
if (!needscan) {
xfs_dir2_data_freeremove(d, dfp, needlogp);
(void)xfs_dir2_data_freeinsert(d, newdup,
diff --git a/fs/xfs/xfs_dir2_data.h b/fs/xfs/xfs_dir2_data.h
index 5e3a7f9ec735..0847cbb53e17 100644
--- a/fs/xfs/xfs_dir2_data.h
+++ b/fs/xfs/xfs_dir2_data.h
@@ -65,8 +65,8 @@ struct xfs_trans;
* The freespace will be formatted as a xfs_dir2_data_unused_t.
*/
typedef struct xfs_dir2_data_free {
- xfs_dir2_data_off_t offset; /* start of freespace */
- xfs_dir2_data_off_t length; /* length of freespace */
+ __be16 offset; /* start of freespace */
+ __be16 length; /* length of freespace */
} xfs_dir2_data_free_t;
/*
@@ -75,7 +75,7 @@ typedef struct xfs_dir2_data_free {
* The code knows that XFS_DIR2_DATA_FD_COUNT is 3.
*/
typedef struct xfs_dir2_data_hdr {
- __uint32_t magic; /* XFS_DIR2_DATA_MAGIC */
+ __be32 magic; /* XFS_DIR2_DATA_MAGIC */
/* or XFS_DIR2_BLOCK_MAGIC */
xfs_dir2_data_free_t bestfree[XFS_DIR2_DATA_FD_COUNT];
} xfs_dir2_data_hdr_t;
@@ -97,10 +97,10 @@ typedef struct xfs_dir2_data_entry {
* Tag appears as the last 2 bytes.
*/
typedef struct xfs_dir2_data_unused {
- __uint16_t freetag; /* XFS_DIR2_DATA_FREE_TAG */
- xfs_dir2_data_off_t length; /* total free length */
+ __be16 freetag; /* XFS_DIR2_DATA_FREE_TAG */
+ __be16 length; /* total free length */
/* variable offset */
- xfs_dir2_data_off_t tag; /* starting offset of us */
+ __be16 tag; /* starting offset of us */
} xfs_dir2_data_unused_t;
typedef union {
@@ -134,12 +134,11 @@ static inline int xfs_dir2_data_entsize(int n)
* Pointer to an entry's tag word.
*/
#define XFS_DIR2_DATA_ENTRY_TAG_P(dep) xfs_dir2_data_entry_tag_p(dep)
-static inline xfs_dir2_data_off_t *
+static inline __be16 *
xfs_dir2_data_entry_tag_p(xfs_dir2_data_entry_t *dep)
{
- return (xfs_dir2_data_off_t *) \
- ((char *)(dep) + XFS_DIR2_DATA_ENTSIZE((dep)->namelen) - \
- (uint)sizeof(xfs_dir2_data_off_t));
+ return (__be16 *)((char *)dep +
+ XFS_DIR2_DATA_ENTSIZE(dep->namelen) - sizeof(__be16));
}
/*
@@ -147,12 +146,11 @@ xfs_dir2_data_entry_tag_p(xfs_dir2_data_entry_t *dep)
*/
#define XFS_DIR2_DATA_UNUSED_TAG_P(dup) \
xfs_dir2_data_unused_tag_p(dup)
-static inline xfs_dir2_data_off_t *
+static inline __be16 *
xfs_dir2_data_unused_tag_p(xfs_dir2_data_unused_t *dup)
{
- return (xfs_dir2_data_off_t *) \
- ((char *)(dup) + INT_GET((dup)->length, ARCH_CONVERT) \
- - (uint)sizeof(xfs_dir2_data_off_t));
+ return (__be16 *)((char *)dup +
+ be16_to_cpu(dup->length) - sizeof(__be16));
}
/*
diff --git a/fs/xfs/xfs_dir2_leaf.c b/fs/xfs/xfs_dir2_leaf.c
index d342b6b55239..08648b18265c 100644
--- a/fs/xfs/xfs_dir2_leaf.c
+++ b/fs/xfs/xfs_dir2_leaf.c
@@ -66,7 +66,7 @@ xfs_dir2_block_to_leaf(
xfs_da_args_t *args, /* operation arguments */
xfs_dabuf_t *dbp) /* input block's buffer */
{
- xfs_dir2_data_off_t *bestsp; /* leaf's bestsp entries */
+ __be16 *bestsp; /* leaf's bestsp entries */
xfs_dablk_t blkno; /* leaf block's bno */
xfs_dir2_block_t *block; /* block structure */
xfs_dir2_leaf_entry_t *blp; /* block's leaf entries */
@@ -111,14 +111,14 @@ xfs_dir2_block_to_leaf(
/*
* Set the counts in the leaf header.
*/
- INT_COPY(leaf->hdr.count, btp->count, ARCH_CONVERT); /* INT_: type change */
- INT_COPY(leaf->hdr.stale, btp->stale, ARCH_CONVERT); /* INT_: type change */
+ leaf->hdr.count = cpu_to_be16(be32_to_cpu(btp->count));
+ leaf->hdr.stale = cpu_to_be16(be32_to_cpu(btp->stale));
/*
* Could compact these but I think we always do the conversion
* after squeezing out stale entries.
*/
- memcpy(leaf->ents, blp, INT_GET(btp->count, ARCH_CONVERT) * sizeof(xfs_dir2_leaf_entry_t));
- xfs_dir2_leaf_log_ents(tp, lbp, 0, INT_GET(leaf->hdr.count, ARCH_CONVERT) - 1);
+ memcpy(leaf->ents, blp, be32_to_cpu(btp->count) * sizeof(xfs_dir2_leaf_entry_t));
+ xfs_dir2_leaf_log_ents(tp, lbp, 0, be16_to_cpu(leaf->hdr.count) - 1);
needscan = 0;
needlog = 1;
/*
@@ -133,7 +133,7 @@ xfs_dir2_block_to_leaf(
/*
* Fix up the block header, make it a data block.
*/
- INT_SET(block->hdr.magic, ARCH_CONVERT, XFS_DIR2_DATA_MAGIC);
+ block->hdr.magic = cpu_to_be32(XFS_DIR2_DATA_MAGIC);
if (needscan)
xfs_dir2_data_freescan(mp, (xfs_dir2_data_t *)block, &needlog,
NULL);
@@ -141,9 +141,9 @@ xfs_dir2_block_to_leaf(
* Set up leaf tail and bests table.
*/
ltp = XFS_DIR2_LEAF_TAIL_P(mp, leaf);
- INT_SET(ltp->bestcount, ARCH_CONVERT, 1);
+ ltp->bestcount = cpu_to_be32(1);
bestsp = XFS_DIR2_LEAF_BESTS_P(ltp);
- INT_COPY(bestsp[0], block->hdr.bestfree[0].length, ARCH_CONVERT);
+ bestsp[0] = block->hdr.bestfree[0].length;
/*
* Log the data header and leaf bests table.
*/
@@ -163,7 +163,7 @@ int /* error */
xfs_dir2_leaf_addname(
xfs_da_args_t *args) /* operation arguments */
{
- xfs_dir2_data_off_t *bestsp; /* freespace table in leaf */
+ __be16 *bestsp; /* freespace table in leaf */
int compact; /* need to compact leaves */
xfs_dir2_data_t *data; /* data block structure */
xfs_dabuf_t *dbp; /* data block buffer */
@@ -187,7 +187,7 @@ xfs_dir2_leaf_addname(
int needbytes; /* leaf block bytes needed */
int needlog; /* need to log data header */
int needscan; /* need to rescan data free */
- xfs_dir2_data_off_t *tagp; /* end of data entry */
+ __be16 *tagp; /* end of data entry */
xfs_trans_t *tp; /* transaction pointer */
xfs_dir2_db_t use_block; /* data block number */
@@ -222,14 +222,14 @@ xfs_dir2_leaf_addname(
* in a data block, improving the lookup of those entries.
*/
for (use_block = -1, lep = &leaf->ents[index];
- index < INT_GET(leaf->hdr.count, ARCH_CONVERT) && INT_GET(lep->hashval, ARCH_CONVERT) == args->hashval;
+ index < be16_to_cpu(leaf->hdr.count) && be32_to_cpu(lep->hashval) == args->hashval;
index++, lep++) {
- if (INT_GET(lep->address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR)
+ if (be32_to_cpu(lep->address) == XFS_DIR2_NULL_DATAPTR)
continue;
- i = XFS_DIR2_DATAPTR_TO_DB(mp, INT_GET(lep->address, ARCH_CONVERT));
- ASSERT(i < INT_GET(ltp->bestcount, ARCH_CONVERT));
- ASSERT(INT_GET(bestsp[i], ARCH_CONVERT) != NULLDATAOFF);
- if (INT_GET(bestsp[i], ARCH_CONVERT) >= length) {
+ i = XFS_DIR2_DATAPTR_TO_DB(mp, be32_to_cpu(lep->address));
+ ASSERT(i < be32_to_cpu(ltp->bestcount));
+ ASSERT(be16_to_cpu(bestsp[i]) != NULLDATAOFF);
+ if (be16_to_cpu(bestsp[i]) >= length) {
use_block = i;
break;
}
@@ -238,13 +238,13 @@ xfs_dir2_leaf_addname(
* Didn't find a block yet, linear search all the data blocks.
*/
if (use_block == -1) {
- for (i = 0; i < INT_GET(ltp->bestcount, ARCH_CONVERT); i++) {
+ for (i = 0; i < be32_to_cpu(ltp->bestcount); i++) {
/*
* Remember a block we see that's missing.
*/
- if (INT_GET(bestsp[i], ARCH_CONVERT) == NULLDATAOFF && use_block == -1)
+ if (be16_to_cpu(bestsp[i]) == NULLDATAOFF && use_block == -1)
use_block = i;
- else if (INT_GET(bestsp[i], ARCH_CONVERT) >= length) {
+ else if (be16_to_cpu(bestsp[i]) >= length) {
use_block = i;
break;
}
@@ -260,21 +260,21 @@ xfs_dir2_leaf_addname(
* Now kill use_block if it refers to a missing block, so we
* can use it as an indication of allocation needed.
*/
- if (use_block != -1 && INT_GET(bestsp[use_block], ARCH_CONVERT) == NULLDATAOFF)
+ if (use_block != -1 && be16_to_cpu(bestsp[use_block]) == NULLDATAOFF)
use_block = -1;
/*
* If we don't have enough free bytes but we can make enough
* by compacting out stale entries, we'll do that.
*/
- if ((char *)bestsp - (char *)&leaf->ents[INT_GET(leaf->hdr.count, ARCH_CONVERT)] < needbytes &&
- INT_GET(leaf->hdr.stale, ARCH_CONVERT) > 1) {
+ if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] < needbytes &&
+ be16_to_cpu(leaf->hdr.stale) > 1) {
compact = 1;
}
/*
* Otherwise if we don't have enough free bytes we need to
* convert to node form.
*/
- else if ((char *)bestsp - (char *)&leaf->ents[INT_GET(leaf->hdr.count, ARCH_CONVERT)] <
+ else if ((char *)bestsp - (char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] <
needbytes) {
/*
* Just checking or no space reservation, give up.
@@ -330,8 +330,8 @@ xfs_dir2_leaf_addname(
* There are stale entries, so we'll need log-low and log-high
* impossibly bad values later.
*/
- else if (INT_GET(leaf->hdr.stale, ARCH_CONVERT)) {
- lfloglow = INT_GET(leaf->hdr.count, ARCH_CONVERT);
+ else if (be16_to_cpu(leaf->hdr.stale)) {
+ lfloglow = be16_to_cpu(leaf->hdr.count);
lfloghigh = -1;
}
/*
@@ -358,13 +358,13 @@ xfs_dir2_leaf_addname(
* If we're adding a new data block on the end we need to
* extend the bests table. Copy it up one entry.
*/
- if (use_block >= INT_GET(ltp->bestcount, ARCH_CONVERT)) {
+ if (use_block >= be32_to_cpu(ltp->bestcount)) {
bestsp--;
memmove(&bestsp[0], &bestsp[1],
- INT_GET(ltp->bestcount, ARCH_CONVERT) * sizeof(bestsp[0]));
- INT_MOD(ltp->bestcount, ARCH_CONVERT, +1);
+ be32_to_cpu(ltp->bestcount) * sizeof(bestsp[0]));
+ be32_add(&ltp->bestcount, 1);
xfs_dir2_leaf_log_tail(tp, lbp);
- xfs_dir2_leaf_log_bests(tp, lbp, 0, INT_GET(ltp->bestcount, ARCH_CONVERT) - 1);
+ xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
}
/*
* If we're filling in a previously empty block just log it.
@@ -372,7 +372,7 @@ xfs_dir2_leaf_addname(
else
xfs_dir2_leaf_log_bests(tp, lbp, use_block, use_block);
data = dbp->data;
- INT_COPY(bestsp[use_block], data->hdr.bestfree[0].length, ARCH_CONVERT);
+ bestsp[use_block] = data->hdr.bestfree[0].length;
grown = 1;
}
/*
@@ -394,8 +394,8 @@ xfs_dir2_leaf_addname(
* Point to the biggest freespace in our data block.
*/
dup = (xfs_dir2_data_unused_t *)
- ((char *)data + INT_GET(data->hdr.bestfree[0].offset, ARCH_CONVERT));
- ASSERT(INT_GET(dup->length, ARCH_CONVERT) >= length);
+ ((char *)data + be16_to_cpu(data->hdr.bestfree[0].offset));
+ ASSERT(be16_to_cpu(dup->length) >= length);
needscan = needlog = 0;
/*
* Mark the initial part of our freespace in use for the new entry.
@@ -411,7 +411,7 @@ xfs_dir2_leaf_addname(
dep->namelen = args->namelen;
memcpy(dep->name, args->name, dep->namelen);
tagp = XFS_DIR2_DATA_ENTRY_TAG_P(dep);
- INT_SET(*tagp, ARCH_CONVERT, (xfs_dir2_data_off_t)((char *)dep - (char *)data));
+ *tagp = cpu_to_be16((char *)dep - (char *)data);
/*
* Need to scan fix up the bestfree table.
*/
@@ -427,8 +427,8 @@ xfs_dir2_leaf_addname(
* If the bests table needs to be changed, do it.
* Log the change unless we've already done that.
*/
- if (INT_GET(bestsp[use_block], ARCH_CONVERT) != INT_GET(data->hdr.bestfree[0].length, ARCH_CONVERT)) {
- INT_COPY(bestsp[use_block], data->hdr.bestfree[0].length, ARCH_CONVERT);
+ if (be16_to_cpu(bestsp[use_block]) != be16_to_cpu(data->hdr.bestfree[0].length)) {
+ bestsp[use_block] = data->hdr.bestfree[0].length;
if (!grown)
xfs_dir2_leaf_log_bests(tp, lbp, use_block, use_block);
}
@@ -440,15 +440,15 @@ xfs_dir2_leaf_addname(
/*
* lep is still good as the index leaf entry.
*/
- if (index < INT_GET(leaf->hdr.count, ARCH_CONVERT))
+ if (index < be16_to_cpu(leaf->hdr.count))
memmove(lep + 1, lep,
- (INT_GET(leaf->hdr.count, ARCH_CONVERT) - index) * sizeof(*lep));
+ (be16_to_cpu(leaf->hdr.count) - index) * sizeof(*lep));
/*
* Record low and high logging indices for the leaf.
*/
lfloglow = index;
- lfloghigh = INT_GET(leaf->hdr.count, ARCH_CONVERT);
- INT_MOD(leaf->hdr.count, ARCH_CONVERT, +1);
+ lfloghigh = be16_to_cpu(leaf->hdr.count);
+ be16_add(&leaf->hdr.count, 1);
}
/*
* There are stale entries.
@@ -468,7 +468,7 @@ xfs_dir2_leaf_addname(
*/
for (lowstale = index - 1;
lowstale >= 0 &&
- INT_GET(leaf->ents[lowstale].address, ARCH_CONVERT) !=
+ be32_to_cpu(leaf->ents[lowstale].address) !=
XFS_DIR2_NULL_DATAPTR;
lowstale--)
continue;
@@ -478,8 +478,8 @@ xfs_dir2_leaf_addname(
* lowstale entry would be better.
*/
for (highstale = index;
- highstale < INT_GET(leaf->hdr.count, ARCH_CONVERT) &&
- INT_GET(leaf->ents[highstale].address, ARCH_CONVERT) !=
+ highstale < be16_to_cpu(leaf->hdr.count) &&
+ be32_to_cpu(leaf->ents[highstale].address) !=
XFS_DIR2_NULL_DATAPTR &&
(lowstale < 0 ||
index - lowstale - 1 >= highstale - index);
@@ -490,10 +490,10 @@ xfs_dir2_leaf_addname(
* If the low one is better, use it.
*/
if (lowstale >= 0 &&
- (highstale == INT_GET(leaf->hdr.count, ARCH_CONVERT) ||
+ (highstale == be16_to_cpu(leaf->hdr.count) ||
index - lowstale - 1 < highstale - index)) {
ASSERT(index - lowstale - 1 >= 0);
- ASSERT(INT_GET(leaf->ents[lowstale].address, ARCH_CONVERT) ==
+ ASSERT(be32_to_cpu(leaf->ents[lowstale].address) ==
XFS_DIR2_NULL_DATAPTR);
/*
* Copy entries up to cover the stale entry
@@ -512,7 +512,7 @@ xfs_dir2_leaf_addname(
*/
else {
ASSERT(highstale - index >= 0);
- ASSERT(INT_GET(leaf->ents[highstale].address, ARCH_CONVERT) ==
+ ASSERT(be32_to_cpu(leaf->ents[highstale].address) ==
XFS_DIR2_NULL_DATAPTR);
/*
* Copy entries down to copver the stale entry
@@ -526,13 +526,14 @@ xfs_dir2_leaf_addname(
lfloglow = MIN(index, lfloglow);
lfloghigh = MAX(highstale, lfloghigh);
}
- INT_MOD(leaf->hdr.stale, ARCH_CONVERT, -1);
+ be16_add(&leaf->hdr.stale, -1);
}
/*
* Fill in the new leaf entry.
*/
- INT_SET(lep->hashval, ARCH_CONVERT, args->hashval);
- INT_SET(lep->address, ARCH_CONVERT, XFS_DIR2_DB_OFF_TO_DATAPTR(mp, use_block, INT_GET(*tagp, ARCH_CONVERT)));
+ lep->hashval = cpu_to_be32(args->hashval);
+ lep->address = cpu_to_be32(XFS_DIR2_DB_OFF_TO_DATAPTR(mp, use_block,
+ be16_to_cpu(*tagp)));
/*
* Log the leaf fields and give up the buffers.
*/
@@ -563,30 +564,30 @@ xfs_dir2_leaf_check(
leaf = bp->data;
mp = dp->i_mount;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAF1_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC);
/*
* This value is not restrictive enough.
* Should factor in the size of the bests table as well.
* We can deduce a value for that from di_size.
*/
- ASSERT(INT_GET(leaf->hdr.count, ARCH_CONVERT) <= XFS_DIR2_MAX_LEAF_ENTS(mp));
+ ASSERT(be16_to_cpu(leaf->hdr.count) <= XFS_DIR2_MAX_LEAF_ENTS(mp));
ltp = XFS_DIR2_LEAF_TAIL_P(mp, leaf);
/*
* Leaves and bests don't overlap.
*/
- ASSERT((char *)&leaf->ents[INT_GET(leaf->hdr.count, ARCH_CONVERT)] <=
+ ASSERT((char *)&leaf->ents[be16_to_cpu(leaf->hdr.count)] <=
(char *)XFS_DIR2_LEAF_BESTS_P(ltp));
/*
* Check hash value order, count stale entries.
*/
- for (i = stale = 0; i < INT_GET(leaf->hdr.count, ARCH_CONVERT); i++) {
- if (i + 1 < INT_GET(leaf->hdr.count, ARCH_CONVERT))
- ASSERT(INT_GET(leaf->ents[i].hashval, ARCH_CONVERT) <=
- INT_GET(leaf->ents[i + 1].hashval, ARCH_CONVERT));
- if (INT_GET(leaf->ents[i].address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR)
+ for (i = stale = 0; i < be16_to_cpu(leaf->hdr.count); i++) {
+ if (i + 1 < be16_to_cpu(leaf->hdr.count))
+ ASSERT(be32_to_cpu(leaf->ents[i].hashval) <=
+ be32_to_cpu(leaf->ents[i + 1].hashval));
+ if (be32_to_cpu(leaf->ents[i].address) == XFS_DIR2_NULL_DATAPTR)
stale++;
}
- ASSERT(INT_GET(leaf->hdr.stale, ARCH_CONVERT) == stale);
+ ASSERT(be16_to_cpu(leaf->hdr.stale) == stale);
}
#endif /* DEBUG */
@@ -611,8 +612,8 @@ xfs_dir2_leaf_compact(
/*
* Compress out the stale entries in place.
*/
- for (from = to = 0, loglow = -1; from < INT_GET(leaf->hdr.count, ARCH_CONVERT); from++) {
- if (INT_GET(leaf->ents[from].address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR)
+ for (from = to = 0, loglow = -1; from < be16_to_cpu(leaf->hdr.count); from++) {
+ if (be32_to_cpu(leaf->ents[from].address) == XFS_DIR2_NULL_DATAPTR)
continue;
/*
* Only actually copy the entries that are different.
@@ -627,8 +628,8 @@ xfs_dir2_leaf_compact(
/*
* Update and log the header, log the leaf entries.
*/
- ASSERT(INT_GET(leaf->hdr.stale, ARCH_CONVERT) == from - to);
- INT_MOD(leaf->hdr.count, ARCH_CONVERT, -(INT_GET(leaf->hdr.stale, ARCH_CONVERT)));
+ ASSERT(be16_to_cpu(leaf->hdr.stale) == from - to);
+ be16_add(&leaf->hdr.count, -(be16_to_cpu(leaf->hdr.stale)));
leaf->hdr.stale = 0;
xfs_dir2_leaf_log_header(args->trans, bp);
if (loglow != -1)
@@ -662,14 +663,14 @@ xfs_dir2_leaf_compact_x1(
int to; /* destination copy index */
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.stale, ARCH_CONVERT) > 1);
+ ASSERT(be16_to_cpu(leaf->hdr.stale) > 1);
index = *indexp;
/*
* Find the first stale entry before our index, if any.
*/
for (lowstale = index - 1;
lowstale >= 0 &&
- INT_GET(leaf->ents[lowstale].address, ARCH_CONVERT) != XFS_DIR2_NULL_DATAPTR;
+ be32_to_cpu(leaf->ents[lowstale].address) != XFS_DIR2_NULL_DATAPTR;
lowstale--)
continue;
/*
@@ -677,8 +678,8 @@ xfs_dir2_leaf_compact_x1(
* Stop if the answer would be worse than lowstale.
*/
for (highstale = index;
- highstale < INT_GET(leaf->hdr.count, ARCH_CONVERT) &&
- INT_GET(leaf->ents[highstale].address, ARCH_CONVERT) != XFS_DIR2_NULL_DATAPTR &&
+ highstale < be16_to_cpu(leaf->hdr.count) &&
+ be32_to_cpu(leaf->ents[highstale].address) != XFS_DIR2_NULL_DATAPTR &&
(lowstale < 0 || index - lowstale > highstale - index);
highstale++)
continue;
@@ -686,7 +687,7 @@ xfs_dir2_leaf_compact_x1(
* Pick the better of lowstale and highstale.
*/
if (lowstale >= 0 &&
- (highstale == INT_GET(leaf->hdr.count, ARCH_CONVERT) ||
+ (highstale == be16_to_cpu(leaf->hdr.count) ||
index - lowstale <= highstale - index))
keepstale = lowstale;
else
@@ -695,14 +696,14 @@ xfs_dir2_leaf_compact_x1(
* Copy the entries in place, removing all the stale entries
* except keepstale.
*/
- for (from = to = 0; from < INT_GET(leaf->hdr.count, ARCH_CONVERT); from++) {
+ for (from = to = 0; from < be16_to_cpu(leaf->hdr.count); from++) {
/*
* Notice the new value of index.
*/
if (index == from)
newindex = to;
if (from != keepstale &&
- INT_GET(leaf->ents[from].address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR) {
+ be32_to_cpu(leaf->ents[from].address) == XFS_DIR2_NULL_DATAPTR) {
if (from == to)
*lowlogp = to;
continue;
@@ -730,8 +731,8 @@ xfs_dir2_leaf_compact_x1(
/*
* Adjust the leaf header values.
*/
- INT_MOD(leaf->hdr.count, ARCH_CONVERT, -(from - to));
- INT_SET(leaf->hdr.stale, ARCH_CONVERT, 1);
+ be16_add(&leaf->hdr.count, -(from - to));
+ leaf->hdr.stale = cpu_to_be16(1);
/*
* Remember the low/high stale value only in the "right"
* direction.
@@ -739,8 +740,8 @@ xfs_dir2_leaf_compact_x1(
if (lowstale >= newindex)
lowstale = -1;
else
- highstale = INT_GET(leaf->hdr.count, ARCH_CONVERT);
- *highlogp = INT_GET(leaf->hdr.count, ARCH_CONVERT) - 1;
+ highstale = be16_to_cpu(leaf->hdr.count);
+ *highlogp = be16_to_cpu(leaf->hdr.count) - 1;
*lowstalep = lowstale;
*highstalep = highstale;
}
@@ -766,7 +767,7 @@ xfs_dir2_leaf_getdents(
xfs_dir2_data_entry_t *dep; /* data entry */
xfs_dir2_data_unused_t *dup; /* unused entry */
int eof; /* reached end of directory */
- int error=0; /* error return value */
+ int error = 0; /* error return value */
int i; /* temporary loop index */
int j; /* temporary loop index */
int length; /* temporary length value */
@@ -778,8 +779,8 @@ xfs_dir2_leaf_getdents(
xfs_mount_t *mp; /* filesystem mount point */
xfs_dir2_off_t newoff; /* new curoff after new blk */
int nmap; /* mappings to ask xfs_bmapi */
- xfs_dir2_put_args_t p; /* formatting arg bundle */
- char *ptr=NULL; /* pointer to current data */
+ xfs_dir2_put_args_t *p; /* formatting arg bundle */
+ char *ptr = NULL; /* pointer to current data */
int ra_current; /* number of read-ahead blks */
int ra_index; /* *map index for read-ahead */
int ra_offset; /* map entry offset for ra */
@@ -797,9 +798,10 @@ xfs_dir2_leaf_getdents(
/*
* Setup formatting arguments.
*/
- p.dbp = dbp;
- p.put = put;
- p.uio = uio;
+ p = kmem_alloc(sizeof(*p), KM_SLEEP);
+ p->dbp = dbp;
+ p->put = put;
+ p->uio = uio;
/*
* Set up to bmap a number of blocks based on the caller's
* buffer size, the directory block size, and the filesystem
@@ -1046,11 +1048,10 @@ xfs_dir2_leaf_getdents(
while ((char *)ptr - (char *)data < byteoff) {
dup = (xfs_dir2_data_unused_t *)ptr;
- if (INT_GET(dup->freetag, ARCH_CONVERT)
+ if (be16_to_cpu(dup->freetag)
== XFS_DIR2_DATA_FREE_TAG) {
- length = INT_GET(dup->length,
- ARCH_CONVERT);
+ length = be16_to_cpu(dup->length);
ptr += length;
continue;
}
@@ -1079,9 +1080,8 @@ xfs_dir2_leaf_getdents(
/*
* No, it's unused, skip over it.
*/
- if (INT_GET(dup->freetag, ARCH_CONVERT)
- == XFS_DIR2_DATA_FREE_TAG) {
- length = INT_GET(dup->length, ARCH_CONVERT);
+ if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
+ length = be16_to_cpu(dup->length);
ptr += length;
curoff += length;
continue;
@@ -1092,24 +1092,24 @@ xfs_dir2_leaf_getdents(
*/
dep = (xfs_dir2_data_entry_t *)ptr;
- p.namelen = dep->namelen;
+ p->namelen = dep->namelen;
- length = XFS_DIR2_DATA_ENTSIZE(p.namelen);
+ length = XFS_DIR2_DATA_ENTSIZE(p->namelen);
- p.cook = XFS_DIR2_BYTE_TO_DATAPTR(mp, curoff + length);
+ p->cook = XFS_DIR2_BYTE_TO_DATAPTR(mp, curoff + length);
- p.ino = INT_GET(dep->inumber, ARCH_CONVERT);
+ p->ino = INT_GET(dep->inumber, ARCH_CONVERT);
#if XFS_BIG_INUMS
- p.ino += mp->m_inoadd;
+ p->ino += mp->m_inoadd;
#endif
- p.name = (char *)dep->name;
+ p->name = (char *)dep->name;
- error = p.put(&p);
+ error = p->put(p);
/*
* Won't fit. Return to caller.
*/
- if (!p.done) {
+ if (!p->done) {
eof = 0;
break;
}
@@ -1129,6 +1129,7 @@ xfs_dir2_leaf_getdents(
else
uio->uio_offset = XFS_DIR2_BYTE_TO_DATAPTR(mp, curoff);
kmem_free(map, map_size * sizeof(*map));
+ kmem_free(p, sizeof(*p));
if (bp)
xfs_da_brelse(tp, bp);
return error;
@@ -1171,7 +1172,7 @@ xfs_dir2_leaf_init(
/*
* Initialize the header.
*/
- INT_SET(leaf->hdr.info.magic, ARCH_CONVERT, magic);
+ leaf->hdr.info.magic = cpu_to_be16(magic);
leaf->hdr.info.forw = 0;
leaf->hdr.info.back = 0;
leaf->hdr.count = 0;
@@ -1201,13 +1202,13 @@ xfs_dir2_leaf_log_bests(
int first, /* first entry to log */
int last) /* last entry to log */
{
- xfs_dir2_data_off_t *firstb; /* pointer to first entry */
- xfs_dir2_data_off_t *lastb; /* pointer to last entry */
+ __be16 *firstb; /* pointer to first entry */
+ __be16 *lastb; /* pointer to last entry */
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_tail_t *ltp; /* leaf tail structure */
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAF1_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC);
ltp = XFS_DIR2_LEAF_TAIL_P(tp->t_mountp, leaf);
firstb = XFS_DIR2_LEAF_BESTS_P(ltp) + first;
lastb = XFS_DIR2_LEAF_BESTS_P(ltp) + last;
@@ -1230,8 +1231,8 @@ xfs_dir2_leaf_log_ents(
xfs_dir2_leaf_t *leaf; /* leaf structure */
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAF1_MAGIC ||
- INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC ||
+ be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
firstlep = &leaf->ents[first];
lastlep = &leaf->ents[last];
xfs_da_log_buf(tp, bp, (uint)((char *)firstlep - (char *)leaf),
@@ -1249,8 +1250,8 @@ xfs_dir2_leaf_log_header(
xfs_dir2_leaf_t *leaf; /* leaf structure */
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAF1_MAGIC ||
- INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC ||
+ be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
xfs_da_log_buf(tp, bp, (uint)((char *)&leaf->hdr - (char *)leaf),
(uint)(sizeof(leaf->hdr) - 1));
}
@@ -1269,7 +1270,7 @@ xfs_dir2_leaf_log_tail(
mp = tp->t_mountp;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAF1_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAF1_MAGIC);
ltp = XFS_DIR2_LEAF_TAIL_P(mp, leaf);
xfs_da_log_buf(tp, bp, (uint)((char *)ltp - (char *)leaf),
(uint)(mp->m_dirblksize - 1));
@@ -1314,7 +1315,7 @@ xfs_dir2_leaf_lookup(
*/
dep = (xfs_dir2_data_entry_t *)
((char *)dbp->data +
- XFS_DIR2_DATAPTR_TO_OFF(dp->i_mount, INT_GET(lep->address, ARCH_CONVERT)));
+ XFS_DIR2_DATAPTR_TO_OFF(dp->i_mount, be32_to_cpu(lep->address)));
/*
* Return the found inode number.
*/
@@ -1373,17 +1374,17 @@ xfs_dir2_leaf_lookup_int(
* looking to match the name.
*/
for (lep = &leaf->ents[index], dbp = NULL, curdb = -1;
- index < INT_GET(leaf->hdr.count, ARCH_CONVERT) && INT_GET(lep->hashval, ARCH_CONVERT) == args->hashval;
+ index < be16_to_cpu(leaf->hdr.count) && be32_to_cpu(lep->hashval) == args->hashval;
lep++, index++) {
/*
* Skip over stale leaf entries.
*/
- if (INT_GET(lep->address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR)
+ if (be32_to_cpu(lep->address) == XFS_DIR2_NULL_DATAPTR)
continue;
/*
* Get the new data block number.
*/
- newdb = XFS_DIR2_DATAPTR_TO_DB(mp, INT_GET(lep->address, ARCH_CONVERT));
+ newdb = XFS_DIR2_DATAPTR_TO_DB(mp, be32_to_cpu(lep->address));
/*
* If it's not the same as the old data block number,
* need to pitch the old one and read the new one.
@@ -1406,7 +1407,7 @@ xfs_dir2_leaf_lookup_int(
*/
dep = (xfs_dir2_data_entry_t *)
((char *)dbp->data +
- XFS_DIR2_DATAPTR_TO_OFF(mp, INT_GET(lep->address, ARCH_CONVERT)));
+ XFS_DIR2_DATAPTR_TO_OFF(mp, be32_to_cpu(lep->address)));
/*
* If it matches then return it.
*/
@@ -1435,7 +1436,7 @@ int /* error */
xfs_dir2_leaf_removename(
xfs_da_args_t *args) /* operation arguments */
{
- xfs_dir2_data_off_t *bestsp; /* leaf block best freespace */
+ __be16 *bestsp; /* leaf block best freespace */
xfs_dir2_data_t *data; /* data block structure */
xfs_dir2_db_t db; /* data block number */
xfs_dabuf_t *dbp; /* data block buffer */
@@ -1471,14 +1472,14 @@ xfs_dir2_leaf_removename(
* Point to the leaf entry, use that to point to the data entry.
*/
lep = &leaf->ents[index];
- db = XFS_DIR2_DATAPTR_TO_DB(mp, INT_GET(lep->address, ARCH_CONVERT));
+ db = XFS_DIR2_DATAPTR_TO_DB(mp, be32_to_cpu(lep->address));
dep = (xfs_dir2_data_entry_t *)
- ((char *)data + XFS_DIR2_DATAPTR_TO_OFF(mp, INT_GET(lep->address, ARCH_CONVERT)));
+ ((char *)data + XFS_DIR2_DATAPTR_TO_OFF(mp, be32_to_cpu(lep->address)));
needscan = needlog = 0;
- oldbest = INT_GET(data->hdr.bestfree[0].length, ARCH_CONVERT);
+ oldbest = be16_to_cpu(data->hdr.bestfree[0].length);
ltp = XFS_DIR2_LEAF_TAIL_P(mp, leaf);
bestsp = XFS_DIR2_LEAF_BESTS_P(ltp);
- ASSERT(INT_GET(bestsp[db], ARCH_CONVERT) == oldbest);
+ ASSERT(be16_to_cpu(bestsp[db]) == oldbest);
/*
* Mark the former data entry unused.
*/
@@ -1488,9 +1489,9 @@ xfs_dir2_leaf_removename(
/*
* We just mark the leaf entry stale by putting a null in it.
*/
- INT_MOD(leaf->hdr.stale, ARCH_CONVERT, +1);
+ be16_add(&leaf->hdr.stale, 1);
xfs_dir2_leaf_log_header(tp, lbp);
- INT_SET(lep->address, ARCH_CONVERT, XFS_DIR2_NULL_DATAPTR);
+ lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
xfs_dir2_leaf_log_ents(tp, lbp, index, index);
/*
* Scan the freespace in the data block again if necessary,
@@ -1504,15 +1505,15 @@ xfs_dir2_leaf_removename(
* If the longest freespace in the data block has changed,
* put the new value in the bests table and log that.
*/
- if (INT_GET(data->hdr.bestfree[0].length, ARCH_CONVERT) != oldbest) {
- INT_COPY(bestsp[db], data->hdr.bestfree[0].length, ARCH_CONVERT);
+ if (be16_to_cpu(data->hdr.bestfree[0].length) != oldbest) {
+ bestsp[db] = data->hdr.bestfree[0].length;
xfs_dir2_leaf_log_bests(tp, lbp, db, db);
}
xfs_dir2_data_check(dp, dbp);
/*
* If the data block is now empty then get rid of the data block.
*/
- if (INT_GET(data->hdr.bestfree[0].length, ARCH_CONVERT) ==
+ if (be16_to_cpu(data->hdr.bestfree[0].length) ==
mp->m_dirblksize - (uint)sizeof(data->hdr)) {
ASSERT(db != mp->m_dirdatablk);
if ((error = xfs_dir2_shrink_inode(args, db, dbp))) {
@@ -1535,12 +1536,12 @@ xfs_dir2_leaf_removename(
* If this is the last data block then compact the
* bests table by getting rid of entries.
*/
- if (db == INT_GET(ltp->bestcount, ARCH_CONVERT) - 1) {
+ if (db == be32_to_cpu(ltp->bestcount) - 1) {
/*
* Look for the last active entry (i).
*/
for (i = db - 1; i > 0; i--) {
- if (INT_GET(bestsp[i], ARCH_CONVERT) != NULLDATAOFF)
+ if (be16_to_cpu(bestsp[i]) != NULLDATAOFF)
break;
}
/*
@@ -1548,12 +1549,12 @@ xfs_dir2_leaf_removename(
* end are removed.
*/
memmove(&bestsp[db - i], bestsp,
- (INT_GET(ltp->bestcount, ARCH_CONVERT) - (db - i)) * sizeof(*bestsp));
- INT_MOD(ltp->bestcount, ARCH_CONVERT, -(db - i));
+ (be32_to_cpu(ltp->bestcount) - (db - i)) * sizeof(*bestsp));
+ be32_add(&ltp->bestcount, -(db - i));
xfs_dir2_leaf_log_tail(tp, lbp);
- xfs_dir2_leaf_log_bests(tp, lbp, 0, INT_GET(ltp->bestcount, ARCH_CONVERT) - 1);
+ xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
} else
- INT_SET(bestsp[db], ARCH_CONVERT, NULLDATAOFF);
+ bestsp[db] = cpu_to_be16(NULLDATAOFF);
}
/*
* If the data block was not the first one, drop it.
@@ -1604,7 +1605,7 @@ xfs_dir2_leaf_replace(
*/
dep = (xfs_dir2_data_entry_t *)
((char *)dbp->data +
- XFS_DIR2_DATAPTR_TO_OFF(dp->i_mount, INT_GET(lep->address, ARCH_CONVERT)));
+ XFS_DIR2_DATAPTR_TO_OFF(dp->i_mount, be32_to_cpu(lep->address)));
ASSERT(args->inumber != INT_GET(dep->inumber, ARCH_CONVERT));
/*
* Put the new inode number in, log it.
@@ -1645,11 +1646,11 @@ xfs_dir2_leaf_search_hash(
* Note, the table cannot be empty, so we have to go through the loop.
* Binary search the leaf entries looking for our hash value.
*/
- for (lep = leaf->ents, low = 0, high = INT_GET(leaf->hdr.count, ARCH_CONVERT) - 1,
+ for (lep = leaf->ents, low = 0, high = be16_to_cpu(leaf->hdr.count) - 1,
hashwant = args->hashval;
low <= high; ) {
mid = (low + high) >> 1;
- if ((hash = INT_GET(lep[mid].hashval, ARCH_CONVERT)) == hashwant)
+ if ((hash = be32_to_cpu(lep[mid].hashval)) == hashwant)
break;
if (hash < hashwant)
low = mid + 1;
@@ -1660,7 +1661,7 @@ xfs_dir2_leaf_search_hash(
* Found one, back up through all the equal hash values.
*/
if (hash == hashwant) {
- while (mid > 0 && INT_GET(lep[mid - 1].hashval, ARCH_CONVERT) == hashwant) {
+ while (mid > 0 && be32_to_cpu(lep[mid - 1].hashval) == hashwant) {
mid--;
}
}
@@ -1682,7 +1683,7 @@ xfs_dir2_leaf_trim_data(
xfs_dabuf_t *lbp, /* leaf buffer */
xfs_dir2_db_t db) /* data block number */
{
- xfs_dir2_data_off_t *bestsp; /* leaf bests table */
+ __be16 *bestsp; /* leaf bests table */
#ifdef DEBUG
xfs_dir2_data_t *data; /* data block structure */
#endif
@@ -1706,7 +1707,7 @@ xfs_dir2_leaf_trim_data(
}
#ifdef DEBUG
data = dbp->data;
- ASSERT(INT_GET(data->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC);
+ ASSERT(be32_to_cpu(data->hdr.magic) == XFS_DIR2_DATA_MAGIC);
#endif
/* this seems to be an error
* data is only valid if DEBUG is defined?
@@ -1715,9 +1716,9 @@ xfs_dir2_leaf_trim_data(
leaf = lbp->data;
ltp = XFS_DIR2_LEAF_TAIL_P(mp, leaf);
- ASSERT(INT_GET(data->hdr.bestfree[0].length, ARCH_CONVERT) ==
+ ASSERT(be16_to_cpu(data->hdr.bestfree[0].length) ==
mp->m_dirblksize - (uint)sizeof(data->hdr));
- ASSERT(db == INT_GET(ltp->bestcount, ARCH_CONVERT) - 1);
+ ASSERT(db == be32_to_cpu(ltp->bestcount) - 1);
/*
* Get rid of the data block.
*/
@@ -1730,10 +1731,10 @@ xfs_dir2_leaf_trim_data(
* Eliminate the last bests entry from the table.
*/
bestsp = XFS_DIR2_LEAF_BESTS_P(ltp);
- INT_MOD(ltp->bestcount, ARCH_CONVERT, -1);
- memmove(&bestsp[1], &bestsp[0], INT_GET(ltp->bestcount, ARCH_CONVERT) * sizeof(*bestsp));
+ be32_add(&ltp->bestcount, -1);
+ memmove(&bestsp[1], &bestsp[0], be32_to_cpu(ltp->bestcount) * sizeof(*bestsp));
xfs_dir2_leaf_log_tail(tp, lbp);
- xfs_dir2_leaf_log_bests(tp, lbp, 0, INT_GET(ltp->bestcount, ARCH_CONVERT) - 1);
+ xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
return 0;
}
@@ -1805,7 +1806,7 @@ xfs_dir2_node_to_leaf(
return 0;
lbp = state->path.blk[0].bp;
leaf = lbp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
/*
* Read the freespace block.
*/
@@ -1814,15 +1815,15 @@ xfs_dir2_node_to_leaf(
return error;
}
free = fbp->data;
- ASSERT(INT_GET(free->hdr.magic, ARCH_CONVERT) == XFS_DIR2_FREE_MAGIC);
+ ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
ASSERT(!free->hdr.firstdb);
/*
* Now see if the leafn and free data will fit in a leaf1.
* If not, release the buffer and give up.
*/
if ((uint)sizeof(leaf->hdr) +
- (INT_GET(leaf->hdr.count, ARCH_CONVERT) - INT_GET(leaf->hdr.stale, ARCH_CONVERT)) * (uint)sizeof(leaf->ents[0]) +
- INT_GET(free->hdr.nvalid, ARCH_CONVERT) * (uint)sizeof(leaf->bests[0]) +
+ (be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale)) * (uint)sizeof(leaf->ents[0]) +
+ be32_to_cpu(free->hdr.nvalid) * (uint)sizeof(leaf->bests[0]) +
(uint)sizeof(leaf->tail) >
mp->m_dirblksize) {
xfs_da_brelse(tp, fbp);
@@ -1832,22 +1833,22 @@ xfs_dir2_node_to_leaf(
* If the leaf has any stale entries in it, compress them out.
* The compact routine will log the header.
*/
- if (INT_GET(leaf->hdr.stale, ARCH_CONVERT))
+ if (be16_to_cpu(leaf->hdr.stale))
xfs_dir2_leaf_compact(args, lbp);
else
xfs_dir2_leaf_log_header(tp, lbp);
- INT_SET(leaf->hdr.info.magic, ARCH_CONVERT, XFS_DIR2_LEAF1_MAGIC);
+ leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAF1_MAGIC);
/*
* Set up the leaf tail from the freespace block.
*/
ltp = XFS_DIR2_LEAF_TAIL_P(mp, leaf);
- INT_COPY(ltp->bestcount, free->hdr.nvalid, ARCH_CONVERT);
+ ltp->bestcount = free->hdr.nvalid;
/*
* Set up the leaf bests table.
*/
memcpy(XFS_DIR2_LEAF_BESTS_P(ltp), free->bests,
- INT_GET(ltp->bestcount, ARCH_CONVERT) * sizeof(leaf->bests[0]));
- xfs_dir2_leaf_log_bests(tp, lbp, 0, INT_GET(ltp->bestcount, ARCH_CONVERT) - 1);
+ be32_to_cpu(ltp->bestcount) * sizeof(leaf->bests[0]));
+ xfs_dir2_leaf_log_bests(tp, lbp, 0, be32_to_cpu(ltp->bestcount) - 1);
xfs_dir2_leaf_log_tail(tp, lbp);
xfs_dir2_leaf_check(dp, lbp);
/*
diff --git a/fs/xfs/xfs_dir2_leaf.h b/fs/xfs/xfs_dir2_leaf.h
index 1393993d61e9..f57ca1162412 100644
--- a/fs/xfs/xfs_dir2_leaf.h
+++ b/fs/xfs/xfs_dir2_leaf.h
@@ -46,23 +46,23 @@ typedef __uint32_t xfs_dir2_dataptr_t;
*/
typedef struct xfs_dir2_leaf_hdr {
xfs_da_blkinfo_t info; /* header for da routines */
- __uint16_t count; /* count of entries */
- __uint16_t stale; /* count of stale entries */
+ __be16 count; /* count of entries */
+ __be16 stale; /* count of stale entries */
} xfs_dir2_leaf_hdr_t;
/*
* Leaf block entry.
*/
typedef struct xfs_dir2_leaf_entry {
- xfs_dahash_t hashval; /* hash value of name */
- xfs_dir2_dataptr_t address; /* address of data entry */
+ __be32 hashval; /* hash value of name */
+ __be32 address; /* address of data entry */
} xfs_dir2_leaf_entry_t;
/*
* Leaf block tail.
*/
typedef struct xfs_dir2_leaf_tail {
- __uint32_t bestcount;
+ __be32 bestcount;
} xfs_dir2_leaf_tail_t;
/*
@@ -105,11 +105,10 @@ xfs_dir2_leaf_tail_p(struct xfs_mount *mp, xfs_dir2_leaf_t *lp)
* Get address of the bests array in the single-leaf block.
*/
#define XFS_DIR2_LEAF_BESTS_P(ltp) xfs_dir2_leaf_bests_p(ltp)
-static inline xfs_dir2_data_off_t *
+static inline __be16 *
xfs_dir2_leaf_bests_p(xfs_dir2_leaf_tail_t *ltp)
{
- return (xfs_dir2_data_off_t *)
- (ltp) - INT_GET((ltp)->bestcount, ARCH_CONVERT);
+ return (__be16 *)ltp - be32_to_cpu(ltp->bestcount);
}
/*
diff --git a/fs/xfs/xfs_dir2_node.c b/fs/xfs/xfs_dir2_node.c
index 641f8633d254..af556f16a0c7 100644
--- a/fs/xfs/xfs_dir2_node.c
+++ b/fs/xfs/xfs_dir2_node.c
@@ -76,7 +76,7 @@ xfs_dir2_free_log_bests(
xfs_dir2_free_t *free; /* freespace structure */
free = bp->data;
- ASSERT(INT_GET(free->hdr.magic, ARCH_CONVERT) == XFS_DIR2_FREE_MAGIC);
+ ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
xfs_da_log_buf(tp, bp,
(uint)((char *)&free->bests[first] - (char *)free),
(uint)((char *)&free->bests[last] - (char *)free +
@@ -94,7 +94,7 @@ xfs_dir2_free_log_header(
xfs_dir2_free_t *free; /* freespace structure */
free = bp->data;
- ASSERT(INT_GET(free->hdr.magic, ARCH_CONVERT) == XFS_DIR2_FREE_MAGIC);
+ ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
xfs_da_log_buf(tp, bp, (uint)((char *)&free->hdr - (char *)free),
(uint)(sizeof(xfs_dir2_free_hdr_t) - 1));
}
@@ -114,14 +114,14 @@ xfs_dir2_leaf_to_node(
xfs_dabuf_t *fbp; /* freespace buffer */
xfs_dir2_db_t fdb; /* freespace block number */
xfs_dir2_free_t *free; /* freespace structure */
- xfs_dir2_data_off_t *from; /* pointer to freespace entry */
+ __be16 *from; /* pointer to freespace entry */
int i; /* leaf freespace index */
xfs_dir2_leaf_t *leaf; /* leaf structure */
xfs_dir2_leaf_tail_t *ltp; /* leaf tail structure */
xfs_mount_t *mp; /* filesystem mount point */
int n; /* count of live freespc ents */
xfs_dir2_data_off_t off; /* freespace entry value */
- xfs_dir2_data_off_t *to; /* pointer to freespace entry */
+ __be16 *to; /* pointer to freespace entry */
xfs_trans_t *tp; /* transaction pointer */
xfs_dir2_trace_args_b("leaf_to_node", args, lbp);
@@ -149,28 +149,28 @@ xfs_dir2_leaf_to_node(
/*
* Initialize the freespace block header.
*/
- INT_SET(free->hdr.magic, ARCH_CONVERT, XFS_DIR2_FREE_MAGIC);
+ free->hdr.magic = cpu_to_be32(XFS_DIR2_FREE_MAGIC);
free->hdr.firstdb = 0;
- ASSERT(INT_GET(ltp->bestcount, ARCH_CONVERT) <= (uint)dp->i_d.di_size / mp->m_dirblksize);
- INT_COPY(free->hdr.nvalid, ltp->bestcount, ARCH_CONVERT);
+ ASSERT(be32_to_cpu(ltp->bestcount) <= (uint)dp->i_d.di_size / mp->m_dirblksize);
+ free->hdr.nvalid = ltp->bestcount;
/*
* Copy freespace entries from the leaf block to the new block.
* Count active entries.
*/
for (i = n = 0, from = XFS_DIR2_LEAF_BESTS_P(ltp), to = free->bests;
- i < INT_GET(ltp->bestcount, ARCH_CONVERT); i++, from++, to++) {
- if ((off = INT_GET(*from, ARCH_CONVERT)) != NULLDATAOFF)
+ i < be32_to_cpu(ltp->bestcount); i++, from++, to++) {
+ if ((off = be16_to_cpu(*from)) != NULLDATAOFF)
n++;
- INT_SET(*to, ARCH_CONVERT, off);
+ *to = cpu_to_be16(off);
}
- INT_SET(free->hdr.nused, ARCH_CONVERT, n);
- INT_SET(leaf->hdr.info.magic, ARCH_CONVERT, XFS_DIR2_LEAFN_MAGIC);
+ free->hdr.nused = cpu_to_be32(n);
+ leaf->hdr.info.magic = cpu_to_be16(XFS_DIR2_LEAFN_MAGIC);
/*
* Log everything.
*/
xfs_dir2_leaf_log_header(tp, lbp);
xfs_dir2_free_log_header(tp, fbp);
- xfs_dir2_free_log_bests(tp, fbp, 0, INT_GET(free->hdr.nvalid, ARCH_CONVERT) - 1);
+ xfs_dir2_free_log_bests(tp, fbp, 0, be32_to_cpu(free->hdr.nvalid) - 1);
xfs_da_buf_done(fbp);
xfs_dir2_leafn_check(dp, lbp);
return 0;
@@ -217,15 +217,15 @@ xfs_dir2_leafn_add(
* a compact.
*/
- if (INT_GET(leaf->hdr.count, ARCH_CONVERT) == XFS_DIR2_MAX_LEAF_ENTS(mp)) {
+ if (be16_to_cpu(leaf->hdr.count) == XFS_DIR2_MAX_LEAF_ENTS(mp)) {
if (!leaf->hdr.stale)
return XFS_ERROR(ENOSPC);
- compact = INT_GET(leaf->hdr.stale, ARCH_CONVERT) > 1;
+ compact = be16_to_cpu(leaf->hdr.stale) > 1;
} else
compact = 0;
- ASSERT(index == 0 || INT_GET(leaf->ents[index - 1].hashval, ARCH_CONVERT) <= args->hashval);
- ASSERT(index == INT_GET(leaf->hdr.count, ARCH_CONVERT) ||
- INT_GET(leaf->ents[index].hashval, ARCH_CONVERT) >= args->hashval);
+ ASSERT(index == 0 || be32_to_cpu(leaf->ents[index - 1].hashval) <= args->hashval);
+ ASSERT(index == be16_to_cpu(leaf->hdr.count) ||
+ be32_to_cpu(leaf->ents[index].hashval) >= args->hashval);
if (args->justcheck)
return 0;
@@ -242,7 +242,7 @@ xfs_dir2_leafn_add(
* Set impossible logging indices for this case.
*/
else if (leaf->hdr.stale) {
- lfloglow = INT_GET(leaf->hdr.count, ARCH_CONVERT);
+ lfloglow = be16_to_cpu(leaf->hdr.count);
lfloghigh = -1;
}
/*
@@ -250,12 +250,12 @@ xfs_dir2_leafn_add(
*/
if (!leaf->hdr.stale) {
lep = &leaf->ents[index];
- if (index < INT_GET(leaf->hdr.count, ARCH_CONVERT))
+ if (index < be16_to_cpu(leaf->hdr.count))
memmove(lep + 1, lep,
- (INT_GET(leaf->hdr.count, ARCH_CONVERT) - index) * sizeof(*lep));
+ (be16_to_cpu(leaf->hdr.count) - index) * sizeof(*lep));
lfloglow = index;
- lfloghigh = INT_GET(leaf->hdr.count, ARCH_CONVERT);
- INT_MOD(leaf->hdr.count, ARCH_CONVERT, +1);
+ lfloghigh = be16_to_cpu(leaf->hdr.count);
+ be16_add(&leaf->hdr.count, 1);
}
/*
* There are stale entries. We'll use one for the new entry.
@@ -271,7 +271,7 @@ xfs_dir2_leafn_add(
*/
for (lowstale = index - 1;
lowstale >= 0 &&
- INT_GET(leaf->ents[lowstale].address, ARCH_CONVERT) !=
+ be32_to_cpu(leaf->ents[lowstale].address) !=
XFS_DIR2_NULL_DATAPTR;
lowstale--)
continue;
@@ -281,8 +281,8 @@ xfs_dir2_leafn_add(
* lowstale already found.
*/
for (highstale = index;
- highstale < INT_GET(leaf->hdr.count, ARCH_CONVERT) &&
- INT_GET(leaf->ents[highstale].address, ARCH_CONVERT) !=
+ highstale < be16_to_cpu(leaf->hdr.count) &&
+ be32_to_cpu(leaf->ents[highstale].address) !=
XFS_DIR2_NULL_DATAPTR &&
(lowstale < 0 ||
index - lowstale - 1 >= highstale - index);
@@ -294,9 +294,9 @@ xfs_dir2_leafn_add(
* Shift entries up toward the stale slot.
*/
if (lowstale >= 0 &&
- (highstale == INT_GET(leaf->hdr.count, ARCH_CONVERT) ||
+ (highstale == be16_to_cpu(leaf->hdr.count) ||
index - lowstale - 1 < highstale - index)) {
- ASSERT(INT_GET(leaf->ents[lowstale].address, ARCH_CONVERT) ==
+ ASSERT(be32_to_cpu(leaf->ents[lowstale].address) ==
XFS_DIR2_NULL_DATAPTR);
ASSERT(index - lowstale - 1 >= 0);
if (index - lowstale - 1 > 0)
@@ -312,7 +312,7 @@ xfs_dir2_leafn_add(
* Shift entries down toward the stale slot.
*/
else {
- ASSERT(INT_GET(leaf->ents[highstale].address, ARCH_CONVERT) ==
+ ASSERT(be32_to_cpu(leaf->ents[highstale].address) ==
XFS_DIR2_NULL_DATAPTR);
ASSERT(highstale - index >= 0);
if (highstale - index > 0)
@@ -323,13 +323,14 @@ xfs_dir2_leafn_add(
lfloglow = MIN(index, lfloglow);
lfloghigh = MAX(highstale, lfloghigh);
}
- INT_MOD(leaf->hdr.stale, ARCH_CONVERT, -1);
+ be16_add(&leaf->hdr.stale, -1);
}
/*
* Insert the new entry, log everything.
*/
- INT_SET(lep->hashval, ARCH_CONVERT, args->hashval);
- INT_SET(lep->address, ARCH_CONVERT, XFS_DIR2_DB_OFF_TO_DATAPTR(mp, args->blkno, args->index));
+ lep->hashval = cpu_to_be32(args->hashval);
+ lep->address = cpu_to_be32(XFS_DIR2_DB_OFF_TO_DATAPTR(mp,
+ args->blkno, args->index));
xfs_dir2_leaf_log_header(tp, bp);
xfs_dir2_leaf_log_ents(tp, bp, lfloglow, lfloghigh);
xfs_dir2_leafn_check(dp, bp);
@@ -352,17 +353,17 @@ xfs_dir2_leafn_check(
leaf = bp->data;
mp = dp->i_mount;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
- ASSERT(INT_GET(leaf->hdr.count, ARCH_CONVERT) <= XFS_DIR2_MAX_LEAF_ENTS(mp));
- for (i = stale = 0; i < INT_GET(leaf->hdr.count, ARCH_CONVERT); i++) {
- if (i + 1 < INT_GET(leaf->hdr.count, ARCH_CONVERT)) {
- ASSERT(INT_GET(leaf->ents[i].hashval, ARCH_CONVERT) <=
- INT_GET(leaf->ents[i + 1].hashval, ARCH_CONVERT));
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.count) <= XFS_DIR2_MAX_LEAF_ENTS(mp));
+ for (i = stale = 0; i < be16_to_cpu(leaf->hdr.count); i++) {
+ if (i + 1 < be16_to_cpu(leaf->hdr.count)) {
+ ASSERT(be32_to_cpu(leaf->ents[i].hashval) <=
+ be32_to_cpu(leaf->ents[i + 1].hashval));
}
- if (INT_GET(leaf->ents[i].address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR)
+ if (be32_to_cpu(leaf->ents[i].address) == XFS_DIR2_NULL_DATAPTR)
stale++;
}
- ASSERT(INT_GET(leaf->hdr.stale, ARCH_CONVERT) == stale);
+ ASSERT(be16_to_cpu(leaf->hdr.stale) == stale);
}
#endif /* DEBUG */
@@ -378,12 +379,12 @@ xfs_dir2_leafn_lasthash(
xfs_dir2_leaf_t *leaf; /* leaf structure */
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
if (count)
- *count = INT_GET(leaf->hdr.count, ARCH_CONVERT);
+ *count = be16_to_cpu(leaf->hdr.count);
if (!leaf->hdr.count)
return 0;
- return INT_GET(leaf->ents[INT_GET(leaf->hdr.count, ARCH_CONVERT) - 1].hashval, ARCH_CONVERT);
+ return be32_to_cpu(leaf->ents[be16_to_cpu(leaf->hdr.count) - 1].hashval);
}
/*
@@ -419,9 +420,9 @@ xfs_dir2_leafn_lookup_int(
tp = args->trans;
mp = dp->i_mount;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
#ifdef __KERNEL__
- ASSERT(INT_GET(leaf->hdr.count, ARCH_CONVERT) > 0);
+ ASSERT(be16_to_cpu(leaf->hdr.count) > 0);
#endif
xfs_dir2_leafn_check(dp, bp);
/*
@@ -443,7 +444,7 @@ xfs_dir2_leafn_lookup_int(
curdb = -1;
length = XFS_DIR2_DATA_ENTSIZE(args->namelen);
if ((free = (curbp ? curbp->data : NULL)))
- ASSERT(INT_GET(free->hdr.magic, ARCH_CONVERT) == XFS_DIR2_FREE_MAGIC);
+ ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
}
/*
* For others, it's a data block buffer, get the block number.
@@ -456,17 +457,17 @@ xfs_dir2_leafn_lookup_int(
* Loop over leaf entries with the right hash value.
*/
for (lep = &leaf->ents[index];
- index < INT_GET(leaf->hdr.count, ARCH_CONVERT) && INT_GET(lep->hashval, ARCH_CONVERT) == args->hashval;
+ index < be16_to_cpu(leaf->hdr.count) && be32_to_cpu(lep->hashval) == args->hashval;
lep++, index++) {
/*
* Skip stale leaf entries.
*/
- if (INT_GET(lep->address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR)
+ if (be32_to_cpu(lep->address) == XFS_DIR2_NULL_DATAPTR)
continue;
/*
* Pull the data block number from the entry.
*/
- newdb = XFS_DIR2_DATAPTR_TO_DB(mp, INT_GET(lep->address, ARCH_CONVERT));
+ newdb = XFS_DIR2_DATAPTR_TO_DB(mp, be32_to_cpu(lep->address));
/*
* For addname, we're looking for a place to put the new entry.
* We want to use a data block with an entry of equal
@@ -506,15 +507,15 @@ xfs_dir2_leafn_lookup_int(
}
curfdb = newfdb;
free = curbp->data;
- ASSERT(INT_GET(free->hdr.magic, ARCH_CONVERT) ==
+ ASSERT(be32_to_cpu(free->hdr.magic) ==
XFS_DIR2_FREE_MAGIC);
- ASSERT((INT_GET(free->hdr.firstdb, ARCH_CONVERT) %
+ ASSERT((be32_to_cpu(free->hdr.firstdb) %
XFS_DIR2_MAX_FREE_BESTS(mp)) ==
0);
- ASSERT(INT_GET(free->hdr.firstdb, ARCH_CONVERT) <= curdb);
+ ASSERT(be32_to_cpu(free->hdr.firstdb) <= curdb);
ASSERT(curdb <
- INT_GET(free->hdr.firstdb, ARCH_CONVERT) +
- INT_GET(free->hdr.nvalid, ARCH_CONVERT));
+ be32_to_cpu(free->hdr.firstdb) +
+ be32_to_cpu(free->hdr.nvalid));
}
/*
* Get the index for our entry.
@@ -523,12 +524,12 @@ xfs_dir2_leafn_lookup_int(
/*
* If it has room, return it.
*/
- if (unlikely(INT_GET(free->bests[fi], ARCH_CONVERT) == NULLDATAOFF)) {
+ if (unlikely(be16_to_cpu(free->bests[fi]) == NULLDATAOFF)) {
XFS_ERROR_REPORT("xfs_dir2_leafn_lookup_int",
XFS_ERRLEVEL_LOW, mp);
return XFS_ERROR(EFSCORRUPTED);
}
- if (INT_GET(free->bests[fi], ARCH_CONVERT) >= length) {
+ if (be16_to_cpu(free->bests[fi]) >= length) {
*indexp = index;
state->extravalid = 1;
state->extrablk.bp = curbp;
@@ -572,7 +573,7 @@ xfs_dir2_leafn_lookup_int(
*/
dep = (xfs_dir2_data_entry_t *)
((char *)curbp->data +
- XFS_DIR2_DATAPTR_TO_OFF(mp, INT_GET(lep->address, ARCH_CONVERT)));
+ XFS_DIR2_DATAPTR_TO_OFF(mp, be32_to_cpu(lep->address)));
/*
* Compare the entry, return it if it matches.
*/
@@ -619,7 +620,7 @@ xfs_dir2_leafn_lookup_int(
* Return the final index, that will be the insertion point.
*/
*indexp = index;
- ASSERT(index == INT_GET(leaf->hdr.count, ARCH_CONVERT) || args->oknoent);
+ ASSERT(index == be16_to_cpu(leaf->hdr.count) || args->oknoent);
return XFS_ERROR(ENOENT);
}
@@ -657,12 +658,12 @@ xfs_dir2_leafn_moveents(
* destination leaf entries, open up a hole in the destination
* to hold the new entries.
*/
- if (start_d < INT_GET(leaf_d->hdr.count, ARCH_CONVERT)) {
+ if (start_d < be16_to_cpu(leaf_d->hdr.count)) {
memmove(&leaf_d->ents[start_d + count], &leaf_d->ents[start_d],
- (INT_GET(leaf_d->hdr.count, ARCH_CONVERT) - start_d) *
+ (be16_to_cpu(leaf_d->hdr.count) - start_d) *
sizeof(xfs_dir2_leaf_entry_t));
xfs_dir2_leaf_log_ents(tp, bp_d, start_d + count,
- count + INT_GET(leaf_d->hdr.count, ARCH_CONVERT) - 1);
+ count + be16_to_cpu(leaf_d->hdr.count) - 1);
}
/*
* If the source has stale leaves, count the ones in the copy range
@@ -672,7 +673,7 @@ xfs_dir2_leafn_moveents(
int i; /* temp leaf index */
for (i = start_s, stale = 0; i < start_s + count; i++) {
- if (INT_GET(leaf_s->ents[i].address, ARCH_CONVERT) == XFS_DIR2_NULL_DATAPTR)
+ if (be32_to_cpu(leaf_s->ents[i].address) == XFS_DIR2_NULL_DATAPTR)
stale++;
}
} else
@@ -687,7 +688,7 @@ xfs_dir2_leafn_moveents(
* If there are source entries after the ones we copied,
* delete the ones we copied by sliding the next ones down.
*/
- if (start_s + count < INT_GET(leaf_s->hdr.count, ARCH_CONVERT)) {
+ if (start_s + count < be16_to_cpu(leaf_s->hdr.count)) {
memmove(&leaf_s->ents[start_s], &leaf_s->ents[start_s + count],
count * sizeof(xfs_dir2_leaf_entry_t));
xfs_dir2_leaf_log_ents(tp, bp_s, start_s, start_s + count - 1);
@@ -695,10 +696,10 @@ xfs_dir2_leafn_moveents(
/*
* Update the headers and log them.
*/
- INT_MOD(leaf_s->hdr.count, ARCH_CONVERT, -(count));
- INT_MOD(leaf_s->hdr.stale, ARCH_CONVERT, -(stale));
- INT_MOD(leaf_d->hdr.count, ARCH_CONVERT, count);
- INT_MOD(leaf_d->hdr.stale, ARCH_CONVERT, stale);
+ be16_add(&leaf_s->hdr.count, -(count));
+ be16_add(&leaf_s->hdr.stale, -(stale));
+ be16_add(&leaf_d->hdr.count, count);
+ be16_add(&leaf_d->hdr.stale, stale);
xfs_dir2_leaf_log_header(tp, bp_s);
xfs_dir2_leaf_log_header(tp, bp_d);
xfs_dir2_leafn_check(args->dp, bp_s);
@@ -719,13 +720,13 @@ xfs_dir2_leafn_order(
leaf1 = leaf1_bp->data;
leaf2 = leaf2_bp->data;
- ASSERT(INT_GET(leaf1->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
- ASSERT(INT_GET(leaf2->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
- if (INT_GET(leaf1->hdr.count, ARCH_CONVERT) > 0 &&
- INT_GET(leaf2->hdr.count, ARCH_CONVERT) > 0 &&
- (INT_GET(leaf2->ents[0].hashval, ARCH_CONVERT) < INT_GET(leaf1->ents[0].hashval, ARCH_CONVERT) ||
- INT_GET(leaf2->ents[INT_GET(leaf2->hdr.count, ARCH_CONVERT) - 1].hashval, ARCH_CONVERT) <
- INT_GET(leaf1->ents[INT_GET(leaf1->hdr.count, ARCH_CONVERT) - 1].hashval, ARCH_CONVERT)))
+ ASSERT(be16_to_cpu(leaf1->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(leaf2->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ if (be16_to_cpu(leaf1->hdr.count) > 0 &&
+ be16_to_cpu(leaf2->hdr.count) > 0 &&
+ (be32_to_cpu(leaf2->ents[0].hashval) < be32_to_cpu(leaf1->ents[0].hashval) ||
+ be32_to_cpu(leaf2->ents[be16_to_cpu(leaf2->hdr.count) - 1].hashval) <
+ be32_to_cpu(leaf1->ents[be16_to_cpu(leaf1->hdr.count) - 1].hashval)))
return 1;
return 0;
}
@@ -768,9 +769,9 @@ xfs_dir2_leafn_rebalance(
}
leaf1 = blk1->bp->data;
leaf2 = blk2->bp->data;
- oldsum = INT_GET(leaf1->hdr.count, ARCH_CONVERT) + INT_GET(leaf2->hdr.count, ARCH_CONVERT);
+ oldsum = be16_to_cpu(leaf1->hdr.count) + be16_to_cpu(leaf2->hdr.count);
#ifdef DEBUG
- oldstale = INT_GET(leaf1->hdr.stale, ARCH_CONVERT) + INT_GET(leaf2->hdr.stale, ARCH_CONVERT);
+ oldstale = be16_to_cpu(leaf1->hdr.stale) + be16_to_cpu(leaf2->hdr.stale);
#endif
mid = oldsum >> 1;
/*
@@ -780,10 +781,10 @@ xfs_dir2_leafn_rebalance(
if (oldsum & 1) {
xfs_dahash_t midhash; /* middle entry hash value */
- if (mid >= INT_GET(leaf1->hdr.count, ARCH_CONVERT))
- midhash = INT_GET(leaf2->ents[mid - INT_GET(leaf1->hdr.count, ARCH_CONVERT)].hashval, ARCH_CONVERT);
+ if (mid >= be16_to_cpu(leaf1->hdr.count))
+ midhash = be32_to_cpu(leaf2->ents[mid - be16_to_cpu(leaf1->hdr.count)].hashval);
else
- midhash = INT_GET(leaf1->ents[mid].hashval, ARCH_CONVERT);
+ midhash = be32_to_cpu(leaf1->ents[mid].hashval);
isleft = args->hashval <= midhash;
}
/*
@@ -797,30 +798,30 @@ xfs_dir2_leafn_rebalance(
* Calculate moved entry count. Positive means left-to-right,
* negative means right-to-left. Then move the entries.
*/
- count = INT_GET(leaf1->hdr.count, ARCH_CONVERT) - mid + (isleft == 0);
+ count = be16_to_cpu(leaf1->hdr.count) - mid + (isleft == 0);
if (count > 0)
xfs_dir2_leafn_moveents(args, blk1->bp,
- INT_GET(leaf1->hdr.count, ARCH_CONVERT) - count, blk2->bp, 0, count);
+ be16_to_cpu(leaf1->hdr.count) - count, blk2->bp, 0, count);
else if (count < 0)
xfs_dir2_leafn_moveents(args, blk2->bp, 0, blk1->bp,
- INT_GET(leaf1->hdr.count, ARCH_CONVERT), count);
- ASSERT(INT_GET(leaf1->hdr.count, ARCH_CONVERT) + INT_GET(leaf2->hdr.count, ARCH_CONVERT) == oldsum);
- ASSERT(INT_GET(leaf1->hdr.stale, ARCH_CONVERT) + INT_GET(leaf2->hdr.stale, ARCH_CONVERT) == oldstale);
+ be16_to_cpu(leaf1->hdr.count), count);
+ ASSERT(be16_to_cpu(leaf1->hdr.count) + be16_to_cpu(leaf2->hdr.count) == oldsum);
+ ASSERT(be16_to_cpu(leaf1->hdr.stale) + be16_to_cpu(leaf2->hdr.stale) == oldstale);
/*
* Mark whether we're inserting into the old or new leaf.
*/
- if (INT_GET(leaf1->hdr.count, ARCH_CONVERT) < INT_GET(leaf2->hdr.count, ARCH_CONVERT))
+ if (be16_to_cpu(leaf1->hdr.count) < be16_to_cpu(leaf2->hdr.count))
state->inleaf = swap;
- else if (INT_GET(leaf1->hdr.count, ARCH_CONVERT) > INT_GET(leaf2->hdr.count, ARCH_CONVERT))
+ else if (be16_to_cpu(leaf1->hdr.count) > be16_to_cpu(leaf2->hdr.count))
state->inleaf = !swap;
else
state->inleaf =
- swap ^ (blk1->index <= INT_GET(leaf1->hdr.count, ARCH_CONVERT));
+ swap ^ (blk1->index <= be16_to_cpu(leaf1->hdr.count));
/*
* Adjust the expected index for insertion.
*/
if (!state->inleaf)
- blk2->index = blk1->index - INT_GET(leaf1->hdr.count, ARCH_CONVERT);
+ blk2->index = blk1->index - be16_to_cpu(leaf1->hdr.count);
/*
* Finally sanity check just to make sure we are not returning a negative index
@@ -867,7 +868,7 @@ xfs_dir2_leafn_remove(
tp = args->trans;
mp = dp->i_mount;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
/*
* Point to the entry we're removing.
*/
@@ -875,17 +876,17 @@ xfs_dir2_leafn_remove(
/*
* Extract the data block and offset from the entry.
*/
- db = XFS_DIR2_DATAPTR_TO_DB(mp, INT_GET(lep->address, ARCH_CONVERT));
+ db = XFS_DIR2_DATAPTR_TO_DB(mp, be32_to_cpu(lep->address));
ASSERT(dblk->blkno == db);
- off = XFS_DIR2_DATAPTR_TO_OFF(mp, INT_GET(lep->address, ARCH_CONVERT));
+ off = XFS_DIR2_DATAPTR_TO_OFF(mp, be32_to_cpu(lep->address));
ASSERT(dblk->index == off);
/*
* Kill the leaf entry by marking it stale.
* Log the leaf block changes.
*/
- INT_MOD(leaf->hdr.stale, ARCH_CONVERT, +1);
+ be16_add(&leaf->hdr.stale, 1);
xfs_dir2_leaf_log_header(tp, bp);
- INT_SET(lep->address, ARCH_CONVERT, XFS_DIR2_NULL_DATAPTR);
+ lep->address = cpu_to_be32(XFS_DIR2_NULL_DATAPTR);
xfs_dir2_leaf_log_ents(tp, bp, index, index);
/*
* Make the data entry free. Keep track of the longest freespace
@@ -894,7 +895,7 @@ xfs_dir2_leafn_remove(
dbp = dblk->bp;
data = dbp->data;
dep = (xfs_dir2_data_entry_t *)((char *)data + off);
- longest = INT_GET(data->hdr.bestfree[0].length, ARCH_CONVERT);
+ longest = be16_to_cpu(data->hdr.bestfree[0].length);
needlog = needscan = 0;
xfs_dir2_data_make_free(tp, dbp, off,
XFS_DIR2_DATA_ENTSIZE(dep->namelen), &needlog, &needscan);
@@ -911,7 +912,7 @@ xfs_dir2_leafn_remove(
* If the longest data block freespace changes, need to update
* the corresponding freeblock entry.
*/
- if (longest < INT_GET(data->hdr.bestfree[0].length, ARCH_CONVERT)) {
+ if (longest < be16_to_cpu(data->hdr.bestfree[0].length)) {
int error; /* error return value */
xfs_dabuf_t *fbp; /* freeblock buffer */
xfs_dir2_db_t fdb; /* freeblock block number */
@@ -929,15 +930,15 @@ xfs_dir2_leafn_remove(
return error;
}
free = fbp->data;
- ASSERT(INT_GET(free->hdr.magic, ARCH_CONVERT) == XFS_DIR2_FREE_MAGIC);
- ASSERT(INT_GET(free->hdr.firstdb, ARCH_CONVERT) ==
+ ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
+ ASSERT(be32_to_cpu(free->hdr.firstdb) ==
XFS_DIR2_MAX_FREE_BESTS(mp) *
(fdb - XFS_DIR2_FREE_FIRSTDB(mp)));
/*
* Calculate which entry we need to fix.
*/
findex = XFS_DIR2_DB_TO_FDINDEX(mp, db);
- longest = INT_GET(data->hdr.bestfree[0].length, ARCH_CONVERT);
+ longest = be16_to_cpu(data->hdr.bestfree[0].length);
/*
* If the data block is now empty we can get rid of it
* (usually).
@@ -969,7 +970,7 @@ xfs_dir2_leafn_remove(
/*
* One less used entry in the free table.
*/
- INT_MOD(free->hdr.nused, ARCH_CONVERT, -1);
+ free->hdr.nused = cpu_to_be32(-1);
xfs_dir2_free_log_header(tp, fbp);
/*
* If this was the last entry in the table, we can
@@ -977,21 +978,21 @@ xfs_dir2_leafn_remove(
* entries at the end referring to non-existent
* data blocks, get those too.
*/
- if (findex == INT_GET(free->hdr.nvalid, ARCH_CONVERT) - 1) {
+ if (findex == be32_to_cpu(free->hdr.nvalid) - 1) {
int i; /* free entry index */
for (i = findex - 1;
- i >= 0 && INT_GET(free->bests[i], ARCH_CONVERT) == NULLDATAOFF;
+ i >= 0 && be16_to_cpu(free->bests[i]) == NULLDATAOFF;
i--)
continue;
- INT_SET(free->hdr.nvalid, ARCH_CONVERT, i + 1);
+ free->hdr.nvalid = cpu_to_be32(i + 1);
logfree = 0;
}
/*
* Not the last entry, just punch it out.
*/
else {
- INT_SET(free->bests[findex], ARCH_CONVERT, NULLDATAOFF);
+ free->bests[findex] = cpu_to_be16(NULLDATAOFF);
logfree = 1;
}
/*
@@ -1017,7 +1018,7 @@ xfs_dir2_leafn_remove(
* the new value.
*/
else {
- INT_SET(free->bests[findex], ARCH_CONVERT, longest);
+ free->bests[findex] = cpu_to_be16(longest);
logfree = 1;
}
/*
@@ -1039,7 +1040,7 @@ xfs_dir2_leafn_remove(
*rval =
((uint)sizeof(leaf->hdr) +
(uint)sizeof(leaf->ents[0]) *
- (INT_GET(leaf->hdr.count, ARCH_CONVERT) - INT_GET(leaf->hdr.stale, ARCH_CONVERT))) <
+ (be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale))) <
mp->m_dir_magicpct;
return 0;
}
@@ -1138,9 +1139,9 @@ xfs_dir2_leafn_toosmall(
*/
blk = &state->path.blk[state->path.active - 1];
info = blk->bp->data;
- ASSERT(INT_GET(info->magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(info->magic) == XFS_DIR2_LEAFN_MAGIC);
leaf = (xfs_dir2_leaf_t *)info;
- count = INT_GET(leaf->hdr.count, ARCH_CONVERT) - INT_GET(leaf->hdr.stale, ARCH_CONVERT);
+ count = be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale);
bytes = (uint)sizeof(leaf->hdr) + count * (uint)sizeof(leaf->ents[0]);
if (bytes > (state->blocksize >> 1)) {
/*
@@ -1160,7 +1161,7 @@ xfs_dir2_leafn_toosmall(
* Make altpath point to the block we want to keep and
* path point to the block we want to drop (this one).
*/
- forward = info->forw;
+ forward = (info->forw != 0);
memcpy(&state->altpath, &state->path, sizeof(state->path));
error = xfs_da_path_shift(state, &state->altpath, forward, 0,
&rval);
@@ -1176,9 +1177,9 @@ xfs_dir2_leafn_toosmall(
* We prefer coalescing with the lower numbered sibling so as
* to shrink a directory over time.
*/
- forward = INT_GET(info->forw, ARCH_CONVERT) < INT_GET(info->back, ARCH_CONVERT);
+ forward = be32_to_cpu(info->forw) < be32_to_cpu(info->back);
for (i = 0, bp = NULL; i < 2; forward = !forward, i++) {
- blkno = forward ?INT_GET( info->forw, ARCH_CONVERT) : INT_GET(info->back, ARCH_CONVERT);
+ blkno = forward ? be32_to_cpu(info->forw) : be32_to_cpu(info->back);
if (blkno == 0)
continue;
/*
@@ -1194,11 +1195,11 @@ xfs_dir2_leafn_toosmall(
* Count bytes in the two blocks combined.
*/
leaf = (xfs_dir2_leaf_t *)info;
- count = INT_GET(leaf->hdr.count, ARCH_CONVERT) - INT_GET(leaf->hdr.stale, ARCH_CONVERT);
+ count = be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale);
bytes = state->blocksize - (state->blocksize >> 2);
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
- count += INT_GET(leaf->hdr.count, ARCH_CONVERT) - INT_GET(leaf->hdr.stale, ARCH_CONVERT);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ count += be16_to_cpu(leaf->hdr.count) - be16_to_cpu(leaf->hdr.stale);
bytes -= count * (uint)sizeof(leaf->ents[0]);
/*
* Fits with at least 25% to spare.
@@ -1256,27 +1257,27 @@ xfs_dir2_leafn_unbalance(
ASSERT(save_blk->magic == XFS_DIR2_LEAFN_MAGIC);
drop_leaf = drop_blk->bp->data;
save_leaf = save_blk->bp->data;
- ASSERT(INT_GET(drop_leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
- ASSERT(INT_GET(save_leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(drop_leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
+ ASSERT(be16_to_cpu(save_leaf->hdr.info.magic) == XFS_DIR2_LEAFN_MAGIC);
/*
* If there are any stale leaf entries, take this opportunity
* to purge them.
*/
- if (INT_GET(drop_leaf->hdr.stale, ARCH_CONVERT))
+ if (drop_leaf->hdr.stale)
xfs_dir2_leaf_compact(args, drop_blk->bp);
- if (INT_GET(save_leaf->hdr.stale, ARCH_CONVERT))
+ if (save_leaf->hdr.stale)
xfs_dir2_leaf_compact(args, save_blk->bp);
/*
* Move the entries from drop to the appropriate end of save.
*/
- drop_blk->hashval = INT_GET(drop_leaf->ents[INT_GET(drop_leaf->hdr.count, ARCH_CONVERT) - 1].hashval, ARCH_CONVERT);
+ drop_blk->hashval = be32_to_cpu(drop_leaf->ents[be16_to_cpu(drop_leaf->hdr.count) - 1].hashval);
if (xfs_dir2_leafn_order(save_blk->bp, drop_blk->bp))
xfs_dir2_leafn_moveents(args, drop_blk->bp, 0, save_blk->bp, 0,
- INT_GET(drop_leaf->hdr.count, ARCH_CONVERT));
+ be16_to_cpu(drop_leaf->hdr.count));
else
xfs_dir2_leafn_moveents(args, drop_blk->bp, 0, save_blk->bp,
- INT_GET(save_leaf->hdr.count, ARCH_CONVERT), INT_GET(drop_leaf->hdr.count, ARCH_CONVERT));
- save_blk->hashval = INT_GET(save_leaf->ents[INT_GET(save_leaf->hdr.count, ARCH_CONVERT) - 1].hashval, ARCH_CONVERT);
+ be16_to_cpu(save_leaf->hdr.count), be16_to_cpu(drop_leaf->hdr.count));
+ save_blk->hashval = be32_to_cpu(save_leaf->ents[be16_to_cpu(save_leaf->hdr.count) - 1].hashval);
xfs_dir2_leafn_check(args->dp, save_blk->bp);
}
@@ -1378,7 +1379,7 @@ xfs_dir2_node_addname_int(
xfs_mount_t *mp; /* filesystem mount point */
int needlog; /* need to log data header */
int needscan; /* need to rescan data frees */
- xfs_dir2_data_off_t *tagp; /* data entry tag pointer */
+ __be16 *tagp; /* data entry tag pointer */
xfs_trans_t *tp; /* transaction pointer */
dp = args->dp;
@@ -1397,7 +1398,7 @@ xfs_dir2_node_addname_int(
*/
ifbno = fblk->blkno;
free = fbp->data;
- ASSERT(INT_GET(free->hdr.magic, ARCH_CONVERT) == XFS_DIR2_FREE_MAGIC);
+ ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
findex = fblk->index;
/*
* This means the free entry showed that the data block had
@@ -1405,10 +1406,10 @@ xfs_dir2_node_addname_int(
* Use that data block.
*/
if (findex >= 0) {
- ASSERT(findex < INT_GET(free->hdr.nvalid, ARCH_CONVERT));
- ASSERT(INT_GET(free->bests[findex], ARCH_CONVERT) != NULLDATAOFF);
- ASSERT(INT_GET(free->bests[findex], ARCH_CONVERT) >= length);
- dbno = INT_GET(free->hdr.firstdb, ARCH_CONVERT) + findex;
+ ASSERT(findex < be32_to_cpu(free->hdr.nvalid));
+ ASSERT(be16_to_cpu(free->bests[findex]) != NULLDATAOFF);
+ ASSERT(be16_to_cpu(free->bests[findex]) >= length);
+ dbno = be32_to_cpu(free->hdr.firstdb) + findex;
}
/*
* The data block looked at didn't have enough room.
@@ -1481,20 +1482,20 @@ xfs_dir2_node_addname_int(
continue;
}
free = fbp->data;
- ASSERT(INT_GET(free->hdr.magic, ARCH_CONVERT) == XFS_DIR2_FREE_MAGIC);
+ ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
findex = 0;
}
/*
* Look at the current free entry. Is it good enough?
*/
- if (INT_GET(free->bests[findex], ARCH_CONVERT) != NULLDATAOFF &&
- INT_GET(free->bests[findex], ARCH_CONVERT) >= length)
- dbno = INT_GET(free->hdr.firstdb, ARCH_CONVERT) + findex;
+ if (be16_to_cpu(free->bests[findex]) != NULLDATAOFF &&
+ be16_to_cpu(free->bests[findex]) >= length)
+ dbno = be32_to_cpu(free->hdr.firstdb) + findex;
else {
/*
* Are we done with the freeblock?
*/
- if (++findex == INT_GET(free->hdr.nvalid, ARCH_CONVERT)) {
+ if (++findex == be32_to_cpu(free->hdr.nvalid)) {
/*
* Drop the block.
*/
@@ -1608,15 +1609,15 @@ xfs_dir2_node_addname_int(
* its first slot as our empty slot.
*/
free = fbp->data;
- INT_SET(free->hdr.magic, ARCH_CONVERT, XFS_DIR2_FREE_MAGIC);
- INT_SET(free->hdr.firstdb, ARCH_CONVERT,
+ free->hdr.magic = cpu_to_be32(XFS_DIR2_FREE_MAGIC);
+ free->hdr.firstdb = cpu_to_be32(
(fbno - XFS_DIR2_FREE_FIRSTDB(mp)) *
XFS_DIR2_MAX_FREE_BESTS(mp));
free->hdr.nvalid = 0;
free->hdr.nused = 0;
} else {
free = fbp->data;
- ASSERT(INT_GET(free->hdr.magic, ARCH_CONVERT) == XFS_DIR2_FREE_MAGIC);
+ ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
}
/*
@@ -1627,20 +1628,20 @@ xfs_dir2_node_addname_int(
* If it's after the end of the current entries in the
* freespace block, extend that table.
*/
- if (findex >= INT_GET(free->hdr.nvalid, ARCH_CONVERT)) {
+ if (findex >= be32_to_cpu(free->hdr.nvalid)) {
ASSERT(findex < XFS_DIR2_MAX_FREE_BESTS(mp));
- INT_SET(free->hdr.nvalid, ARCH_CONVERT, findex + 1);
+ free->hdr.nvalid = cpu_to_be32(findex + 1);
/*
* Tag new entry so nused will go up.
*/
- INT_SET(free->bests[findex], ARCH_CONVERT, NULLDATAOFF);
+ free->bests[findex] = cpu_to_be16(NULLDATAOFF);
}
/*
* If this entry was for an empty data block
* (this should always be true) then update the header.
*/
- if (INT_GET(free->bests[findex], ARCH_CONVERT) == NULLDATAOFF) {
- INT_MOD(free->hdr.nused, ARCH_CONVERT, +1);
+ if (be16_to_cpu(free->bests[findex]) == NULLDATAOFF) {
+ be32_add(&free->hdr.nused, 1);
xfs_dir2_free_log_header(tp, fbp);
}
/*
@@ -1649,7 +1650,7 @@ xfs_dir2_node_addname_int(
* change again.
*/
data = dbp->data;
- INT_COPY(free->bests[findex], data->hdr.bestfree[0].length, ARCH_CONVERT);
+ free->bests[findex] = data->hdr.bestfree[0].length;
logfree = 1;
}
/*
@@ -1677,12 +1678,12 @@ xfs_dir2_node_addname_int(
data = dbp->data;
logfree = 0;
}
- ASSERT(INT_GET(data->hdr.bestfree[0].length, ARCH_CONVERT) >= length);
+ ASSERT(be16_to_cpu(data->hdr.bestfree[0].length) >= length);
/*
* Point to the existing unused space.
*/
dup = (xfs_dir2_data_unused_t *)
- ((char *)data + INT_GET(data->hdr.bestfree[0].offset, ARCH_CONVERT));
+ ((char *)data + be16_to_cpu(data->hdr.bestfree[0].offset));
needscan = needlog = 0;
/*
* Mark the first part of the unused space, inuse for us.
@@ -1698,7 +1699,7 @@ xfs_dir2_node_addname_int(
dep->namelen = args->namelen;
memcpy(dep->name, args->name, dep->namelen);
tagp = XFS_DIR2_DATA_ENTRY_TAG_P(dep);
- INT_SET(*tagp, ARCH_CONVERT, (xfs_dir2_data_off_t)((char *)dep - (char *)data));
+ *tagp = cpu_to_be16((char *)dep - (char *)data);
xfs_dir2_data_log_entry(tp, dbp, dep);
/*
* Rescan the block for bestfree if needed.
@@ -1713,8 +1714,8 @@ xfs_dir2_node_addname_int(
/*
* If the freespace entry is now wrong, update it.
*/
- if (INT_GET(free->bests[findex], ARCH_CONVERT) != INT_GET(data->hdr.bestfree[0].length, ARCH_CONVERT)) {
- INT_COPY(free->bests[findex], data->hdr.bestfree[0].length, ARCH_CONVERT);
+ if (be16_to_cpu(free->bests[findex]) != be16_to_cpu(data->hdr.bestfree[0].length)) {
+ free->bests[findex] = data->hdr.bestfree[0].length;
logfree = 1;
}
/*
@@ -1731,7 +1732,7 @@ xfs_dir2_node_addname_int(
* Return the data block and offset in args, then drop the data block.
*/
args->blkno = (xfs_dablk_t)dbno;
- args->index = INT_GET(*tagp, ARCH_CONVERT);
+ args->index = be16_to_cpu(*tagp);
xfs_da_buf_done(dbp);
return 0;
}
@@ -1900,10 +1901,10 @@ xfs_dir2_node_replace(
* Point to the data entry.
*/
data = state->extrablk.bp->data;
- ASSERT(INT_GET(data->hdr.magic, ARCH_CONVERT) == XFS_DIR2_DATA_MAGIC);
+ ASSERT(be32_to_cpu(data->hdr.magic) == XFS_DIR2_DATA_MAGIC);
dep = (xfs_dir2_data_entry_t *)
((char *)data +
- XFS_DIR2_DATAPTR_TO_OFF(state->mp, INT_GET(lep->address, ARCH_CONVERT)));
+ XFS_DIR2_DATAPTR_TO_OFF(state->mp, be32_to_cpu(lep->address)));
ASSERT(inum != INT_GET(dep->inumber, ARCH_CONVERT));
/*
* Fill in the new inode number and log the entry.
@@ -1966,11 +1967,11 @@ xfs_dir2_node_trim_free(
return 0;
}
free = bp->data;
- ASSERT(INT_GET(free->hdr.magic, ARCH_CONVERT) == XFS_DIR2_FREE_MAGIC);
+ ASSERT(be32_to_cpu(free->hdr.magic) == XFS_DIR2_FREE_MAGIC);
/*
* If there are used entries, there's nothing to do.
*/
- if (INT_GET(free->hdr.nused, ARCH_CONVERT) > 0) {
+ if (be32_to_cpu(free->hdr.nused) > 0) {
xfs_da_brelse(tp, bp);
*rvalp = 0;
return 0;
diff --git a/fs/xfs/xfs_dir2_node.h b/fs/xfs/xfs_dir2_node.h
index 0ab8fbd59512..c7c870ee7857 100644
--- a/fs/xfs/xfs_dir2_node.h
+++ b/fs/xfs/xfs_dir2_node.h
@@ -41,15 +41,15 @@ struct xfs_trans;
#define XFS_DIR2_FREE_MAGIC 0x58443246 /* XD2F */
typedef struct xfs_dir2_free_hdr {
- __uint32_t magic; /* XFS_DIR2_FREE_MAGIC */
- __int32_t firstdb; /* db of first entry */
- __int32_t nvalid; /* count of valid entries */
- __int32_t nused; /* count of used entries */
+ __be32 magic; /* XFS_DIR2_FREE_MAGIC */
+ __be32 firstdb; /* db of first entry */
+ __be32 nvalid; /* count of valid entries */
+ __be32 nused; /* count of used entries */
} xfs_dir2_free_hdr_t;
typedef struct xfs_dir2_free {
xfs_dir2_free_hdr_t hdr; /* block header */
- xfs_dir2_data_off_t bests[1]; /* best free counts */
+ __be16 bests[1]; /* best free counts */
/* unused entries are -1 */
} xfs_dir2_free_t;
diff --git a/fs/xfs/xfs_dir2_sf.c b/fs/xfs/xfs_dir2_sf.c
index ec8e7476c8b7..d98a41d1fe63 100644
--- a/fs/xfs/xfs_dir2_sf.c
+++ b/fs/xfs/xfs_dir2_sf.c
@@ -98,8 +98,8 @@ xfs_dir2_block_sfsize(
/*
* Iterate over the block's data entries by using the leaf pointers.
*/
- for (i = 0; i < INT_GET(btp->count, ARCH_CONVERT); i++) {
- if ((addr = INT_GET(blp[i].address, ARCH_CONVERT)) == XFS_DIR2_NULL_DATAPTR)
+ for (i = 0; i < be32_to_cpu(btp->count); i++) {
+ if ((addr = be32_to_cpu(blp[i].address)) == XFS_DIR2_NULL_DATAPTR)
continue;
/*
* Calculate the pointer to the entry at hand.
@@ -220,8 +220,8 @@ xfs_dir2_block_to_sf(
* If it's unused, just skip over it.
*/
dup = (xfs_dir2_data_unused_t *)ptr;
- if (INT_GET(dup->freetag, ARCH_CONVERT) == XFS_DIR2_DATA_FREE_TAG) {
- ptr += INT_GET(dup->length, ARCH_CONVERT);
+ if (be16_to_cpu(dup->freetag) == XFS_DIR2_DATA_FREE_TAG) {
+ ptr += be16_to_cpu(dup->length);
continue;
}
dep = (xfs_dir2_data_entry_t *)ptr;
diff --git a/fs/xfs/xfs_dir_leaf.c b/fs/xfs/xfs_dir_leaf.c
index e83074016abb..ee88751c3be6 100644
--- a/fs/xfs/xfs_dir_leaf.c
+++ b/fs/xfs/xfs_dir_leaf.c
@@ -176,7 +176,7 @@ xfs_dir_shortform_addname(xfs_da_args_t *args)
ASSERT(dp->i_df.if_u1.if_data != NULL);
sf = (xfs_dir_shortform_t *)dp->i_df.if_u1.if_data;
sfe = &sf->list[0];
- for (i = INT_GET(sf->hdr.count, ARCH_CONVERT)-1; i >= 0; i--) {
+ for (i = sf->hdr.count-1; i >= 0; i--) {
if (sfe->namelen == args->namelen &&
args->name[0] == sfe->name[0] &&
memcmp(args->name, sfe->name, args->namelen) == 0)
@@ -193,7 +193,7 @@ xfs_dir_shortform_addname(xfs_da_args_t *args)
XFS_DIR_SF_PUT_DIRINO(&args->inumber, &sfe->inumber);
sfe->namelen = args->namelen;
memcpy(sfe->name, args->name, sfe->namelen);
- INT_MOD(sf->hdr.count, ARCH_CONVERT, +1);
+ sf->hdr.count++;
dp->i_d.di_size += size;
xfs_trans_log_inode(args->trans, dp, XFS_ILOG_CORE | XFS_ILOG_DDATA);
@@ -227,7 +227,7 @@ xfs_dir_shortform_removename(xfs_da_args_t *args)
base = sizeof(xfs_dir_sf_hdr_t);
sf = (xfs_dir_shortform_t *)dp->i_df.if_u1.if_data;
sfe = &sf->list[0];
- for (i = INT_GET(sf->hdr.count, ARCH_CONVERT)-1; i >= 0; i--) {
+ for (i = sf->hdr.count-1; i >= 0; i--) {
size = XFS_DIR_SF_ENTSIZE_BYENTRY(sfe);
if (sfe->namelen == args->namelen &&
sfe->name[0] == args->name[0] &&
@@ -245,7 +245,7 @@ xfs_dir_shortform_removename(xfs_da_args_t *args)
memmove(&((char *)sf)[base], &((char *)sf)[base+size],
dp->i_d.di_size - (base+size));
}
- INT_MOD(sf->hdr.count, ARCH_CONVERT, -1);
+ sf->hdr.count--;
xfs_idata_realloc(dp, -size, XFS_DATA_FORK);
dp->i_d.di_size -= size;
@@ -288,7 +288,7 @@ xfs_dir_shortform_lookup(xfs_da_args_t *args)
return(XFS_ERROR(EEXIST));
}
sfe = &sf->list[0];
- for (i = INT_GET(sf->hdr.count, ARCH_CONVERT)-1; i >= 0; i--) {
+ for (i = sf->hdr.count-1; i >= 0; i--) {
if (sfe->namelen == args->namelen &&
sfe->name[0] == args->name[0] &&
memcmp(args->name, sfe->name, args->namelen) == 0) {
@@ -375,7 +375,7 @@ xfs_dir_shortform_to_leaf(xfs_da_args_t *iargs)
goto out;
sfe = &sf->list[0];
- for (i = 0; i < INT_GET(sf->hdr.count, ARCH_CONVERT); i++) {
+ for (i = 0; i < sf->hdr.count; i++) {
args.name = (char *)(sfe->name);
args.namelen = sfe->namelen;
args.hashval = xfs_da_hashname((char *)(sfe->name),
@@ -428,7 +428,7 @@ xfs_dir_shortform_getdents(xfs_inode_t *dp, uio_t *uio, int *eofp,
sf = (xfs_dir_shortform_t *)dp->i_df.if_u1.if_data;
cookhash = XFS_DA_COOKIE_HASH(mp, uio->uio_offset);
want_entno = XFS_DA_COOKIE_ENTRY(mp, uio->uio_offset);
- nsbuf = INT_GET(sf->hdr.count, ARCH_CONVERT) + 2;
+ nsbuf = sf->hdr.count + 2;
sbsize = (nsbuf + 1) * sizeof(*sbuf);
sbp = sbuf = kmem_alloc(sbsize, KM_SLEEP);
@@ -460,8 +460,7 @@ xfs_dir_shortform_getdents(xfs_inode_t *dp, uio_t *uio, int *eofp,
/*
* Scan the directory data for the rest of the entries.
*/
- for (i = 0, sfe = &sf->list[0];
- i < INT_GET(sf->hdr.count, ARCH_CONVERT); i++) {
+ for (i = 0, sfe = &sf->list[0]; i < sf->hdr.count; i++) {
if (unlikely(
((char *)sfe < (char *)sf) ||
@@ -600,7 +599,7 @@ xfs_dir_shortform_replace(xfs_da_args_t *args)
}
ASSERT(args->namelen != 1 || args->name[0] != '.');
sfe = &sf->list[0];
- for (i = INT_GET(sf->hdr.count, ARCH_CONVERT)-1; i >= 0; i--) {
+ for (i = sf->hdr.count-1; i >= 0; i--) {
if (sfe->namelen == args->namelen &&
sfe->name[0] == args->name[0] &&
memcmp(args->name, sfe->name, args->namelen) == 0) {
@@ -644,7 +643,7 @@ xfs_dir_leaf_to_shortform(xfs_da_args_t *iargs)
ASSERT(bp != NULL);
memcpy(tmpbuffer, bp->data, XFS_LBSIZE(dp->i_mount));
leaf = (xfs_dir_leafblock_t *)tmpbuffer;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
memset(bp->data, 0, XFS_LBSIZE(dp->i_mount));
/*
@@ -742,11 +741,13 @@ xfs_dir_leaf_to_node(xfs_da_args_t *args)
}
node = bp1->data;
leaf = bp2->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
- INT_SET(node->btree[0].hashval, ARCH_CONVERT, INT_GET(leaf->entries[ INT_GET(leaf->hdr.count, ARCH_CONVERT)-1 ].hashval, ARCH_CONVERT));
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
+ node->btree[0].hashval = cpu_to_be32(
+ INT_GET(leaf->entries[
+ INT_GET(leaf->hdr.count, ARCH_CONVERT)-1].hashval, ARCH_CONVERT));
xfs_da_buf_done(bp2);
- INT_SET(node->btree[0].before, ARCH_CONVERT, blkno);
- INT_SET(node->hdr.count, ARCH_CONVERT, 1);
+ node->btree[0].before = cpu_to_be32(blkno);
+ node->hdr.count = cpu_to_be16(1);
xfs_da_log_buf(args->trans, bp1,
XFS_DA_LOGRANGE(node, &node->btree[0], sizeof(node->btree[0])));
xfs_da_buf_done(bp1);
@@ -781,7 +782,7 @@ xfs_dir_leaf_create(xfs_da_args_t *args, xfs_dablk_t blkno, xfs_dabuf_t **bpp)
leaf = bp->data;
memset((char *)leaf, 0, XFS_LBSIZE(dp->i_mount));
hdr = &leaf->hdr;
- INT_SET(hdr->info.magic, ARCH_CONVERT, XFS_DIR_LEAF_MAGIC);
+ hdr->info.magic = cpu_to_be16(XFS_DIR_LEAF_MAGIC);
INT_SET(hdr->firstused, ARCH_CONVERT, XFS_LBSIZE(dp->i_mount));
if (!hdr->firstused)
INT_SET(hdr->firstused, ARCH_CONVERT, XFS_LBSIZE(dp->i_mount) - 1);
@@ -860,7 +861,7 @@ xfs_dir_leaf_add(xfs_dabuf_t *bp, xfs_da_args_t *args, int index)
int tablesize, entsize, sum, i, tmp, error;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
ASSERT((index >= 0) && (index <= INT_GET(leaf->hdr.count, ARCH_CONVERT)));
hdr = &leaf->hdr;
entsize = XFS_DIR_LEAF_ENTSIZE_BYNAME(args->namelen);
@@ -940,7 +941,7 @@ xfs_dir_leaf_add_work(xfs_dabuf_t *bp, xfs_da_args_t *args, int index,
int tmp, i;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
hdr = &leaf->hdr;
ASSERT((mapindex >= 0) && (mapindex < XFS_DIR_LEAF_MAPSIZE));
ASSERT((index >= 0) && (index <= INT_GET(hdr->count, ARCH_CONVERT)));
@@ -1097,8 +1098,8 @@ xfs_dir_leaf_rebalance(xfs_da_state_t *state, xfs_da_state_blk_t *blk1,
ASSERT(blk2->magic == XFS_DIR_LEAF_MAGIC);
leaf1 = blk1->bp->data;
leaf2 = blk2->bp->data;
- ASSERT(INT_GET(leaf1->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
- ASSERT(INT_GET(leaf2->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf1->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf2->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
/*
* Check ordering of blocks, reverse if it makes things simpler.
@@ -1325,7 +1326,7 @@ xfs_dir_leaf_toosmall(xfs_da_state_t *state, int *action)
*/
blk = &state->path.blk[ state->path.active-1 ];
info = blk->bp->data;
- ASSERT(INT_GET(info->magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(info->magic) == XFS_DIR_LEAF_MAGIC);
leaf = (xfs_dir_leafblock_t *)info;
count = INT_GET(leaf->hdr.count, ARCH_CONVERT);
bytes = (uint)sizeof(xfs_dir_leaf_hdr_t) +
@@ -1348,7 +1349,7 @@ xfs_dir_leaf_toosmall(xfs_da_state_t *state, int *action)
* Make altpath point to the block we want to keep and
* path point to the block we want to drop (this one).
*/
- forward = info->forw;
+ forward = (info->forw != 0);
memcpy(&state->altpath, &state->path, sizeof(state->path));
error = xfs_da_path_shift(state, &state->altpath, forward,
0, &retval);
@@ -1369,12 +1370,12 @@ xfs_dir_leaf_toosmall(xfs_da_state_t *state, int *action)
* We prefer coalescing with the lower numbered sibling so as
* to shrink a directory over time.
*/
- forward = (INT_GET(info->forw, ARCH_CONVERT) < INT_GET(info->back, ARCH_CONVERT)); /* start with smaller blk num */
+ forward = (be32_to_cpu(info->forw) < be32_to_cpu(info->back)); /* start with smaller blk num */
for (i = 0; i < 2; forward = !forward, i++) {
if (forward)
- blkno = INT_GET(info->forw, ARCH_CONVERT);
+ blkno = be32_to_cpu(info->forw);
else
- blkno = INT_GET(info->back, ARCH_CONVERT);
+ blkno = be32_to_cpu(info->back);
if (blkno == 0)
continue;
error = xfs_da_read_buf(state->args->trans, state->args->dp,
@@ -1389,7 +1390,7 @@ xfs_dir_leaf_toosmall(xfs_da_state_t *state, int *action)
bytes = state->blocksize - (state->blocksize>>2);
bytes -= INT_GET(leaf->hdr.namebytes, ARCH_CONVERT);
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
count += INT_GET(leaf->hdr.count, ARCH_CONVERT);
bytes -= INT_GET(leaf->hdr.namebytes, ARCH_CONVERT);
bytes -= count * ((uint)sizeof(xfs_dir_leaf_name_t) - 1);
@@ -1447,7 +1448,7 @@ xfs_dir_leaf_remove(xfs_trans_t *trans, xfs_dabuf_t *bp, int index)
xfs_mount_t *mp;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
hdr = &leaf->hdr;
mp = trans->t_mountp;
ASSERT((INT_GET(hdr->count, ARCH_CONVERT) > 0) && (INT_GET(hdr->count, ARCH_CONVERT) < (XFS_LBSIZE(mp)/8)));
@@ -1599,8 +1600,8 @@ xfs_dir_leaf_unbalance(xfs_da_state_t *state, xfs_da_state_blk_t *drop_blk,
ASSERT(save_blk->magic == XFS_DIR_LEAF_MAGIC);
drop_leaf = drop_blk->bp->data;
save_leaf = save_blk->bp->data;
- ASSERT(INT_GET(drop_leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
- ASSERT(INT_GET(save_leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(drop_leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(save_leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
drop_hdr = &drop_leaf->hdr;
save_hdr = &save_leaf->hdr;
@@ -1695,7 +1696,7 @@ xfs_dir_leaf_lookup_int(xfs_dabuf_t *bp, xfs_da_args_t *args, int *index)
xfs_dahash_t hashval;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
ASSERT(INT_GET(leaf->hdr.count, ARCH_CONVERT) < (XFS_LBSIZE(args->dp->i_mount)/8));
/*
@@ -1782,8 +1783,8 @@ xfs_dir_leaf_moveents(xfs_dir_leafblock_t *leaf_s, int start_s,
/*
* Set up environment.
*/
- ASSERT(INT_GET(leaf_s->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
- ASSERT(INT_GET(leaf_d->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf_s->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf_d->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
hdr_s = &leaf_s->hdr;
hdr_d = &leaf_d->hdr;
ASSERT((INT_GET(hdr_s->count, ARCH_CONVERT) > 0) && (INT_GET(hdr_s->count, ARCH_CONVERT) < (XFS_LBSIZE(mp)/8)));
@@ -1883,8 +1884,8 @@ xfs_dir_leaf_order(xfs_dabuf_t *leaf1_bp, xfs_dabuf_t *leaf2_bp)
leaf1 = leaf1_bp->data;
leaf2 = leaf2_bp->data;
- ASSERT((INT_GET(leaf1->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC) &&
- (INT_GET(leaf2->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC));
+ ASSERT((be16_to_cpu(leaf1->hdr.info.magic) == XFS_DIR_LEAF_MAGIC) &&
+ (be16_to_cpu(leaf2->hdr.info.magic) == XFS_DIR_LEAF_MAGIC));
if ((INT_GET(leaf1->hdr.count, ARCH_CONVERT) > 0) && (INT_GET(leaf2->hdr.count, ARCH_CONVERT) > 0) &&
((INT_GET(leaf2->entries[ 0 ].hashval, ARCH_CONVERT) <
INT_GET(leaf1->entries[ 0 ].hashval, ARCH_CONVERT)) ||
@@ -1904,7 +1905,7 @@ xfs_dir_leaf_lasthash(xfs_dabuf_t *bp, int *count)
xfs_dir_leafblock_t *leaf;
leaf = bp->data;
- ASSERT(INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) == XFS_DIR_LEAF_MAGIC);
+ ASSERT(be16_to_cpu(leaf->hdr.info.magic) == XFS_DIR_LEAF_MAGIC);
if (count)
*count = INT_GET(leaf->hdr.count, ARCH_CONVERT);
if (!leaf->hdr.count)
@@ -1940,7 +1941,7 @@ xfs_dir_leaf_getdents_int(
mp = dp->i_mount;
leaf = bp->data;
- if (INT_GET(leaf->hdr.info.magic, ARCH_CONVERT) != XFS_DIR_LEAF_MAGIC) {
+ if (be16_to_cpu(leaf->hdr.info.magic) != XFS_DIR_LEAF_MAGIC) {
*eobp = 1;
return XFS_ERROR(ENOENT); /* XXX wrong code */
}
@@ -1992,7 +1993,7 @@ xfs_dir_leaf_getdents_int(
if (i == INT_GET(leaf->hdr.count, ARCH_CONVERT)) {
xfs_dir_trace_g_du("leaf: hash not found", dp, uio);
- if (!INT_GET(leaf->hdr.info.forw, ARCH_CONVERT))
+ if (!leaf->hdr.info.forw)
uio->uio_offset =
XFS_DA_MAKE_COOKIE(mp, 0, 0, XFS_DA_MAXHASH);
/*
@@ -2047,8 +2048,7 @@ xfs_dir_leaf_getdents_int(
xfs_dir_trace_g_duc("leaf: middle cookie ",
dp, uio, p.cook.o);
- } else if ((thishash = INT_GET(leaf->hdr.info.forw,
- ARCH_CONVERT))) {
+ } else if ((thishash = be32_to_cpu(leaf->hdr.info.forw))) {
xfs_dabuf_t *bp2;
xfs_dir_leafblock_t *leaf2;
@@ -2064,9 +2064,9 @@ xfs_dir_leaf_getdents_int(
leaf2 = bp2->data;
if (unlikely(
- (INT_GET(leaf2->hdr.info.magic, ARCH_CONVERT)
+ (be16_to_cpu(leaf2->hdr.info.magic)
!= XFS_DIR_LEAF_MAGIC)
- || (INT_GET(leaf2->hdr.info.back, ARCH_CONVERT)
+ || (be32_to_cpu(leaf2->hdr.info.back)
!= bno))) { /* GROT */
XFS_CORRUPTION_ERROR("xfs_dir_leaf_getdents_int(3)",
XFS_ERRLEVEL_LOW, mp,
diff --git a/fs/xfs/xfs_dir_sf.h b/fs/xfs/xfs_dir_sf.h
index fe44c6f4d560..5b20b4d3f57d 100644
--- a/fs/xfs/xfs_dir_sf.h
+++ b/fs/xfs/xfs_dir_sf.h
@@ -35,19 +35,21 @@ typedef struct { __uint8_t i[sizeof(xfs_ino_t)]; } xfs_dir_ino_t;
* and the elements much be memcpy'd out into a work area to get correct
* alignment for the inode number fields.
*/
+typedef struct xfs_dir_sf_hdr { /* constant-structure header block */
+ xfs_dir_ino_t parent; /* parent dir inode number */
+ __uint8_t count; /* count of active entries */
+} xfs_dir_sf_hdr_t;
+
+typedef struct xfs_dir_sf_entry {
+ xfs_dir_ino_t inumber; /* referenced inode number */
+ __uint8_t namelen; /* actual length of name (no NULL) */
+ __uint8_t name[1]; /* name */
+} xfs_dir_sf_entry_t;
+
typedef struct xfs_dir_shortform {
- struct xfs_dir_sf_hdr { /* constant-structure header block */
- xfs_dir_ino_t parent; /* parent dir inode number */
- __uint8_t count; /* count of active entries */
- } hdr;
- struct xfs_dir_sf_entry {
- xfs_dir_ino_t inumber; /* referenced inode number */
- __uint8_t namelen; /* actual length of name (no NULL) */
- __uint8_t name[1]; /* name */
- } list[1]; /* variable sized array */
+ xfs_dir_sf_hdr_t hdr;
+ xfs_dir_sf_entry_t list[1]; /* variable sized array */
} xfs_dir_shortform_t;
-typedef struct xfs_dir_sf_hdr xfs_dir_sf_hdr_t;
-typedef struct xfs_dir_sf_entry xfs_dir_sf_entry_t;
/*
* We generate this then sort it, so that readdirs are returned in
diff --git a/fs/xfs/xfs_dmapi.h b/fs/xfs/xfs_dmapi.h
index b4c7f2bc55a0..00b1540f8108 100644
--- a/fs/xfs/xfs_dmapi.h
+++ b/fs/xfs/xfs_dmapi.h
@@ -191,14 +191,4 @@ typedef enum {
extern struct bhv_vfsops xfs_dmops;
-#ifdef CONFIG_XFS_DMAPI
-void xfs_dm_init(struct file_system_type *);
-void xfs_dm_exit(struct file_system_type *);
-#define XFS_DM_INIT(fstype) xfs_dm_init(fstype)
-#define XFS_DM_EXIT(fstype) xfs_dm_exit(fstype)
-#else
-#define XFS_DM_INIT(fstype)
-#define XFS_DM_EXIT(fstype)
-#endif
-
#endif /* __XFS_DMAPI_H__ */
diff --git a/fs/xfs/xfs_fsops.c b/fs/xfs/xfs_fsops.c
index b4d971b01588..56caa88713ab 100644
--- a/fs/xfs/xfs_fsops.c
+++ b/fs/xfs/xfs_fsops.c
@@ -462,6 +462,7 @@ xfs_fs_counts(
{
unsigned long s;
+ xfs_icsb_sync_counters_lazy(mp);
s = XFS_SB_LOCK(mp);
cnt->freedata = mp->m_sb.sb_fdblocks;
cnt->freertx = mp->m_sb.sb_frextents;
diff --git a/fs/xfs/xfs_ialloc.c b/fs/xfs/xfs_ialloc.c
index 8f3fae1aa98a..0024892841a3 100644
--- a/fs/xfs/xfs_ialloc.c
+++ b/fs/xfs/xfs_ialloc.c
@@ -138,8 +138,6 @@ xfs_ialloc_ag_alloc(
int version; /* inode version number to use */
int isaligned; /* inode allocation at stripe unit */
/* boundary */
- xfs_dinode_core_t dic; /* a dinode_core to copy to new */
- /* inodes */
args.tp = tp;
args.mp = tp->t_mountp;
@@ -250,10 +248,6 @@ xfs_ialloc_ag_alloc(
else
version = XFS_DINODE_VERSION_1;
- memset(&dic, 0, sizeof(xfs_dinode_core_t));
- INT_SET(dic.di_magic, ARCH_CONVERT, XFS_DINODE_MAGIC);
- INT_SET(dic.di_version, ARCH_CONVERT, version);
-
for (j = 0; j < nbufs; j++) {
/*
* Get the block.
@@ -266,12 +260,13 @@ xfs_ialloc_ag_alloc(
ASSERT(fbuf);
ASSERT(!XFS_BUF_GETERROR(fbuf));
/*
- * Loop over the inodes in this buffer.
+ * Set initial values for the inodes in this buffer.
*/
-
+ xfs_biozero(fbuf, 0, ninodes << args.mp->m_sb.sb_inodelog);
for (i = 0; i < ninodes; i++) {
free = XFS_MAKE_IPTR(args.mp, fbuf, i);
- memcpy(&(free->di_core), &dic, sizeof(xfs_dinode_core_t));
+ INT_SET(free->di_core.di_magic, ARCH_CONVERT, XFS_DINODE_MAGIC);
+ INT_SET(free->di_core.di_version, ARCH_CONVERT, version);
INT_SET(free->di_next_unlinked, ARCH_CONVERT, NULLAGINO);
xfs_ialloc_log_di(tp, fbuf, i,
XFS_DI_CORE_BITS | XFS_DI_NEXT_UNLINKED);
diff --git a/fs/xfs/xfs_iget.c b/fs/xfs/xfs_iget.c
index 8e380a1fb79b..3ce35a6f700b 100644
--- a/fs/xfs/xfs_iget.c
+++ b/fs/xfs/xfs_iget.c
@@ -258,7 +258,7 @@ again:
goto finish_inode;
} else if (vp != inode_vp) {
- struct inode *inode = LINVFS_GET_IP(inode_vp);
+ struct inode *inode = vn_to_inode(inode_vp);
/* The inode is being torn down, pause and
* try again.
@@ -495,7 +495,7 @@ retry:
if ((inode = iget_locked(XFS_MTOVFS(mp)->vfs_super, ino))) {
xfs_inode_t *ip;
- vp = LINVFS_GET_VP(inode);
+ vp = vn_from_inode(inode);
if (inode->i_state & I_NEW) {
vn_initialize(inode);
error = xfs_iget_core(vp, mp, tp, ino, flags,
@@ -617,7 +617,7 @@ xfs_iput_new(xfs_inode_t *ip,
uint lock_flags)
{
vnode_t *vp = XFS_ITOV(ip);
- struct inode *inode = LINVFS_GET_IP(vp);
+ struct inode *inode = vn_to_inode(vp);
vn_trace_entry(vp, "xfs_iput_new", (inst_t *)__return_address);
diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 1d7f5a7e063e..88a517fad07b 100644
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -76,16 +76,18 @@ STATIC int xfs_iformat_btree(xfs_inode_t *, xfs_dinode_t *, int);
*/
STATIC void
xfs_validate_extents(
- xfs_bmbt_rec_t *ep,
+ xfs_ifork_t *ifp,
int nrecs,
int disk,
xfs_exntfmt_t fmt)
{
+ xfs_bmbt_rec_t *ep;
xfs_bmbt_irec_t irec;
xfs_bmbt_rec_t rec;
int i;
for (i = 0; i < nrecs; i++) {
+ ep = xfs_iext_get_ext(ifp, i);
rec.l0 = get_unaligned((__uint64_t*)&ep->l0);
rec.l1 = get_unaligned((__uint64_t*)&ep->l1);
if (disk)
@@ -94,11 +96,10 @@ xfs_validate_extents(
xfs_bmbt_get_all(&rec, &irec);
if (fmt == XFS_EXTFMT_NOSTATE)
ASSERT(irec.br_state == XFS_EXT_NORM);
- ep++;
}
}
#else /* DEBUG */
-#define xfs_validate_extents(ep, nrecs, disk, fmt)
+#define xfs_validate_extents(ifp, nrecs, disk, fmt)
#endif /* DEBUG */
/*
@@ -252,7 +253,8 @@ xfs_itobp(
xfs_inode_t *ip,
xfs_dinode_t **dipp,
xfs_buf_t **bpp,
- xfs_daddr_t bno)
+ xfs_daddr_t bno,
+ uint imap_flags)
{
xfs_buf_t *bp;
int error;
@@ -268,10 +270,9 @@ xfs_itobp(
* inode on disk.
*/
imap.im_blkno = bno;
- error = xfs_imap(mp, tp, ip->i_ino, &imap, XFS_IMAP_LOOKUP);
- if (error != 0) {
+ if ((error = xfs_imap(mp, tp, ip->i_ino, &imap,
+ XFS_IMAP_LOOKUP | imap_flags)))
return error;
- }
/*
* If the inode number maps to a block outside the bounds
@@ -335,9 +336,10 @@ xfs_itobp(
* (if DEBUG kernel) or the first inode in the buffer, otherwise.
*/
#ifdef DEBUG
- ni = BBTOB(imap.im_len) >> mp->m_sb.sb_inodelog;
+ ni = (imap_flags & XFS_IMAP_BULKSTAT) ? 0 :
+ (BBTOB(imap.im_len) >> mp->m_sb.sb_inodelog);
#else
- ni = 1;
+ ni = (imap_flags & XFS_IMAP_BULKSTAT) ? 0 : 1;
#endif
for (i = 0; i < ni; i++) {
int di_ok;
@@ -504,7 +506,7 @@ xfs_iformat(
switch (INT_GET(dip->di_core.di_aformat, ARCH_CONVERT)) {
case XFS_DINODE_FMT_LOCAL:
atp = (xfs_attr_shortform_t *)XFS_DFORK_APTR(dip);
- size = (int)INT_GET(atp->hdr.totsize, ARCH_CONVERT);
+ size = be16_to_cpu(atp->hdr.totsize);
error = xfs_iformat_local(ip, dip, XFS_ATTR_FORK, size);
break;
case XFS_DINODE_FMT_EXTENTS:
@@ -597,7 +599,6 @@ xfs_iformat_extents(
xfs_bmbt_rec_t *ep, *dp;
xfs_ifork_t *ifp;
int nex;
- int real_size;
int size;
int i;
@@ -619,23 +620,20 @@ xfs_iformat_extents(
return XFS_ERROR(EFSCORRUPTED);
}
- real_size = 0;
+ ifp->if_real_bytes = 0;
if (nex == 0)
ifp->if_u1.if_extents = NULL;
else if (nex <= XFS_INLINE_EXTS)
ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext;
- else {
- ifp->if_u1.if_extents = kmem_alloc(size, KM_SLEEP);
- ASSERT(ifp->if_u1.if_extents != NULL);
- real_size = size;
- }
+ else
+ xfs_iext_add(ifp, 0, nex);
+
ifp->if_bytes = size;
- ifp->if_real_bytes = real_size;
if (size) {
dp = (xfs_bmbt_rec_t *) XFS_DFORK_PTR(dip, whichfork);
- xfs_validate_extents(dp, nex, 1, XFS_EXTFMT_INODE(ip));
- ep = ifp->if_u1.if_extents;
- for (i = 0; i < nex; i++, ep++, dp++) {
+ xfs_validate_extents(ifp, nex, 1, XFS_EXTFMT_INODE(ip));
+ for (i = 0; i < nex; i++, dp++) {
+ ep = xfs_iext_get_ext(ifp, i);
ep->l0 = INT_GET(get_unaligned((__uint64_t*)&dp->l0),
ARCH_CONVERT);
ep->l1 = INT_GET(get_unaligned((__uint64_t*)&dp->l1),
@@ -646,7 +644,7 @@ xfs_iformat_extents(
if (whichfork != XFS_DATA_FORK ||
XFS_EXTFMT_INODE(ip) == XFS_EXTFMT_NOSTATE)
if (unlikely(xfs_check_nostate_extents(
- ifp->if_u1.if_extents, nex))) {
+ ifp, 0, nex))) {
XFS_ERROR_REPORT("xfs_iformat_extents(2)",
XFS_ERRLEVEL_LOW,
ip->i_mount);
@@ -871,9 +869,8 @@ xfs_iread(
* return NULL as well. Set i_blkno to 0 so that xfs_itobp() will
* know that this is a new incore inode.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &bp, bno);
-
- if (error != 0) {
+ error = xfs_itobp(mp, tp, ip, &dip, &bp, bno, 0);
+ if (error) {
kmem_zone_free(xfs_inode_zone, ip);
return error;
}
@@ -1015,6 +1012,7 @@ xfs_iread_extents(
{
int error;
xfs_ifork_t *ifp;
+ xfs_extnum_t nextents;
size_t size;
if (unlikely(XFS_IFORK_FORMAT(ip, whichfork) != XFS_DINODE_FMT_BTREE)) {
@@ -1022,26 +1020,24 @@ xfs_iread_extents(
ip->i_mount);
return XFS_ERROR(EFSCORRUPTED);
}
- size = XFS_IFORK_NEXTENTS(ip, whichfork) * (uint)sizeof(xfs_bmbt_rec_t);
+ nextents = XFS_IFORK_NEXTENTS(ip, whichfork);
+ size = nextents * sizeof(xfs_bmbt_rec_t);
ifp = XFS_IFORK_PTR(ip, whichfork);
+
/*
* We know that the size is valid (it's checked in iformat_btree)
*/
- ifp->if_u1.if_extents = kmem_alloc(size, KM_SLEEP);
- ASSERT(ifp->if_u1.if_extents != NULL);
ifp->if_lastex = NULLEXTNUM;
- ifp->if_bytes = ifp->if_real_bytes = (int)size;
+ ifp->if_bytes = ifp->if_real_bytes = 0;
ifp->if_flags |= XFS_IFEXTENTS;
+ xfs_iext_add(ifp, 0, nextents);
error = xfs_bmap_read_extents(tp, ip, whichfork);
if (error) {
- kmem_free(ifp->if_u1.if_extents, size);
- ifp->if_u1.if_extents = NULL;
- ifp->if_bytes = ifp->if_real_bytes = 0;
+ xfs_iext_destroy(ifp);
ifp->if_flags &= ~XFS_IFEXTENTS;
return error;
}
- xfs_validate_extents((xfs_bmbt_rec_t *)ifp->if_u1.if_extents,
- XFS_IFORK_NEXTENTS(ip, whichfork), 0, XFS_EXTFMT_INODE(ip));
+ xfs_validate_extents(ifp, nextents, 0, XFS_EXTFMT_INODE(ip));
return 0;
}
@@ -1376,10 +1372,10 @@ xfs_itrunc_trace(
(void*)(unsigned long)((toss_finish >> 32) & 0xffffffff),
(void*)(unsigned long)(toss_finish & 0xffffffff),
(void*)(unsigned long)current_cpu(),
- (void*)0,
- (void*)0,
- (void*)0,
- (void*)0);
+ (void*)(unsigned long)current_pid(),
+ (void*)NULL,
+ (void*)NULL,
+ (void*)NULL);
}
#else
#define xfs_itrunc_trace(tag, ip, flag, new_size, toss_start, toss_finish)
@@ -1397,6 +1393,16 @@ xfs_itrunc_trace(
* calling into the buffer/page cache code and we can't hold the
* inode lock when we do so.
*
+ * We need to wait for any direct I/Os in flight to complete before we
+ * proceed with the truncate. This is needed to prevent the extents
+ * being read or written by the direct I/Os from being removed while the
+ * I/O is in flight as there is no other method of synchronising
+ * direct I/O with the truncate operation. Also, because we hold
+ * the IOLOCK in exclusive mode, we prevent new direct I/Os from being
+ * started until the truncate completes and drops the lock. Essentially,
+ * the vn_iowait() call forms an I/O barrier that provides strict ordering
+ * between direct I/Os and the truncate operation.
+ *
* The flags parameter can have either the value XFS_ITRUNC_DEFINITE
* or XFS_ITRUNC_MAYBE. The XFS_ITRUNC_MAYBE value should be used
* in the case that the caller is locking things out of order and
@@ -1424,6 +1430,9 @@ xfs_itruncate_start(
mp = ip->i_mount;
vp = XFS_ITOV(ip);
+
+ vn_iowait(vp); /* wait for the completion of any pending DIOs */
+
/*
* Call VOP_TOSS_PAGES() or VOP_FLUSHINVAL_PAGES() to get rid of pages and buffers
* overlapping the region being removed. We have to use
@@ -1899,7 +1908,7 @@ xfs_iunlink(
* Here we put the head pointer into our next pointer,
* and then we fall through to point the head at us.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0);
if (error) {
return error;
}
@@ -2008,7 +2017,7 @@ xfs_iunlink_remove(
* of dealing with the buffer when there is no need to
* change it.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0);
if (error) {
cmn_err(CE_WARN,
"xfs_iunlink_remove: xfs_itobp() returned an error %d on %s. Returning error.",
@@ -2070,7 +2079,7 @@ xfs_iunlink_remove(
* Now last_ibp points to the buffer previous to us on
* the unlinked list. Pull us from the list.
*/
- error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0);
+ error = xfs_itobp(mp, tp, ip, &dip, &ibp, 0, 0);
if (error) {
cmn_err(CE_WARN,
"xfs_iunlink_remove: xfs_itobp() returned an error %d on %s. Returning error.",
@@ -2476,92 +2485,6 @@ xfs_iroot_realloc(
/*
- * This is called when the amount of space needed for if_extents
- * is increased or decreased. The change in size is indicated by
- * the number of extents that need to be added or deleted in the
- * ext_diff parameter.
- *
- * If the amount of space needed has decreased below the size of the
- * inline buffer, then switch to using the inline buffer. Otherwise,
- * use kmem_realloc() or kmem_alloc() to adjust the size of the buffer
- * to what is needed.
- *
- * ip -- the inode whose if_extents area is changing
- * ext_diff -- the change in the number of extents, positive or negative,
- * requested for the if_extents array.
- */
-void
-xfs_iext_realloc(
- xfs_inode_t *ip,
- int ext_diff,
- int whichfork)
-{
- int byte_diff;
- xfs_ifork_t *ifp;
- int new_size;
- uint rnew_size;
-
- if (ext_diff == 0) {
- return;
- }
-
- ifp = XFS_IFORK_PTR(ip, whichfork);
- byte_diff = ext_diff * (uint)sizeof(xfs_bmbt_rec_t);
- new_size = (int)ifp->if_bytes + byte_diff;
- ASSERT(new_size >= 0);
-
- if (new_size == 0) {
- if (ifp->if_u1.if_extents != ifp->if_u2.if_inline_ext) {
- ASSERT(ifp->if_real_bytes != 0);
- kmem_free(ifp->if_u1.if_extents, ifp->if_real_bytes);
- }
- ifp->if_u1.if_extents = NULL;
- rnew_size = 0;
- } else if (new_size <= sizeof(ifp->if_u2.if_inline_ext)) {
- /*
- * If the valid extents can fit in if_inline_ext,
- * copy them from the malloc'd vector and free it.
- */
- if (ifp->if_u1.if_extents != ifp->if_u2.if_inline_ext) {
- /*
- * For now, empty files are format EXTENTS,
- * so the if_extents pointer is null.
- */
- if (ifp->if_u1.if_extents) {
- memcpy(ifp->if_u2.if_inline_ext,
- ifp->if_u1.if_extents, new_size);
- kmem_free(ifp->if_u1.if_extents,
- ifp->if_real_bytes);
- }
- ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext;
- }
- rnew_size = 0;
- } else {
- rnew_size = new_size;
- if ((rnew_size & (rnew_size - 1)) != 0)
- rnew_size = xfs_iroundup(rnew_size);
- /*
- * Stuck with malloc/realloc.
- */
- if (ifp->if_u1.if_extents == ifp->if_u2.if_inline_ext) {
- ifp->if_u1.if_extents = (xfs_bmbt_rec_t *)
- kmem_alloc(rnew_size, KM_SLEEP);
- memcpy(ifp->if_u1.if_extents, ifp->if_u2.if_inline_ext,
- sizeof(ifp->if_u2.if_inline_ext));
- } else if (rnew_size != ifp->if_real_bytes) {
- ifp->if_u1.if_extents = (xfs_bmbt_rec_t *)
- kmem_realloc(ifp->if_u1.if_extents,
- rnew_size,
- ifp->if_real_bytes,
- KM_NOFS);
- }
- }
- ifp->if_real_bytes = rnew_size;
- ifp->if_bytes = new_size;
-}
-
-
-/*
* This is called when the amount of space needed for if_data
* is increased or decreased. The change in size is indicated by
* the number of bytes that need to be added or deleted in the
@@ -2720,12 +2643,11 @@ xfs_idestroy_fork(
ifp->if_real_bytes = 0;
}
} else if ((ifp->if_flags & XFS_IFEXTENTS) &&
- (ifp->if_u1.if_extents != NULL) &&
- (ifp->if_u1.if_extents != ifp->if_u2.if_inline_ext)) {
+ ((ifp->if_flags & XFS_IFEXTIREC) ||
+ ((ifp->if_u1.if_extents != NULL) &&
+ (ifp->if_u1.if_extents != ifp->if_u2.if_inline_ext)))) {
ASSERT(ifp->if_real_bytes != 0);
- kmem_free(ifp->if_u1.if_extents, ifp->if_real_bytes);
- ifp->if_u1.if_extents = NULL;
- ifp->if_real_bytes = 0;
+ xfs_iext_destroy(ifp);
}
ASSERT(ifp->if_u1.if_extents == NULL ||
ifp->if_u1.if_extents == ifp->if_u2.if_inline_ext);
@@ -2814,7 +2736,7 @@ xfs_iunpin(
/* make sync come back and flush this inode */
if (vp) {
- struct inode *inode = LINVFS_GET_IP(vp);
+ struct inode *inode = vn_to_inode(vp);
if (!(inode->i_state & I_NEW))
mark_inode_dirty_sync(inode);
@@ -2902,16 +2824,15 @@ xfs_iextents_copy(
* the delayed ones. There must be at least one
* non-delayed extent.
*/
- ep = ifp->if_u1.if_extents;
dest_ep = buffer;
copied = 0;
for (i = 0; i < nrecs; i++) {
+ ep = xfs_iext_get_ext(ifp, i);
start_block = xfs_bmbt_get_startblock(ep);
if (ISNULLSTARTBLOCK(start_block)) {
/*
* It's a delayed allocation extent, so skip it.
*/
- ep++;
continue;
}
@@ -2921,11 +2842,10 @@ xfs_iextents_copy(
put_unaligned(INT_GET(ep->l1, ARCH_CONVERT),
(__uint64_t*)&dest_ep->l1);
dest_ep++;
- ep++;
copied++;
}
ASSERT(copied != 0);
- xfs_validate_extents(buffer, copied, 1, XFS_EXTFMT_INODE(ip));
+ xfs_validate_extents(ifp, copied, 1, XFS_EXTFMT_INODE(ip));
return (copied * (uint)sizeof(xfs_bmbt_rec_t));
}
@@ -2995,8 +2915,10 @@ xfs_iflush_fork(
case XFS_DINODE_FMT_EXTENTS:
ASSERT((ifp->if_flags & XFS_IFEXTENTS) ||
!(iip->ili_format.ilf_fields & extflag[whichfork]));
- ASSERT((ifp->if_u1.if_extents != NULL) || (ifp->if_bytes == 0));
- ASSERT((ifp->if_u1.if_extents == NULL) || (ifp->if_bytes > 0));
+ ASSERT((xfs_iext_get_ext(ifp, 0) != NULL) ||
+ (ifp->if_bytes == 0));
+ ASSERT((xfs_iext_get_ext(ifp, 0) == NULL) ||
+ (ifp->if_bytes > 0));
if ((iip->ili_format.ilf_fields & extflag[whichfork]) &&
(ifp->if_bytes > 0)) {
ASSERT(XFS_IFORK_NEXTENTS(ip, whichfork) > 0);
@@ -3114,8 +3036,8 @@ xfs_iflush(
/*
* Get the buffer containing the on-disk inode.
*/
- error = xfs_itobp(mp, NULL, ip, &dip, &bp, 0);
- if (error != 0) {
+ error = xfs_itobp(mp, NULL, ip, &dip, &bp, 0, 0);
+ if (error) {
xfs_ifunlock(ip);
return error;
}
@@ -3610,7 +3532,7 @@ xfs_iaccess(
{
int error;
mode_t orgmode = mode;
- struct inode *inode = LINVFS_GET_IP(XFS_ITOV(ip));
+ struct inode *inode = vn_to_inode(XFS_ITOV(ip));
if (mode & S_IWUSR) {
umode_t imode = inode->i_mode;
@@ -3704,3 +3626,1100 @@ xfs_ilock_trace(xfs_inode_t *ip, int lock, unsigned int lockflags, inst_t *ra)
NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL,NULL);
}
#endif
+
+/*
+ * Return a pointer to the extent record at file index idx.
+ */
+xfs_bmbt_rec_t *
+xfs_iext_get_ext(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_extnum_t idx) /* index of target extent */
+{
+ ASSERT(idx >= 0);
+ if ((ifp->if_flags & XFS_IFEXTIREC) && (idx == 0)) {
+ return ifp->if_u1.if_ext_irec->er_extbuf;
+ } else if (ifp->if_flags & XFS_IFEXTIREC) {
+ xfs_ext_irec_t *erp; /* irec pointer */
+ int erp_idx = 0; /* irec index */
+ xfs_extnum_t page_idx = idx; /* ext index in target list */
+
+ erp = xfs_iext_idx_to_irec(ifp, &page_idx, &erp_idx, 0);
+ return &erp->er_extbuf[page_idx];
+ } else if (ifp->if_bytes) {
+ return &ifp->if_u1.if_extents[idx];
+ } else {
+ return NULL;
+ }
+}
+
+/*
+ * Insert new item(s) into the extent records for incore inode
+ * fork 'ifp'. 'count' new items are inserted at index 'idx'.
+ */
+void
+xfs_iext_insert(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_extnum_t idx, /* starting index of new items */
+ xfs_extnum_t count, /* number of inserted items */
+ xfs_bmbt_irec_t *new) /* items to insert */
+{
+ xfs_bmbt_rec_t *ep; /* extent record pointer */
+ xfs_extnum_t i; /* extent record index */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTENTS);
+ xfs_iext_add(ifp, idx, count);
+ for (i = idx; i < idx + count; i++, new++) {
+ ep = xfs_iext_get_ext(ifp, i);
+ xfs_bmbt_set_all(ep, new);
+ }
+}
+
+/*
+ * This is called when the amount of space required for incore file
+ * extents needs to be increased. The ext_diff parameter stores the
+ * number of new extents being added and the idx parameter contains
+ * the extent index where the new extents will be added. If the new
+ * extents are being appended, then we just need to (re)allocate and
+ * initialize the space. Otherwise, if the new extents are being
+ * inserted into the middle of the existing entries, a bit more work
+ * is required to make room for the new extents to be inserted. The
+ * caller is responsible for filling in the new extent entries upon
+ * return.
+ */
+void
+xfs_iext_add(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_extnum_t idx, /* index to begin adding exts */
+ int ext_diff) /* nubmer of extents to add */
+{
+ int byte_diff; /* new bytes being added */
+ int new_size; /* size of extents after adding */
+ xfs_extnum_t nextents; /* number of extents in file */
+
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ ASSERT((idx >= 0) && (idx <= nextents));
+ byte_diff = ext_diff * sizeof(xfs_bmbt_rec_t);
+ new_size = ifp->if_bytes + byte_diff;
+ /*
+ * If the new number of extents (nextents + ext_diff)
+ * fits inside the inode, then continue to use the inline
+ * extent buffer.
+ */
+ if (nextents + ext_diff <= XFS_INLINE_EXTS) {
+ if (idx < nextents) {
+ memmove(&ifp->if_u2.if_inline_ext[idx + ext_diff],
+ &ifp->if_u2.if_inline_ext[idx],
+ (nextents - idx) * sizeof(xfs_bmbt_rec_t));
+ memset(&ifp->if_u2.if_inline_ext[idx], 0, byte_diff);
+ }
+ ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext;
+ ifp->if_real_bytes = 0;
+ ifp->if_lastex = nextents + ext_diff;
+ }
+ /*
+ * Otherwise use a linear (direct) extent list.
+ * If the extents are currently inside the inode,
+ * xfs_iext_realloc_direct will switch us from
+ * inline to direct extent allocation mode.
+ */
+ else if (nextents + ext_diff <= XFS_LINEAR_EXTS) {
+ xfs_iext_realloc_direct(ifp, new_size);
+ if (idx < nextents) {
+ memmove(&ifp->if_u1.if_extents[idx + ext_diff],
+ &ifp->if_u1.if_extents[idx],
+ (nextents - idx) * sizeof(xfs_bmbt_rec_t));
+ memset(&ifp->if_u1.if_extents[idx], 0, byte_diff);
+ }
+ }
+ /* Indirection array */
+ else {
+ xfs_ext_irec_t *erp;
+ int erp_idx = 0;
+ int page_idx = idx;
+
+ ASSERT(nextents + ext_diff > XFS_LINEAR_EXTS);
+ if (ifp->if_flags & XFS_IFEXTIREC) {
+ erp = xfs_iext_idx_to_irec(ifp, &page_idx, &erp_idx, 1);
+ } else {
+ xfs_iext_irec_init(ifp);
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ erp = ifp->if_u1.if_ext_irec;
+ }
+ /* Extents fit in target extent page */
+ if (erp && erp->er_extcount + ext_diff <= XFS_LINEAR_EXTS) {
+ if (page_idx < erp->er_extcount) {
+ memmove(&erp->er_extbuf[page_idx + ext_diff],
+ &erp->er_extbuf[page_idx],
+ (erp->er_extcount - page_idx) *
+ sizeof(xfs_bmbt_rec_t));
+ memset(&erp->er_extbuf[page_idx], 0, byte_diff);
+ }
+ erp->er_extcount += ext_diff;
+ xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, ext_diff);
+ }
+ /* Insert a new extent page */
+ else if (erp) {
+ xfs_iext_add_indirect_multi(ifp,
+ erp_idx, page_idx, ext_diff);
+ }
+ /*
+ * If extent(s) are being appended to the last page in
+ * the indirection array and the new extent(s) don't fit
+ * in the page, then erp is NULL and erp_idx is set to
+ * the next index needed in the indirection array.
+ */
+ else {
+ int count = ext_diff;
+
+ while (count) {
+ erp = xfs_iext_irec_new(ifp, erp_idx);
+ erp->er_extcount = count;
+ count -= MIN(count, (int)XFS_LINEAR_EXTS);
+ if (count) {
+ erp_idx++;
+ }
+ }
+ }
+ }
+ ifp->if_bytes = new_size;
+}
+
+/*
+ * This is called when incore extents are being added to the indirection
+ * array and the new extents do not fit in the target extent list. The
+ * erp_idx parameter contains the irec index for the target extent list
+ * in the indirection array, and the idx parameter contains the extent
+ * index within the list. The number of extents being added is stored
+ * in the count parameter.
+ *
+ * |-------| |-------|
+ * | | | | idx - number of extents before idx
+ * | idx | | count |
+ * | | | | count - number of extents being inserted at idx
+ * |-------| |-------|
+ * | count | | nex2 | nex2 - number of extents after idx + count
+ * |-------| |-------|
+ */
+void
+xfs_iext_add_indirect_multi(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ int erp_idx, /* target extent irec index */
+ xfs_extnum_t idx, /* index within target list */
+ int count) /* new extents being added */
+{
+ int byte_diff; /* new bytes being added */
+ xfs_ext_irec_t *erp; /* pointer to irec entry */
+ xfs_extnum_t ext_diff; /* number of extents to add */
+ xfs_extnum_t ext_cnt; /* new extents still needed */
+ xfs_extnum_t nex2; /* extents after idx + count */
+ xfs_bmbt_rec_t *nex2_ep = NULL; /* temp list for nex2 extents */
+ int nlists; /* number of irec's (lists) */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ erp = &ifp->if_u1.if_ext_irec[erp_idx];
+ nex2 = erp->er_extcount - idx;
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+
+ /*
+ * Save second part of target extent list
+ * (all extents past */
+ if (nex2) {
+ byte_diff = nex2 * sizeof(xfs_bmbt_rec_t);
+ nex2_ep = (xfs_bmbt_rec_t *) kmem_alloc(byte_diff, KM_SLEEP);
+ memmove(nex2_ep, &erp->er_extbuf[idx], byte_diff);
+ erp->er_extcount -= nex2;
+ xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, -nex2);
+ memset(&erp->er_extbuf[idx], 0, byte_diff);
+ }
+
+ /*
+ * Add the new extents to the end of the target
+ * list, then allocate new irec record(s) and
+ * extent buffer(s) as needed to store the rest
+ * of the new extents.
+ */
+ ext_cnt = count;
+ ext_diff = MIN(ext_cnt, (int)XFS_LINEAR_EXTS - erp->er_extcount);
+ if (ext_diff) {
+ erp->er_extcount += ext_diff;
+ xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, ext_diff);
+ ext_cnt -= ext_diff;
+ }
+ while (ext_cnt) {
+ erp_idx++;
+ erp = xfs_iext_irec_new(ifp, erp_idx);
+ ext_diff = MIN(ext_cnt, (int)XFS_LINEAR_EXTS);
+ erp->er_extcount = ext_diff;
+ xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, ext_diff);
+ ext_cnt -= ext_diff;
+ }
+
+ /* Add nex2 extents back to indirection array */
+ if (nex2) {
+ xfs_extnum_t ext_avail;
+ int i;
+
+ byte_diff = nex2 * sizeof(xfs_bmbt_rec_t);
+ ext_avail = XFS_LINEAR_EXTS - erp->er_extcount;
+ i = 0;
+ /*
+ * If nex2 extents fit in the current page, append
+ * nex2_ep after the new extents.
+ */
+ if (nex2 <= ext_avail) {
+ i = erp->er_extcount;
+ }
+ /*
+ * Otherwise, check if space is available in the
+ * next page.
+ */
+ else if ((erp_idx < nlists - 1) &&
+ (nex2 <= (ext_avail = XFS_LINEAR_EXTS -
+ ifp->if_u1.if_ext_irec[erp_idx+1].er_extcount))) {
+ erp_idx++;
+ erp++;
+ /* Create a hole for nex2 extents */
+ memmove(&erp->er_extbuf[nex2], erp->er_extbuf,
+ erp->er_extcount * sizeof(xfs_bmbt_rec_t));
+ }
+ /*
+ * Final choice, create a new extent page for
+ * nex2 extents.
+ */
+ else {
+ erp_idx++;
+ erp = xfs_iext_irec_new(ifp, erp_idx);
+ }
+ memmove(&erp->er_extbuf[i], nex2_ep, byte_diff);
+ kmem_free(nex2_ep, byte_diff);
+ erp->er_extcount += nex2;
+ xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, nex2);
+ }
+}
+
+/*
+ * This is called when the amount of space required for incore file
+ * extents needs to be decreased. The ext_diff parameter stores the
+ * number of extents to be removed and the idx parameter contains
+ * the extent index where the extents will be removed from.
+ *
+ * If the amount of space needed has decreased below the linear
+ * limit, XFS_IEXT_BUFSZ, then switch to using the contiguous
+ * extent array. Otherwise, use kmem_realloc() to adjust the
+ * size to what is needed.
+ */
+void
+xfs_iext_remove(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_extnum_t idx, /* index to begin removing exts */
+ int ext_diff) /* number of extents to remove */
+{
+ xfs_extnum_t nextents; /* number of extents in file */
+ int new_size; /* size of extents after removal */
+
+ ASSERT(ext_diff > 0);
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ new_size = (nextents - ext_diff) * sizeof(xfs_bmbt_rec_t);
+
+ if (new_size == 0) {
+ xfs_iext_destroy(ifp);
+ } else if (ifp->if_flags & XFS_IFEXTIREC) {
+ xfs_iext_remove_indirect(ifp, idx, ext_diff);
+ } else if (ifp->if_real_bytes) {
+ xfs_iext_remove_direct(ifp, idx, ext_diff);
+ } else {
+ xfs_iext_remove_inline(ifp, idx, ext_diff);
+ }
+ ifp->if_bytes = new_size;
+}
+
+/*
+ * This removes ext_diff extents from the inline buffer, beginning
+ * at extent index idx.
+ */
+void
+xfs_iext_remove_inline(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_extnum_t idx, /* index to begin removing exts */
+ int ext_diff) /* number of extents to remove */
+{
+ int nextents; /* number of extents in file */
+
+ ASSERT(!(ifp->if_flags & XFS_IFEXTIREC));
+ ASSERT(idx < XFS_INLINE_EXTS);
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ ASSERT(((nextents - ext_diff) > 0) &&
+ (nextents - ext_diff) < XFS_INLINE_EXTS);
+
+ if (idx + ext_diff < nextents) {
+ memmove(&ifp->if_u2.if_inline_ext[idx],
+ &ifp->if_u2.if_inline_ext[idx + ext_diff],
+ (nextents - (idx + ext_diff)) *
+ sizeof(xfs_bmbt_rec_t));
+ memset(&ifp->if_u2.if_inline_ext[nextents - ext_diff],
+ 0, ext_diff * sizeof(xfs_bmbt_rec_t));
+ } else {
+ memset(&ifp->if_u2.if_inline_ext[idx], 0,
+ ext_diff * sizeof(xfs_bmbt_rec_t));
+ }
+}
+
+/*
+ * This removes ext_diff extents from a linear (direct) extent list,
+ * beginning at extent index idx. If the extents are being removed
+ * from the end of the list (ie. truncate) then we just need to re-
+ * allocate the list to remove the extra space. Otherwise, if the
+ * extents are being removed from the middle of the existing extent
+ * entries, then we first need to move the extent records beginning
+ * at idx + ext_diff up in the list to overwrite the records being
+ * removed, then remove the extra space via kmem_realloc.
+ */
+void
+xfs_iext_remove_direct(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_extnum_t idx, /* index to begin removing exts */
+ int ext_diff) /* number of extents to remove */
+{
+ xfs_extnum_t nextents; /* number of extents in file */
+ int new_size; /* size of extents after removal */
+
+ ASSERT(!(ifp->if_flags & XFS_IFEXTIREC));
+ new_size = ifp->if_bytes -
+ (ext_diff * sizeof(xfs_bmbt_rec_t));
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+
+ if (new_size == 0) {
+ xfs_iext_destroy(ifp);
+ return;
+ }
+ /* Move extents up in the list (if needed) */
+ if (idx + ext_diff < nextents) {
+ memmove(&ifp->if_u1.if_extents[idx],
+ &ifp->if_u1.if_extents[idx + ext_diff],
+ (nextents - (idx + ext_diff)) *
+ sizeof(xfs_bmbt_rec_t));
+ }
+ memset(&ifp->if_u1.if_extents[nextents - ext_diff],
+ 0, ext_diff * sizeof(xfs_bmbt_rec_t));
+ /*
+ * Reallocate the direct extent list. If the extents
+ * will fit inside the inode then xfs_iext_realloc_direct
+ * will switch from direct to inline extent allocation
+ * mode for us.
+ */
+ xfs_iext_realloc_direct(ifp, new_size);
+ ifp->if_bytes = new_size;
+}
+
+/*
+ * This is called when incore extents are being removed from the
+ * indirection array and the extents being removed span multiple extent
+ * buffers. The idx parameter contains the file extent index where we
+ * want to begin removing extents, and the count parameter contains
+ * how many extents need to be removed.
+ *
+ * |-------| |-------|
+ * | nex1 | | | nex1 - number of extents before idx
+ * |-------| | count |
+ * | | | | count - number of extents being removed at idx
+ * | count | |-------|
+ * | | | nex2 | nex2 - number of extents after idx + count
+ * |-------| |-------|
+ */
+void
+xfs_iext_remove_indirect(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_extnum_t idx, /* index to begin removing extents */
+ int count) /* number of extents to remove */
+{
+ xfs_ext_irec_t *erp; /* indirection array pointer */
+ int erp_idx = 0; /* indirection array index */
+ xfs_extnum_t ext_cnt; /* extents left to remove */
+ xfs_extnum_t ext_diff; /* extents to remove in current list */
+ xfs_extnum_t nex1; /* number of extents before idx */
+ xfs_extnum_t nex2; /* extents after idx + count */
+ int nlists; /* entries in indirecton array */
+ int page_idx = idx; /* index in target extent list */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ erp = xfs_iext_idx_to_irec(ifp, &page_idx, &erp_idx, 0);
+ ASSERT(erp != NULL);
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ nex1 = page_idx;
+ ext_cnt = count;
+ while (ext_cnt) {
+ nex2 = MAX((erp->er_extcount - (nex1 + ext_cnt)), 0);
+ ext_diff = MIN(ext_cnt, (erp->er_extcount - nex1));
+ /*
+ * Check for deletion of entire list;
+ * xfs_iext_irec_remove() updates extent offsets.
+ */
+ if (ext_diff == erp->er_extcount) {
+ xfs_iext_irec_remove(ifp, erp_idx);
+ ext_cnt -= ext_diff;
+ nex1 = 0;
+ if (ext_cnt) {
+ ASSERT(erp_idx < ifp->if_real_bytes /
+ XFS_IEXT_BUFSZ);
+ erp = &ifp->if_u1.if_ext_irec[erp_idx];
+ nex1 = 0;
+ continue;
+ } else {
+ break;
+ }
+ }
+ /* Move extents up (if needed) */
+ if (nex2) {
+ memmove(&erp->er_extbuf[nex1],
+ &erp->er_extbuf[nex1 + ext_diff],
+ nex2 * sizeof(xfs_bmbt_rec_t));
+ }
+ /* Zero out rest of page */
+ memset(&erp->er_extbuf[nex1 + nex2], 0, (XFS_IEXT_BUFSZ -
+ ((nex1 + nex2) * sizeof(xfs_bmbt_rec_t))));
+ /* Update remaining counters */
+ erp->er_extcount -= ext_diff;
+ xfs_iext_irec_update_extoffs(ifp, erp_idx + 1, -ext_diff);
+ ext_cnt -= ext_diff;
+ nex1 = 0;
+ erp_idx++;
+ erp++;
+ }
+ ifp->if_bytes -= count * sizeof(xfs_bmbt_rec_t);
+ xfs_iext_irec_compact(ifp);
+}
+
+/*
+ * Create, destroy, or resize a linear (direct) block of extents.
+ */
+void
+xfs_iext_realloc_direct(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ int new_size) /* new size of extents */
+{
+ int rnew_size; /* real new size of extents */
+
+ rnew_size = new_size;
+
+ ASSERT(!(ifp->if_flags & XFS_IFEXTIREC) ||
+ ((new_size >= 0) && (new_size <= XFS_IEXT_BUFSZ) &&
+ (new_size != ifp->if_real_bytes)));
+
+ /* Free extent records */
+ if (new_size == 0) {
+ xfs_iext_destroy(ifp);
+ }
+ /* Resize direct extent list and zero any new bytes */
+ else if (ifp->if_real_bytes) {
+ /* Check if extents will fit inside the inode */
+ if (new_size <= XFS_INLINE_EXTS * sizeof(xfs_bmbt_rec_t)) {
+ xfs_iext_direct_to_inline(ifp, new_size /
+ (uint)sizeof(xfs_bmbt_rec_t));
+ ifp->if_bytes = new_size;
+ return;
+ }
+ if ((new_size & (new_size - 1)) != 0) {
+ rnew_size = xfs_iroundup(new_size);
+ }
+ if (rnew_size != ifp->if_real_bytes) {
+ ifp->if_u1.if_extents = (xfs_bmbt_rec_t *)
+ kmem_realloc(ifp->if_u1.if_extents,
+ rnew_size,
+ ifp->if_real_bytes,
+ KM_SLEEP);
+ }
+ if (rnew_size > ifp->if_real_bytes) {
+ memset(&ifp->if_u1.if_extents[ifp->if_bytes /
+ (uint)sizeof(xfs_bmbt_rec_t)], 0,
+ rnew_size - ifp->if_real_bytes);
+ }
+ }
+ /*
+ * Switch from the inline extent buffer to a direct
+ * extent list. Be sure to include the inline extent
+ * bytes in new_size.
+ */
+ else {
+ new_size += ifp->if_bytes;
+ if ((new_size & (new_size - 1)) != 0) {
+ rnew_size = xfs_iroundup(new_size);
+ }
+ xfs_iext_inline_to_direct(ifp, rnew_size);
+ }
+ ifp->if_real_bytes = rnew_size;
+ ifp->if_bytes = new_size;
+}
+
+/*
+ * Switch from linear (direct) extent records to inline buffer.
+ */
+void
+xfs_iext_direct_to_inline(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_extnum_t nextents) /* number of extents in file */
+{
+ ASSERT(ifp->if_flags & XFS_IFEXTENTS);
+ ASSERT(nextents <= XFS_INLINE_EXTS);
+ /*
+ * The inline buffer was zeroed when we switched
+ * from inline to direct extent allocation mode,
+ * so we don't need to clear it here.
+ */
+ memcpy(ifp->if_u2.if_inline_ext, ifp->if_u1.if_extents,
+ nextents * sizeof(xfs_bmbt_rec_t));
+ kmem_free(ifp->if_u1.if_extents, KM_SLEEP);
+ ifp->if_u1.if_extents = ifp->if_u2.if_inline_ext;
+ ifp->if_real_bytes = 0;
+}
+
+/*
+ * Switch from inline buffer to linear (direct) extent records.
+ * new_size should already be rounded up to the next power of 2
+ * by the caller (when appropriate), so use new_size as it is.
+ * However, since new_size may be rounded up, we can't update
+ * if_bytes here. It is the caller's responsibility to update
+ * if_bytes upon return.
+ */
+void
+xfs_iext_inline_to_direct(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ int new_size) /* number of extents in file */
+{
+ ifp->if_u1.if_extents = (xfs_bmbt_rec_t *)
+ kmem_alloc(new_size, KM_SLEEP);
+ memset(ifp->if_u1.if_extents, 0, new_size);
+ if (ifp->if_bytes) {
+ memcpy(ifp->if_u1.if_extents, ifp->if_u2.if_inline_ext,
+ ifp->if_bytes);
+ memset(ifp->if_u2.if_inline_ext, 0, XFS_INLINE_EXTS *
+ sizeof(xfs_bmbt_rec_t));
+ }
+ ifp->if_real_bytes = new_size;
+}
+
+/*
+ * Resize an extent indirection array to new_size bytes.
+ */
+void
+xfs_iext_realloc_indirect(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ int new_size) /* new indirection array size */
+{
+ int nlists; /* number of irec's (ex lists) */
+ int size; /* current indirection array size */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ size = nlists * sizeof(xfs_ext_irec_t);
+ ASSERT(ifp->if_real_bytes);
+ ASSERT((new_size >= 0) && (new_size != size));
+ if (new_size == 0) {
+ xfs_iext_destroy(ifp);
+ } else {
+ ifp->if_u1.if_ext_irec = (xfs_ext_irec_t *)
+ kmem_realloc(ifp->if_u1.if_ext_irec,
+ new_size, size, KM_SLEEP);
+ }
+}
+
+/*
+ * Switch from indirection array to linear (direct) extent allocations.
+ */
+void
+xfs_iext_indirect_to_direct(
+ xfs_ifork_t *ifp) /* inode fork pointer */
+{
+ xfs_bmbt_rec_t *ep; /* extent record pointer */
+ xfs_extnum_t nextents; /* number of extents in file */
+ int size; /* size of file extents */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ ASSERT(nextents <= XFS_LINEAR_EXTS);
+ size = nextents * sizeof(xfs_bmbt_rec_t);
+
+ xfs_iext_irec_compact_full(ifp);
+ ASSERT(ifp->if_real_bytes == XFS_IEXT_BUFSZ);
+
+ ep = ifp->if_u1.if_ext_irec->er_extbuf;
+ kmem_free(ifp->if_u1.if_ext_irec, sizeof(xfs_ext_irec_t));
+ ifp->if_flags &= ~XFS_IFEXTIREC;
+ ifp->if_u1.if_extents = ep;
+ ifp->if_bytes = size;
+ if (nextents < XFS_LINEAR_EXTS) {
+ xfs_iext_realloc_direct(ifp, size);
+ }
+}
+
+/*
+ * Free incore file extents.
+ */
+void
+xfs_iext_destroy(
+ xfs_ifork_t *ifp) /* inode fork pointer */
+{
+ if (ifp->if_flags & XFS_IFEXTIREC) {
+ int erp_idx;
+ int nlists;
+
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ for (erp_idx = nlists - 1; erp_idx >= 0 ; erp_idx--) {
+ xfs_iext_irec_remove(ifp, erp_idx);
+ }
+ ifp->if_flags &= ~XFS_IFEXTIREC;
+ } else if (ifp->if_real_bytes) {
+ kmem_free(ifp->if_u1.if_extents, ifp->if_real_bytes);
+ } else if (ifp->if_bytes) {
+ memset(ifp->if_u2.if_inline_ext, 0, XFS_INLINE_EXTS *
+ sizeof(xfs_bmbt_rec_t));
+ }
+ ifp->if_u1.if_extents = NULL;
+ ifp->if_real_bytes = 0;
+ ifp->if_bytes = 0;
+}
+
+/*
+ * Return a pointer to the extent record for file system block bno.
+ */
+xfs_bmbt_rec_t * /* pointer to found extent record */
+xfs_iext_bno_to_ext(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_fileoff_t bno, /* block number to search for */
+ xfs_extnum_t *idxp) /* index of target extent */
+{
+ xfs_bmbt_rec_t *base; /* pointer to first extent */
+ xfs_filblks_t blockcount = 0; /* number of blocks in extent */
+ xfs_bmbt_rec_t *ep = NULL; /* pointer to target extent */
+ xfs_ext_irec_t *erp = NULL; /* indirection array pointer */
+ int high; /* upper boundry in search */
+ xfs_extnum_t idx = 0; /* index of target extent */
+ int low; /* lower boundry in search */
+ xfs_extnum_t nextents; /* number of file extents */
+ xfs_fileoff_t startoff = 0; /* start offset of extent */
+
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ if (nextents == 0) {
+ *idxp = 0;
+ return NULL;
+ }
+ low = 0;
+ if (ifp->if_flags & XFS_IFEXTIREC) {
+ /* Find target extent list */
+ int erp_idx = 0;
+ erp = xfs_iext_bno_to_irec(ifp, bno, &erp_idx);
+ base = erp->er_extbuf;
+ high = erp->er_extcount - 1;
+ } else {
+ base = ifp->if_u1.if_extents;
+ high = nextents - 1;
+ }
+ /* Binary search extent records */
+ while (low <= high) {
+ idx = (low + high) >> 1;
+ ep = base + idx;
+ startoff = xfs_bmbt_get_startoff(ep);
+ blockcount = xfs_bmbt_get_blockcount(ep);
+ if (bno < startoff) {
+ high = idx - 1;
+ } else if (bno >= startoff + blockcount) {
+ low = idx + 1;
+ } else {
+ /* Convert back to file-based extent index */
+ if (ifp->if_flags & XFS_IFEXTIREC) {
+ idx += erp->er_extoff;
+ }
+ *idxp = idx;
+ return ep;
+ }
+ }
+ /* Convert back to file-based extent index */
+ if (ifp->if_flags & XFS_IFEXTIREC) {
+ idx += erp->er_extoff;
+ }
+ if (bno >= startoff + blockcount) {
+ if (++idx == nextents) {
+ ep = NULL;
+ } else {
+ ep = xfs_iext_get_ext(ifp, idx);
+ }
+ }
+ *idxp = idx;
+ return ep;
+}
+
+/*
+ * Return a pointer to the indirection array entry containing the
+ * extent record for filesystem block bno. Store the index of the
+ * target irec in *erp_idxp.
+ */
+xfs_ext_irec_t * /* pointer to found extent record */
+xfs_iext_bno_to_irec(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_fileoff_t bno, /* block number to search for */
+ int *erp_idxp) /* irec index of target ext list */
+{
+ xfs_ext_irec_t *erp = NULL; /* indirection array pointer */
+ xfs_ext_irec_t *erp_next; /* next indirection array entry */
+ int erp_idx; /* indirection array index */
+ int nlists; /* number of extent irec's (lists) */
+ int high; /* binary search upper limit */
+ int low; /* binary search lower limit */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ erp_idx = 0;
+ low = 0;
+ high = nlists - 1;
+ while (low <= high) {
+ erp_idx = (low + high) >> 1;
+ erp = &ifp->if_u1.if_ext_irec[erp_idx];
+ erp_next = erp_idx < nlists - 1 ? erp + 1 : NULL;
+ if (bno < xfs_bmbt_get_startoff(erp->er_extbuf)) {
+ high = erp_idx - 1;
+ } else if (erp_next && bno >=
+ xfs_bmbt_get_startoff(erp_next->er_extbuf)) {
+ low = erp_idx + 1;
+ } else {
+ break;
+ }
+ }
+ *erp_idxp = erp_idx;
+ return erp;
+}
+
+/*
+ * Return a pointer to the indirection array entry containing the
+ * extent record at file extent index *idxp. Store the index of the
+ * target irec in *erp_idxp and store the page index of the target
+ * extent record in *idxp.
+ */
+xfs_ext_irec_t *
+xfs_iext_idx_to_irec(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ xfs_extnum_t *idxp, /* extent index (file -> page) */
+ int *erp_idxp, /* pointer to target irec */
+ int realloc) /* new bytes were just added */
+{
+ xfs_ext_irec_t *prev; /* pointer to previous irec */
+ xfs_ext_irec_t *erp = NULL; /* pointer to current irec */
+ int erp_idx; /* indirection array index */
+ int nlists; /* number of irec's (ex lists) */
+ int high; /* binary search upper limit */
+ int low; /* binary search lower limit */
+ xfs_extnum_t page_idx = *idxp; /* extent index in target list */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ ASSERT(page_idx >= 0 && page_idx <=
+ ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t));
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ erp_idx = 0;
+ low = 0;
+ high = nlists - 1;
+
+ /* Binary search extent irec's */
+ while (low <= high) {
+ erp_idx = (low + high) >> 1;
+ erp = &ifp->if_u1.if_ext_irec[erp_idx];
+ prev = erp_idx > 0 ? erp - 1 : NULL;
+ if (page_idx < erp->er_extoff || (page_idx == erp->er_extoff &&
+ realloc && prev && prev->er_extcount < XFS_LINEAR_EXTS)) {
+ high = erp_idx - 1;
+ } else if (page_idx > erp->er_extoff + erp->er_extcount ||
+ (page_idx == erp->er_extoff + erp->er_extcount &&
+ !realloc)) {
+ low = erp_idx + 1;
+ } else if (page_idx == erp->er_extoff + erp->er_extcount &&
+ erp->er_extcount == XFS_LINEAR_EXTS) {
+ ASSERT(realloc);
+ page_idx = 0;
+ erp_idx++;
+ erp = erp_idx < nlists ? erp + 1 : NULL;
+ break;
+ } else {
+ page_idx -= erp->er_extoff;
+ break;
+ }
+ }
+ *idxp = page_idx;
+ *erp_idxp = erp_idx;
+ return(erp);
+}
+
+/*
+ * Allocate and initialize an indirection array once the space needed
+ * for incore extents increases above XFS_IEXT_BUFSZ.
+ */
+void
+xfs_iext_irec_init(
+ xfs_ifork_t *ifp) /* inode fork pointer */
+{
+ xfs_ext_irec_t *erp; /* indirection array pointer */
+ xfs_extnum_t nextents; /* number of extents in file */
+
+ ASSERT(!(ifp->if_flags & XFS_IFEXTIREC));
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+ ASSERT(nextents <= XFS_LINEAR_EXTS);
+
+ erp = (xfs_ext_irec_t *)
+ kmem_alloc(sizeof(xfs_ext_irec_t), KM_SLEEP);
+
+ if (nextents == 0) {
+ ifp->if_u1.if_extents = (xfs_bmbt_rec_t *)
+ kmem_alloc(XFS_IEXT_BUFSZ, KM_SLEEP);
+ } else if (!ifp->if_real_bytes) {
+ xfs_iext_inline_to_direct(ifp, XFS_IEXT_BUFSZ);
+ } else if (ifp->if_real_bytes < XFS_IEXT_BUFSZ) {
+ xfs_iext_realloc_direct(ifp, XFS_IEXT_BUFSZ);
+ }
+ erp->er_extbuf = ifp->if_u1.if_extents;
+ erp->er_extcount = nextents;
+ erp->er_extoff = 0;
+
+ ifp->if_flags |= XFS_IFEXTIREC;
+ ifp->if_real_bytes = XFS_IEXT_BUFSZ;
+ ifp->if_bytes = nextents * sizeof(xfs_bmbt_rec_t);
+ ifp->if_u1.if_ext_irec = erp;
+
+ return;
+}
+
+/*
+ * Allocate and initialize a new entry in the indirection array.
+ */
+xfs_ext_irec_t *
+xfs_iext_irec_new(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ int erp_idx) /* index for new irec */
+{
+ xfs_ext_irec_t *erp; /* indirection array pointer */
+ int i; /* loop counter */
+ int nlists; /* number of irec's (ex lists) */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+
+ /* Resize indirection array */
+ xfs_iext_realloc_indirect(ifp, ++nlists *
+ sizeof(xfs_ext_irec_t));
+ /*
+ * Move records down in the array so the
+ * new page can use erp_idx.
+ */
+ erp = ifp->if_u1.if_ext_irec;
+ for (i = nlists - 1; i > erp_idx; i--) {
+ memmove(&erp[i], &erp[i-1], sizeof(xfs_ext_irec_t));
+ }
+ ASSERT(i == erp_idx);
+
+ /* Initialize new extent record */
+ erp = ifp->if_u1.if_ext_irec;
+ erp[erp_idx].er_extbuf = (xfs_bmbt_rec_t *)
+ kmem_alloc(XFS_IEXT_BUFSZ, KM_SLEEP);
+ ifp->if_real_bytes = nlists * XFS_IEXT_BUFSZ;
+ memset(erp[erp_idx].er_extbuf, 0, XFS_IEXT_BUFSZ);
+ erp[erp_idx].er_extcount = 0;
+ erp[erp_idx].er_extoff = erp_idx > 0 ?
+ erp[erp_idx-1].er_extoff + erp[erp_idx-1].er_extcount : 0;
+ return (&erp[erp_idx]);
+}
+
+/*
+ * Remove a record from the indirection array.
+ */
+void
+xfs_iext_irec_remove(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ int erp_idx) /* irec index to remove */
+{
+ xfs_ext_irec_t *erp; /* indirection array pointer */
+ int i; /* loop counter */
+ int nlists; /* number of irec's (ex lists) */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ erp = &ifp->if_u1.if_ext_irec[erp_idx];
+ if (erp->er_extbuf) {
+ xfs_iext_irec_update_extoffs(ifp, erp_idx + 1,
+ -erp->er_extcount);
+ kmem_free(erp->er_extbuf, XFS_IEXT_BUFSZ);
+ }
+ /* Compact extent records */
+ erp = ifp->if_u1.if_ext_irec;
+ for (i = erp_idx; i < nlists - 1; i++) {
+ memmove(&erp[i], &erp[i+1], sizeof(xfs_ext_irec_t));
+ }
+ /*
+ * Manually free the last extent record from the indirection
+ * array. A call to xfs_iext_realloc_indirect() with a size
+ * of zero would result in a call to xfs_iext_destroy() which
+ * would in turn call this function again, creating a nasty
+ * infinite loop.
+ */
+ if (--nlists) {
+ xfs_iext_realloc_indirect(ifp,
+ nlists * sizeof(xfs_ext_irec_t));
+ } else {
+ kmem_free(ifp->if_u1.if_ext_irec,
+ sizeof(xfs_ext_irec_t));
+ }
+ ifp->if_real_bytes = nlists * XFS_IEXT_BUFSZ;
+}
+
+/*
+ * This is called to clean up large amounts of unused memory allocated
+ * by the indirection array. Before compacting anything though, verify
+ * that the indirection array is still needed and switch back to the
+ * linear extent list (or even the inline buffer) if possible. The
+ * compaction policy is as follows:
+ *
+ * Full Compaction: Extents fit into a single page (or inline buffer)
+ * Full Compaction: Extents occupy less than 10% of allocated space
+ * Partial Compaction: Extents occupy > 10% and < 50% of allocated space
+ * No Compaction: Extents occupy at least 50% of allocated space
+ */
+void
+xfs_iext_irec_compact(
+ xfs_ifork_t *ifp) /* inode fork pointer */
+{
+ xfs_extnum_t nextents; /* number of extents in file */
+ int nlists; /* number of irec's (ex lists) */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ nextents = ifp->if_bytes / (uint)sizeof(xfs_bmbt_rec_t);
+
+ if (nextents == 0) {
+ xfs_iext_destroy(ifp);
+ } else if (nextents <= XFS_INLINE_EXTS) {
+ xfs_iext_indirect_to_direct(ifp);
+ xfs_iext_direct_to_inline(ifp, nextents);
+ } else if (nextents <= XFS_LINEAR_EXTS) {
+ xfs_iext_indirect_to_direct(ifp);
+ } else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 3) {
+ xfs_iext_irec_compact_full(ifp);
+ } else if (nextents < (nlists * XFS_LINEAR_EXTS) >> 1) {
+ xfs_iext_irec_compact_pages(ifp);
+ }
+}
+
+/*
+ * Combine extents from neighboring extent pages.
+ */
+void
+xfs_iext_irec_compact_pages(
+ xfs_ifork_t *ifp) /* inode fork pointer */
+{
+ xfs_ext_irec_t *erp, *erp_next;/* pointers to irec entries */
+ int erp_idx = 0; /* indirection array index */
+ int nlists; /* number of irec's (ex lists) */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ while (erp_idx < nlists - 1) {
+ erp = &ifp->if_u1.if_ext_irec[erp_idx];
+ erp_next = erp + 1;
+ if (erp_next->er_extcount <=
+ (XFS_LINEAR_EXTS - erp->er_extcount)) {
+ memmove(&erp->er_extbuf[erp->er_extcount],
+ erp_next->er_extbuf, erp_next->er_extcount *
+ sizeof(xfs_bmbt_rec_t));
+ erp->er_extcount += erp_next->er_extcount;
+ /*
+ * Free page before removing extent record
+ * so er_extoffs don't get modified in
+ * xfs_iext_irec_remove.
+ */
+ kmem_free(erp_next->er_extbuf, XFS_IEXT_BUFSZ);
+ erp_next->er_extbuf = NULL;
+ xfs_iext_irec_remove(ifp, erp_idx + 1);
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ } else {
+ erp_idx++;
+ }
+ }
+}
+
+/*
+ * Fully compact the extent records managed by the indirection array.
+ */
+void
+xfs_iext_irec_compact_full(
+ xfs_ifork_t *ifp) /* inode fork pointer */
+{
+ xfs_bmbt_rec_t *ep, *ep_next; /* extent record pointers */
+ xfs_ext_irec_t *erp, *erp_next; /* extent irec pointers */
+ int erp_idx = 0; /* extent irec index */
+ int ext_avail; /* empty entries in ex list */
+ int ext_diff; /* number of exts to add */
+ int nlists; /* number of irec's (ex lists) */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ erp = ifp->if_u1.if_ext_irec;
+ ep = &erp->er_extbuf[erp->er_extcount];
+ erp_next = erp + 1;
+ ep_next = erp_next->er_extbuf;
+ while (erp_idx < nlists - 1) {
+ ext_avail = XFS_LINEAR_EXTS - erp->er_extcount;
+ ext_diff = MIN(ext_avail, erp_next->er_extcount);
+ memcpy(ep, ep_next, ext_diff * sizeof(xfs_bmbt_rec_t));
+ erp->er_extcount += ext_diff;
+ erp_next->er_extcount -= ext_diff;
+ /* Remove next page */
+ if (erp_next->er_extcount == 0) {
+ /*
+ * Free page before removing extent record
+ * so er_extoffs don't get modified in
+ * xfs_iext_irec_remove.
+ */
+ kmem_free(erp_next->er_extbuf,
+ erp_next->er_extcount * sizeof(xfs_bmbt_rec_t));
+ erp_next->er_extbuf = NULL;
+ xfs_iext_irec_remove(ifp, erp_idx + 1);
+ erp = &ifp->if_u1.if_ext_irec[erp_idx];
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ /* Update next page */
+ } else {
+ /* Move rest of page up to become next new page */
+ memmove(erp_next->er_extbuf, ep_next,
+ erp_next->er_extcount * sizeof(xfs_bmbt_rec_t));
+ ep_next = erp_next->er_extbuf;
+ memset(&ep_next[erp_next->er_extcount], 0,
+ (XFS_LINEAR_EXTS - erp_next->er_extcount) *
+ sizeof(xfs_bmbt_rec_t));
+ }
+ if (erp->er_extcount == XFS_LINEAR_EXTS) {
+ erp_idx++;
+ if (erp_idx < nlists)
+ erp = &ifp->if_u1.if_ext_irec[erp_idx];
+ else
+ break;
+ }
+ ep = &erp->er_extbuf[erp->er_extcount];
+ erp_next = erp + 1;
+ ep_next = erp_next->er_extbuf;
+ }
+}
+
+/*
+ * This is called to update the er_extoff field in the indirection
+ * array when extents have been added or removed from one of the
+ * extent lists. erp_idx contains the irec index to begin updating
+ * at and ext_diff contains the number of extents that were added
+ * or removed.
+ */
+void
+xfs_iext_irec_update_extoffs(
+ xfs_ifork_t *ifp, /* inode fork pointer */
+ int erp_idx, /* irec index to update */
+ int ext_diff) /* number of new extents */
+{
+ int i; /* loop counter */
+ int nlists; /* number of irec's (ex lists */
+
+ ASSERT(ifp->if_flags & XFS_IFEXTIREC);
+ nlists = ifp->if_real_bytes / XFS_IEXT_BUFSZ;
+ for (i = erp_idx; i < nlists; i++) {
+ ifp->if_u1.if_ext_irec[i].er_extoff += ext_diff;
+ }
+}
diff --git a/fs/xfs/xfs_inode.h b/fs/xfs/xfs_inode.h
index 1cfbcf18ce86..39ef9c36ea55 100644
--- a/fs/xfs/xfs_inode.h
+++ b/fs/xfs/xfs_inode.h
@@ -25,10 +25,37 @@
#define XFS_ATTR_FORK 1
/*
+ * The following xfs_ext_irec_t struct introduces a second (top) level
+ * to the in-core extent allocation scheme. These structs are allocated
+ * in a contiguous block, creating an indirection array where each entry
+ * (irec) contains a pointer to a buffer of in-core extent records which
+ * it manages. Each extent buffer is 4k in size, since 4k is the system
+ * page size on Linux i386 and systems with larger page sizes don't seem
+ * to gain much, if anything, by using their native page size as the
+ * extent buffer size. Also, using 4k extent buffers everywhere provides
+ * a consistent interface for CXFS across different platforms.
+ *
+ * There is currently no limit on the number of irec's (extent lists)
+ * allowed, so heavily fragmented files may require an indirection array
+ * which spans multiple system pages of memory. The number of extents
+ * which would require this amount of contiguous memory is very large
+ * and should not cause problems in the foreseeable future. However,
+ * if the memory needed for the contiguous array ever becomes a problem,
+ * it is possible that a third level of indirection may be required.
+ */
+typedef struct xfs_ext_irec {
+ xfs_bmbt_rec_t *er_extbuf; /* block of extent records */
+ xfs_extnum_t er_extoff; /* extent offset in file */
+ xfs_extnum_t er_extcount; /* number of extents in page/block */
+} xfs_ext_irec_t;
+
+/*
* File incore extent information, present for each of data & attr forks.
*/
-#define XFS_INLINE_EXTS 2
-#define XFS_INLINE_DATA 32
+#define XFS_IEXT_BUFSZ 4096
+#define XFS_LINEAR_EXTS (XFS_IEXT_BUFSZ / (uint)sizeof(xfs_bmbt_rec_t))
+#define XFS_INLINE_EXTS 2
+#define XFS_INLINE_DATA 32
typedef struct xfs_ifork {
int if_bytes; /* bytes in if_u1 */
int if_real_bytes; /* bytes allocated in if_u1 */
@@ -39,6 +66,7 @@ typedef struct xfs_ifork {
xfs_extnum_t if_lastex; /* last if_extents used */
union {
xfs_bmbt_rec_t *if_extents; /* linear map file exts */
+ xfs_ext_irec_t *if_ext_irec; /* irec map file exts */
char *if_data; /* inline file data */
} if_u1;
union {
@@ -61,20 +89,16 @@ typedef struct xfs_ifork {
/*
* Per-fork incore inode flags.
*/
-#define XFS_IFINLINE 0x0001 /* Inline data is read in */
-#define XFS_IFEXTENTS 0x0002 /* All extent pointers are read in */
-#define XFS_IFBROOT 0x0004 /* i_broot points to the bmap b-tree root */
+#define XFS_IFINLINE 0x01 /* Inline data is read in */
+#define XFS_IFEXTENTS 0x02 /* All extent pointers are read in */
+#define XFS_IFBROOT 0x04 /* i_broot points to the bmap b-tree root */
+#define XFS_IFEXTIREC 0x08 /* Indirection array of extent blocks */
/*
- * Flags for xfs_imap() and xfs_dilocate().
+ * Flags for xfs_itobp(), xfs_imap() and xfs_dilocate().
*/
-#define XFS_IMAP_LOOKUP 0x1
-
-/*
- * Maximum number of extent pointers in if_u1.if_extents.
- */
-#define XFS_MAX_INCORE_EXTENTS 32768
-
+#define XFS_IMAP_LOOKUP 0x1
+#define XFS_IMAP_BULKSTAT 0x2
#ifdef __KERNEL__
struct bhv_desc;
@@ -398,7 +422,7 @@ int xfs_finish_reclaim_all(struct xfs_mount *, int);
*/
int xfs_itobp(struct xfs_mount *, struct xfs_trans *,
xfs_inode_t *, xfs_dinode_t **, struct xfs_buf **,
- xfs_daddr_t);
+ xfs_daddr_t, uint);
int xfs_iread(struct xfs_mount *, struct xfs_trans *, xfs_ino_t,
xfs_inode_t **, xfs_daddr_t);
int xfs_iread_extents(struct xfs_trans *, xfs_inode_t *, int);
@@ -440,6 +464,32 @@ xfs_inode_t *xfs_vtoi(struct vnode *vp);
void xfs_synchronize_atime(xfs_inode_t *);
+xfs_bmbt_rec_t *xfs_iext_get_ext(xfs_ifork_t *, xfs_extnum_t);
+void xfs_iext_insert(xfs_ifork_t *, xfs_extnum_t, xfs_extnum_t,
+ xfs_bmbt_irec_t *);
+void xfs_iext_add(xfs_ifork_t *, xfs_extnum_t, int);
+void xfs_iext_add_indirect_multi(xfs_ifork_t *, int, xfs_extnum_t, int);
+void xfs_iext_remove(xfs_ifork_t *, xfs_extnum_t, int);
+void xfs_iext_remove_inline(xfs_ifork_t *, xfs_extnum_t, int);
+void xfs_iext_remove_direct(xfs_ifork_t *, xfs_extnum_t, int);
+void xfs_iext_remove_indirect(xfs_ifork_t *, xfs_extnum_t, int);
+void xfs_iext_realloc_direct(xfs_ifork_t *, int);
+void xfs_iext_realloc_indirect(xfs_ifork_t *, int);
+void xfs_iext_indirect_to_direct(xfs_ifork_t *);
+void xfs_iext_direct_to_inline(xfs_ifork_t *, xfs_extnum_t);
+void xfs_iext_inline_to_direct(xfs_ifork_t *, int);
+void xfs_iext_destroy(xfs_ifork_t *);
+xfs_bmbt_rec_t *xfs_iext_bno_to_ext(xfs_ifork_t *, xfs_fileoff_t, int *);
+xfs_ext_irec_t *xfs_iext_bno_to_irec(xfs_ifork_t *, xfs_fileoff_t, int *);
+xfs_ext_irec_t *xfs_iext_idx_to_irec(xfs_ifork_t *, xfs_extnum_t *, int *, int);
+void xfs_iext_irec_init(xfs_ifork_t *);
+xfs_ext_irec_t *xfs_iext_irec_new(xfs_ifork_t *, int);
+void xfs_iext_irec_remove(xfs_ifork_t *, int);
+void xfs_iext_irec_compact(xfs_ifork_t *);
+void xfs_iext_irec_compact_pages(xfs_ifork_t *);
+void xfs_iext_irec_compact_full(xfs_ifork_t *);
+void xfs_iext_irec_update_extoffs(xfs_ifork_t *, int, int);
+
#define xfs_ipincount(ip) ((unsigned int) atomic_read(&ip->i_pincount))
#ifdef DEBUG
diff --git a/fs/xfs/xfs_iomap.c b/fs/xfs/xfs_iomap.c
index 788917f355c4..d5dfedcb8922 100644
--- a/fs/xfs/xfs_iomap.c
+++ b/fs/xfs/xfs_iomap.c
@@ -76,7 +76,7 @@ xfs_iomap_enter_trace(
(void *)((unsigned long)count),
(void *)((unsigned long)((io->io_new_size >> 32) & 0xffffffff)),
(void *)((unsigned long)(io->io_new_size & 0xffffffff)),
- (void *)NULL,
+ (void *)((unsigned long)current_pid()),
(void *)NULL,
(void *)NULL,
(void *)NULL,
diff --git a/fs/xfs/xfs_itable.c b/fs/xfs/xfs_itable.c
index c59450e1be40..32247b6bfee7 100644
--- a/fs/xfs/xfs_itable.c
+++ b/fs/xfs/xfs_itable.c
@@ -562,7 +562,8 @@ xfs_bulkstat(
if (bp)
xfs_buf_relse(bp);
error = xfs_itobp(mp, NULL, ip,
- &dip, &bp, bno);
+ &dip, &bp, bno,
+ XFS_IMAP_BULKSTAT);
if (!error)
clustidx = ip->i_boffset / mp->m_sb.sb_inodesize;
kmem_zone_free(xfs_inode_zone, ip);
@@ -570,6 +571,8 @@ xfs_bulkstat(
mp, XFS_ERRTAG_BULKSTAT_READ_CHUNK,
XFS_RANDOM_BULKSTAT_READ_CHUNK)) {
bp = NULL;
+ ubleft = 0;
+ rval = error;
break;
}
}
diff --git a/fs/xfs/xfs_log_recover.c b/fs/xfs/xfs_log_recover.c
index 7d46cbd6a07a..add13f507ed2 100644
--- a/fs/xfs/xfs_log_recover.c
+++ b/fs/xfs/xfs_log_recover.c
@@ -3249,7 +3249,7 @@ xlog_recover_process_iunlinks(
* next inode in the bucket.
*/
error = xfs_itobp(mp, NULL, ip, &dip,
- &ibp, 0);
+ &ibp, 0, 0);
ASSERT(error || (dip != NULL));
}
diff --git a/fs/xfs/xfs_mount.c b/fs/xfs/xfs_mount.c
index 62188ea392c7..20e8abc16d18 100644
--- a/fs/xfs/xfs_mount.c
+++ b/fs/xfs/xfs_mount.c
@@ -51,11 +51,32 @@ STATIC int xfs_uuid_mount(xfs_mount_t *);
STATIC void xfs_uuid_unmount(xfs_mount_t *mp);
STATIC void xfs_unmountfs_wait(xfs_mount_t *);
+
+#ifdef HAVE_PERCPU_SB
+STATIC void xfs_icsb_destroy_counters(xfs_mount_t *);
+STATIC void xfs_icsb_balance_counter(xfs_mount_t *, xfs_sb_field_t, int);
+STATIC void xfs_icsb_sync_counters(xfs_mount_t *);
+STATIC int xfs_icsb_modify_counters(xfs_mount_t *, xfs_sb_field_t,
+ int, int);
+STATIC int xfs_icsb_modify_counters_locked(xfs_mount_t *, xfs_sb_field_t,
+ int, int);
+STATIC int xfs_icsb_disable_counter(xfs_mount_t *, xfs_sb_field_t);
+
+#else
+
+#define xfs_icsb_destroy_counters(mp) do { } while (0)
+#define xfs_icsb_balance_counter(mp, a, b) do { } while (0)
+#define xfs_icsb_sync_counters(mp) do { } while (0)
+#define xfs_icsb_modify_counters(mp, a, b, c) do { } while (0)
+#define xfs_icsb_modify_counters_locked(mp, a, b, c) do { } while (0)
+
+#endif
+
static const struct {
- short offset;
- short type; /* 0 = integer
- * 1 = binary / string (no translation)
- */
+ short offset;
+ short type; /* 0 = integer
+ * 1 = binary / string (no translation)
+ */
} xfs_sb_info[] = {
{ offsetof(xfs_sb_t, sb_magicnum), 0 },
{ offsetof(xfs_sb_t, sb_blocksize), 0 },
@@ -113,7 +134,11 @@ xfs_mount_init(void)
{
xfs_mount_t *mp;
- mp = kmem_zalloc(sizeof(*mp), KM_SLEEP);
+ mp = kmem_zalloc(sizeof(xfs_mount_t), KM_SLEEP);
+
+ if (xfs_icsb_init_counters(mp)) {
+ mp->m_flags |= XFS_MOUNT_NO_PERCPU_SB;
+ }
AIL_LOCKINIT(&mp->m_ail_lock, "xfs_ail");
spinlock_init(&mp->m_sb_lock, "xfs_sb");
@@ -136,8 +161,8 @@ xfs_mount_init(void)
*/
void
xfs_mount_free(
- xfs_mount_t *mp,
- int remove_bhv)
+ xfs_mount_t *mp,
+ int remove_bhv)
{
if (mp->m_ihash)
xfs_ihash_free(mp);
@@ -177,6 +202,7 @@ xfs_mount_free(
VFS_REMOVEBHV(vfsp, &mp->m_bhv);
}
+ xfs_icsb_destroy_counters(mp);
kmem_free(mp, sizeof(xfs_mount_t));
}
@@ -242,9 +268,12 @@ xfs_mount_validate_sb(
sbp->sb_blocklog > XFS_MAX_BLOCKSIZE_LOG ||
sbp->sb_inodesize < XFS_DINODE_MIN_SIZE ||
sbp->sb_inodesize > XFS_DINODE_MAX_SIZE ||
+ sbp->sb_inodelog < XFS_DINODE_MIN_LOG ||
+ sbp->sb_inodelog > XFS_DINODE_MAX_LOG ||
+ (sbp->sb_blocklog - sbp->sb_inodelog != sbp->sb_inopblog) ||
(sbp->sb_rextsize * sbp->sb_blocksize > XFS_MAX_RTEXTSIZE) ||
(sbp->sb_rextsize * sbp->sb_blocksize < XFS_MIN_RTEXTSIZE) ||
- sbp->sb_imax_pct > 100)) {
+ (sbp->sb_imax_pct > 100 || sbp->sb_imax_pct < 1))) {
cmn_err(CE_WARN, "XFS: SB sanity check 1 failed");
XFS_CORRUPTION_ERROR("xfs_mount_validate_sb(3)",
XFS_ERRLEVEL_LOW, mp, sbp);
@@ -527,6 +556,10 @@ xfs_readsb(xfs_mount_t *mp)
ASSERT(XFS_BUF_VALUSEMA(bp) <= 0);
}
+ xfs_icsb_balance_counter(mp, XFS_SBS_ICOUNT, 0);
+ xfs_icsb_balance_counter(mp, XFS_SBS_IFREE, 0);
+ xfs_icsb_balance_counter(mp, XFS_SBS_FDBLOCKS, 0);
+
mp->m_sb_bp = bp;
xfs_buf_relse(bp);
ASSERT(XFS_BUF_VALUSEMA(bp) > 0);
@@ -1154,6 +1187,9 @@ xfs_unmountfs_writesb(xfs_mount_t *mp)
sbp = xfs_getsb(mp, 0);
if (!(XFS_MTOVFS(mp)->vfs_flag & VFS_RDONLY ||
XFS_FORCED_SHUTDOWN(mp))) {
+
+ xfs_icsb_sync_counters(mp);
+
/*
* mark shared-readonly if desired
*/
@@ -1227,7 +1263,6 @@ xfs_mod_sb(xfs_trans_t *tp, __int64_t fields)
xfs_trans_log_buf(tp, bp, first, last);
}
-
/*
* xfs_mod_incore_sb_unlocked() is a utility routine common used to apply
* a delta to a specified field in the in-core superblock. Simply
@@ -1237,7 +1272,7 @@ xfs_mod_sb(xfs_trans_t *tp, __int64_t fields)
*
* The SB_LOCK must be held when this routine is called.
*/
-STATIC int
+int
xfs_mod_incore_sb_unlocked(xfs_mount_t *mp, xfs_sb_field_t field,
int delta, int rsvd)
{
@@ -1406,9 +1441,26 @@ xfs_mod_incore_sb(xfs_mount_t *mp, xfs_sb_field_t field, int delta, int rsvd)
unsigned long s;
int status;
- s = XFS_SB_LOCK(mp);
- status = xfs_mod_incore_sb_unlocked(mp, field, delta, rsvd);
- XFS_SB_UNLOCK(mp, s);
+ /* check for per-cpu counters */
+ switch (field) {
+#ifdef HAVE_PERCPU_SB
+ case XFS_SBS_ICOUNT:
+ case XFS_SBS_IFREE:
+ case XFS_SBS_FDBLOCKS:
+ if (!(mp->m_flags & XFS_MOUNT_NO_PERCPU_SB)) {
+ status = xfs_icsb_modify_counters(mp, field,
+ delta, rsvd);
+ break;
+ }
+ /* FALLTHROUGH */
+#endif
+ default:
+ s = XFS_SB_LOCK(mp);
+ status = xfs_mod_incore_sb_unlocked(mp, field, delta, rsvd);
+ XFS_SB_UNLOCK(mp, s);
+ break;
+ }
+
return status;
}
@@ -1445,8 +1497,26 @@ xfs_mod_incore_sb_batch(xfs_mount_t *mp, xfs_mod_sb_t *msb, uint nmsb, int rsvd)
* from the loop so we'll fall into the undo loop
* below.
*/
- status = xfs_mod_incore_sb_unlocked(mp, msbp->msb_field,
- msbp->msb_delta, rsvd);
+ switch (msbp->msb_field) {
+#ifdef HAVE_PERCPU_SB
+ case XFS_SBS_ICOUNT:
+ case XFS_SBS_IFREE:
+ case XFS_SBS_FDBLOCKS:
+ if (!(mp->m_flags & XFS_MOUNT_NO_PERCPU_SB)) {
+ status = xfs_icsb_modify_counters_locked(mp,
+ msbp->msb_field,
+ msbp->msb_delta, rsvd);
+ break;
+ }
+ /* FALLTHROUGH */
+#endif
+ default:
+ status = xfs_mod_incore_sb_unlocked(mp,
+ msbp->msb_field,
+ msbp->msb_delta, rsvd);
+ break;
+ }
+
if (status != 0) {
break;
}
@@ -1463,8 +1533,28 @@ xfs_mod_incore_sb_batch(xfs_mount_t *mp, xfs_mod_sb_t *msb, uint nmsb, int rsvd)
if (status != 0) {
msbp--;
while (msbp >= msb) {
- status = xfs_mod_incore_sb_unlocked(mp,
- msbp->msb_field, -(msbp->msb_delta), rsvd);
+ switch (msbp->msb_field) {
+#ifdef HAVE_PERCPU_SB
+ case XFS_SBS_ICOUNT:
+ case XFS_SBS_IFREE:
+ case XFS_SBS_FDBLOCKS:
+ if (!(mp->m_flags & XFS_MOUNT_NO_PERCPU_SB)) {
+ status =
+ xfs_icsb_modify_counters_locked(mp,
+ msbp->msb_field,
+ -(msbp->msb_delta),
+ rsvd);
+ break;
+ }
+ /* FALLTHROUGH */
+#endif
+ default:
+ status = xfs_mod_incore_sb_unlocked(mp,
+ msbp->msb_field,
+ -(msbp->msb_delta),
+ rsvd);
+ break;
+ }
ASSERT(status == 0);
msbp--;
}
@@ -1577,3 +1667,525 @@ xfs_mount_log_sbunit(
xfs_mod_sb(tp, fields);
xfs_trans_commit(tp, 0, NULL);
}
+
+
+#ifdef HAVE_PERCPU_SB
+/*
+ * Per-cpu incore superblock counters
+ *
+ * Simple concept, difficult implementation
+ *
+ * Basically, replace the incore superblock counters with a distributed per cpu
+ * counter for contended fields (e.g. free block count).
+ *
+ * Difficulties arise in that the incore sb is used for ENOSPC checking, and
+ * hence needs to be accurately read when we are running low on space. Hence
+ * there is a method to enable and disable the per-cpu counters based on how
+ * much "stuff" is available in them.
+ *
+ * Basically, a counter is enabled if there is enough free resource to justify
+ * running a per-cpu fast-path. If the per-cpu counter runs out (i.e. a local
+ * ENOSPC), then we disable the counters to synchronise all callers and
+ * re-distribute the available resources.
+ *
+ * If, once we redistributed the available resources, we still get a failure,
+ * we disable the per-cpu counter and go through the slow path.
+ *
+ * The slow path is the current xfs_mod_incore_sb() function. This means that
+ * when we disable a per-cpu counter, we need to drain it's resources back to
+ * the global superblock. We do this after disabling the counter to prevent
+ * more threads from queueing up on the counter.
+ *
+ * Essentially, this means that we still need a lock in the fast path to enable
+ * synchronisation between the global counters and the per-cpu counters. This
+ * is not a problem because the lock will be local to a CPU almost all the time
+ * and have little contention except when we get to ENOSPC conditions.
+ *
+ * Basically, this lock becomes a barrier that enables us to lock out the fast
+ * path while we do things like enabling and disabling counters and
+ * synchronising the counters.
+ *
+ * Locking rules:
+ *
+ * 1. XFS_SB_LOCK() before picking up per-cpu locks
+ * 2. per-cpu locks always picked up via for_each_online_cpu() order
+ * 3. accurate counter sync requires XFS_SB_LOCK + per cpu locks
+ * 4. modifying per-cpu counters requires holding per-cpu lock
+ * 5. modifying global counters requires holding XFS_SB_LOCK
+ * 6. enabling or disabling a counter requires holding the XFS_SB_LOCK
+ * and _none_ of the per-cpu locks.
+ *
+ * Disabled counters are only ever re-enabled by a balance operation
+ * that results in more free resources per CPU than a given threshold.
+ * To ensure counters don't remain disabled, they are rebalanced when
+ * the global resource goes above a higher threshold (i.e. some hysteresis
+ * is present to prevent thrashing).
+ */
+
+/*
+ * hot-plug CPU notifier support.
+ *
+ * We cannot use the hotcpu_register() function because it does
+ * not allow notifier instances. We need a notifier per filesystem
+ * as we need to be able to identify the filesystem to balance
+ * the counters out. This is acheived by having a notifier block
+ * embedded in the xfs_mount_t and doing pointer magic to get the
+ * mount pointer from the notifier block address.
+ */
+STATIC int
+xfs_icsb_cpu_notify(
+ struct notifier_block *nfb,
+ unsigned long action,
+ void *hcpu)
+{
+ xfs_icsb_cnts_t *cntp;
+ xfs_mount_t *mp;
+ int s;
+
+ mp = (xfs_mount_t *)container_of(nfb, xfs_mount_t, m_icsb_notifier);
+ cntp = (xfs_icsb_cnts_t *)
+ per_cpu_ptr(mp->m_sb_cnts, (unsigned long)hcpu);
+ switch (action) {
+ case CPU_UP_PREPARE:
+ /* Easy Case - initialize the area and locks, and
+ * then rebalance when online does everything else for us. */
+ memset(cntp, 0, sizeof(xfs_icsb_cnts_t));
+ break;
+ case CPU_ONLINE:
+ xfs_icsb_balance_counter(mp, XFS_SBS_ICOUNT, 0);
+ xfs_icsb_balance_counter(mp, XFS_SBS_IFREE, 0);
+ xfs_icsb_balance_counter(mp, XFS_SBS_FDBLOCKS, 0);
+ break;
+ case CPU_DEAD:
+ /* Disable all the counters, then fold the dead cpu's
+ * count into the total on the global superblock and
+ * re-enable the counters. */
+ s = XFS_SB_LOCK(mp);
+ xfs_icsb_disable_counter(mp, XFS_SBS_ICOUNT);
+ xfs_icsb_disable_counter(mp, XFS_SBS_IFREE);
+ xfs_icsb_disable_counter(mp, XFS_SBS_FDBLOCKS);
+
+ mp->m_sb.sb_icount += cntp->icsb_icount;
+ mp->m_sb.sb_ifree += cntp->icsb_ifree;
+ mp->m_sb.sb_fdblocks += cntp->icsb_fdblocks;
+
+ memset(cntp, 0, sizeof(xfs_icsb_cnts_t));
+
+ xfs_icsb_balance_counter(mp, XFS_SBS_ICOUNT, XFS_ICSB_SB_LOCKED);
+ xfs_icsb_balance_counter(mp, XFS_SBS_IFREE, XFS_ICSB_SB_LOCKED);
+ xfs_icsb_balance_counter(mp, XFS_SBS_FDBLOCKS, XFS_ICSB_SB_LOCKED);
+ XFS_SB_UNLOCK(mp, s);
+ break;
+ }
+
+ return NOTIFY_OK;
+}
+
+int
+xfs_icsb_init_counters(
+ xfs_mount_t *mp)
+{
+ xfs_icsb_cnts_t *cntp;
+ int i;
+
+ mp->m_sb_cnts = alloc_percpu(xfs_icsb_cnts_t);
+ if (mp->m_sb_cnts == NULL)
+ return -ENOMEM;
+
+ mp->m_icsb_notifier.notifier_call = xfs_icsb_cpu_notify;
+ mp->m_icsb_notifier.priority = 0;
+ register_cpu_notifier(&mp->m_icsb_notifier);
+
+ for_each_online_cpu(i) {
+ cntp = (xfs_icsb_cnts_t *)per_cpu_ptr(mp->m_sb_cnts, i);
+ memset(cntp, 0, sizeof(xfs_icsb_cnts_t));
+ }
+ /*
+ * start with all counters disabled so that the
+ * initial balance kicks us off correctly
+ */
+ mp->m_icsb_counters = -1;
+ return 0;
+}
+
+STATIC void
+xfs_icsb_destroy_counters(
+ xfs_mount_t *mp)
+{
+ if (mp->m_sb_cnts) {
+ unregister_cpu_notifier(&mp->m_icsb_notifier);
+ free_percpu(mp->m_sb_cnts);
+ }
+}
+
+STATIC inline void
+xfs_icsb_lock_cntr(
+ xfs_icsb_cnts_t *icsbp)
+{
+ while (test_and_set_bit(XFS_ICSB_FLAG_LOCK, &icsbp->icsb_flags)) {
+ ndelay(1000);
+ }
+}
+
+STATIC inline void
+xfs_icsb_unlock_cntr(
+ xfs_icsb_cnts_t *icsbp)
+{
+ clear_bit(XFS_ICSB_FLAG_LOCK, &icsbp->icsb_flags);
+}
+
+
+STATIC inline void
+xfs_icsb_lock_all_counters(
+ xfs_mount_t *mp)
+{
+ xfs_icsb_cnts_t *cntp;
+ int i;
+
+ for_each_online_cpu(i) {
+ cntp = (xfs_icsb_cnts_t *)per_cpu_ptr(mp->m_sb_cnts, i);
+ xfs_icsb_lock_cntr(cntp);
+ }
+}
+
+STATIC inline void
+xfs_icsb_unlock_all_counters(
+ xfs_mount_t *mp)
+{
+ xfs_icsb_cnts_t *cntp;
+ int i;
+
+ for_each_online_cpu(i) {
+ cntp = (xfs_icsb_cnts_t *)per_cpu_ptr(mp->m_sb_cnts, i);
+ xfs_icsb_unlock_cntr(cntp);
+ }
+}
+
+STATIC void
+xfs_icsb_count(
+ xfs_mount_t *mp,
+ xfs_icsb_cnts_t *cnt,
+ int flags)
+{
+ xfs_icsb_cnts_t *cntp;
+ int i;
+
+ memset(cnt, 0, sizeof(xfs_icsb_cnts_t));
+
+ if (!(flags & XFS_ICSB_LAZY_COUNT))
+ xfs_icsb_lock_all_counters(mp);
+
+ for_each_online_cpu(i) {
+ cntp = (xfs_icsb_cnts_t *)per_cpu_ptr(mp->m_sb_cnts, i);
+ cnt->icsb_icount += cntp->icsb_icount;
+ cnt->icsb_ifree += cntp->icsb_ifree;
+ cnt->icsb_fdblocks += cntp->icsb_fdblocks;
+ }
+
+ if (!(flags & XFS_ICSB_LAZY_COUNT))
+ xfs_icsb_unlock_all_counters(mp);
+}
+
+STATIC int
+xfs_icsb_counter_disabled(
+ xfs_mount_t *mp,
+ xfs_sb_field_t field)
+{
+ ASSERT((field >= XFS_SBS_ICOUNT) && (field <= XFS_SBS_FDBLOCKS));
+ return test_bit(field, &mp->m_icsb_counters);
+}
+
+STATIC int
+xfs_icsb_disable_counter(
+ xfs_mount_t *mp,
+ xfs_sb_field_t field)
+{
+ xfs_icsb_cnts_t cnt;
+
+ ASSERT((field >= XFS_SBS_ICOUNT) && (field <= XFS_SBS_FDBLOCKS));
+
+ xfs_icsb_lock_all_counters(mp);
+ if (!test_and_set_bit(field, &mp->m_icsb_counters)) {
+ /* drain back to superblock */
+
+ xfs_icsb_count(mp, &cnt, XFS_ICSB_SB_LOCKED|XFS_ICSB_LAZY_COUNT);
+ switch(field) {
+ case XFS_SBS_ICOUNT:
+ mp->m_sb.sb_icount = cnt.icsb_icount;
+ break;
+ case XFS_SBS_IFREE:
+ mp->m_sb.sb_ifree = cnt.icsb_ifree;
+ break;
+ case XFS_SBS_FDBLOCKS:
+ mp->m_sb.sb_fdblocks = cnt.icsb_fdblocks;
+ break;
+ default:
+ BUG();
+ }
+ }
+
+ xfs_icsb_unlock_all_counters(mp);
+
+ return 0;
+}
+
+STATIC void
+xfs_icsb_enable_counter(
+ xfs_mount_t *mp,
+ xfs_sb_field_t field,
+ uint64_t count,
+ uint64_t resid)
+{
+ xfs_icsb_cnts_t *cntp;
+ int i;
+
+ ASSERT((field >= XFS_SBS_ICOUNT) && (field <= XFS_SBS_FDBLOCKS));
+
+ xfs_icsb_lock_all_counters(mp);
+ for_each_online_cpu(i) {
+ cntp = per_cpu_ptr(mp->m_sb_cnts, i);
+ switch (field) {
+ case XFS_SBS_ICOUNT:
+ cntp->icsb_icount = count + resid;
+ break;
+ case XFS_SBS_IFREE:
+ cntp->icsb_ifree = count + resid;
+ break;
+ case XFS_SBS_FDBLOCKS:
+ cntp->icsb_fdblocks = count + resid;
+ break;
+ default:
+ BUG();
+ break;
+ }
+ resid = 0;
+ }
+ clear_bit(field, &mp->m_icsb_counters);
+ xfs_icsb_unlock_all_counters(mp);
+}
+
+STATIC void
+xfs_icsb_sync_counters_int(
+ xfs_mount_t *mp,
+ int flags)
+{
+ xfs_icsb_cnts_t cnt;
+ int s;
+
+ /* Pass 1: lock all counters */
+ if ((flags & XFS_ICSB_SB_LOCKED) == 0)
+ s = XFS_SB_LOCK(mp);
+
+ xfs_icsb_count(mp, &cnt, flags);
+
+ /* Step 3: update mp->m_sb fields */
+ if (!xfs_icsb_counter_disabled(mp, XFS_SBS_ICOUNT))
+ mp->m_sb.sb_icount = cnt.icsb_icount;
+ if (!xfs_icsb_counter_disabled(mp, XFS_SBS_IFREE))
+ mp->m_sb.sb_ifree = cnt.icsb_ifree;
+ if (!xfs_icsb_counter_disabled(mp, XFS_SBS_FDBLOCKS))
+ mp->m_sb.sb_fdblocks = cnt.icsb_fdblocks;
+
+ if ((flags & XFS_ICSB_SB_LOCKED) == 0)
+ XFS_SB_UNLOCK(mp, s);
+}
+
+/*
+ * Accurate update of per-cpu counters to incore superblock
+ */
+STATIC void
+xfs_icsb_sync_counters(
+ xfs_mount_t *mp)
+{
+ xfs_icsb_sync_counters_int(mp, 0);
+}
+
+/*
+ * lazy addition used for things like df, background sb syncs, etc
+ */
+void
+xfs_icsb_sync_counters_lazy(
+ xfs_mount_t *mp)
+{
+ xfs_icsb_sync_counters_int(mp, XFS_ICSB_LAZY_COUNT);
+}
+
+/*
+ * Balance and enable/disable counters as necessary.
+ *
+ * Thresholds for re-enabling counters are somewhat magic.
+ * inode counts are chosen to be the same number as single
+ * on disk allocation chunk per CPU, and free blocks is
+ * something far enough zero that we aren't going thrash
+ * when we get near ENOSPC.
+ */
+#define XFS_ICSB_INO_CNTR_REENABLE 64
+#define XFS_ICSB_FDBLK_CNTR_REENABLE 512
+STATIC void
+xfs_icsb_balance_counter(
+ xfs_mount_t *mp,
+ xfs_sb_field_t field,
+ int flags)
+{
+ uint64_t count, resid = 0;
+ int weight = num_online_cpus();
+ int s;
+
+ if (!(flags & XFS_ICSB_SB_LOCKED))
+ s = XFS_SB_LOCK(mp);
+
+ /* disable counter and sync counter */
+ xfs_icsb_disable_counter(mp, field);
+
+ /* update counters - first CPU gets residual*/
+ switch (field) {
+ case XFS_SBS_ICOUNT:
+ count = mp->m_sb.sb_icount;
+ resid = do_div(count, weight);
+ if (count < XFS_ICSB_INO_CNTR_REENABLE)
+ goto out;
+ break;
+ case XFS_SBS_IFREE:
+ count = mp->m_sb.sb_ifree;
+ resid = do_div(count, weight);
+ if (count < XFS_ICSB_INO_CNTR_REENABLE)
+ goto out;
+ break;
+ case XFS_SBS_FDBLOCKS:
+ count = mp->m_sb.sb_fdblocks;
+ resid = do_div(count, weight);
+ if (count < XFS_ICSB_FDBLK_CNTR_REENABLE)
+ goto out;
+ break;
+ default:
+ BUG();
+ break;
+ }
+
+ xfs_icsb_enable_counter(mp, field, count, resid);
+out:
+ if (!(flags & XFS_ICSB_SB_LOCKED))
+ XFS_SB_UNLOCK(mp, s);
+}
+
+STATIC int
+xfs_icsb_modify_counters_int(
+ xfs_mount_t *mp,
+ xfs_sb_field_t field,
+ int delta,
+ int rsvd,
+ int flags)
+{
+ xfs_icsb_cnts_t *icsbp;
+ long long lcounter; /* long counter for 64 bit fields */
+ int cpu, s, locked = 0;
+ int ret = 0, balance_done = 0;
+
+again:
+ cpu = get_cpu();
+ icsbp = (xfs_icsb_cnts_t *)per_cpu_ptr(mp->m_sb_cnts, cpu),
+ xfs_icsb_lock_cntr(icsbp);
+ if (unlikely(xfs_icsb_counter_disabled(mp, field)))
+ goto slow_path;
+
+ switch (field) {
+ case XFS_SBS_ICOUNT:
+ lcounter = icsbp->icsb_icount;
+ lcounter += delta;
+ if (unlikely(lcounter < 0))
+ goto slow_path;
+ icsbp->icsb_icount = lcounter;
+ break;
+
+ case XFS_SBS_IFREE:
+ lcounter = icsbp->icsb_ifree;
+ lcounter += delta;
+ if (unlikely(lcounter < 0))
+ goto slow_path;
+ icsbp->icsb_ifree = lcounter;
+ break;
+
+ case XFS_SBS_FDBLOCKS:
+ BUG_ON((mp->m_resblks - mp->m_resblks_avail) != 0);
+
+ lcounter = icsbp->icsb_fdblocks;
+ lcounter += delta;
+ if (unlikely(lcounter < 0))
+ goto slow_path;
+ icsbp->icsb_fdblocks = lcounter;
+ break;
+ default:
+ BUG();
+ break;
+ }
+ xfs_icsb_unlock_cntr(icsbp);
+ put_cpu();
+ if (locked)
+ XFS_SB_UNLOCK(mp, s);
+ return 0;
+
+ /*
+ * The slow path needs to be run with the SBLOCK
+ * held so that we prevent other threads from
+ * attempting to run this path at the same time.
+ * this provides exclusion for the balancing code,
+ * and exclusive fallback if the balance does not
+ * provide enough resources to continue in an unlocked
+ * manner.
+ */
+slow_path:
+ xfs_icsb_unlock_cntr(icsbp);
+ put_cpu();
+
+ /* need to hold superblock incase we need
+ * to disable a counter */
+ if (!(flags & XFS_ICSB_SB_LOCKED)) {
+ s = XFS_SB_LOCK(mp);
+ locked = 1;
+ flags |= XFS_ICSB_SB_LOCKED;
+ }
+ if (!balance_done) {
+ xfs_icsb_balance_counter(mp, field, flags);
+ balance_done = 1;
+ goto again;
+ } else {
+ /*
+ * we might not have enough on this local
+ * cpu to allocate for a bulk request.
+ * We need to drain this field from all CPUs
+ * and disable the counter fastpath
+ */
+ xfs_icsb_disable_counter(mp, field);
+ }
+
+ ret = xfs_mod_incore_sb_unlocked(mp, field, delta, rsvd);
+
+ if (locked)
+ XFS_SB_UNLOCK(mp, s);
+ return ret;
+}
+
+STATIC int
+xfs_icsb_modify_counters(
+ xfs_mount_t *mp,
+ xfs_sb_field_t field,
+ int delta,
+ int rsvd)
+{
+ return xfs_icsb_modify_counters_int(mp, field, delta, rsvd, 0);
+}
+
+/*
+ * Called when superblock is already locked
+ */
+STATIC int
+xfs_icsb_modify_counters_locked(
+ xfs_mount_t *mp,
+ xfs_sb_field_t field,
+ int delta,
+ int rsvd)
+{
+ return xfs_icsb_modify_counters_int(mp, field, delta,
+ rsvd, XFS_ICSB_SB_LOCKED);
+}
+#endif
diff --git a/fs/xfs/xfs_mount.h b/fs/xfs/xfs_mount.h
index cd3cf9613a00..ebd73960e9db 100644
--- a/fs/xfs/xfs_mount.h
+++ b/fs/xfs/xfs_mount.h
@@ -267,6 +267,34 @@ typedef struct xfs_ioops {
#define XFS_IODONE(vfsp) \
(*(mp)->m_io_ops.xfs_iodone)(vfsp)
+#ifdef HAVE_PERCPU_SB
+
+/*
+ * Valid per-cpu incore superblock counters. Note that if you add new counters,
+ * you may need to define new counter disabled bit field descriptors as there
+ * are more possible fields in the superblock that can fit in a bitfield on a
+ * 32 bit platform. The XFS_SBS_* values for the current current counters just
+ * fit.
+ */
+typedef struct xfs_icsb_cnts {
+ uint64_t icsb_fdblocks;
+ uint64_t icsb_ifree;
+ uint64_t icsb_icount;
+ unsigned long icsb_flags;
+} xfs_icsb_cnts_t;
+
+#define XFS_ICSB_FLAG_LOCK (1 << 0) /* counter lock bit */
+
+#define XFS_ICSB_SB_LOCKED (1 << 0) /* sb already locked */
+#define XFS_ICSB_LAZY_COUNT (1 << 1) /* accuracy not needed */
+
+extern int xfs_icsb_init_counters(struct xfs_mount *);
+extern void xfs_icsb_sync_counters_lazy(struct xfs_mount *);
+
+#else
+#define xfs_icsb_init_counters(mp) (0)
+#define xfs_icsb_sync_counters_lazy(mp) do { } while (0)
+#endif
typedef struct xfs_mount {
bhv_desc_t m_bhv; /* vfs xfs behavior */
@@ -372,6 +400,11 @@ typedef struct xfs_mount {
struct xfs_qmops m_qm_ops; /* vector of XQM ops */
struct xfs_ioops m_io_ops; /* vector of I/O ops */
atomic_t m_active_trans; /* number trans frozen */
+#ifdef HAVE_PERCPU_SB
+ xfs_icsb_cnts_t *m_sb_cnts; /* per-cpu superblock counters */
+ unsigned long m_icsb_counters; /* disabled per-cpu counters */
+ struct notifier_block m_icsb_notifier; /* hotplug cpu notifier */
+#endif
} xfs_mount_t;
/*
@@ -386,8 +419,6 @@ typedef struct xfs_mount {
#define XFS_MOUNT_FS_SHUTDOWN (1ULL << 4) /* atomic stop of all filesystem
operations, typically for
disk errors in metadata */
-#define XFS_MOUNT_NOATIME (1ULL << 5) /* don't modify inode access
- times on reads */
#define XFS_MOUNT_RETERR (1ULL << 6) /* return alignment errors to
user */
#define XFS_MOUNT_NOALIGN (1ULL << 7) /* turn off stripe alignment
@@ -411,6 +442,8 @@ typedef struct xfs_mount {
#define XFS_MOUNT_DIRSYNC (1ULL << 21) /* synchronous directory ops */
#define XFS_MOUNT_COMPAT_IOSIZE (1ULL << 22) /* don't report large preferred
* I/O size in stat() */
+#define XFS_MOUNT_NO_PERCPU_SB (1ULL << 23) /* don't use per-cpu superblock
+ counters */
/*
@@ -473,11 +506,6 @@ xfs_preferred_iosize(xfs_mount_t *mp)
#define XFS_SHUTDOWN_REMOTE_REQ 0x10 /* Shutdown came from remote cell */
/*
- * xflags for xfs_syncsub
- */
-#define XFS_XSYNC_RELOC 0x01
-
-/*
* Flags for xfs_mountfs
*/
#define XFS_MFSI_SECOND 0x01 /* Secondary mount -- skip stuff */
@@ -548,6 +576,8 @@ extern void xfs_unmountfs_close(xfs_mount_t *, struct cred *);
extern int xfs_unmountfs_writesb(xfs_mount_t *);
extern int xfs_unmount_flush(xfs_mount_t *, int);
extern int xfs_mod_incore_sb(xfs_mount_t *, xfs_sb_field_t, int, int);
+extern int xfs_mod_incore_sb_unlocked(xfs_mount_t *, xfs_sb_field_t,
+ int, int);
extern int xfs_mod_incore_sb_batch(xfs_mount_t *, xfs_mod_sb_t *,
uint, int);
extern struct xfs_buf *xfs_getsb(xfs_mount_t *, int);
diff --git a/fs/xfs/xfs_rw.h b/fs/xfs/xfs_rw.h
index de85eefb7966..e63795644478 100644
--- a/fs/xfs/xfs_rw.h
+++ b/fs/xfs/xfs_rw.h
@@ -89,6 +89,7 @@ extern void xfs_ioerror_alert(char *func, struct xfs_mount *mp,
*/
extern int xfs_rwlock(bhv_desc_t *bdp, vrwlock_t write_lock);
extern void xfs_rwunlock(bhv_desc_t *bdp, vrwlock_t write_lock);
+extern int xfs_setattr(bhv_desc_t *bdp, vattr_t *vap, int flags, cred_t *credp);
extern int xfs_change_file_space(bhv_desc_t *bdp, int cmd, xfs_flock64_t *bf,
xfs_off_t offset, cred_t *credp, int flags);
extern int xfs_set_dmattrs(bhv_desc_t *bdp, u_int evmask, u_int16_t state,
diff --git a/fs/xfs/xfs_trans.c b/fs/xfs/xfs_trans.c
index d3d714e6b32a..2918956553a5 100644
--- a/fs/xfs/xfs_trans.c
+++ b/fs/xfs/xfs_trans.c
@@ -55,10 +55,141 @@ STATIC void xfs_trans_committed(xfs_trans_t *, int);
STATIC void xfs_trans_chunk_committed(xfs_log_item_chunk_t *, xfs_lsn_t, int);
STATIC void xfs_trans_free(xfs_trans_t *);
-kmem_zone_t *xfs_trans_zone;
+kmem_zone_t *xfs_trans_zone;
/*
+ * Reservation functions here avoid a huge stack in xfs_trans_init
+ * due to register overflow from temporaries in the calculations.
+ */
+
+STATIC uint
+xfs_calc_write_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_WRITE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_itruncate_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_ITRUNCATE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_rename_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_RENAME_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_link_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_LINK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_remove_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_REMOVE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_symlink_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_SYMLINK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_create_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_CREATE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_mkdir_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_MKDIR_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_ifree_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_IFREE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_ichange_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_ICHANGE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_growdata_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_GROWDATA_LOG_RES(mp);
+}
+
+STATIC uint
+xfs_calc_growrtalloc_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_GROWRTALLOC_LOG_RES(mp);
+}
+
+STATIC uint
+xfs_calc_growrtzero_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_GROWRTZERO_LOG_RES(mp);
+}
+
+STATIC uint
+xfs_calc_growrtfree_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_GROWRTFREE_LOG_RES(mp);
+}
+
+STATIC uint
+xfs_calc_swrite_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_SWRITE_LOG_RES(mp);
+}
+
+STATIC uint
+xfs_calc_writeid_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_WRITEID_LOG_RES(mp);
+}
+
+STATIC uint
+xfs_calc_addafork_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_ADDAFORK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_attrinval_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_ATTRINVAL_LOG_RES(mp);
+}
+
+STATIC uint
+xfs_calc_attrset_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_ATTRSET_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_attrrm_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_ATTRRM_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp);
+}
+
+STATIC uint
+xfs_calc_clear_agi_bucket_reservation(xfs_mount_t *mp)
+{
+ return XFS_CALC_CLEAR_AGI_BUCKET_LOG_RES(mp);
+}
+
+/*
* Initialize the precomputed transaction reservation values
* in the mount structure.
*/
@@ -69,39 +200,27 @@ xfs_trans_init(
xfs_trans_reservations_t *resp;
resp = &(mp->m_reservations);
- resp->tr_write =
- (uint)(XFS_CALC_WRITE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_itruncate =
- (uint)(XFS_CALC_ITRUNCATE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_rename =
- (uint)(XFS_CALC_RENAME_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_link = (uint)XFS_CALC_LINK_LOG_RES(mp);
- resp->tr_remove =
- (uint)(XFS_CALC_REMOVE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_symlink =
- (uint)(XFS_CALC_SYMLINK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_create =
- (uint)(XFS_CALC_CREATE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_mkdir =
- (uint)(XFS_CALC_MKDIR_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_ifree =
- (uint)(XFS_CALC_IFREE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_ichange =
- (uint)(XFS_CALC_ICHANGE_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_growdata = (uint)XFS_CALC_GROWDATA_LOG_RES(mp);
- resp->tr_swrite = (uint)XFS_CALC_SWRITE_LOG_RES(mp);
- resp->tr_writeid = (uint)XFS_CALC_WRITEID_LOG_RES(mp);
- resp->tr_addafork =
- (uint)(XFS_CALC_ADDAFORK_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_attrinval = (uint)XFS_CALC_ATTRINVAL_LOG_RES(mp);
- resp->tr_attrset =
- (uint)(XFS_CALC_ATTRSET_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_attrrm =
- (uint)(XFS_CALC_ATTRRM_LOG_RES(mp) + XFS_DQUOT_LOGRES(mp));
- resp->tr_clearagi = (uint)XFS_CALC_CLEAR_AGI_BUCKET_LOG_RES(mp);
- resp->tr_growrtalloc = (uint)XFS_CALC_GROWRTALLOC_LOG_RES(mp);
- resp->tr_growrtzero = (uint)XFS_CALC_GROWRTZERO_LOG_RES(mp);
- resp->tr_growrtfree = (uint)XFS_CALC_GROWRTFREE_LOG_RES(mp);
+ resp->tr_write = xfs_calc_write_reservation(mp);
+ resp->tr_itruncate = xfs_calc_itruncate_reservation(mp);
+ resp->tr_rename = xfs_calc_rename_reservation(mp);
+ resp->tr_link = xfs_calc_link_reservation(mp);
+ resp->tr_remove = xfs_calc_remove_reservation(mp);
+ resp->tr_symlink = xfs_calc_symlink_reservation(mp);
+ resp->tr_create = xfs_calc_create_reservation(mp);
+ resp->tr_mkdir = xfs_calc_mkdir_reservation(mp);
+ resp->tr_ifree = xfs_calc_ifree_reservation(mp);
+ resp->tr_ichange = xfs_calc_ichange_reservation(mp);
+ resp->tr_growdata = xfs_calc_growdata_reservation(mp);
+ resp->tr_swrite = xfs_calc_swrite_reservation(mp);
+ resp->tr_writeid = xfs_calc_writeid_reservation(mp);
+ resp->tr_addafork = xfs_calc_addafork_reservation(mp);
+ resp->tr_attrinval = xfs_calc_attrinval_reservation(mp);
+ resp->tr_attrset = xfs_calc_attrset_reservation(mp);
+ resp->tr_attrrm = xfs_calc_attrrm_reservation(mp);
+ resp->tr_clearagi = xfs_calc_clear_agi_bucket_reservation(mp);
+ resp->tr_growrtalloc = xfs_calc_growrtalloc_reservation(mp);
+ resp->tr_growrtzero = xfs_calc_growrtzero_reservation(mp);
+ resp->tr_growrtfree = xfs_calc_growrtfree_reservation(mp);
}
/*
diff --git a/fs/xfs/xfs_trans.h b/fs/xfs/xfs_trans.h
index d77901c07f63..e48befa4e337 100644
--- a/fs/xfs/xfs_trans.h
+++ b/fs/xfs/xfs_trans.h
@@ -380,7 +380,7 @@ typedef struct xfs_trans {
xfs_trans_header_t t_header; /* header for in-log trans */
unsigned int t_busy_free; /* busy descs free */
xfs_log_busy_chunk_t t_busy; /* busy/async free blocks */
- xfs_pflags_t t_pflags; /* saved pflags state */
+ unsigned long t_pflags; /* saved process flags state */
} xfs_trans_t;
#endif /* __KERNEL__ */
diff --git a/fs/xfs/xfs_vfsops.c b/fs/xfs/xfs_vfsops.c
index b6ad370fab3d..d4ec4dfaf19c 100644
--- a/fs/xfs/xfs_vfsops.c
+++ b/fs/xfs/xfs_vfsops.c
@@ -55,7 +55,7 @@
#include "xfs_clnt.h"
#include "xfs_fsops.h"
-STATIC int xfs_sync(bhv_desc_t *, int, cred_t *);
+STATIC int xfs_sync(bhv_desc_t *, int, cred_t *);
int
xfs_init(void)
@@ -77,11 +77,12 @@ xfs_init(void)
"xfs_bmap_free_item");
xfs_btree_cur_zone = kmem_zone_init(sizeof(xfs_btree_cur_t),
"xfs_btree_cur");
- xfs_inode_zone = kmem_zone_init(sizeof(xfs_inode_t), "xfs_inode");
xfs_trans_zone = kmem_zone_init(sizeof(xfs_trans_t), "xfs_trans");
xfs_da_state_zone =
kmem_zone_init(sizeof(xfs_da_state_t), "xfs_da_state");
xfs_dabuf_zone = kmem_zone_init(sizeof(xfs_dabuf_t), "xfs_dabuf");
+ xfs_ifork_zone = kmem_zone_init(sizeof(xfs_ifork_t), "xfs_ifork");
+ xfs_acl_zone_init(xfs_acl_zone, "xfs_acl");
/*
* The size of the zone allocated buf log item is the maximum
@@ -93,17 +94,30 @@ xfs_init(void)
(((XFS_MAX_BLOCKSIZE / XFS_BLI_CHUNK) /
NBWORD) * sizeof(int))),
"xfs_buf_item");
- xfs_efd_zone = kmem_zone_init((sizeof(xfs_efd_log_item_t) +
- ((XFS_EFD_MAX_FAST_EXTENTS - 1) * sizeof(xfs_extent_t))),
+ xfs_efd_zone =
+ kmem_zone_init((sizeof(xfs_efd_log_item_t) +
+ ((XFS_EFD_MAX_FAST_EXTENTS - 1) *
+ sizeof(xfs_extent_t))),
"xfs_efd_item");
- xfs_efi_zone = kmem_zone_init((sizeof(xfs_efi_log_item_t) +
- ((XFS_EFI_MAX_FAST_EXTENTS - 1) * sizeof(xfs_extent_t))),
+ xfs_efi_zone =
+ kmem_zone_init((sizeof(xfs_efi_log_item_t) +
+ ((XFS_EFI_MAX_FAST_EXTENTS - 1) *
+ sizeof(xfs_extent_t))),
"xfs_efi_item");
- xfs_ifork_zone = kmem_zone_init(sizeof(xfs_ifork_t), "xfs_ifork");
- xfs_ili_zone = kmem_zone_init(sizeof(xfs_inode_log_item_t), "xfs_ili");
- xfs_chashlist_zone = kmem_zone_init(sizeof(xfs_chashlist_t),
- "xfs_chashlist");
- xfs_acl_zone_init(xfs_acl_zone, "xfs_acl");
+
+ /*
+ * These zones warrant special memory allocator hints
+ */
+ xfs_inode_zone =
+ kmem_zone_init_flags(sizeof(xfs_inode_t), "xfs_inode",
+ KM_ZONE_HWALIGN | KM_ZONE_RECLAIM |
+ KM_ZONE_SPREAD, NULL);
+ xfs_ili_zone =
+ kmem_zone_init_flags(sizeof(xfs_inode_log_item_t), "xfs_ili",
+ KM_ZONE_SPREAD, NULL);
+ xfs_chashlist_zone =
+ kmem_zone_init_flags(sizeof(xfs_chashlist_t), "xfs_chashlist",
+ KM_ZONE_SPREAD, NULL);
/*
* Allocate global trace buffers.
@@ -176,18 +190,18 @@ xfs_cleanup(void)
ktrace_free(xfs_alloc_trace_buf);
#endif
- kmem_cache_destroy(xfs_bmap_free_item_zone);
- kmem_cache_destroy(xfs_btree_cur_zone);
- kmem_cache_destroy(xfs_inode_zone);
- kmem_cache_destroy(xfs_trans_zone);
- kmem_cache_destroy(xfs_da_state_zone);
- kmem_cache_destroy(xfs_dabuf_zone);
- kmem_cache_destroy(xfs_buf_item_zone);
- kmem_cache_destroy(xfs_efd_zone);
- kmem_cache_destroy(xfs_efi_zone);
- kmem_cache_destroy(xfs_ifork_zone);
- kmem_cache_destroy(xfs_ili_zone);
- kmem_cache_destroy(xfs_chashlist_zone);
+ kmem_zone_destroy(xfs_bmap_free_item_zone);
+ kmem_zone_destroy(xfs_btree_cur_zone);
+ kmem_zone_destroy(xfs_inode_zone);
+ kmem_zone_destroy(xfs_trans_zone);
+ kmem_zone_destroy(xfs_da_state_zone);
+ kmem_zone_destroy(xfs_dabuf_zone);
+ kmem_zone_destroy(xfs_buf_item_zone);
+ kmem_zone_destroy(xfs_efd_zone);
+ kmem_zone_destroy(xfs_efi_zone);
+ kmem_zone_destroy(xfs_ifork_zone);
+ kmem_zone_destroy(xfs_ili_zone);
+ kmem_zone_destroy(xfs_chashlist_zone);
}
/*
@@ -258,8 +272,6 @@ xfs_start_flags(
mp->m_inoadd = XFS_INO64_OFFSET;
}
#endif
- if (ap->flags & XFSMNT_NOATIME)
- mp->m_flags |= XFS_MOUNT_NOATIME;
if (ap->flags & XFSMNT_RETERR)
mp->m_flags |= XFS_MOUNT_RETERR;
if (ap->flags & XFSMNT_NOALIGN)
@@ -620,7 +632,7 @@ xfs_quiesce_fs(
xfs_mount_t *mp)
{
int count = 0, pincount;
-
+
xfs_refcache_purge_mp(mp);
xfs_flush_buftarg(mp->m_ddev_targp, 0);
xfs_finish_reclaim_all(mp, 0);
@@ -631,7 +643,7 @@ xfs_quiesce_fs(
* meta data (typically directory updates).
* Which then must be flushed and logged before
* we can write the unmount record.
- */
+ */
do {
xfs_syncsub(mp, SYNC_REMOUNT|SYNC_ATTR|SYNC_WAIT, 0, NULL);
pincount = xfs_flush_buftarg(mp->m_ddev_targp, 1);
@@ -654,11 +666,6 @@ xfs_mntupdate(
xfs_mount_t *mp = XFS_BHVTOM(bdp);
int error;
- if (args->flags & XFSMNT_NOATIME)
- mp->m_flags |= XFS_MOUNT_NOATIME;
- else
- mp->m_flags &= ~XFS_MOUNT_NOATIME;
-
if (args->flags & XFSMNT_BARRIER)
mp->m_flags |= XFS_MOUNT_BARRIER;
else
@@ -814,6 +821,7 @@ xfs_statvfs(
statp->f_type = XFS_SB_MAGIC;
+ xfs_icsb_sync_counters_lazy(mp);
s = XFS_SB_LOCK(mp);
statp->f_bsize = sbp->sb_blocksize;
lsize = sbp->sb_logstart ? sbp->sb_logblocks : 0;
@@ -1221,7 +1229,7 @@ xfs_sync_inodes(
xfs_iunlock(ip, XFS_ILOCK_SHARED);
error = xfs_itobp(mp, NULL, ip,
- &dip, &bp, 0);
+ &dip, &bp, 0, 0);
if (!error) {
xfs_buf_relse(bp);
} else {
@@ -1690,10 +1698,7 @@ xfs_parseargs(
int iosize;
args->flags2 |= XFSMNT2_COMPAT_IOSIZE;
-
-#if 0 /* XXX: off by default, until some remaining issues ironed out */
- args->flags |= XFSMNT_IDELETE; /* default to on */
-#endif
+ args->flags |= XFSMNT_IDELETE;
if (!options)
goto done;
@@ -1903,7 +1908,6 @@ xfs_showargs(
{ XFS_MOUNT_NOUUID, "," MNTOPT_NOUUID },
{ XFS_MOUNT_NORECOVERY, "," MNTOPT_NORECOVERY },
{ XFS_MOUNT_OSYNCISOSYNC, "," MNTOPT_OSYNCISOSYNC },
- { XFS_MOUNT_IDELETE, "," MNTOPT_NOIKEEP },
{ 0, NULL }
};
struct proc_xfs_info *xfs_infop;
@@ -1939,6 +1943,8 @@ xfs_showargs(
seq_printf(m, "," MNTOPT_SWIDTH "=%d",
(int)XFS_FSB_TO_BB(mp, mp->m_swidth));
+ if (!(mp->m_flags & XFS_MOUNT_IDELETE))
+ seq_printf(m, "," MNTOPT_IKEEP);
if (!(mp->m_flags & XFS_MOUNT_COMPAT_IOSIZE))
seq_printf(m, "," MNTOPT_LARGEIO);
if (mp->m_flags & XFS_MOUNT_BARRIER)
diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c
index eaab355f5a89..a478f42e63ff 100644
--- a/fs/xfs/xfs_vnodeops.c
+++ b/fs/xfs/xfs_vnodeops.c
@@ -615,6 +615,7 @@ xfs_setattr(
code = xfs_igrow_start(ip, vap->va_size, credp);
}
xfs_iunlock(ip, XFS_ILOCK_EXCL);
+ vn_iowait(vp); /* wait for the completion of any pending DIOs */
if (!code)
code = xfs_itruncate_data(ip, vap->va_size);
if (code) {
@@ -1556,7 +1557,7 @@ xfs_release(
if ((error = xfs_inactive_free_eofblocks(mp, ip)))
return error;
/* Update linux inode block count after free above */
- LINVFS_GET_IP(vp)->i_blocks = XFS_FSB_TO_BB(mp,
+ vn_to_inode(vp)->i_blocks = XFS_FSB_TO_BB(mp,
ip->i_d.di_nblocks + ip->i_delayed_blks);
}
}
@@ -1637,7 +1638,7 @@ xfs_inactive(
if ((error = xfs_inactive_free_eofblocks(mp, ip)))
return VN_INACTIVE_CACHE;
/* Update linux inode block count after free above */
- LINVFS_GET_IP(vp)->i_blocks = XFS_FSB_TO_BB(mp,
+ vn_to_inode(vp)->i_blocks = XFS_FSB_TO_BB(mp,
ip->i_d.di_nblocks + ip->i_delayed_blks);
}
goto out;
@@ -3186,7 +3187,7 @@ xfs_rmdir(
/* Fall through to std_return with error = 0 or the errno
* from xfs_trans_commit. */
-std_return:
+ std_return:
if (DM_EVENT_ENABLED(dir_vp->v_vfsp, dp, DM_EVENT_POSTREMOVE)) {
(void) XFS_SEND_NAMESP(mp, DM_EVENT_POSTREMOVE,
dir_vp, DM_RIGHT_NULL,
@@ -3196,12 +3197,12 @@ std_return:
}
return error;
-error1:
+ error1:
xfs_bmap_cancel(&free_list);
cancel_flags |= XFS_TRANS_ABORT;
/* FALLTHROUGH */
-error_return:
+ error_return:
xfs_trans_cancel(tp, cancel_flags);
goto std_return;
}
@@ -4310,8 +4311,10 @@ xfs_free_file_space(
ASSERT(attr_flags & ATTR_NOLOCK ? attr_flags & ATTR_DMI : 1);
if (attr_flags & ATTR_NOLOCK)
need_iolock = 0;
- if (need_iolock)
+ if (need_iolock) {
xfs_ilock(ip, XFS_IOLOCK_EXCL);
+ vn_iowait(vp); /* wait for the completion of any pending DIOs */
+ }
rounding = MAX((__uint8_t)(1 << mp->m_sb.sb_blocklog),
(__uint8_t)NBPP);
diff --git a/include/asm-alpha/mmu_context.h b/include/asm-alpha/mmu_context.h
index 6f92482cc96c..0c017fc181c1 100644
--- a/include/asm-alpha/mmu_context.h
+++ b/include/asm-alpha/mmu_context.h
@@ -231,9 +231,8 @@ init_new_context(struct task_struct *tsk, struct mm_struct *mm)
{
int i;
- for (i = 0; i < NR_CPUS; i++)
- if (cpu_online(i))
- mm->context[i] = 0;
+ for_each_online_cpu(i)
+ mm->context[i] = 0;
if (tsk != current)
task_thread_info(tsk)->pcb.ptbr
= ((unsigned long)mm->pgd - IDENT_ADDR) >> PAGE_SHIFT;
diff --git a/include/asm-alpha/topology.h b/include/asm-alpha/topology.h
index eb740e280d9c..420ccde6b916 100644
--- a/include/asm-alpha/topology.h
+++ b/include/asm-alpha/topology.h
@@ -27,8 +27,8 @@ static inline cpumask_t node_to_cpumask(int node)
cpumask_t node_cpu_mask = CPU_MASK_NONE;
int cpu;
- for(cpu = 0; cpu < NR_CPUS; cpu++) {
- if (cpu_online(cpu) && (cpu_to_node(cpu) == node))
+ for_each_online_cpu(cpu) {
+ if (cpu_to_node(cpu) == node)
cpu_set(cpu, node_cpu_mask);
}
diff --git a/include/asm-generic/bug.h b/include/asm-generic/bug.h
index 400c2b41896e..1a565a9d2fa7 100644
--- a/include/asm-generic/bug.h
+++ b/include/asm-generic/bug.h
@@ -7,7 +7,7 @@
#ifdef CONFIG_BUG
#ifndef HAVE_ARCH_BUG
#define BUG() do { \
- printk("kernel BUG at %s:%d!\n", __FILE__, __LINE__); \
+ printk("BUG: failure at %s:%d/%s()!\n", __FILE__, __LINE__, __FUNCTION__); \
panic("BUG!"); \
} while (0)
#endif
@@ -19,7 +19,7 @@
#ifndef HAVE_ARCH_WARN_ON
#define WARN_ON(condition) do { \
if (unlikely((condition)!=0)) { \
- printk("Badness in %s at %s:%d\n", __FUNCTION__, __FILE__, __LINE__); \
+ printk("BUG: warning at %s:%d/%s()\n", __FILE__, __LINE__, __FUNCTION__); \
dump_stack(); \
} \
} while (0)
diff --git a/include/asm-generic/percpu.h b/include/asm-generic/percpu.h
index 9044aeb37828..78cf45547e31 100644
--- a/include/asm-generic/percpu.h
+++ b/include/asm-generic/percpu.h
@@ -19,10 +19,9 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
#define percpu_modcopy(pcpudst, src, size) \
do { \
unsigned int __i; \
- for (__i = 0; __i < NR_CPUS; __i++) \
- if (cpu_possible(__i)) \
- memcpy((pcpudst)+__per_cpu_offset[__i], \
- (src), (size)); \
+ for_each_cpu(__i) \
+ memcpy((pcpudst)+__per_cpu_offset[__i], \
+ (src), (size)); \
} while (0)
#else /* ! SMP */
diff --git a/include/asm-i386/alternative.h b/include/asm-i386/alternative.h
new file mode 100644
index 000000000000..e201decea0c9
--- /dev/null
+++ b/include/asm-i386/alternative.h
@@ -0,0 +1,129 @@
+#ifndef _I386_ALTERNATIVE_H
+#define _I386_ALTERNATIVE_H
+
+#ifdef __KERNEL__
+
+struct alt_instr {
+ u8 *instr; /* original instruction */
+ u8 *replacement;
+ u8 cpuid; /* cpuid bit set for replacement */
+ u8 instrlen; /* length of original instruction */
+ u8 replacementlen; /* length of new instruction, <= instrlen */
+ u8 pad;
+};
+
+extern void apply_alternatives(struct alt_instr *start, struct alt_instr *end);
+
+struct module;
+extern void alternatives_smp_module_add(struct module *mod, char *name,
+ void *locks, void *locks_end,
+ void *text, void *text_end);
+extern void alternatives_smp_module_del(struct module *mod);
+extern void alternatives_smp_switch(int smp);
+
+#endif
+
+/*
+ * Alternative instructions for different CPU types or capabilities.
+ *
+ * This allows to use optimized instructions even on generic binary
+ * kernels.
+ *
+ * length of oldinstr must be longer or equal the length of newinstr
+ * It can be padded with nops as needed.
+ *
+ * For non barrier like inlines please define new variants
+ * without volatile and memory clobber.
+ */
+#define alternative(oldinstr, newinstr, feature) \
+ asm volatile ("661:\n\t" oldinstr "\n662:\n" \
+ ".section .altinstructions,\"a\"\n" \
+ " .align 4\n" \
+ " .long 661b\n" /* label */ \
+ " .long 663f\n" /* new instruction */ \
+ " .byte %c0\n" /* feature bit */ \
+ " .byte 662b-661b\n" /* sourcelen */ \
+ " .byte 664f-663f\n" /* replacementlen */ \
+ ".previous\n" \
+ ".section .altinstr_replacement,\"ax\"\n" \
+ "663:\n\t" newinstr "\n664:\n" /* replacement */\
+ ".previous" :: "i" (feature) : "memory")
+
+/*
+ * Alternative inline assembly with input.
+ *
+ * Pecularities:
+ * No memory clobber here.
+ * Argument numbers start with 1.
+ * Best is to use constraints that are fixed size (like (%1) ... "r")
+ * If you use variable sized constraints like "m" or "g" in the
+ * replacement maake sure to pad to the worst case length.
+ */
+#define alternative_input(oldinstr, newinstr, feature, input...) \
+ asm volatile ("661:\n\t" oldinstr "\n662:\n" \
+ ".section .altinstructions,\"a\"\n" \
+ " .align 4\n" \
+ " .long 661b\n" /* label */ \
+ " .long 663f\n" /* new instruction */ \
+ " .byte %c0\n" /* feature bit */ \
+ " .byte 662b-661b\n" /* sourcelen */ \
+ " .byte 664f-663f\n" /* replacementlen */ \
+ ".previous\n" \
+ ".section .altinstr_replacement,\"ax\"\n" \
+ "663:\n\t" newinstr "\n664:\n" /* replacement */\
+ ".previous" :: "i" (feature), ##input)
+
+/*
+ * Alternative inline assembly for SMP.
+ *
+ * alternative_smp() takes two versions (SMP first, UP second) and is
+ * for more complex stuff such as spinlocks.
+ *
+ * The LOCK_PREFIX macro defined here replaces the LOCK and
+ * LOCK_PREFIX macros used everywhere in the source tree.
+ *
+ * SMP alternatives use the same data structures as the other
+ * alternatives and the X86_FEATURE_UP flag to indicate the case of a
+ * UP system running a SMP kernel. The existing apply_alternatives()
+ * works fine for patching a SMP kernel for UP.
+ *
+ * The SMP alternative tables can be kept after boot and contain both
+ * UP and SMP versions of the instructions to allow switching back to
+ * SMP at runtime, when hotplugging in a new CPU, which is especially
+ * useful in virtualized environments.
+ *
+ * The very common lock prefix is handled as special case in a
+ * separate table which is a pure address list without replacement ptr
+ * and size information. That keeps the table sizes small.
+ */
+
+#ifdef CONFIG_SMP
+#define alternative_smp(smpinstr, upinstr, args...) \
+ asm volatile ("661:\n\t" smpinstr "\n662:\n" \
+ ".section .smp_altinstructions,\"a\"\n" \
+ " .align 4\n" \
+ " .long 661b\n" /* label */ \
+ " .long 663f\n" /* new instruction */ \
+ " .byte 0x68\n" /* X86_FEATURE_UP */ \
+ " .byte 662b-661b\n" /* sourcelen */ \
+ " .byte 664f-663f\n" /* replacementlen */ \
+ ".previous\n" \
+ ".section .smp_altinstr_replacement,\"awx\"\n" \
+ "663:\n\t" upinstr "\n" /* replacement */ \
+ "664:\n\t.fill 662b-661b,1,0x42\n" /* space for original */ \
+ ".previous" : args)
+
+#define LOCK_PREFIX \
+ ".section .smp_locks,\"a\"\n" \
+ " .align 4\n" \
+ " .long 661f\n" /* address */ \
+ ".previous\n" \
+ "661:\n\tlock; "
+
+#else /* ! CONFIG_SMP */
+#define alternative_smp(smpinstr, upinstr, args...) \
+ asm volatile (upinstr : args)
+#define LOCK_PREFIX ""
+#endif
+
+#endif /* _I386_ALTERNATIVE_H */
diff --git a/include/asm-i386/arch_hooks.h b/include/asm-i386/arch_hooks.h
index 28b96a6fb9fa..238cf4275b96 100644
--- a/include/asm-i386/arch_hooks.h
+++ b/include/asm-i386/arch_hooks.h
@@ -24,4 +24,7 @@ extern void trap_init_hook(void);
extern void time_init_hook(void);
extern void mca_nmi_hook(void);
+extern int setup_early_printk(char *);
+extern void early_printk(const char *fmt, ...) __attribute__((format(printf,1,2)));
+
#endif
diff --git a/include/asm-i386/atomic.h b/include/asm-i386/atomic.h
index de649d3aa2d4..22d80ece95cb 100644
--- a/include/asm-i386/atomic.h
+++ b/include/asm-i386/atomic.h
@@ -10,12 +10,6 @@
* resource counting etc..
*/
-#ifdef CONFIG_SMP
-#define LOCK "lock ; "
-#else
-#define LOCK ""
-#endif
-
/*
* Make sure gcc doesn't try to be clever and move things around
* on us. We need to use _exactly_ the address the user gave us,
@@ -52,7 +46,7 @@ typedef struct { volatile int counter; } atomic_t;
static __inline__ void atomic_add(int i, atomic_t *v)
{
__asm__ __volatile__(
- LOCK "addl %1,%0"
+ LOCK_PREFIX "addl %1,%0"
:"=m" (v->counter)
:"ir" (i), "m" (v->counter));
}
@@ -67,7 +61,7 @@ static __inline__ void atomic_add(int i, atomic_t *v)
static __inline__ void atomic_sub(int i, atomic_t *v)
{
__asm__ __volatile__(
- LOCK "subl %1,%0"
+ LOCK_PREFIX "subl %1,%0"
:"=m" (v->counter)
:"ir" (i), "m" (v->counter));
}
@@ -86,7 +80,7 @@ static __inline__ int atomic_sub_and_test(int i, atomic_t *v)
unsigned char c;
__asm__ __volatile__(
- LOCK "subl %2,%0; sete %1"
+ LOCK_PREFIX "subl %2,%0; sete %1"
:"=m" (v->counter), "=qm" (c)
:"ir" (i), "m" (v->counter) : "memory");
return c;
@@ -101,7 +95,7 @@ static __inline__ int atomic_sub_and_test(int i, atomic_t *v)
static __inline__ void atomic_inc(atomic_t *v)
{
__asm__ __volatile__(
- LOCK "incl %0"
+ LOCK_PREFIX "incl %0"
:"=m" (v->counter)
:"m" (v->counter));
}
@@ -115,7 +109,7 @@ static __inline__ void atomic_inc(atomic_t *v)
static __inline__ void atomic_dec(atomic_t *v)
{
__asm__ __volatile__(
- LOCK "decl %0"
+ LOCK_PREFIX "decl %0"
:"=m" (v->counter)
:"m" (v->counter));
}
@@ -133,7 +127,7 @@ static __inline__ int atomic_dec_and_test(atomic_t *v)
unsigned char c;
__asm__ __volatile__(
- LOCK "decl %0; sete %1"
+ LOCK_PREFIX "decl %0; sete %1"
:"=m" (v->counter), "=qm" (c)
:"m" (v->counter) : "memory");
return c != 0;
@@ -152,7 +146,7 @@ static __inline__ int atomic_inc_and_test(atomic_t *v)
unsigned char c;
__asm__ __volatile__(
- LOCK "incl %0; sete %1"
+ LOCK_PREFIX "incl %0; sete %1"
:"=m" (v->counter), "=qm" (c)
:"m" (v->counter) : "memory");
return c != 0;
@@ -172,7 +166,7 @@ static __inline__ int atomic_add_negative(int i, atomic_t *v)
unsigned char c;
__asm__ __volatile__(
- LOCK "addl %2,%0; sets %1"
+ LOCK_PREFIX "addl %2,%0; sets %1"
:"=m" (v->counter), "=qm" (c)
:"ir" (i), "m" (v->counter) : "memory");
return c;
@@ -195,7 +189,7 @@ static __inline__ int atomic_add_return(int i, atomic_t *v)
/* Modern 486+ processor */
__i = i;
__asm__ __volatile__(
- LOCK "xaddl %0, %1;"
+ LOCK_PREFIX "xaddl %0, %1;"
:"=r"(i)
:"m"(v->counter), "0"(i));
return i + __i;
@@ -231,8 +225,14 @@ static __inline__ int atomic_sub_return(int i, atomic_t *v)
({ \
int c, old; \
c = atomic_read(v); \
- while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+ for (;;) { \
+ if (unlikely(c == (u))) \
+ break; \
+ old = atomic_cmpxchg((v), c, c + (a)); \
+ if (likely(old == c)) \
+ break; \
c = old; \
+ } \
c != (u); \
})
#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
@@ -242,11 +242,11 @@ static __inline__ int atomic_sub_return(int i, atomic_t *v)
/* These are x86-specific, used by some header files */
#define atomic_clear_mask(mask, addr) \
-__asm__ __volatile__(LOCK "andl %0,%1" \
+__asm__ __volatile__(LOCK_PREFIX "andl %0,%1" \
: : "r" (~(mask)),"m" (*addr) : "memory")
#define atomic_set_mask(mask, addr) \
-__asm__ __volatile__(LOCK "orl %0,%1" \
+__asm__ __volatile__(LOCK_PREFIX "orl %0,%1" \
: : "r" (mask),"m" (*(addr)) : "memory")
/* Atomic operations are already serializing on x86 */
diff --git a/include/asm-i386/bitops.h b/include/asm-i386/bitops.h
index 88e6ca248cd7..7d20b95edb3b 100644
--- a/include/asm-i386/bitops.h
+++ b/include/asm-i386/bitops.h
@@ -7,6 +7,7 @@
#include <linux/config.h>
#include <linux/compiler.h>
+#include <asm/alternative.h>
/*
* These have to be done with inline assembly: that way the bit-setting
@@ -16,12 +17,6 @@
* bit 0 is the LSB of addr; bit 32 is the LSB of (addr+1).
*/
-#ifdef CONFIG_SMP
-#define LOCK_PREFIX "lock ; "
-#else
-#define LOCK_PREFIX ""
-#endif
-
#define ADDR (*(volatile long *) addr)
/**
diff --git a/include/asm-i386/cache.h b/include/asm-i386/cache.h
index 615911e5bd24..ca15c9c665cf 100644
--- a/include/asm-i386/cache.h
+++ b/include/asm-i386/cache.h
@@ -10,4 +10,6 @@
#define L1_CACHE_SHIFT (CONFIG_X86_L1_CACHE_SHIFT)
#define L1_CACHE_BYTES (1 << L1_CACHE_SHIFT)
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
#endif
diff --git a/include/asm-i386/cpufeature.h b/include/asm-i386/cpufeature.h
index c4ec2a4d8fdf..5c0b5876b931 100644
--- a/include/asm-i386/cpufeature.h
+++ b/include/asm-i386/cpufeature.h
@@ -70,6 +70,7 @@
#define X86_FEATURE_P3 (3*32+ 6) /* P3 */
#define X86_FEATURE_P4 (3*32+ 7) /* P4 */
#define X86_FEATURE_CONSTANT_TSC (3*32+ 8) /* TSC ticks at a constant rate */
+#define X86_FEATURE_UP (3*32+ 9) /* smp kernel running on up */
/* Intel-defined CPU features, CPUID level 0x00000001 (ecx), word 4 */
#define X86_FEATURE_XMM3 (4*32+ 0) /* Streaming SIMD Extensions-3 */
diff --git a/include/asm-i386/mach-default/do_timer.h b/include/asm-i386/mach-default/do_timer.h
index 56211414fc95..6312c3e79814 100644
--- a/include/asm-i386/mach-default/do_timer.h
+++ b/include/asm-i386/mach-default/do_timer.h
@@ -18,7 +18,7 @@ static inline void do_timer_interrupt_hook(struct pt_regs *regs)
{
do_timer(regs);
#ifndef CONFIG_SMP
- update_process_times(user_mode(regs));
+ update_process_times(user_mode_vm(regs));
#endif
/*
* In the SMP case we use the local APIC timer interrupt to do the
diff --git a/include/asm-i386/mach-es7000/mach_mpparse.h b/include/asm-i386/mach-es7000/mach_mpparse.h
index 4a0637a3e208..99f66be240be 100644
--- a/include/asm-i386/mach-es7000/mach_mpparse.h
+++ b/include/asm-i386/mach-es7000/mach_mpparse.h
@@ -30,7 +30,8 @@ static inline int mps_oem_check(struct mp_config_table *mpc, char *oem,
return 0;
}
-static inline int es7000_check_dsdt()
+#ifdef CONFIG_ACPI
+static inline int es7000_check_dsdt(void)
{
struct acpi_table_header *header = NULL;
if(!acpi_get_table_header_early(ACPI_DSDT, &header))
@@ -54,6 +55,11 @@ static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id)
}
return 0;
}
-
+#else
+static inline int acpi_madt_oem_check(char *oem_id, char *oem_table_id)
+{
+ return 0;
+}
+#endif
#endif /* __ASM_MACH_MPPARSE_H */
diff --git a/include/asm-i386/mach-visws/do_timer.h b/include/asm-i386/mach-visws/do_timer.h
index 92d638fc8b11..95568e6ca91c 100644
--- a/include/asm-i386/mach-visws/do_timer.h
+++ b/include/asm-i386/mach-visws/do_timer.h
@@ -11,7 +11,7 @@ static inline void do_timer_interrupt_hook(struct pt_regs *regs)
do_timer(regs);
#ifndef CONFIG_SMP
- update_process_times(user_mode(regs));
+ update_process_times(user_mode_vm(regs));
#endif
/*
* In the SMP case we use the local APIC timer interrupt to do the
diff --git a/include/asm-i386/mach-voyager/do_timer.h b/include/asm-i386/mach-voyager/do_timer.h
index ae510e5d0d78..eaf518098981 100644
--- a/include/asm-i386/mach-voyager/do_timer.h
+++ b/include/asm-i386/mach-voyager/do_timer.h
@@ -5,7 +5,7 @@ static inline void do_timer_interrupt_hook(struct pt_regs *regs)
{
do_timer(regs);
#ifndef CONFIG_SMP
- update_process_times(user_mode(regs));
+ update_process_times(user_mode_vm(regs));
#endif
voyager_timer_interrupt(regs);
diff --git a/include/asm-i386/mpspec.h b/include/asm-i386/mpspec.h
index 64a0b8e6afeb..62113d3bfdc2 100644
--- a/include/asm-i386/mpspec.h
+++ b/include/asm-i386/mpspec.h
@@ -22,7 +22,6 @@ extern int mp_bus_id_to_type [MAX_MP_BUSSES];
extern int mp_irq_entries;
extern struct mpc_config_intsrc mp_irqs [MAX_IRQ_SOURCES];
extern int mpc_default_type;
-extern int mp_bus_id_to_pci_bus [MAX_MP_BUSSES];
extern unsigned long mp_lapic_addr;
extern int pic_mode;
extern int using_apic_timer;
diff --git a/include/asm-i386/mtrr.h b/include/asm-i386/mtrr.h
index 5b6ceda68c5f..64cf937c7e33 100644
--- a/include/asm-i386/mtrr.h
+++ b/include/asm-i386/mtrr.h
@@ -25,6 +25,7 @@
#include <linux/config.h>
#include <linux/ioctl.h>
+#include <linux/errno.h>
#define MTRR_IOCTL_BASE 'M'
diff --git a/include/asm-i386/mutex.h b/include/asm-i386/mutex.h
index 9b2199e829f3..05a538531229 100644
--- a/include/asm-i386/mutex.h
+++ b/include/asm-i386/mutex.h
@@ -9,6 +9,8 @@
#ifndef _ASM_MUTEX_H
#define _ASM_MUTEX_H
+#include "asm/alternative.h"
+
/**
* __mutex_fastpath_lock - try to take the lock by moving the count
* from 1 to a 0 value
@@ -27,7 +29,7 @@ do { \
typecheck_fn(fastcall void (*)(atomic_t *), fail_fn); \
\
__asm__ __volatile__( \
- LOCK " decl (%%eax) \n" \
+ LOCK_PREFIX " decl (%%eax) \n" \
" js 2f \n" \
"1: \n" \
\
@@ -83,7 +85,7 @@ do { \
typecheck_fn(fastcall void (*)(atomic_t *), fail_fn); \
\
__asm__ __volatile__( \
- LOCK " incl (%%eax) \n" \
+ LOCK_PREFIX " incl (%%eax) \n" \
" jle 2f \n" \
"1: \n" \
\
diff --git a/include/asm-i386/pgtable-2level.h b/include/asm-i386/pgtable-2level.h
index 74ef721b534d..27bde973abc7 100644
--- a/include/asm-i386/pgtable-2level.h
+++ b/include/asm-i386/pgtable-2level.h
@@ -61,4 +61,6 @@ static inline int pte_exec_kernel(pte_t pte)
#define __pte_to_swp_entry(pte) ((swp_entry_t) { (pte).pte_low })
#define __swp_entry_to_pte(x) ((pte_t) { (x).val })
+void vmalloc_sync_all(void);
+
#endif /* _I386_PGTABLE_2LEVEL_H */
diff --git a/include/asm-i386/pgtable-3level.h b/include/asm-i386/pgtable-3level.h
index f1a8b454920a..36a5aa63cbbf 100644
--- a/include/asm-i386/pgtable-3level.h
+++ b/include/asm-i386/pgtable-3level.h
@@ -152,4 +152,6 @@ static inline pmd_t pfn_pmd(unsigned long page_nr, pgprot_t pgprot)
#define __pmd_free_tlb(tlb, x) do { } while (0)
+#define vmalloc_sync_all() ((void)0)
+
#endif /* _I386_PGTABLE_3LEVEL_H */
diff --git a/include/asm-i386/rwlock.h b/include/asm-i386/rwlock.h
index b57cc7afdf7e..94f00195d543 100644
--- a/include/asm-i386/rwlock.h
+++ b/include/asm-i386/rwlock.h
@@ -21,21 +21,23 @@
#define RW_LOCK_BIAS_STR "0x01000000"
#define __build_read_lock_ptr(rw, helper) \
- asm volatile(LOCK "subl $1,(%0)\n\t" \
- "jns 1f\n" \
- "call " helper "\n\t" \
- "1:\n" \
- ::"a" (rw) : "memory")
+ alternative_smp("lock; subl $1,(%0)\n\t" \
+ "jns 1f\n" \
+ "call " helper "\n\t" \
+ "1:\n", \
+ "subl $1,(%0)\n\t", \
+ :"a" (rw) : "memory")
#define __build_read_lock_const(rw, helper) \
- asm volatile(LOCK "subl $1,%0\n\t" \
- "jns 1f\n" \
- "pushl %%eax\n\t" \
- "leal %0,%%eax\n\t" \
- "call " helper "\n\t" \
- "popl %%eax\n\t" \
- "1:\n" \
- :"=m" (*(volatile int *)rw) : : "memory")
+ alternative_smp("lock; subl $1,%0\n\t" \
+ "jns 1f\n" \
+ "pushl %%eax\n\t" \
+ "leal %0,%%eax\n\t" \
+ "call " helper "\n\t" \
+ "popl %%eax\n\t" \
+ "1:\n", \
+ "subl $1,%0\n\t", \
+ "=m" (*(volatile int *)rw) : : "memory")
#define __build_read_lock(rw, helper) do { \
if (__builtin_constant_p(rw)) \
@@ -45,21 +47,23 @@
} while (0)
#define __build_write_lock_ptr(rw, helper) \
- asm volatile(LOCK "subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
- "jz 1f\n" \
- "call " helper "\n\t" \
- "1:\n" \
- ::"a" (rw) : "memory")
+ alternative_smp("lock; subl $" RW_LOCK_BIAS_STR ",(%0)\n\t" \
+ "jz 1f\n" \
+ "call " helper "\n\t" \
+ "1:\n", \
+ "subl $" RW_LOCK_BIAS_STR ",(%0)\n\t", \
+ :"a" (rw) : "memory")
#define __build_write_lock_const(rw, helper) \
- asm volatile(LOCK "subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
- "jz 1f\n" \
- "pushl %%eax\n\t" \
- "leal %0,%%eax\n\t" \
- "call " helper "\n\t" \
- "popl %%eax\n\t" \
- "1:\n" \
- :"=m" (*(volatile int *)rw) : : "memory")
+ alternative_smp("lock; subl $" RW_LOCK_BIAS_STR ",%0\n\t" \
+ "jz 1f\n" \
+ "pushl %%eax\n\t" \
+ "leal %0,%%eax\n\t" \
+ "call " helper "\n\t" \
+ "popl %%eax\n\t" \
+ "1:\n", \
+ "subl $" RW_LOCK_BIAS_STR ",%0\n\t", \
+ "=m" (*(volatile int *)rw) : : "memory")
#define __build_write_lock(rw, helper) do { \
if (__builtin_constant_p(rw)) \
diff --git a/include/asm-i386/semaphore.h b/include/asm-i386/semaphore.h
index 6a42b2142fd6..f7a0f310c524 100644
--- a/include/asm-i386/semaphore.h
+++ b/include/asm-i386/semaphore.h
@@ -99,7 +99,7 @@ static inline void down(struct semaphore * sem)
might_sleep();
__asm__ __volatile__(
"# atomic down operation\n\t"
- LOCK "decl %0\n\t" /* --sem->count */
+ LOCK_PREFIX "decl %0\n\t" /* --sem->count */
"js 2f\n"
"1:\n"
LOCK_SECTION_START("")
@@ -123,7 +123,7 @@ static inline int down_interruptible(struct semaphore * sem)
might_sleep();
__asm__ __volatile__(
"# atomic interruptible down operation\n\t"
- LOCK "decl %1\n\t" /* --sem->count */
+ LOCK_PREFIX "decl %1\n\t" /* --sem->count */
"js 2f\n\t"
"xorl %0,%0\n"
"1:\n"
@@ -148,7 +148,7 @@ static inline int down_trylock(struct semaphore * sem)
__asm__ __volatile__(
"# atomic interruptible down operation\n\t"
- LOCK "decl %1\n\t" /* --sem->count */
+ LOCK_PREFIX "decl %1\n\t" /* --sem->count */
"js 2f\n\t"
"xorl %0,%0\n"
"1:\n"
@@ -173,7 +173,7 @@ static inline void up(struct semaphore * sem)
{
__asm__ __volatile__(
"# atomic up operation\n\t"
- LOCK "incl %0\n\t" /* ++sem->count */
+ LOCK_PREFIX "incl %0\n\t" /* ++sem->count */
"jle 2f\n"
"1:\n"
LOCK_SECTION_START("")
diff --git a/include/asm-i386/spinlock.h b/include/asm-i386/spinlock.h
index 23604350cdf4..d76b7693cf1d 100644
--- a/include/asm-i386/spinlock.h
+++ b/include/asm-i386/spinlock.h
@@ -35,31 +35,41 @@
#define __raw_spin_lock_string_flags \
"\n1:\t" \
"lock ; decb %0\n\t" \
- "jns 4f\n\t" \
+ "jns 5f\n" \
"2:\t" \
"testl $0x200, %1\n\t" \
- "jz 3f\n\t" \
- "sti\n\t" \
+ "jz 4f\n\t" \
+ "sti\n" \
"3:\t" \
"rep;nop\n\t" \
"cmpb $0, %0\n\t" \
"jle 3b\n\t" \
"cli\n\t" \
"jmp 1b\n" \
- "4:\n\t"
+ "4:\t" \
+ "rep;nop\n\t" \
+ "cmpb $0, %0\n\t" \
+ "jg 1b\n\t" \
+ "jmp 4b\n" \
+ "5:\n\t"
+
+#define __raw_spin_lock_string_up \
+ "\n\tdecb %0"
static inline void __raw_spin_lock(raw_spinlock_t *lock)
{
- __asm__ __volatile__(
- __raw_spin_lock_string
- :"=m" (lock->slock) : : "memory");
+ alternative_smp(
+ __raw_spin_lock_string,
+ __raw_spin_lock_string_up,
+ "=m" (lock->slock) : : "memory");
}
static inline void __raw_spin_lock_flags(raw_spinlock_t *lock, unsigned long flags)
{
- __asm__ __volatile__(
- __raw_spin_lock_string_flags
- :"=m" (lock->slock) : "r" (flags) : "memory");
+ alternative_smp(
+ __raw_spin_lock_string_flags,
+ __raw_spin_lock_string_up,
+ "=m" (lock->slock) : "r" (flags) : "memory");
}
static inline int __raw_spin_trylock(raw_spinlock_t *lock)
@@ -178,12 +188,12 @@ static inline int __raw_write_trylock(raw_rwlock_t *lock)
static inline void __raw_read_unlock(raw_rwlock_t *rw)
{
- asm volatile("lock ; incl %0" :"=m" (rw->lock) : : "memory");
+ asm volatile(LOCK_PREFIX "incl %0" :"=m" (rw->lock) : : "memory");
}
static inline void __raw_write_unlock(raw_rwlock_t *rw)
{
- asm volatile("lock ; addl $" RW_LOCK_BIAS_STR ", %0"
+ asm volatile(LOCK_PREFIX "addl $" RW_LOCK_BIAS_STR ", %0"
: "=m" (rw->lock) : : "memory");
}
diff --git a/include/asm-i386/system.h b/include/asm-i386/system.h
index 399145a247f2..d0d8d7448d88 100644
--- a/include/asm-i386/system.h
+++ b/include/asm-i386/system.h
@@ -352,67 +352,6 @@ static inline unsigned long long __cmpxchg64(volatile void *ptr, unsigned long l
#endif
-#ifdef __KERNEL__
-struct alt_instr {
- __u8 *instr; /* original instruction */
- __u8 *replacement;
- __u8 cpuid; /* cpuid bit set for replacement */
- __u8 instrlen; /* length of original instruction */
- __u8 replacementlen; /* length of new instruction, <= instrlen */
- __u8 pad;
-};
-#endif
-
-/*
- * Alternative instructions for different CPU types or capabilities.
- *
- * This allows to use optimized instructions even on generic binary
- * kernels.
- *
- * length of oldinstr must be longer or equal the length of newinstr
- * It can be padded with nops as needed.
- *
- * For non barrier like inlines please define new variants
- * without volatile and memory clobber.
- */
-#define alternative(oldinstr, newinstr, feature) \
- asm volatile ("661:\n\t" oldinstr "\n662:\n" \
- ".section .altinstructions,\"a\"\n" \
- " .align 4\n" \
- " .long 661b\n" /* label */ \
- " .long 663f\n" /* new instruction */ \
- " .byte %c0\n" /* feature bit */ \
- " .byte 662b-661b\n" /* sourcelen */ \
- " .byte 664f-663f\n" /* replacementlen */ \
- ".previous\n" \
- ".section .altinstr_replacement,\"ax\"\n" \
- "663:\n\t" newinstr "\n664:\n" /* replacement */ \
- ".previous" :: "i" (feature) : "memory")
-
-/*
- * Alternative inline assembly with input.
- *
- * Pecularities:
- * No memory clobber here.
- * Argument numbers start with 1.
- * Best is to use constraints that are fixed size (like (%1) ... "r")
- * If you use variable sized constraints like "m" or "g" in the
- * replacement maake sure to pad to the worst case length.
- */
-#define alternative_input(oldinstr, newinstr, feature, input...) \
- asm volatile ("661:\n\t" oldinstr "\n662:\n" \
- ".section .altinstructions,\"a\"\n" \
- " .align 4\n" \
- " .long 661b\n" /* label */ \
- " .long 663f\n" /* new instruction */ \
- " .byte %c0\n" /* feature bit */ \
- " .byte 662b-661b\n" /* sourcelen */ \
- " .byte 664f-663f\n" /* replacementlen */ \
- ".previous\n" \
- ".section .altinstr_replacement,\"ax\"\n" \
- "663:\n\t" newinstr "\n664:\n" /* replacement */ \
- ".previous" :: "i" (feature), ##input)
-
/*
* Force strict CPU ordering.
* And yes, this is required on UP too when we're talking
@@ -558,5 +497,6 @@ static inline void sched_cacheflush(void)
}
extern unsigned long arch_align_stack(unsigned long sp);
+extern void free_init_pages(char *what, unsigned long begin, unsigned long end);
#endif
diff --git a/include/asm-i386/uaccess.h b/include/asm-i386/uaccess.h
index 3f1337c34208..371457b1ceb6 100644
--- a/include/asm-i386/uaccess.h
+++ b/include/asm-i386/uaccess.h
@@ -197,13 +197,15 @@ extern void __put_user_8(void);
#define put_user(x,ptr) \
({ int __ret_pu; \
+ __typeof__(*(ptr)) __pu_val; \
__chk_user_ptr(ptr); \
+ __pu_val = x; \
switch(sizeof(*(ptr))) { \
- case 1: __put_user_1(x, ptr); break; \
- case 2: __put_user_2(x, ptr); break; \
- case 4: __put_user_4(x, ptr); break; \
- case 8: __put_user_8(x, ptr); break; \
- default:__put_user_X(x, ptr); break; \
+ case 1: __put_user_1(__pu_val, ptr); break; \
+ case 2: __put_user_2(__pu_val, ptr); break; \
+ case 4: __put_user_4(__pu_val, ptr); break; \
+ case 8: __put_user_8(__pu_val, ptr); break; \
+ default:__put_user_X(__pu_val, ptr); break; \
} \
__ret_pu; \
})
diff --git a/include/asm-i386/unistd.h b/include/asm-i386/unistd.h
index dc81a55dd94d..d8afd0e3b81a 100644
--- a/include/asm-i386/unistd.h
+++ b/include/asm-i386/unistd.h
@@ -347,9 +347,9 @@ __syscall_return(type,__res); \
type name(type1 arg1) \
{ \
long __res; \
-__asm__ volatile ("int $0x80" \
+__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
: "=a" (__res) \
- : "0" (__NR_##name),"b" ((long)(arg1)) : "memory"); \
+ : "0" (__NR_##name),"ri" ((long)(arg1)) : "memory"); \
__syscall_return(type,__res); \
}
@@ -357,9 +357,10 @@ __syscall_return(type,__res); \
type name(type1 arg1,type2 arg2) \
{ \
long __res; \
-__asm__ volatile ("int $0x80" \
+__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
: "=a" (__res) \
- : "0" (__NR_##name),"b" ((long)(arg1)),"c" ((long)(arg2)) : "memory"); \
+ : "0" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)) \
+ : "memory"); \
__syscall_return(type,__res); \
}
@@ -367,9 +368,9 @@ __syscall_return(type,__res); \
type name(type1 arg1,type2 arg2,type3 arg3) \
{ \
long __res; \
-__asm__ volatile ("int $0x80" \
+__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
: "=a" (__res) \
- : "0" (__NR_##name),"b" ((long)(arg1)),"c" ((long)(arg2)), \
+ : "0" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)), \
"d" ((long)(arg3)) : "memory"); \
__syscall_return(type,__res); \
}
@@ -378,9 +379,9 @@ __syscall_return(type,__res); \
type name (type1 arg1, type2 arg2, type3 arg3, type4 arg4) \
{ \
long __res; \
-__asm__ volatile ("int $0x80" \
+__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; int $0x80 ; pop %%ebx" \
: "=a" (__res) \
- : "0" (__NR_##name),"b" ((long)(arg1)),"c" ((long)(arg2)), \
+ : "0" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)), \
"d" ((long)(arg3)),"S" ((long)(arg4)) : "memory"); \
__syscall_return(type,__res); \
}
@@ -390,10 +391,12 @@ __syscall_return(type,__res); \
type name (type1 arg1,type2 arg2,type3 arg3,type4 arg4,type5 arg5) \
{ \
long __res; \
-__asm__ volatile ("int $0x80" \
+__asm__ volatile ("push %%ebx ; movl %2,%%ebx ; movl %1,%%eax ; " \
+ "int $0x80 ; pop %%ebx" \
: "=a" (__res) \
- : "0" (__NR_##name),"b" ((long)(arg1)),"c" ((long)(arg2)), \
- "d" ((long)(arg3)),"S" ((long)(arg4)),"D" ((long)(arg5)) : "memory"); \
+ : "i" (__NR_##name),"ri" ((long)(arg1)),"c" ((long)(arg2)), \
+ "d" ((long)(arg3)),"S" ((long)(arg4)),"D" ((long)(arg5)) \
+ : "memory"); \
__syscall_return(type,__res); \
}
@@ -402,11 +405,14 @@ __syscall_return(type,__res); \
type name (type1 arg1,type2 arg2,type3 arg3,type4 arg4,type5 arg5,type6 arg6) \
{ \
long __res; \
-__asm__ volatile ("push %%ebp ; movl %%eax,%%ebp ; movl %1,%%eax ; int $0x80 ; pop %%ebp" \
+ struct { long __a1; long __a6; } __s = { (long)arg1, (long)arg6 }; \
+__asm__ volatile ("push %%ebp ; push %%ebx ; movl 4(%2),%%ebp ; " \
+ "movl 0(%2),%%ebx ; movl %1,%%eax ; int $0x80 ; " \
+ "pop %%ebx ; pop %%ebp" \
: "=a" (__res) \
- : "i" (__NR_##name),"b" ((long)(arg1)),"c" ((long)(arg2)), \
- "d" ((long)(arg3)),"S" ((long)(arg4)),"D" ((long)(arg5)), \
- "0" ((long)(arg6)) : "memory"); \
+ : "i" (__NR_##name),"0" ((long)(&__s)),"c" ((long)(arg2)), \
+ "d" ((long)(arg3)),"S" ((long)(arg4)),"D" ((long)(arg5)) \
+ : "memory"); \
__syscall_return(type,__res); \
}
diff --git a/include/asm-ia64/atomic.h b/include/asm-ia64/atomic.h
index d3e0dfa99e1f..569ec7574baf 100644
--- a/include/asm-ia64/atomic.h
+++ b/include/asm-ia64/atomic.h
@@ -95,8 +95,14 @@ ia64_atomic64_sub (__s64 i, atomic64_t *v)
({ \
int c, old; \
c = atomic_read(v); \
- while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+ for (;;) { \
+ if (unlikely(c == (u))) \
+ break; \
+ old = atomic_cmpxchg((v), c, c + (a)); \
+ if (likely(old == c)) \
+ break; \
c = old; \
+ } \
c != (u); \
})
#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
diff --git a/include/asm-ia64/cache.h b/include/asm-ia64/cache.h
index 40dd25195d65..f0a104db8f20 100644
--- a/include/asm-ia64/cache.h
+++ b/include/asm-ia64/cache.h
@@ -25,4 +25,6 @@
# define SMP_CACHE_BYTES (1 << 3)
#endif
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
#endif /* _ASM_IA64_CACHE_H */
diff --git a/include/asm-m68k/atomic.h b/include/asm-m68k/atomic.h
index 862e497c2645..732d696d31a6 100644
--- a/include/asm-m68k/atomic.h
+++ b/include/asm-m68k/atomic.h
@@ -175,8 +175,14 @@ static inline void atomic_set_mask(unsigned long mask, unsigned long *v)
({ \
int c, old; \
c = atomic_read(v); \
- while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+ for (;;) { \
+ if (unlikely(c == (u))) \
+ break; \
+ old = atomic_cmpxchg((v), c, c + (a)); \
+ if (likely(old == c)) \
+ break; \
c = old; \
+ } \
c != (u); \
})
#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
diff --git a/include/asm-parisc/cache.h b/include/asm-parisc/cache.h
index 93f179f13ce8..ae50f8e12eed 100644
--- a/include/asm-parisc/cache.h
+++ b/include/asm-parisc/cache.h
@@ -29,6 +29,8 @@
#define SMP_CACHE_BYTES L1_CACHE_BYTES
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
extern void flush_data_cache_local(void *); /* flushes local data-cache only */
extern void flush_instruction_cache_local(void *); /* flushes local code-cache only */
#ifdef CONFIG_SMP
diff --git a/include/asm-powerpc/percpu.h b/include/asm-powerpc/percpu.h
index e31922c50e53..464301cd0d03 100644
--- a/include/asm-powerpc/percpu.h
+++ b/include/asm-powerpc/percpu.h
@@ -27,10 +27,9 @@
#define percpu_modcopy(pcpudst, src, size) \
do { \
unsigned int __i; \
- for (__i = 0; __i < NR_CPUS; __i++) \
- if (cpu_possible(__i)) \
- memcpy((pcpudst)+__per_cpu_offset(__i), \
- (src), (size)); \
+ for_each_cpu(__i) \
+ memcpy((pcpudst)+__per_cpu_offset(__i), \
+ (src), (size)); \
} while (0)
extern void setup_per_cpu_areas(void);
diff --git a/include/asm-s390/atomic.h b/include/asm-s390/atomic.h
index be6fefe223d6..de1d9926aa60 100644
--- a/include/asm-s390/atomic.h
+++ b/include/asm-s390/atomic.h
@@ -89,10 +89,15 @@ static __inline__ int atomic_cmpxchg(atomic_t *v, int old, int new)
static __inline__ int atomic_add_unless(atomic_t *v, int a, int u)
{
int c, old;
-
c = atomic_read(v);
- while (c != u && (old = atomic_cmpxchg(v, c, c + a)) != c)
+ for (;;) {
+ if (unlikely(c == u))
+ break;
+ old = atomic_cmpxchg(v, c, c + a);
+ if (likely(old == c))
+ break;
c = old;
+ }
return c != u;
}
@@ -167,10 +172,15 @@ static __inline__ int atomic64_add_unless(atomic64_t *v,
long long a, long long u)
{
long long c, old;
-
c = atomic64_read(v);
- while (c != u && (old = atomic64_cmpxchg(v, c, c + a)) != c)
+ for (;;) {
+ if (unlikely(c == u))
+ break;
+ old = atomic64_cmpxchg(v, c, c + a);
+ if (likely(old == c))
+ break;
c = old;
+ }
return c != u;
}
diff --git a/include/asm-s390/percpu.h b/include/asm-s390/percpu.h
index 123fcaca295e..e10ed87094f0 100644
--- a/include/asm-s390/percpu.h
+++ b/include/asm-s390/percpu.h
@@ -46,10 +46,9 @@ extern unsigned long __per_cpu_offset[NR_CPUS];
#define percpu_modcopy(pcpudst, src, size) \
do { \
unsigned int __i; \
- for (__i = 0; __i < NR_CPUS; __i++) \
- if (cpu_possible(__i)) \
- memcpy((pcpudst)+__per_cpu_offset[__i], \
- (src), (size)); \
+ for_each_cpu(__i) \
+ memcpy((pcpudst)+__per_cpu_offset[__i], \
+ (src), (size)); \
} while (0)
#else /* ! SMP */
diff --git a/include/asm-sparc64/atomic.h b/include/asm-sparc64/atomic.h
index 25256bdc8aae..468eb48d8142 100644
--- a/include/asm-sparc64/atomic.h
+++ b/include/asm-sparc64/atomic.h
@@ -78,9 +78,15 @@ extern int atomic64_sub_ret(int, atomic64_t *);
({ \
int c, old; \
c = atomic_read(v); \
- while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+ for (;;) { \
+ if (unlikely(c == (u))) \
+ break; \
+ old = atomic_cmpxchg((v), c, c + (a)); \
+ if (likely(old == c)) \
+ break; \
c = old; \
- c != (u); \
+ } \
+ likely(c != (u)); \
})
#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
diff --git a/include/asm-sparc64/cache.h b/include/asm-sparc64/cache.h
index f7d35a2ae9b8..e9df17acedde 100644
--- a/include/asm-sparc64/cache.h
+++ b/include/asm-sparc64/cache.h
@@ -13,4 +13,6 @@
#define SMP_CACHE_BYTES_SHIFT 6
#define SMP_CACHE_BYTES (1 << SMP_CACHE_BYTES_SHIFT) /* L2 cache line size. */
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
#endif
diff --git a/include/asm-sparc64/percpu.h b/include/asm-sparc64/percpu.h
index aea4e51e7cd1..82032e159a76 100644
--- a/include/asm-sparc64/percpu.h
+++ b/include/asm-sparc64/percpu.h
@@ -26,10 +26,9 @@ register unsigned long __local_per_cpu_offset asm("g5");
#define percpu_modcopy(pcpudst, src, size) \
do { \
unsigned int __i; \
- for (__i = 0; __i < NR_CPUS; __i++) \
- if (cpu_possible(__i)) \
- memcpy((pcpudst)+__per_cpu_offset(__i), \
- (src), (size)); \
+ for_each_cpu(__i) \
+ memcpy((pcpudst)+__per_cpu_offset(__i), \
+ (src), (size)); \
} while (0)
#else /* ! SMP */
diff --git a/include/asm-um/alternative.h b/include/asm-um/alternative.h
new file mode 100644
index 000000000000..b6434396bd42
--- /dev/null
+++ b/include/asm-um/alternative.h
@@ -0,0 +1,6 @@
+#ifndef __UM_ALTERNATIVE_H
+#define __UM_ALTERNATIVE_H
+
+#include "asm/arch/alternative.h"
+
+#endif
diff --git a/include/asm-x86_64/atomic.h b/include/asm-x86_64/atomic.h
index 4b5cd553e772..cecbf7baa6aa 100644
--- a/include/asm-x86_64/atomic.h
+++ b/include/asm-x86_64/atomic.h
@@ -405,8 +405,14 @@ static __inline__ long atomic64_sub_return(long i, atomic64_t *v)
({ \
int c, old; \
c = atomic_read(v); \
- while (c != (u) && (old = atomic_cmpxchg((v), c, c + (a))) != c) \
+ for (;;) { \
+ if (unlikely(c == (u))) \
+ break; \
+ old = atomic_cmpxchg((v), c, c + (a)); \
+ if (likely(old == c)) \
+ break; \
c = old; \
+ } \
c != (u); \
})
#define atomic_inc_not_zero(v) atomic_add_unless((v), 1, 0)
diff --git a/include/asm-x86_64/cache.h b/include/asm-x86_64/cache.h
index 263f0a211ed7..c8043a16152e 100644
--- a/include/asm-x86_64/cache.h
+++ b/include/asm-x86_64/cache.h
@@ -20,6 +20,8 @@
__attribute__((__section__(".data.page_aligned")))
#endif
+#define __read_mostly __attribute__((__section__(".data.read_mostly")))
+
#endif
#endif
diff --git a/include/asm-x86_64/percpu.h b/include/asm-x86_64/percpu.h
index 29a6b0408f75..4405b4adeaba 100644
--- a/include/asm-x86_64/percpu.h
+++ b/include/asm-x86_64/percpu.h
@@ -26,10 +26,9 @@
#define percpu_modcopy(pcpudst, src, size) \
do { \
unsigned int __i; \
- for (__i = 0; __i < NR_CPUS; __i++) \
- if (cpu_possible(__i)) \
- memcpy((pcpudst)+__per_cpu_offset(__i), \
- (src), (size)); \
+ for_each_cpu(__i) \
+ memcpy((pcpudst)+__per_cpu_offset(__i), \
+ (src), (size)); \
} while (0)
extern void setup_per_cpu_areas(void);
diff --git a/include/linux/cache.h b/include/linux/cache.h
index d22e632f41fb..cc4b3aafad9a 100644
--- a/include/linux/cache.h
+++ b/include/linux/cache.h
@@ -13,9 +13,7 @@
#define SMP_CACHE_BYTES L1_CACHE_BYTES
#endif
-#if defined(CONFIG_X86) || defined(CONFIG_SPARC64) || defined(CONFIG_IA64) || defined(CONFIG_PARISC)
-#define __read_mostly __attribute__((__section__(".data.read_mostly")))
-#else
+#ifndef __read_mostly
#define __read_mostly
#endif
diff --git a/include/linux/cdrom.h b/include/linux/cdrom.h
index b68fdf1f3156..3c9b0bc05123 100644
--- a/include/linux/cdrom.h
+++ b/include/linux/cdrom.h
@@ -378,7 +378,6 @@ struct cdrom_generic_command
#define CDC_MEDIA_CHANGED 0x80 /* media changed */
#define CDC_PLAY_AUDIO 0x100 /* audio functions */
#define CDC_RESET 0x200 /* hard reset device */
-#define CDC_IOCTLS 0x400 /* driver has non-standard ioctls */
#define CDC_DRIVE_STATUS 0x800 /* driver implements drive status */
#define CDC_GENERIC_PACKET 0x1000 /* driver implements generic packets */
#define CDC_CD_R 0x2000 /* drive is a CD-R */
@@ -974,9 +973,7 @@ struct cdrom_device_ops {
int (*reset) (struct cdrom_device_info *);
/* play stuff */
int (*audio_ioctl) (struct cdrom_device_info *,unsigned int, void *);
- /* dev-specific */
- int (*dev_ioctl) (struct cdrom_device_info *,
- unsigned int, unsigned long);
+
/* driver specifications */
const int capability; /* capability flags */
int n_minors; /* number of active minor devices */
diff --git a/include/linux/eventpoll.h b/include/linux/eventpoll.h
index 1289f0ec4c00..1e4bdfcf83a2 100644
--- a/include/linux/eventpoll.h
+++ b/include/linux/eventpoll.h
@@ -52,7 +52,12 @@ struct file;
#ifdef CONFIG_EPOLL
/* Used to initialize the epoll bits inside the "struct file" */
-void eventpoll_init_file(struct file *file);
+static inline void eventpoll_init_file(struct file *file)
+{
+ INIT_LIST_HEAD(&file->f_ep_links);
+ spin_lock_init(&file->f_ep_lock);
+}
+
/* Used to release the epoll bits inside the "struct file" */
void eventpoll_release_file(struct file *file);
@@ -85,7 +90,6 @@ static inline void eventpoll_release(struct file *file)
eventpoll_release_file(file);
}
-
#else
static inline void eventpoll_init_file(struct file *file) {}
diff --git a/include/linux/ext3_fs.h b/include/linux/ext3_fs.h
index c0272d73ab20..e7239f2f97a1 100644
--- a/include/linux/ext3_fs.h
+++ b/include/linux/ext3_fs.h
@@ -772,9 +772,12 @@ extern unsigned long ext3_count_free (struct buffer_head *, unsigned);
/* inode.c */
-extern int ext3_forget(handle_t *, int, struct inode *, struct buffer_head *, int);
-extern struct buffer_head * ext3_getblk (handle_t *, struct inode *, long, int, int *);
-extern struct buffer_head * ext3_bread (handle_t *, struct inode *, int, int, int *);
+int ext3_forget(handle_t *, int, struct inode *, struct buffer_head *, int);
+struct buffer_head * ext3_getblk (handle_t *, struct inode *, long, int, int *);
+struct buffer_head * ext3_bread (handle_t *, struct inode *, int, int, int *);
+int ext3_get_block_handle(handle_t *handle, struct inode *inode,
+ sector_t iblock, struct buffer_head *bh_result, int create,
+ int extend_disksize);
extern void ext3_read_inode (struct inode *);
extern int ext3_write_inode (struct inode *, int);
diff --git a/include/linux/ext3_fs_i.h b/include/linux/ext3_fs_i.h
index e71dd98dbcae..7abf90147180 100644
--- a/include/linux/ext3_fs_i.h
+++ b/include/linux/ext3_fs_i.h
@@ -19,6 +19,7 @@
#include <linux/rwsem.h>
#include <linux/rbtree.h>
#include <linux/seqlock.h>
+#include <linux/mutex.h>
struct ext3_reserve_window {
__u32 _rsv_start; /* First byte reserved */
@@ -122,16 +123,16 @@ struct ext3_inode_info {
__u16 i_extra_isize;
/*
- * truncate_sem is for serialising ext3_truncate() against
+ * truncate_mutex is for serialising ext3_truncate() against
* ext3_getblock(). In the 2.4 ext2 design, great chunks of inode's
* data tree are chopped off during truncate. We can't do that in
* ext3 because whenever we perform intermediate commits during
* truncate, the inode and all the metadata blocks *must* be in a
* consistent state which allows truncation of the orphans to restart
* during recovery. Hence we must fix the get_block-vs-truncate race
- * by other means, so we have truncate_sem.
+ * by other means, so we have truncate_mutex.
*/
- struct semaphore truncate_sem;
+ struct mutex truncate_mutex;
struct inode vfs_inode;
};
diff --git a/include/linux/file.h b/include/linux/file.h
index 9901b850f2e4..9f7c2513866f 100644
--- a/include/linux/file.h
+++ b/include/linux/file.h
@@ -10,6 +10,7 @@
#include <linux/compiler.h>
#include <linux/spinlock.h>
#include <linux/rcupdate.h>
+#include <linux/types.h>
/*
* The default fd array needs to be at least BITS_PER_LONG,
@@ -17,10 +18,22 @@
*/
#define NR_OPEN_DEFAULT BITS_PER_LONG
+/*
+ * The embedded_fd_set is a small fd_set,
+ * suitable for most tasks (which open <= BITS_PER_LONG files)
+ */
+struct embedded_fd_set {
+ unsigned long fds_bits[1];
+};
+
+/*
+ * More than this number of fds: we use a separately allocated fd_set
+ */
+#define EMBEDDED_FD_SET_SIZE (BITS_PER_BYTE * sizeof(struct embedded_fd_set))
+
struct fdtable {
unsigned int max_fds;
int max_fdset;
- int next_fd;
struct file ** fd; /* current fd array */
fd_set *close_on_exec;
fd_set *open_fds;
@@ -33,13 +46,20 @@ struct fdtable {
* Open file table structure
*/
struct files_struct {
+ /*
+ * read mostly part
+ */
atomic_t count;
struct fdtable *fdt;
struct fdtable fdtab;
- fd_set close_on_exec_init;
- fd_set open_fds_init;
+ /*
+ * written part on a separate cache line in SMP
+ */
+ spinlock_t file_lock ____cacheline_aligned_in_smp;
+ int next_fd;
+ struct embedded_fd_set close_on_exec_init;
+ struct embedded_fd_set open_fds_init;
struct file * fd_array[NR_OPEN_DEFAULT];
- spinlock_t file_lock; /* Protects concurrent writers. Nests inside tsk->alloc_lock */
};
#define files_fdtable(files) (rcu_dereference((files)->fdt))
diff --git a/include/linux/fs.h b/include/linux/fs.h
index 128d0082522c..f9c9dea636d0 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -397,8 +397,8 @@ struct block_device {
dev_t bd_dev; /* not a kdev_t - it's a search key */
struct inode * bd_inode; /* will die */
int bd_openers;
- struct semaphore bd_sem; /* open/close mutex */
- struct semaphore bd_mount_sem; /* mount mutex */
+ struct mutex bd_mutex; /* open/close mutex */
+ struct mutex bd_mount_mutex; /* mount mutex */
struct list_head bd_inodes;
void * bd_holder;
int bd_holders;
@@ -509,7 +509,7 @@ struct inode {
#ifdef CONFIG_INOTIFY
struct list_head inotify_watches; /* watches on this inode */
- struct semaphore inotify_sem; /* protects the watches list */
+ struct mutex inotify_mutex; /* protects the watches list */
#endif
unsigned long i_state;
@@ -847,7 +847,7 @@ struct super_block {
* The next field is for VFS *only*. No filesystems have any business
* even looking at it. You had been warned.
*/
- struct semaphore s_vfs_rename_sem; /* Kludge */
+ struct mutex s_vfs_rename_mutex; /* Kludge */
/* Granuality of c/m/atime in ns.
Cannot be worse than a second */
@@ -1115,6 +1115,18 @@ static inline void mark_inode_dirty_sync(struct inode *inode)
__mark_inode_dirty(inode, I_DIRTY_SYNC);
}
+static inline void inode_inc_link_count(struct inode *inode)
+{
+ inode->i_nlink++;
+ mark_inode_dirty(inode);
+}
+
+static inline void inode_dec_link_count(struct inode *inode)
+{
+ inode->i_nlink--;
+ mark_inode_dirty(inode);
+}
+
extern void touch_atime(struct vfsmount *mnt, struct dentry *dentry);
static inline void file_accessed(struct file *file)
{
@@ -1534,7 +1546,7 @@ extern void destroy_inode(struct inode *);
extern struct inode *new_inode(struct super_block *);
extern int remove_suid(struct dentry *);
extern void remove_dquot_ref(struct super_block *, int, struct list_head *);
-extern struct semaphore iprune_sem;
+extern struct mutex iprune_mutex;
extern void __insert_inode_hash(struct inode *, unsigned long hashval);
extern void remove_inode_hash(struct inode *);
diff --git a/include/linux/generic_serial.h b/include/linux/generic_serial.h
index 0abe9d9a0069..652611a4bdcd 100644
--- a/include/linux/generic_serial.h
+++ b/include/linux/generic_serial.h
@@ -12,6 +12,8 @@
#ifndef GENERIC_SERIAL_H
#define GENERIC_SERIAL_H
+#include <linux/mutex.h>
+
struct real_driver {
void (*disable_tx_interrupts) (void *);
void (*enable_tx_interrupts) (void *);
@@ -34,7 +36,7 @@ struct gs_port {
int xmit_head;
int xmit_tail;
int xmit_cnt;
- struct semaphore port_write_sem;
+ struct mutex port_write_mutex;
int flags;
wait_queue_head_t open_wait;
wait_queue_head_t close_wait;
diff --git a/include/linux/genhd.h b/include/linux/genhd.h
index eef5ccdcd731..fd647fde5ec1 100644
--- a/include/linux/genhd.h
+++ b/include/linux/genhd.h
@@ -149,22 +149,16 @@ struct disk_attribute {
({ \
typeof(gendiskp->dkstats->field) res = 0; \
int i; \
- for (i=0; i < NR_CPUS; i++) { \
- if (!cpu_possible(i)) \
- continue; \
+ for_each_cpu(i) \
res += per_cpu_ptr(gendiskp->dkstats, i)->field; \
- } \
res; \
})
static inline void disk_stat_set_all(struct gendisk *gendiskp, int value) {
int i;
- for (i=0; i < NR_CPUS; i++) {
- if (cpu_possible(i)) {
- memset(per_cpu_ptr(gendiskp->dkstats, i), value,
- sizeof (struct disk_stats));
- }
- }
+ for_each_cpu(i)
+ memset(per_cpu_ptr(gendiskp->dkstats, i), value,
+ sizeof (struct disk_stats));
}
#else
diff --git a/include/linux/init_task.h b/include/linux/init_task.h
index dcfd2ecccb5d..92146f3b7423 100644
--- a/include/linux/init_task.h
+++ b/include/linux/init_task.h
@@ -7,11 +7,10 @@
#define INIT_FDTABLE \
{ \
.max_fds = NR_OPEN_DEFAULT, \
- .max_fdset = __FD_SETSIZE, \
- .next_fd = 0, \
+ .max_fdset = EMBEDDED_FD_SET_SIZE, \
.fd = &init_files.fd_array[0], \
- .close_on_exec = &init_files.close_on_exec_init, \
- .open_fds = &init_files.open_fds_init, \
+ .close_on_exec = (fd_set *)&init_files.close_on_exec_init, \
+ .open_fds = (fd_set *)&init_files.open_fds_init, \
.rcu = RCU_HEAD_INIT, \
.free_files = NULL, \
.next = NULL, \
@@ -20,9 +19,10 @@
#define INIT_FILES \
{ \
.count = ATOMIC_INIT(1), \
- .file_lock = SPIN_LOCK_UNLOCKED, \
.fdt = &init_files.fdtab, \
.fdtab = INIT_FDTABLE, \
+ .file_lock = SPIN_LOCK_UNLOCKED, \
+ .next_fd = 0, \
.close_on_exec_init = { { 0, } }, \
.open_fds_init = { { 0, } }, \
.fd_array = { NULL, } \
diff --git a/include/linux/jbd.h b/include/linux/jbd.h
index 41ee79962bb2..2ccbfb6340ba 100644
--- a/include/linux/jbd.h
+++ b/include/linux/jbd.h
@@ -28,6 +28,7 @@
#include <linux/journal-head.h>
#include <linux/stddef.h>
#include <linux/bit_spinlock.h>
+#include <linux/mutex.h>
#include <asm/semaphore.h>
#endif
@@ -575,7 +576,7 @@ struct transaction_s
* @j_wait_checkpoint: Wait queue to trigger checkpointing
* @j_wait_commit: Wait queue to trigger commit
* @j_wait_updates: Wait queue to wait for updates to complete
- * @j_checkpoint_sem: Semaphore for locking against concurrent checkpoints
+ * @j_checkpoint_mutex: Mutex for locking against concurrent checkpoints
* @j_head: Journal head - identifies the first unused block in the journal
* @j_tail: Journal tail - identifies the oldest still-used block in the
* journal.
@@ -645,7 +646,7 @@ struct journal_s
int j_barrier_count;
/* The barrier lock itself */
- struct semaphore j_barrier;
+ struct mutex j_barrier;
/*
* Transactions: The current running transaction...
@@ -687,7 +688,7 @@ struct journal_s
wait_queue_head_t j_wait_updates;
/* Semaphore for locking against concurrent checkpoints */
- struct semaphore j_checkpoint_sem;
+ struct mutex j_checkpoint_mutex;
/*
* Journal head: identifies the first unused block in the journal.
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 3b507bf05d09..bb6e7ddee2fd 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -91,6 +91,9 @@ extern struct notifier_block *panic_notifier_list;
extern long (*panic_blink)(long time);
NORET_TYPE void panic(const char * fmt, ...)
__attribute__ ((NORET_AND format (printf, 1, 2)));
+extern void oops_enter(void);
+extern void oops_exit(void);
+extern int oops_may_print(void);
fastcall NORET_TYPE void do_exit(long error_code)
ATTRIB_NORET;
NORET_TYPE void complete_and_exit(struct completion *, long)
diff --git a/include/linux/kprobes.h b/include/linux/kprobes.h
index 669756bc20a2..778adc0fa640 100644
--- a/include/linux/kprobes.h
+++ b/include/linux/kprobes.h
@@ -36,6 +36,7 @@
#include <linux/percpu.h>
#include <linux/spinlock.h>
#include <linux/rcupdate.h>
+#include <linux/mutex.h>
#ifdef CONFIG_KPROBES
#include <asm/kprobes.h>
@@ -152,7 +153,7 @@ struct kretprobe_instance {
};
extern spinlock_t kretprobe_lock;
-extern struct semaphore kprobe_mutex;
+extern struct mutex kprobe_mutex;
extern int arch_prepare_kprobe(struct kprobe *p);
extern void arch_arm_kprobe(struct kprobe *p);
extern void arch_disarm_kprobe(struct kprobe *p);
diff --git a/include/linux/loop.h b/include/linux/loop.h
index f96506782ebe..e76c7611d6cc 100644
--- a/include/linux/loop.h
+++ b/include/linux/loop.h
@@ -17,6 +17,7 @@
#include <linux/bio.h>
#include <linux/blkdev.h>
#include <linux/spinlock.h>
+#include <linux/mutex.h>
/* Possible states of device */
enum {
@@ -60,7 +61,7 @@ struct loop_device {
int lo_state;
struct completion lo_done;
struct completion lo_bh_done;
- struct semaphore lo_ctl_mutex;
+ struct mutex lo_ctl_mutex;
int lo_pending;
request_queue_t *lo_queue;
diff --git a/include/linux/msdos_fs.h b/include/linux/msdos_fs.h
index 8bcd9450d926..779e6a5744c7 100644
--- a/include/linux/msdos_fs.h
+++ b/include/linux/msdos_fs.h
@@ -184,6 +184,7 @@ struct fat_slot_info {
#include <linux/string.h>
#include <linux/nls.h>
#include <linux/fs.h>
+#include <linux/mutex.h>
struct fat_mount_options {
uid_t fs_uid;
@@ -226,7 +227,7 @@ struct msdos_sb_info {
unsigned long max_cluster; /* maximum cluster number */
unsigned long root_cluster; /* first cluster of the root directory */
unsigned long fsinfo_sector; /* sector number of FAT32 fsinfo */
- struct semaphore fat_lock;
+ struct mutex fat_lock;
unsigned int prev_free; /* previously allocated cluster number */
unsigned int free_clusters; /* -1 if undefined */
struct fat_mount_options options;
diff --git a/include/linux/nbd.h b/include/linux/nbd.h
index f95d51fae733..a6ce409ec6fc 100644
--- a/include/linux/nbd.h
+++ b/include/linux/nbd.h
@@ -38,6 +38,7 @@ enum {
#ifdef __KERNEL__
#include <linux/wait.h>
+#include <linux/mutex.h>
/* values for flags field */
#define NBD_READ_ONLY 0x0001
@@ -57,7 +58,7 @@ struct nbd_device {
struct request *active_req;
wait_queue_head_t active_wq;
- struct semaphore tx_lock;
+ struct mutex tx_lock;
struct gendisk *disk;
int blksize;
u64 bytesize;
diff --git a/include/linux/ncp_fs_i.h b/include/linux/ncp_fs_i.h
index 415be1ec6f98..bdb4c8ae6924 100644
--- a/include/linux/ncp_fs_i.h
+++ b/include/linux/ncp_fs_i.h
@@ -19,7 +19,7 @@ struct ncp_inode_info {
__le32 DosDirNum;
__u8 volNumber;
__le32 nwattr;
- struct semaphore open_sem;
+ struct mutex open_mutex;
atomic_t opened;
int access;
int flags;
diff --git a/include/linux/ncp_fs_sb.h b/include/linux/ncp_fs_sb.h
index cf858eb80f0b..b089d9506283 100644
--- a/include/linux/ncp_fs_sb.h
+++ b/include/linux/ncp_fs_sb.h
@@ -11,6 +11,7 @@
#include <linux/types.h>
#include <linux/ncp_mount.h>
#include <linux/net.h>
+#include <linux/mutex.h>
#ifdef __KERNEL__
@@ -51,7 +52,7 @@ struct ncp_server {
receive replies */
int lock; /* To prevent mismatch in protocols. */
- struct semaphore sem;
+ struct mutex mutex;
int current_size; /* for packet preparation */
int has_subfunction;
@@ -96,7 +97,7 @@ struct ncp_server {
struct {
struct work_struct tq; /* STREAM/DGRAM: data/error ready */
struct ncp_request_reply* creq; /* STREAM/DGRAM: awaiting reply from this request */
- struct semaphore creq_sem; /* DGRAM only: lock accesses to rcv.creq */
+ struct mutex creq_mutex; /* DGRAM only: lock accesses to rcv.creq */
unsigned int state; /* STREAM only: receiver state */
struct {
diff --git a/include/linux/pm.h b/include/linux/pm.h
index 5be87ba3b7ac..6df2585c0169 100644
--- a/include/linux/pm.h
+++ b/include/linux/pm.h
@@ -188,6 +188,8 @@ extern void device_power_up(void);
extern void device_resume(void);
#ifdef CONFIG_PM
+extern suspend_disk_method_t pm_disk_mode;
+
extern int device_suspend(pm_message_t state);
#define device_set_wakeup_enable(dev,val) \
@@ -215,7 +217,6 @@ static inline int dpm_runtime_suspend(struct device * dev, pm_message_t state)
static inline void dpm_runtime_resume(struct device * dev)
{
-
}
#endif
diff --git a/include/linux/profile.h b/include/linux/profile.h
index 026969a5595c..1f2fea6640a4 100644
--- a/include/linux/profile.h
+++ b/include/linux/profile.h
@@ -14,6 +14,7 @@
struct proc_dir_entry;
struct pt_regs;
+struct notifier_block;
/* init basic kernel profiler */
void __init profile_init(void);
@@ -32,7 +33,6 @@ enum profile_type {
#ifdef CONFIG_PROFILING
-struct notifier_block;
struct task_struct;
struct mm_struct;
diff --git a/include/linux/quota.h b/include/linux/quota.h
index f33aeb22c26a..8dc2d04a103f 100644
--- a/include/linux/quota.h
+++ b/include/linux/quota.h
@@ -38,6 +38,7 @@
#include <linux/errno.h>
#include <linux/types.h>
#include <linux/spinlock.h>
+#include <linux/mutex.h>
#define __DQUOT_VERSION__ "dquot_6.5.1"
#define __DQUOT_NUM_VERSION__ 6*10000+5*100+1
@@ -215,7 +216,7 @@ struct dquot {
struct list_head dq_inuse; /* List of all quotas */
struct list_head dq_free; /* Free list element */
struct list_head dq_dirty; /* List of dirty dquots */
- struct semaphore dq_lock; /* dquot IO lock */
+ struct mutex dq_lock; /* dquot IO lock */
atomic_t dq_count; /* Use count */
wait_queue_head_t dq_wait_unused; /* Wait queue for dquot to become unused */
struct super_block *dq_sb; /* superblock this applies to */
@@ -285,8 +286,8 @@ struct quota_format_type {
struct quota_info {
unsigned int flags; /* Flags for diskquotas on this device */
- struct semaphore dqio_sem; /* lock device while I/O in progress */
- struct semaphore dqonoff_sem; /* Serialize quotaon & quotaoff */
+ struct mutex dqio_mutex; /* lock device while I/O in progress */
+ struct mutex dqonoff_mutex; /* Serialize quotaon & quotaoff */
struct rw_semaphore dqptr_sem; /* serialize ops using quota_info struct, pointers from inode to dquots */
struct inode *files[MAXQUOTAS]; /* inodes of quotafiles */
struct mem_dqinfo info[MAXQUOTAS]; /* Information for each quota type */
diff --git a/include/linux/raid/raid1.h b/include/linux/raid/raid1.h
index 9d5494aaac0f..3009c813d83d 100644
--- a/include/linux/raid/raid1.h
+++ b/include/linux/raid/raid1.h
@@ -130,6 +130,6 @@ struct r1bio_s {
* with failure when last write completes (and all failed).
* Record that bi_end_io was called with this flag...
*/
-#define R1BIO_Returned 4
+#define R1BIO_Returned 6
#endif
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index c2ec6c77874e..5673008b61e1 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -113,8 +113,6 @@ struct rcu_data {
DECLARE_PER_CPU(struct rcu_data, rcu_data);
DECLARE_PER_CPU(struct rcu_data, rcu_bh_data);
-extern struct rcu_ctrlblk rcu_ctrlblk;
-extern struct rcu_ctrlblk rcu_bh_ctrlblk;
/*
* Increment the quiescent state counter.
diff --git a/include/linux/seq_file.h b/include/linux/seq_file.h
index 850a974ee505..b95f6eb7254c 100644
--- a/include/linux/seq_file.h
+++ b/include/linux/seq_file.h
@@ -4,7 +4,7 @@
#include <linux/types.h>
#include <linux/string.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
struct seq_operations;
struct file;
@@ -19,7 +19,7 @@ struct seq_file {
size_t count;
loff_t index;
loff_t version;
- struct semaphore sem;
+ struct mutex lock;
struct seq_operations *op;
void *private;
};
diff --git a/include/linux/swap.h b/include/linux/swap.h
index 12415dd94451..54eac8a39a4c 100644
--- a/include/linux/swap.h
+++ b/include/linux/swap.h
@@ -234,14 +234,15 @@ extern struct page * read_swap_cache_async(swp_entry_t, struct vm_area_struct *v
/* linux/mm/swapfile.c */
extern long total_swap_pages;
extern unsigned int nr_swapfiles;
-extern struct swap_info_struct swap_info[];
extern void si_swapinfo(struct sysinfo *);
extern swp_entry_t get_swap_page(void);
-extern swp_entry_t get_swap_page_of_type(int type);
+extern swp_entry_t get_swap_page_of_type(int);
extern int swap_duplicate(swp_entry_t);
extern int valid_swaphandles(swp_entry_t, unsigned long *);
extern void swap_free(swp_entry_t);
extern void free_swap_and_cache(swp_entry_t);
+extern int swap_type_of(dev_t);
+extern unsigned int count_swap_pages(int, int);
extern sector_t map_swap_page(struct swap_info_struct *, pgoff_t);
extern struct swap_info_struct *get_swap_info_struct(unsigned);
extern int can_share_swap_page(struct page *);
diff --git a/include/linux/tty.h b/include/linux/tty.h
index f45cd74e6f24..f13f49afe198 100644
--- a/include/linux/tty.h
+++ b/include/linux/tty.h
@@ -24,6 +24,7 @@
#include <linux/tty_driver.h>
#include <linux/tty_ldisc.h>
#include <linux/screen_info.h>
+#include <linux/mutex.h>
#include <asm/system.h>
@@ -231,8 +232,8 @@ struct tty_struct {
int canon_data;
unsigned long canon_head;
unsigned int canon_column;
- struct semaphore atomic_read;
- struct semaphore atomic_write;
+ struct mutex atomic_read_lock;
+ struct mutex atomic_write_lock;
unsigned char *write_buf;
int write_cnt;
spinlock_t read_lock;
@@ -319,8 +320,7 @@ extern void tty_ldisc_put(int);
extern void tty_wakeup(struct tty_struct *tty);
extern void tty_ldisc_flush(struct tty_struct *tty);
-struct semaphore;
-extern struct semaphore tty_sem;
+extern struct mutex tty_mutex;
/* n_tty.c */
extern struct tty_ldisc tty_ldisc_N_TTY;
diff --git a/include/linux/tty_flip.h b/include/linux/tty_flip.h
index 222faf97d5f9..0c6169fff366 100644
--- a/include/linux/tty_flip.h
+++ b/include/linux/tty_flip.h
@@ -7,14 +7,8 @@ extern int tty_insert_flip_string_flags(struct tty_struct *tty, unsigned char *c
extern int tty_prepare_flip_string(struct tty_struct *tty, unsigned char **chars, size_t size);
extern int tty_prepare_flip_string_flags(struct tty_struct *tty, unsigned char **chars, char **flags, size_t size);
-#ifdef INCLUDE_INLINE_FUNCS
-#define _INLINE_ extern
-#else
-#define _INLINE_ static __inline__
-#endif
-
-_INLINE_ int tty_insert_flip_char(struct tty_struct *tty,
- unsigned char ch, char flag)
+static inline int tty_insert_flip_char(struct tty_struct *tty,
+ unsigned char ch, char flag)
{
struct tty_buffer *tb = tty->buf.tail;
if (tb && tb->active && tb->used < tb->size) {
@@ -25,7 +19,7 @@ _INLINE_ int tty_insert_flip_char(struct tty_struct *tty,
return tty_insert_flip_string_flags(tty, &ch, &flag, 1);
}
-_INLINE_ void tty_schedule_flip(struct tty_struct *tty)
+static inline void tty_schedule_flip(struct tty_struct *tty)
{
unsigned long flags;
spin_lock_irqsave(&tty->buf.lock, flags);
diff --git a/include/linux/udf_fs_sb.h b/include/linux/udf_fs_sb.h
index b15ff2e99c91..80ae9ef940dc 100644
--- a/include/linux/udf_fs_sb.h
+++ b/include/linux/udf_fs_sb.h
@@ -13,7 +13,7 @@
#ifndef _UDF_FS_SB_H
#define _UDF_FS_SB_H 1
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
#pragma pack(1)
@@ -111,7 +111,7 @@ struct udf_sb_info
/* VAT inode */
struct inode *s_vat;
- struct semaphore s_alloc_sem;
+ struct mutex s_alloc_mutex;
};
#endif /* _UDF_FS_SB_H */
diff --git a/include/linux/vt_kern.h b/include/linux/vt_kern.h
index fab5aed8ca31..530ae3f4248c 100644
--- a/include/linux/vt_kern.h
+++ b/include/linux/vt_kern.h
@@ -73,6 +73,11 @@ int con_copy_unimap(struct vc_data *dst_vc, struct vc_data *src_vc);
int vt_waitactive(int vt);
void change_console(struct vc_data *new_vc);
void reset_vc(struct vc_data *vc);
+#ifdef CONFIG_VT
+int is_console_suspend_safe(void);
+#else
+static inline int is_console_suspend_safe(void) { return 1; }
+#endif
/*
* vc_screen.c shares this temporary buffer with the console write code so that
diff --git a/init/do_mounts_initrd.c b/init/do_mounts_initrd.c
index a05cabd0fd10..405f9031af87 100644
--- a/init/do_mounts_initrd.c
+++ b/init/do_mounts_initrd.c
@@ -56,6 +56,7 @@ static void __init handle_initrd(void)
sys_chroot(".");
mount_devfs_fs ();
+ current->flags |= PF_NOFREEZE;
pid = kernel_thread(do_linuxrc, "/linuxrc", SIGCHLD);
if (pid > 0) {
while (pid != sys_wait4(-1, NULL, 0, NULL))
diff --git a/init/main.c b/init/main.c
index 4c194c47395f..2714e0e7cfec 100644
--- a/init/main.c
+++ b/init/main.c
@@ -325,7 +325,7 @@ static inline void smp_prepare_cpus(unsigned int maxcpus) { }
#else
#ifdef __GENERIC_PER_CPU
-unsigned long __per_cpu_offset[NR_CPUS];
+unsigned long __per_cpu_offset[NR_CPUS] __read_mostly;
EXPORT_SYMBOL(__per_cpu_offset);
@@ -333,6 +333,7 @@ static void __init setup_per_cpu_areas(void)
{
unsigned long size, i;
char *ptr;
+ unsigned long nr_possible_cpus = num_possible_cpus();
/* Copy section for each CPU (we discard the original) */
size = ALIGN(__per_cpu_end - __per_cpu_start, SMP_CACHE_BYTES);
@@ -340,12 +341,12 @@ static void __init setup_per_cpu_areas(void)
if (size < PERCPU_ENOUGH_ROOM)
size = PERCPU_ENOUGH_ROOM;
#endif
+ ptr = alloc_bootmem(size * nr_possible_cpus);
- ptr = alloc_bootmem(size * NR_CPUS);
-
- for (i = 0; i < NR_CPUS; i++, ptr += size) {
+ for_each_cpu(i) {
__per_cpu_offset[i] = ptr - __per_cpu_start;
memcpy(ptr, __per_cpu_start, __per_cpu_end - __per_cpu_start);
+ ptr += size;
}
}
#endif /* !__GENERIC_PER_CPU */
@@ -438,6 +439,15 @@ void __init parse_early_param(void)
* Activate the first processor.
*/
+static void __init boot_cpu_init(void)
+{
+ int cpu = smp_processor_id();
+ /* Mark the boot cpu "present", "online" etc for SMP and UP case */
+ cpu_set(cpu, cpu_online_map);
+ cpu_set(cpu, cpu_present_map);
+ cpu_set(cpu, cpu_possible_map);
+}
+
asmlinkage void __init start_kernel(void)
{
char * command_line;
@@ -447,17 +457,13 @@ asmlinkage void __init start_kernel(void)
* enable them
*/
lock_kernel();
+ boot_cpu_init();
page_address_init();
printk(KERN_NOTICE);
printk(linux_banner);
setup_arch(&command_line);
setup_per_cpu_areas();
-
- /*
- * Mark the boot cpu "online" so that it can call console drivers in
- * printk() and can access its per-cpu storage.
- */
- smp_prepare_boot_cpu();
+ smp_prepare_boot_cpu(); /* arch-specific boot-cpu hooks */
/*
* Set up the scheduler prior starting any interrupts (such as the
diff --git a/kernel/cpuset.c b/kernel/cpuset.c
index 12815d3f1a05..c86ee051b734 100644
--- a/kernel/cpuset.c
+++ b/kernel/cpuset.c
@@ -53,7 +53,7 @@
#include <asm/uaccess.h>
#include <asm/atomic.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
#define CPUSET_SUPER_MAGIC 0x27e0eb
@@ -168,63 +168,57 @@ static struct vfsmount *cpuset_mount;
static struct super_block *cpuset_sb;
/*
- * We have two global cpuset semaphores below. They can nest.
- * It is ok to first take manage_sem, then nest callback_sem. We also
+ * We have two global cpuset mutexes below. They can nest.
+ * It is ok to first take manage_mutex, then nest callback_mutex. We also
* require taking task_lock() when dereferencing a tasks cpuset pointer.
* See "The task_lock() exception", at the end of this comment.
*
- * A task must hold both semaphores to modify cpusets. If a task
- * holds manage_sem, then it blocks others wanting that semaphore,
- * ensuring that it is the only task able to also acquire callback_sem
+ * A task must hold both mutexes to modify cpusets. If a task
+ * holds manage_mutex, then it blocks others wanting that mutex,
+ * ensuring that it is the only task able to also acquire callback_mutex
* and be able to modify cpusets. It can perform various checks on
* the cpuset structure first, knowing nothing will change. It can
- * also allocate memory while just holding manage_sem. While it is
+ * also allocate memory while just holding manage_mutex. While it is
* performing these checks, various callback routines can briefly
- * acquire callback_sem to query cpusets. Once it is ready to make
- * the changes, it takes callback_sem, blocking everyone else.
+ * acquire callback_mutex to query cpusets. Once it is ready to make
+ * the changes, it takes callback_mutex, blocking everyone else.
*
* Calls to the kernel memory allocator can not be made while holding
- * callback_sem, as that would risk double tripping on callback_sem
+ * callback_mutex, as that would risk double tripping on callback_mutex
* from one of the callbacks into the cpuset code from within
* __alloc_pages().
*
- * If a task is only holding callback_sem, then it has read-only
+ * If a task is only holding callback_mutex, then it has read-only
* access to cpusets.
*
* The task_struct fields mems_allowed and mems_generation may only
* be accessed in the context of that task, so require no locks.
*
* Any task can increment and decrement the count field without lock.
- * So in general, code holding manage_sem or callback_sem can't rely
+ * So in general, code holding manage_mutex or callback_mutex can't rely
* on the count field not changing. However, if the count goes to
- * zero, then only attach_task(), which holds both semaphores, can
+ * zero, then only attach_task(), which holds both mutexes, can
* increment it again. Because a count of zero means that no tasks
* are currently attached, therefore there is no way a task attached
* to that cpuset can fork (the other way to increment the count).
- * So code holding manage_sem or callback_sem can safely assume that
+ * So code holding manage_mutex or callback_mutex can safely assume that
* if the count is zero, it will stay zero. Similarly, if a task
- * holds manage_sem or callback_sem on a cpuset with zero count, it
+ * holds manage_mutex or callback_mutex on a cpuset with zero count, it
* knows that the cpuset won't be removed, as cpuset_rmdir() needs
- * both of those semaphores.
- *
- * A possible optimization to improve parallelism would be to make
- * callback_sem a R/W semaphore (rwsem), allowing the callback routines
- * to proceed in parallel, with read access, until the holder of
- * manage_sem needed to take this rwsem for exclusive write access
- * and modify some cpusets.
+ * both of those mutexes.
*
* The cpuset_common_file_write handler for operations that modify
- * the cpuset hierarchy holds manage_sem across the entire operation,
+ * the cpuset hierarchy holds manage_mutex across the entire operation,
* single threading all such cpuset modifications across the system.
*
- * The cpuset_common_file_read() handlers only hold callback_sem across
+ * The cpuset_common_file_read() handlers only hold callback_mutex across
* small pieces of code, such as when reading out possibly multi-word
* cpumasks and nodemasks.
*
* The fork and exit callbacks cpuset_fork() and cpuset_exit(), don't
- * (usually) take either semaphore. These are the two most performance
+ * (usually) take either mutex. These are the two most performance
* critical pieces of code here. The exception occurs on cpuset_exit(),
- * when a task in a notify_on_release cpuset exits. Then manage_sem
+ * when a task in a notify_on_release cpuset exits. Then manage_mutex
* is taken, and if the cpuset count is zero, a usermode call made
* to /sbin/cpuset_release_agent with the name of the cpuset (path
* relative to the root of cpuset file system) as the argument.
@@ -242,9 +236,9 @@ static struct super_block *cpuset_sb;
*
* The need for this exception arises from the action of attach_task(),
* which overwrites one tasks cpuset pointer with another. It does
- * so using both semaphores, however there are several performance
+ * so using both mutexes, however there are several performance
* critical places that need to reference task->cpuset without the
- * expense of grabbing a system global semaphore. Therefore except as
+ * expense of grabbing a system global mutex. Therefore except as
* noted below, when dereferencing or, as in attach_task(), modifying
* a tasks cpuset pointer we use task_lock(), which acts on a spinlock
* (task->alloc_lock) already in the task_struct routinely used for
@@ -256,8 +250,8 @@ static struct super_block *cpuset_sb;
* the routine cpuset_update_task_memory_state().
*/
-static DECLARE_MUTEX(manage_sem);
-static DECLARE_MUTEX(callback_sem);
+static DEFINE_MUTEX(manage_mutex);
+static DEFINE_MUTEX(callback_mutex);
/*
* A couple of forward declarations required, due to cyclic reference loop:
@@ -432,7 +426,7 @@ static inline struct cftype *__d_cft(struct dentry *dentry)
}
/*
- * Call with manage_sem held. Writes path of cpuset into buf.
+ * Call with manage_mutex held. Writes path of cpuset into buf.
* Returns 0 on success, -errno on error.
*/
@@ -484,11 +478,11 @@ static int cpuset_path(const struct cpuset *cs, char *buf, int buflen)
* status of the /sbin/cpuset_release_agent task, so no sense holding
* our caller up for that.
*
- * When we had only one cpuset semaphore, we had to call this
+ * When we had only one cpuset mutex, we had to call this
* without holding it, to avoid deadlock when call_usermodehelper()
* allocated memory. With two locks, we could now call this while
- * holding manage_sem, but we still don't, so as to minimize
- * the time manage_sem is held.
+ * holding manage_mutex, but we still don't, so as to minimize
+ * the time manage_mutex is held.
*/
static void cpuset_release_agent(const char *pathbuf)
@@ -520,15 +514,15 @@ static void cpuset_release_agent(const char *pathbuf)
* cs is notify_on_release() and now both the user count is zero and
* the list of children is empty, prepare cpuset path in a kmalloc'd
* buffer, to be returned via ppathbuf, so that the caller can invoke
- * cpuset_release_agent() with it later on, once manage_sem is dropped.
- * Call here with manage_sem held.
+ * cpuset_release_agent() with it later on, once manage_mutex is dropped.
+ * Call here with manage_mutex held.
*
* This check_for_release() routine is responsible for kmalloc'ing
* pathbuf. The above cpuset_release_agent() is responsible for
* kfree'ing pathbuf. The caller of these routines is responsible
* for providing a pathbuf pointer, initialized to NULL, then
- * calling check_for_release() with manage_sem held and the address
- * of the pathbuf pointer, then dropping manage_sem, then calling
+ * calling check_for_release() with manage_mutex held and the address
+ * of the pathbuf pointer, then dropping manage_mutex, then calling
* cpuset_release_agent() with pathbuf, as set by check_for_release().
*/
@@ -559,7 +553,7 @@ static void check_for_release(struct cpuset *cs, char **ppathbuf)
* One way or another, we guarantee to return some non-empty subset
* of cpu_online_map.
*
- * Call with callback_sem held.
+ * Call with callback_mutex held.
*/
static void guarantee_online_cpus(const struct cpuset *cs, cpumask_t *pmask)
@@ -583,7 +577,7 @@ static void guarantee_online_cpus(const struct cpuset *cs, cpumask_t *pmask)
* One way or another, we guarantee to return some non-empty subset
* of node_online_map.
*
- * Call with callback_sem held.
+ * Call with callback_mutex held.
*/
static void guarantee_online_mems(const struct cpuset *cs, nodemask_t *pmask)
@@ -608,12 +602,12 @@ static void guarantee_online_mems(const struct cpuset *cs, nodemask_t *pmask)
* current->cpuset if a task has its memory placement changed.
* Do not call this routine if in_interrupt().
*
- * Call without callback_sem or task_lock() held. May be called
- * with or without manage_sem held. Doesn't need task_lock to guard
+ * Call without callback_mutex or task_lock() held. May be called
+ * with or without manage_mutex held. Doesn't need task_lock to guard
* against another task changing a non-NULL cpuset pointer to NULL,
* as that is only done by a task on itself, and if the current task
* is here, it is not simultaneously in the exit code NULL'ing its
- * cpuset pointer. This routine also might acquire callback_sem and
+ * cpuset pointer. This routine also might acquire callback_mutex and
* current->mm->mmap_sem during call.
*
* Reading current->cpuset->mems_generation doesn't need task_lock
@@ -658,13 +652,13 @@ void cpuset_update_task_memory_state(void)
}
if (my_cpusets_mem_gen != tsk->cpuset_mems_generation) {
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
task_lock(tsk);
cs = tsk->cpuset; /* Maybe changed when task not locked */
guarantee_online_mems(cs, &tsk->mems_allowed);
tsk->cpuset_mems_generation = cs->mems_generation;
task_unlock(tsk);
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
mpol_rebind_task(tsk, &tsk->mems_allowed);
}
}
@@ -674,7 +668,7 @@ void cpuset_update_task_memory_state(void)
*
* One cpuset is a subset of another if all its allowed CPUs and
* Memory Nodes are a subset of the other, and its exclusive flags
- * are only set if the other's are set. Call holding manage_sem.
+ * are only set if the other's are set. Call holding manage_mutex.
*/
static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q)
@@ -692,7 +686,7 @@ static int is_cpuset_subset(const struct cpuset *p, const struct cpuset *q)
* If we replaced the flag and mask values of the current cpuset
* (cur) with those values in the trial cpuset (trial), would
* our various subset and exclusive rules still be valid? Presumes
- * manage_sem held.
+ * manage_mutex held.
*
* 'cur' is the address of an actual, in-use cpuset. Operations
* such as list traversal that depend on the actual address of the
@@ -746,7 +740,7 @@ static int validate_change(const struct cpuset *cur, const struct cpuset *trial)
* exclusive child cpusets
* Build these two partitions by calling partition_sched_domains
*
- * Call with manage_sem held. May nest a call to the
+ * Call with manage_mutex held. May nest a call to the
* lock_cpu_hotplug()/unlock_cpu_hotplug() pair.
*/
@@ -792,7 +786,7 @@ static void update_cpu_domains(struct cpuset *cur)
}
/*
- * Call with manage_sem held. May take callback_sem during call.
+ * Call with manage_mutex held. May take callback_mutex during call.
*/
static int update_cpumask(struct cpuset *cs, char *buf)
@@ -811,9 +805,9 @@ static int update_cpumask(struct cpuset *cs, char *buf)
if (retval < 0)
return retval;
cpus_unchanged = cpus_equal(cs->cpus_allowed, trialcs.cpus_allowed);
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
cs->cpus_allowed = trialcs.cpus_allowed;
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
if (is_cpu_exclusive(cs) && !cpus_unchanged)
update_cpu_domains(cs);
return 0;
@@ -827,7 +821,7 @@ static int update_cpumask(struct cpuset *cs, char *buf)
* the cpuset is marked 'memory_migrate', migrate the tasks
* pages to the new memory.
*
- * Call with manage_sem held. May take callback_sem during call.
+ * Call with manage_mutex held. May take callback_mutex during call.
* Will take tasklist_lock, scan tasklist for tasks in cpuset cs,
* lock each such tasks mm->mmap_sem, scan its vma's and rebind
* their mempolicies to the cpusets new mems_allowed.
@@ -862,11 +856,11 @@ static int update_nodemask(struct cpuset *cs, char *buf)
if (retval < 0)
goto done;
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
cs->mems_allowed = trialcs.mems_allowed;
atomic_inc(&cpuset_mems_generation);
cs->mems_generation = atomic_read(&cpuset_mems_generation);
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
set_cpuset_being_rebound(cs); /* causes mpol_copy() rebind */
@@ -922,7 +916,7 @@ static int update_nodemask(struct cpuset *cs, char *buf)
* tasklist_lock. Forks can happen again now - the mpol_copy()
* cpuset_being_rebound check will catch such forks, and rebind
* their vma mempolicies too. Because we still hold the global
- * cpuset manage_sem, we know that no other rebind effort will
+ * cpuset manage_mutex, we know that no other rebind effort will
* be contending for the global variable cpuset_being_rebound.
* It's ok if we rebind the same mm twice; mpol_rebind_mm()
* is idempotent. Also migrate pages in each mm to new nodes.
@@ -948,7 +942,7 @@ done:
}
/*
- * Call with manage_sem held.
+ * Call with manage_mutex held.
*/
static int update_memory_pressure_enabled(struct cpuset *cs, char *buf)
@@ -967,7 +961,7 @@ static int update_memory_pressure_enabled(struct cpuset *cs, char *buf)
* cs: the cpuset to update
* buf: the buffer where we read the 0 or 1
*
- * Call with manage_sem held.
+ * Call with manage_mutex held.
*/
static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf)
@@ -989,12 +983,12 @@ static int update_flag(cpuset_flagbits_t bit, struct cpuset *cs, char *buf)
return err;
cpu_exclusive_changed =
(is_cpu_exclusive(cs) != is_cpu_exclusive(&trialcs));
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
if (turning_on)
set_bit(bit, &cs->flags);
else
clear_bit(bit, &cs->flags);
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
if (cpu_exclusive_changed)
update_cpu_domains(cs);
@@ -1104,7 +1098,7 @@ static int fmeter_getrate(struct fmeter *fmp)
* writing the path of the old cpuset in 'ppathbuf' if it needs to be
* notified on release.
*
- * Call holding manage_sem. May take callback_sem and task_lock of
+ * Call holding manage_mutex. May take callback_mutex and task_lock of
* the task 'pid' during call.
*/
@@ -1144,13 +1138,13 @@ static int attach_task(struct cpuset *cs, char *pidbuf, char **ppathbuf)
get_task_struct(tsk);
}
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
task_lock(tsk);
oldcs = tsk->cpuset;
if (!oldcs) {
task_unlock(tsk);
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
put_task_struct(tsk);
return -ESRCH;
}
@@ -1164,7 +1158,7 @@ static int attach_task(struct cpuset *cs, char *pidbuf, char **ppathbuf)
from = oldcs->mems_allowed;
to = cs->mems_allowed;
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
mm = get_task_mm(tsk);
if (mm) {
@@ -1221,7 +1215,7 @@ static ssize_t cpuset_common_file_write(struct file *file, const char __user *us
}
buffer[nbytes] = 0; /* nul-terminate */
- down(&manage_sem);
+ mutex_lock(&manage_mutex);
if (is_removed(cs)) {
retval = -ENODEV;
@@ -1264,7 +1258,7 @@ static ssize_t cpuset_common_file_write(struct file *file, const char __user *us
if (retval == 0)
retval = nbytes;
out2:
- up(&manage_sem);
+ mutex_unlock(&manage_mutex);
cpuset_release_agent(pathbuf);
out1:
kfree(buffer);
@@ -1304,9 +1298,9 @@ static int cpuset_sprintf_cpulist(char *page, struct cpuset *cs)
{
cpumask_t mask;
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
mask = cs->cpus_allowed;
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
return cpulist_scnprintf(page, PAGE_SIZE, mask);
}
@@ -1315,9 +1309,9 @@ static int cpuset_sprintf_memlist(char *page, struct cpuset *cs)
{
nodemask_t mask;
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
mask = cs->mems_allowed;
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
return nodelist_scnprintf(page, PAGE_SIZE, mask);
}
@@ -1598,7 +1592,7 @@ static int pid_array_to_buf(char *buf, int sz, pid_t *a, int npids)
* Handle an open on 'tasks' file. Prepare a buffer listing the
* process id's of tasks currently attached to the cpuset being opened.
*
- * Does not require any specific cpuset semaphores, and does not take any.
+ * Does not require any specific cpuset mutexes, and does not take any.
*/
static int cpuset_tasks_open(struct inode *unused, struct file *file)
{
@@ -1754,7 +1748,7 @@ static int cpuset_populate_dir(struct dentry *cs_dentry)
* name: name of the new cpuset. Will be strcpy'ed.
* mode: mode to set on new inode
*
- * Must be called with the semaphore on the parent inode held
+ * Must be called with the mutex on the parent inode held
*/
static long cpuset_create(struct cpuset *parent, const char *name, int mode)
@@ -1766,7 +1760,7 @@ static long cpuset_create(struct cpuset *parent, const char *name, int mode)
if (!cs)
return -ENOMEM;
- down(&manage_sem);
+ mutex_lock(&manage_mutex);
cpuset_update_task_memory_state();
cs->flags = 0;
if (notify_on_release(parent))
@@ -1782,28 +1776,28 @@ static long cpuset_create(struct cpuset *parent, const char *name, int mode)
cs->parent = parent;
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
list_add(&cs->sibling, &cs->parent->children);
number_of_cpusets++;
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
err = cpuset_create_dir(cs, name, mode);
if (err < 0)
goto err;
/*
- * Release manage_sem before cpuset_populate_dir() because it
+ * Release manage_mutex before cpuset_populate_dir() because it
* will down() this new directory's i_mutex and if we race with
* another mkdir, we might deadlock.
*/
- up(&manage_sem);
+ mutex_unlock(&manage_mutex);
err = cpuset_populate_dir(cs->dentry);
/* If err < 0, we have a half-filled directory - oh well ;) */
return 0;
err:
list_del(&cs->sibling);
- up(&manage_sem);
+ mutex_unlock(&manage_mutex);
kfree(cs);
return err;
}
@@ -1825,18 +1819,18 @@ static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry)
/* the vfs holds both inode->i_mutex already */
- down(&manage_sem);
+ mutex_lock(&manage_mutex);
cpuset_update_task_memory_state();
if (atomic_read(&cs->count) > 0) {
- up(&manage_sem);
+ mutex_unlock(&manage_mutex);
return -EBUSY;
}
if (!list_empty(&cs->children)) {
- up(&manage_sem);
+ mutex_unlock(&manage_mutex);
return -EBUSY;
}
parent = cs->parent;
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
set_bit(CS_REMOVED, &cs->flags);
if (is_cpu_exclusive(cs))
update_cpu_domains(cs);
@@ -1848,10 +1842,10 @@ static int cpuset_rmdir(struct inode *unused_dir, struct dentry *dentry)
cpuset_d_remove_dir(d);
dput(d);
number_of_cpusets--;
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
if (list_empty(&parent->children))
check_for_release(parent, &pathbuf);
- up(&manage_sem);
+ mutex_unlock(&manage_mutex);
cpuset_release_agent(pathbuf);
return 0;
}
@@ -1960,19 +1954,19 @@ void cpuset_fork(struct task_struct *child)
* Description: Detach cpuset from @tsk and release it.
*
* Note that cpusets marked notify_on_release force every task in
- * them to take the global manage_sem semaphore when exiting.
+ * them to take the global manage_mutex mutex when exiting.
* This could impact scaling on very large systems. Be reluctant to
* use notify_on_release cpusets where very high task exit scaling
* is required on large systems.
*
* Don't even think about derefencing 'cs' after the cpuset use count
- * goes to zero, except inside a critical section guarded by manage_sem
- * or callback_sem. Otherwise a zero cpuset use count is a license to
+ * goes to zero, except inside a critical section guarded by manage_mutex
+ * or callback_mutex. Otherwise a zero cpuset use count is a license to
* any other task to nuke the cpuset immediately, via cpuset_rmdir().
*
- * This routine has to take manage_sem, not callback_sem, because
- * it is holding that semaphore while calling check_for_release(),
- * which calls kmalloc(), so can't be called holding callback__sem().
+ * This routine has to take manage_mutex, not callback_mutex, because
+ * it is holding that mutex while calling check_for_release(),
+ * which calls kmalloc(), so can't be called holding callback_mutex().
*
* We don't need to task_lock() this reference to tsk->cpuset,
* because tsk is already marked PF_EXITING, so attach_task() won't
@@ -2022,10 +2016,10 @@ void cpuset_exit(struct task_struct *tsk)
if (notify_on_release(cs)) {
char *pathbuf = NULL;
- down(&manage_sem);
+ mutex_lock(&manage_mutex);
if (atomic_dec_and_test(&cs->count))
check_for_release(cs, &pathbuf);
- up(&manage_sem);
+ mutex_unlock(&manage_mutex);
cpuset_release_agent(pathbuf);
} else {
atomic_dec(&cs->count);
@@ -2046,11 +2040,11 @@ cpumask_t cpuset_cpus_allowed(struct task_struct *tsk)
{
cpumask_t mask;
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
task_lock(tsk);
guarantee_online_cpus(tsk->cpuset, &mask);
task_unlock(tsk);
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
return mask;
}
@@ -2074,11 +2068,11 @@ nodemask_t cpuset_mems_allowed(struct task_struct *tsk)
{
nodemask_t mask;
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
task_lock(tsk);
guarantee_online_mems(tsk->cpuset, &mask);
task_unlock(tsk);
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
return mask;
}
@@ -2104,7 +2098,7 @@ int cpuset_zonelist_valid_mems_allowed(struct zonelist *zl)
/*
* nearest_exclusive_ancestor() - Returns the nearest mem_exclusive
- * ancestor to the specified cpuset. Call holding callback_sem.
+ * ancestor to the specified cpuset. Call holding callback_mutex.
* If no ancestor is mem_exclusive (an unusual configuration), then
* returns the root cpuset.
*/
@@ -2131,12 +2125,12 @@ static const struct cpuset *nearest_exclusive_ancestor(const struct cpuset *cs)
* GFP_KERNEL allocations are not so marked, so can escape to the
* nearest mem_exclusive ancestor cpuset.
*
- * Scanning up parent cpusets requires callback_sem. The __alloc_pages()
+ * Scanning up parent cpusets requires callback_mutex. The __alloc_pages()
* routine only calls here with __GFP_HARDWALL bit _not_ set if
* it's a GFP_KERNEL allocation, and all nodes in the current tasks
* mems_allowed came up empty on the first pass over the zonelist.
* So only GFP_KERNEL allocations, if all nodes in the cpuset are
- * short of memory, might require taking the callback_sem semaphore.
+ * short of memory, might require taking the callback_mutex mutex.
*
* The first loop over the zonelist in mm/page_alloc.c:__alloc_pages()
* calls here with __GFP_HARDWALL always set in gfp_mask, enforcing
@@ -2171,31 +2165,31 @@ int __cpuset_zone_allowed(struct zone *z, gfp_t gfp_mask)
return 1;
/* Not hardwall and node outside mems_allowed: scan up cpusets */
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
task_lock(current);
cs = nearest_exclusive_ancestor(current->cpuset);
task_unlock(current);
allowed = node_isset(node, cs->mems_allowed);
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
return allowed;
}
/**
* cpuset_lock - lock out any changes to cpuset structures
*
- * The out of memory (oom) code needs to lock down cpusets
+ * The out of memory (oom) code needs to mutex_lock cpusets
* from being changed while it scans the tasklist looking for a
- * task in an overlapping cpuset. Expose callback_sem via this
+ * task in an overlapping cpuset. Expose callback_mutex via this
* cpuset_lock() routine, so the oom code can lock it, before
* locking the task list. The tasklist_lock is a spinlock, so
- * must be taken inside callback_sem.
+ * must be taken inside callback_mutex.
*/
void cpuset_lock(void)
{
- down(&callback_sem);
+ mutex_lock(&callback_mutex);
}
/**
@@ -2206,7 +2200,7 @@ void cpuset_lock(void)
void cpuset_unlock(void)
{
- up(&callback_sem);
+ mutex_unlock(&callback_mutex);
}
/**
@@ -2218,7 +2212,7 @@ void cpuset_unlock(void)
* determine if task @p's memory usage might impact the memory
* available to the current task.
*
- * Call while holding callback_sem.
+ * Call while holding callback_mutex.
**/
int cpuset_excl_nodes_overlap(const struct task_struct *p)
@@ -2289,7 +2283,7 @@ void __cpuset_memory_pressure_bump(void)
* - Used for /proc/<pid>/cpuset.
* - No need to task_lock(tsk) on this tsk->cpuset reference, as it
* doesn't really matter if tsk->cpuset changes after we read it,
- * and we take manage_sem, keeping attach_task() from changing it
+ * and we take manage_mutex, keeping attach_task() from changing it
* anyway.
*/
@@ -2305,7 +2299,7 @@ static int proc_cpuset_show(struct seq_file *m, void *v)
return -ENOMEM;
tsk = m->private;
- down(&manage_sem);
+ mutex_lock(&manage_mutex);
cs = tsk->cpuset;
if (!cs) {
retval = -EINVAL;
@@ -2318,7 +2312,7 @@ static int proc_cpuset_show(struct seq_file *m, void *v)
seq_puts(m, buf);
seq_putc(m, '\n');
out:
- up(&manage_sem);
+ mutex_unlock(&manage_mutex);
kfree(buf);
return retval;
}
diff --git a/kernel/exit.c b/kernel/exit.c
index d1e8d500a7e1..8037405e136e 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -345,9 +345,9 @@ void daemonize(const char *name, ...)
exit_mm(current);
set_special_pids(1, 1);
- down(&tty_sem);
+ mutex_lock(&tty_mutex);
current->signal->tty = NULL;
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
/* Block and flush all signals */
sigfillset(&blocked);
diff --git a/kernel/fork.c b/kernel/fork.c
index 9bd7b65ee418..c79ae0b19a49 100644
--- a/kernel/fork.c
+++ b/kernel/fork.c
@@ -607,12 +607,12 @@ static struct files_struct *alloc_files(void)
atomic_set(&newf->count, 1);
spin_lock_init(&newf->file_lock);
+ newf->next_fd = 0;
fdt = &newf->fdtab;
- fdt->next_fd = 0;
fdt->max_fds = NR_OPEN_DEFAULT;
- fdt->max_fdset = __FD_SETSIZE;
- fdt->close_on_exec = &newf->close_on_exec_init;
- fdt->open_fds = &newf->open_fds_init;
+ fdt->max_fdset = EMBEDDED_FD_SET_SIZE;
+ fdt->close_on_exec = (fd_set *)&newf->close_on_exec_init;
+ fdt->open_fds = (fd_set *)&newf->open_fds_init;
fdt->fd = &newf->fd_array[0];
INIT_RCU_HEAD(&fdt->rcu);
fdt->free_files = NULL;
diff --git a/kernel/kprobes.c b/kernel/kprobes.c
index fef1af8a73ce..1fb9f753ef60 100644
--- a/kernel/kprobes.c
+++ b/kernel/kprobes.c
@@ -48,7 +48,7 @@
static struct hlist_head kprobe_table[KPROBE_TABLE_SIZE];
static struct hlist_head kretprobe_inst_table[KPROBE_TABLE_SIZE];
-DECLARE_MUTEX(kprobe_mutex); /* Protects kprobe_table */
+DEFINE_MUTEX(kprobe_mutex); /* Protects kprobe_table */
DEFINE_SPINLOCK(kretprobe_lock); /* Protects kretprobe_inst_table */
static DEFINE_PER_CPU(struct kprobe *, kprobe_instance) = NULL;
@@ -460,7 +460,7 @@ static int __kprobes __register_kprobe(struct kprobe *p,
}
p->nmissed = 0;
- down(&kprobe_mutex);
+ mutex_lock(&kprobe_mutex);
old_p = get_kprobe(p->addr);
if (old_p) {
ret = register_aggr_kprobe(old_p, p);
@@ -477,7 +477,7 @@ static int __kprobes __register_kprobe(struct kprobe *p,
arch_arm_kprobe(p);
out:
- up(&kprobe_mutex);
+ mutex_unlock(&kprobe_mutex);
if (ret && probed_mod)
module_put(probed_mod);
@@ -496,10 +496,10 @@ void __kprobes unregister_kprobe(struct kprobe *p)
struct kprobe *old_p, *list_p;
int cleanup_p;
- down(&kprobe_mutex);
+ mutex_lock(&kprobe_mutex);
old_p = get_kprobe(p->addr);
if (unlikely(!old_p)) {
- up(&kprobe_mutex);
+ mutex_unlock(&kprobe_mutex);
return;
}
if (p != old_p) {
@@ -507,7 +507,7 @@ void __kprobes unregister_kprobe(struct kprobe *p)
if (list_p == p)
/* kprobe p is a valid probe */
goto valid_p;
- up(&kprobe_mutex);
+ mutex_unlock(&kprobe_mutex);
return;
}
valid_p:
@@ -523,7 +523,7 @@ valid_p:
cleanup_p = 0;
}
- up(&kprobe_mutex);
+ mutex_unlock(&kprobe_mutex);
synchronize_sched();
if (p->mod_refcounted &&
diff --git a/kernel/kthread.c b/kernel/kthread.c
index e75950a1092c..6a5373868a98 100644
--- a/kernel/kthread.c
+++ b/kernel/kthread.c
@@ -12,6 +12,7 @@
#include <linux/unistd.h>
#include <linux/file.h>
#include <linux/module.h>
+#include <linux/mutex.h>
#include <asm/semaphore.h>
/*
@@ -41,7 +42,7 @@ struct kthread_stop_info
/* Thread stopping is done by setthing this var: lock serializes
* multiple kthread_stop calls. */
-static DECLARE_MUTEX(kthread_stop_lock);
+static DEFINE_MUTEX(kthread_stop_lock);
static struct kthread_stop_info kthread_stop_info;
int kthread_should_stop(void)
@@ -173,7 +174,7 @@ int kthread_stop_sem(struct task_struct *k, struct semaphore *s)
{
int ret;
- down(&kthread_stop_lock);
+ mutex_lock(&kthread_stop_lock);
/* It could exit after stop_info.k set, but before wake_up_process. */
get_task_struct(k);
@@ -194,7 +195,7 @@ int kthread_stop_sem(struct task_struct *k, struct semaphore *s)
wait_for_completion(&kthread_stop_info.done);
kthread_stop_info.k = NULL;
ret = kthread_stop_info.err;
- up(&kthread_stop_lock);
+ mutex_unlock(&kthread_stop_lock);
return ret;
}
diff --git a/kernel/module.c b/kernel/module.c
index 77764f22f021..fb404299082e 100644
--- a/kernel/module.c
+++ b/kernel/module.c
@@ -39,6 +39,7 @@
#include <linux/device.h>
#include <linux/string.h>
#include <linux/sched.h>
+#include <linux/mutex.h>
#include <asm/uaccess.h>
#include <asm/semaphore.h>
#include <asm/cacheflush.h>
@@ -60,18 +61,18 @@
static DEFINE_SPINLOCK(modlist_lock);
/* List of modules, protected by module_mutex AND modlist_lock */
-static DECLARE_MUTEX(module_mutex);
+static DEFINE_MUTEX(module_mutex);
static LIST_HEAD(modules);
-static DECLARE_MUTEX(notify_mutex);
+static DEFINE_MUTEX(notify_mutex);
static struct notifier_block * module_notify_list;
int register_module_notifier(struct notifier_block * nb)
{
int err;
- down(&notify_mutex);
+ mutex_lock(&notify_mutex);
err = notifier_chain_register(&module_notify_list, nb);
- up(&notify_mutex);
+ mutex_unlock(&notify_mutex);
return err;
}
EXPORT_SYMBOL(register_module_notifier);
@@ -79,9 +80,9 @@ EXPORT_SYMBOL(register_module_notifier);
int unregister_module_notifier(struct notifier_block * nb)
{
int err;
- down(&notify_mutex);
+ mutex_lock(&notify_mutex);
err = notifier_chain_unregister(&module_notify_list, nb);
- up(&notify_mutex);
+ mutex_unlock(&notify_mutex);
return err;
}
EXPORT_SYMBOL(unregister_module_notifier);
@@ -601,7 +602,7 @@ static void free_module(struct module *mod);
static void wait_for_zero_refcount(struct module *mod)
{
/* Since we might sleep for some time, drop the semaphore first */
- up(&module_mutex);
+ mutex_unlock(&module_mutex);
for (;;) {
DEBUGP("Looking at refcount...\n");
set_current_state(TASK_UNINTERRUPTIBLE);
@@ -610,7 +611,7 @@ static void wait_for_zero_refcount(struct module *mod)
schedule();
}
current->state = TASK_RUNNING;
- down(&module_mutex);
+ mutex_lock(&module_mutex);
}
asmlinkage long
@@ -627,7 +628,7 @@ sys_delete_module(const char __user *name_user, unsigned int flags)
return -EFAULT;
name[MODULE_NAME_LEN-1] = '\0';
- if (down_interruptible(&module_mutex) != 0)
+ if (mutex_lock_interruptible(&module_mutex) != 0)
return -EINTR;
mod = find_module(name);
@@ -676,14 +677,14 @@ sys_delete_module(const char __user *name_user, unsigned int flags)
/* Final destruction now noone is using it. */
if (mod->exit != NULL) {
- up(&module_mutex);
+ mutex_unlock(&module_mutex);
mod->exit();
- down(&module_mutex);
+ mutex_lock(&module_mutex);
}
free_module(mod);
out:
- up(&module_mutex);
+ mutex_unlock(&module_mutex);
return ret;
}
@@ -1972,13 +1973,13 @@ sys_init_module(void __user *umod,
return -EPERM;
/* Only one module load at a time, please */
- if (down_interruptible(&module_mutex) != 0)
+ if (mutex_lock_interruptible(&module_mutex) != 0)
return -EINTR;
/* Do all the hard work */
mod = load_module(umod, len, uargs);
if (IS_ERR(mod)) {
- up(&module_mutex);
+ mutex_unlock(&module_mutex);
return PTR_ERR(mod);
}
@@ -1987,11 +1988,11 @@ sys_init_module(void __user *umod,
stop_machine_run(__link_module, mod, NR_CPUS);
/* Drop lock so they can recurse */
- up(&module_mutex);
+ mutex_unlock(&module_mutex);
- down(&notify_mutex);
+ mutex_lock(&notify_mutex);
notifier_call_chain(&module_notify_list, MODULE_STATE_COMING, mod);
- up(&notify_mutex);
+ mutex_unlock(&notify_mutex);
/* Start the module */
if (mod->init != NULL)
@@ -2006,15 +2007,15 @@ sys_init_module(void __user *umod,
mod->name);
else {
module_put(mod);
- down(&module_mutex);
+ mutex_lock(&module_mutex);
free_module(mod);
- up(&module_mutex);
+ mutex_unlock(&module_mutex);
}
return ret;
}
/* Now it's a first class citizen! */
- down(&module_mutex);
+ mutex_lock(&module_mutex);
mod->state = MODULE_STATE_LIVE;
/* Drop initial reference. */
module_put(mod);
@@ -2022,7 +2023,7 @@ sys_init_module(void __user *umod,
mod->module_init = NULL;
mod->init_size = 0;
mod->init_text_size = 0;
- up(&module_mutex);
+ mutex_unlock(&module_mutex);
return 0;
}
@@ -2112,7 +2113,7 @@ struct module *module_get_kallsym(unsigned int symnum,
{
struct module *mod;
- down(&module_mutex);
+ mutex_lock(&module_mutex);
list_for_each_entry(mod, &modules, list) {
if (symnum < mod->num_symtab) {
*value = mod->symtab[symnum].st_value;
@@ -2120,12 +2121,12 @@ struct module *module_get_kallsym(unsigned int symnum,
strncpy(namebuf,
mod->strtab + mod->symtab[symnum].st_name,
127);
- up(&module_mutex);
+ mutex_unlock(&module_mutex);
return mod;
}
symnum -= mod->num_symtab;
}
- up(&module_mutex);
+ mutex_unlock(&module_mutex);
return NULL;
}
@@ -2168,7 +2169,7 @@ static void *m_start(struct seq_file *m, loff_t *pos)
struct list_head *i;
loff_t n = 0;
- down(&module_mutex);
+ mutex_lock(&module_mutex);
list_for_each(i, &modules) {
if (n++ == *pos)
break;
@@ -2189,7 +2190,7 @@ static void *m_next(struct seq_file *m, void *p, loff_t *pos)
static void m_stop(struct seq_file *m, void *p)
{
- up(&module_mutex);
+ mutex_unlock(&module_mutex);
}
static int m_show(struct seq_file *m, void *p)
diff --git a/kernel/panic.c b/kernel/panic.c
index 126dc43f1c74..acd95adddb93 100644
--- a/kernel/panic.c
+++ b/kernel/panic.c
@@ -20,10 +20,13 @@
#include <linux/nmi.h>
#include <linux/kexec.h>
-int panic_timeout;
int panic_on_oops;
int tainted;
+static int pause_on_oops;
+static int pause_on_oops_flag;
+static DEFINE_SPINLOCK(pause_on_oops_lock);
+int panic_timeout;
EXPORT_SYMBOL(panic_timeout);
struct notifier_block *panic_notifier_list;
@@ -174,3 +177,95 @@ void add_taint(unsigned flag)
tainted |= flag;
}
EXPORT_SYMBOL(add_taint);
+
+static int __init pause_on_oops_setup(char *str)
+{
+ pause_on_oops = simple_strtoul(str, NULL, 0);
+ return 1;
+}
+__setup("pause_on_oops=", pause_on_oops_setup);
+
+static void spin_msec(int msecs)
+{
+ int i;
+
+ for (i = 0; i < msecs; i++) {
+ touch_nmi_watchdog();
+ mdelay(1);
+ }
+}
+
+/*
+ * It just happens that oops_enter() and oops_exit() are identically
+ * implemented...
+ */
+static void do_oops_enter_exit(void)
+{
+ unsigned long flags;
+ static int spin_counter;
+
+ if (!pause_on_oops)
+ return;
+
+ spin_lock_irqsave(&pause_on_oops_lock, flags);
+ if (pause_on_oops_flag == 0) {
+ /* This CPU may now print the oops message */
+ pause_on_oops_flag = 1;
+ } else {
+ /* We need to stall this CPU */
+ if (!spin_counter) {
+ /* This CPU gets to do the counting */
+ spin_counter = pause_on_oops;
+ do {
+ spin_unlock(&pause_on_oops_lock);
+ spin_msec(MSEC_PER_SEC);
+ spin_lock(&pause_on_oops_lock);
+ } while (--spin_counter);
+ pause_on_oops_flag = 0;
+ } else {
+ /* This CPU waits for a different one */
+ while (spin_counter) {
+ spin_unlock(&pause_on_oops_lock);
+ spin_msec(1);
+ spin_lock(&pause_on_oops_lock);
+ }
+ }
+ }
+ spin_unlock_irqrestore(&pause_on_oops_lock, flags);
+}
+
+/*
+ * Return true if the calling CPU is allowed to print oops-related info. This
+ * is a bit racy..
+ */
+int oops_may_print(void)
+{
+ return pause_on_oops_flag == 0;
+}
+
+/*
+ * Called when the architecture enters its oops handler, before it prints
+ * anything. If this is the first CPU to oops, and it's oopsing the first time
+ * then let it proceed.
+ *
+ * This is all enabled by the pause_on_oops kernel boot option. We do all this
+ * to ensure that oopses don't scroll off the screen. It has the side-effect
+ * of preventing later-oopsing CPUs from mucking up the display, too.
+ *
+ * It turns out that the CPU which is allowed to print ends up pausing for the
+ * right duration, whereas all the other CPUs pause for twice as long: once in
+ * oops_enter(), once in oops_exit().
+ */
+void oops_enter(void)
+{
+ do_oops_enter_exit();
+}
+
+/*
+ * Called when the architecture exits its oops handler, after printing
+ * everything.
+ */
+void oops_exit(void)
+{
+ do_oops_enter_exit();
+}
diff --git a/kernel/posix-timers.c b/kernel/posix-timers.c
index fa895fc2ecf5..9944379360b5 100644
--- a/kernel/posix-timers.c
+++ b/kernel/posix-timers.c
@@ -35,6 +35,7 @@
#include <linux/interrupt.h>
#include <linux/slab.h>
#include <linux/time.h>
+#include <linux/mutex.h>
#include <asm/uaccess.h>
#include <asm/semaphore.h>
diff --git a/kernel/power/Makefile b/kernel/power/Makefile
index 04be7d0d96a7..8d0af3d37a4b 100644
--- a/kernel/power/Makefile
+++ b/kernel/power/Makefile
@@ -5,7 +5,7 @@ endif
obj-y := main.o process.o console.o
obj-$(CONFIG_PM_LEGACY) += pm.o
-obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o disk.o snapshot.o
+obj-$(CONFIG_SOFTWARE_SUSPEND) += swsusp.o disk.o snapshot.o swap.o user.o
obj-$(CONFIG_SUSPEND_SMP) += smp.o
diff --git a/kernel/power/disk.c b/kernel/power/disk.c
index 0b43847dc980..81d4d982f3f0 100644
--- a/kernel/power/disk.c
+++ b/kernel/power/disk.c
@@ -22,17 +22,6 @@
#include "power.h"
-extern suspend_disk_method_t pm_disk_mode;
-
-extern int swsusp_shrink_memory(void);
-extern int swsusp_suspend(void);
-extern int swsusp_write(struct pbe *pblist, unsigned int nr_pages);
-extern int swsusp_check(void);
-extern int swsusp_read(struct pbe **pblist_ptr);
-extern void swsusp_close(void);
-extern int swsusp_resume(void);
-
-
static int noresume = 0;
char resume_file[256] = CONFIG_PM_STD_PARTITION;
dev_t swsusp_resume_device;
@@ -70,10 +59,6 @@ static void power_down(suspend_disk_method_t mode)
while(1);
}
-
-static int in_suspend __nosavedata = 0;
-
-
static inline void platform_finish(void)
{
if (pm_disk_mode == PM_DISK_PLATFORM) {
@@ -87,7 +72,6 @@ static int prepare_processes(void)
int error;
pm_prepare_console();
- sys_sync();
disable_nonboot_cpus();
if (freeze_processes()) {
@@ -145,7 +129,7 @@ int pm_suspend_disk(void)
if (in_suspend) {
device_resume();
pr_debug("PM: writing image.\n");
- error = swsusp_write(pagedir_nosave, nr_copy_pages);
+ error = swsusp_write();
if (!error)
power_down(pm_disk_mode);
else {
@@ -216,7 +200,7 @@ static int software_resume(void)
pr_debug("PM: Reading swsusp image.\n");
- if ((error = swsusp_read(&pagedir_nosave))) {
+ if ((error = swsusp_read())) {
swsusp_free();
goto Thaw;
}
diff --git a/kernel/power/main.c b/kernel/power/main.c
index 9cb235cba4a9..ee371f50ccaa 100644
--- a/kernel/power/main.c
+++ b/kernel/power/main.c
@@ -103,7 +103,7 @@ static int suspend_prepare(suspend_state_t state)
}
-static int suspend_enter(suspend_state_t state)
+int suspend_enter(suspend_state_t state)
{
int error = 0;
unsigned long flags;
diff --git a/kernel/power/pm.c b/kernel/power/pm.c
index 33c508e857dd..0f6908cce1dd 100644
--- a/kernel/power/pm.c
+++ b/kernel/power/pm.c
@@ -25,6 +25,7 @@
#include <linux/pm.h>
#include <linux/pm_legacy.h>
#include <linux/interrupt.h>
+#include <linux/mutex.h>
int pm_active;
@@ -40,7 +41,7 @@ int pm_active;
* until a resume but that will be fine.
*/
-static DECLARE_MUTEX(pm_devs_lock);
+static DEFINE_MUTEX(pm_devs_lock);
static LIST_HEAD(pm_devs);
/**
@@ -67,9 +68,9 @@ struct pm_dev *pm_register(pm_dev_t type,
dev->id = id;
dev->callback = callback;
- down(&pm_devs_lock);
+ mutex_lock(&pm_devs_lock);
list_add(&dev->entry, &pm_devs);
- up(&pm_devs_lock);
+ mutex_unlock(&pm_devs_lock);
}
return dev;
}
@@ -85,9 +86,9 @@ struct pm_dev *pm_register(pm_dev_t type,
void pm_unregister(struct pm_dev *dev)
{
if (dev) {
- down(&pm_devs_lock);
+ mutex_lock(&pm_devs_lock);
list_del(&dev->entry);
- up(&pm_devs_lock);
+ mutex_unlock(&pm_devs_lock);
kfree(dev);
}
@@ -118,7 +119,7 @@ void pm_unregister_all(pm_callback callback)
if (!callback)
return;
- down(&pm_devs_lock);
+ mutex_lock(&pm_devs_lock);
entry = pm_devs.next;
while (entry != &pm_devs) {
struct pm_dev *dev = list_entry(entry, struct pm_dev, entry);
@@ -126,7 +127,7 @@ void pm_unregister_all(pm_callback callback)
if (dev->callback == callback)
__pm_unregister(dev);
}
- up(&pm_devs_lock);
+ mutex_unlock(&pm_devs_lock);
}
/**
@@ -234,7 +235,7 @@ int pm_send_all(pm_request_t rqst, void *data)
{
struct list_head *entry;
- down(&pm_devs_lock);
+ mutex_lock(&pm_devs_lock);
entry = pm_devs.next;
while (entry != &pm_devs) {
struct pm_dev *dev = list_entry(entry, struct pm_dev, entry);
@@ -246,13 +247,13 @@ int pm_send_all(pm_request_t rqst, void *data)
*/
if (rqst == PM_SUSPEND)
pm_undo_all(dev);
- up(&pm_devs_lock);
+ mutex_unlock(&pm_devs_lock);
return status;
}
}
entry = entry->next;
}
- up(&pm_devs_lock);
+ mutex_unlock(&pm_devs_lock);
return 0;
}
diff --git a/kernel/power/power.h b/kernel/power/power.h
index 388dba680841..f06f12f21767 100644
--- a/kernel/power/power.h
+++ b/kernel/power/power.h
@@ -8,6 +8,7 @@ struct swsusp_info {
int cpus;
unsigned long image_pages;
unsigned long pages;
+ unsigned long size;
} __attribute__((aligned(PAGE_SIZE)));
@@ -37,21 +38,79 @@ extern struct subsystem power_subsys;
/* References to section boundaries */
extern const void __nosave_begin, __nosave_end;
-extern unsigned int nr_copy_pages;
extern struct pbe *pagedir_nosave;
/* Preferred image size in bytes (default 500 MB) */
extern unsigned long image_size;
+extern int in_suspend;
+extern dev_t swsusp_resume_device;
extern asmlinkage int swsusp_arch_suspend(void);
extern asmlinkage int swsusp_arch_resume(void);
extern unsigned int count_data_pages(void);
-extern void free_pagedir(struct pbe *pblist);
-extern void release_eaten_pages(void);
-extern struct pbe *alloc_pagedir(unsigned nr_pages, gfp_t gfp_mask, int safe_needed);
+
+struct snapshot_handle {
+ loff_t offset;
+ unsigned int page;
+ unsigned int page_offset;
+ unsigned int prev;
+ struct pbe *pbe;
+ void *buffer;
+ unsigned int buf_offset;
+};
+
+#define data_of(handle) ((handle).buffer + (handle).buf_offset)
+
+extern int snapshot_read_next(struct snapshot_handle *handle, size_t count);
+extern int snapshot_write_next(struct snapshot_handle *handle, size_t count);
+int snapshot_image_loaded(struct snapshot_handle *handle);
+
+#define SNAPSHOT_IOC_MAGIC '3'
+#define SNAPSHOT_FREEZE _IO(SNAPSHOT_IOC_MAGIC, 1)
+#define SNAPSHOT_UNFREEZE _IO(SNAPSHOT_IOC_MAGIC, 2)
+#define SNAPSHOT_ATOMIC_SNAPSHOT _IOW(SNAPSHOT_IOC_MAGIC, 3, void *)
+#define SNAPSHOT_ATOMIC_RESTORE _IO(SNAPSHOT_IOC_MAGIC, 4)
+#define SNAPSHOT_FREE _IO(SNAPSHOT_IOC_MAGIC, 5)
+#define SNAPSHOT_SET_IMAGE_SIZE _IOW(SNAPSHOT_IOC_MAGIC, 6, unsigned long)
+#define SNAPSHOT_AVAIL_SWAP _IOR(SNAPSHOT_IOC_MAGIC, 7, void *)
+#define SNAPSHOT_GET_SWAP_PAGE _IOR(SNAPSHOT_IOC_MAGIC, 8, void *)
+#define SNAPSHOT_FREE_SWAP_PAGES _IO(SNAPSHOT_IOC_MAGIC, 9)
+#define SNAPSHOT_SET_SWAP_FILE _IOW(SNAPSHOT_IOC_MAGIC, 10, unsigned int)
+#define SNAPSHOT_S2RAM _IO(SNAPSHOT_IOC_MAGIC, 11)
+#define SNAPSHOT_IOC_MAXNR 11
+
+/**
+ * The bitmap is used for tracing allocated swap pages
+ *
+ * The entire bitmap consists of a number of bitmap_page
+ * structures linked with the help of the .next member.
+ * Thus each page can be allocated individually, so we only
+ * need to make 0-order memory allocations to create
+ * the bitmap.
+ */
+
+#define BITMAP_PAGE_SIZE (PAGE_SIZE - sizeof(void *))
+#define BITMAP_PAGE_CHUNKS (BITMAP_PAGE_SIZE / sizeof(long))
+#define BITS_PER_CHUNK (sizeof(long) * 8)
+#define BITMAP_PAGE_BITS (BITMAP_PAGE_CHUNKS * BITS_PER_CHUNK)
+
+struct bitmap_page {
+ unsigned long chunks[BITMAP_PAGE_CHUNKS];
+ struct bitmap_page *next;
+};
+
+extern void free_bitmap(struct bitmap_page *bitmap);
+extern struct bitmap_page *alloc_bitmap(unsigned int nr_bits);
+extern unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap);
+extern void free_all_swap_pages(int swap, struct bitmap_page *bitmap);
+
+extern int swsusp_check(void);
+extern int swsusp_shrink_memory(void);
extern void swsusp_free(void);
-extern int alloc_data_pages(struct pbe *pblist, gfp_t gfp_mask, int safe_needed);
-extern unsigned int snapshot_nr_pages(void);
-extern struct pbe *snapshot_pblist(void);
-extern void snapshot_pblist_set(struct pbe *pblist);
+extern int swsusp_suspend(void);
+extern int swsusp_resume(void);
+extern int swsusp_read(void);
+extern int swsusp_write(void);
+extern void swsusp_close(void);
+extern int suspend_enter(suspend_state_t state);
diff --git a/kernel/power/process.c b/kernel/power/process.c
index 28de118f7a0b..8ac7c35fad77 100644
--- a/kernel/power/process.c
+++ b/kernel/power/process.c
@@ -12,11 +12,12 @@
#include <linux/interrupt.h>
#include <linux/suspend.h>
#include <linux/module.h>
+#include <linux/syscalls.h>
/*
* Timeout for stopping processes
*/
-#define TIMEOUT (6 * HZ)
+#define TIMEOUT (20 * HZ)
static inline int freezeable(struct task_struct * p)
@@ -54,38 +55,62 @@ void refrigerator(void)
current->state = save;
}
+static inline void freeze_process(struct task_struct *p)
+{
+ unsigned long flags;
+
+ if (!freezing(p)) {
+ freeze(p);
+ spin_lock_irqsave(&p->sighand->siglock, flags);
+ signal_wake_up(p, 0);
+ spin_unlock_irqrestore(&p->sighand->siglock, flags);
+ }
+}
+
/* 0 = success, else # of processes that we failed to stop */
int freeze_processes(void)
{
- int todo;
+ int todo, nr_user, user_frozen;
unsigned long start_time;
struct task_struct *g, *p;
unsigned long flags;
printk( "Stopping tasks: " );
start_time = jiffies;
+ user_frozen = 0;
do {
- todo = 0;
+ nr_user = todo = 0;
read_lock(&tasklist_lock);
do_each_thread(g, p) {
if (!freezeable(p))
continue;
if (frozen(p))
continue;
-
- freeze(p);
- spin_lock_irqsave(&p->sighand->siglock, flags);
- signal_wake_up(p, 0);
- spin_unlock_irqrestore(&p->sighand->siglock, flags);
- todo++;
+ if (p->mm && !(p->flags & PF_BORROWED_MM)) {
+ /* The task is a user-space one.
+ * Freeze it unless there's a vfork completion
+ * pending
+ */
+ if (!p->vfork_done)
+ freeze_process(p);
+ nr_user++;
+ } else {
+ /* Freeze only if the user space is frozen */
+ if (user_frozen)
+ freeze_process(p);
+ todo++;
+ }
} while_each_thread(g, p);
read_unlock(&tasklist_lock);
+ todo += nr_user;
+ if (!user_frozen && !nr_user) {
+ sys_sync();
+ start_time = jiffies;
+ }
+ user_frozen = !nr_user;
yield(); /* Yield is okay here */
- if (todo && time_after(jiffies, start_time + TIMEOUT)) {
- printk( "\n" );
- printk(KERN_ERR " stopping tasks failed (%d tasks remaining)\n", todo );
+ if (todo && time_after(jiffies, start_time + TIMEOUT))
break;
- }
} while(todo);
/* This does not unfreeze processes that are already frozen
@@ -94,8 +119,14 @@ int freeze_processes(void)
* but it cleans up leftover PF_FREEZE requests.
*/
if (todo) {
+ printk( "\n" );
+ printk(KERN_ERR " stopping tasks timed out "
+ "after %d seconds (%d tasks remaining):\n",
+ TIMEOUT / HZ, todo);
read_lock(&tasklist_lock);
- do_each_thread(g, p)
+ do_each_thread(g, p) {
+ if (freezeable(p) && !frozen(p))
+ printk(KERN_ERR " %s\n", p->comm);
if (freezing(p)) {
pr_debug(" clean up: %s\n", p->comm);
p->flags &= ~PF_FREEZE;
@@ -103,7 +134,7 @@ int freeze_processes(void)
recalc_sigpending_tsk(p);
spin_unlock_irqrestore(&p->sighand->siglock, flags);
}
- while_each_thread(g, p);
+ } while_each_thread(g, p);
read_unlock(&tasklist_lock);
return todo;
}
diff --git a/kernel/power/snapshot.c b/kernel/power/snapshot.c
index 8d5a5986d621..c5863d02c89e 100644
--- a/kernel/power/snapshot.c
+++ b/kernel/power/snapshot.c
@@ -10,6 +10,7 @@
*/
+#include <linux/version.h>
#include <linux/module.h>
#include <linux/mm.h>
#include <linux/suspend.h>
@@ -34,7 +35,9 @@
#include "power.h"
struct pbe *pagedir_nosave;
-unsigned int nr_copy_pages;
+static unsigned int nr_copy_pages;
+static unsigned int nr_meta_pages;
+static unsigned long *buffer;
#ifdef CONFIG_HIGHMEM
unsigned int count_highmem_pages(void)
@@ -80,7 +83,7 @@ static int save_highmem_zone(struct zone *zone)
void *kaddr;
unsigned long pfn = zone_pfn + zone->zone_start_pfn;
- if (!(pfn%1000))
+ if (!(pfn%10000))
printk(".");
if (!pfn_valid(pfn))
continue;
@@ -119,13 +122,15 @@ int save_highmem(void)
struct zone *zone;
int res = 0;
- pr_debug("swsusp: Saving Highmem\n");
+ pr_debug("swsusp: Saving Highmem");
+ drain_local_pages();
for_each_zone (zone) {
if (is_highmem(zone))
res = save_highmem_zone(zone);
if (res)
return res;
}
+ printk("\n");
return 0;
}
@@ -235,7 +240,7 @@ static void copy_data_pages(struct pbe *pblist)
* free_pagedir - free pages allocated with alloc_pagedir()
*/
-void free_pagedir(struct pbe *pblist)
+static void free_pagedir(struct pbe *pblist)
{
struct pbe *pbe;
@@ -301,7 +306,7 @@ struct eaten_page {
static struct eaten_page *eaten_pages = NULL;
-void release_eaten_pages(void)
+static void release_eaten_pages(void)
{
struct eaten_page *p, *q;
@@ -376,7 +381,6 @@ struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask, int safe_needed
if (!nr_pages)
return NULL;
- pr_debug("alloc_pagedir(): nr_pages = %d\n", nr_pages);
pblist = alloc_image_page(gfp_mask, safe_needed);
/* FIXME: rewrite this ugly loop */
for (pbe = pblist, num = PBES_PER_PAGE; pbe && num < nr_pages;
@@ -388,7 +392,7 @@ struct pbe *alloc_pagedir(unsigned int nr_pages, gfp_t gfp_mask, int safe_needed
free_pagedir(pblist);
pblist = NULL;
} else
- create_pbe_list(pblist, nr_pages);
+ create_pbe_list(pblist, nr_pages);
return pblist;
}
@@ -414,6 +418,10 @@ void swsusp_free(void)
}
}
}
+ nr_copy_pages = 0;
+ nr_meta_pages = 0;
+ pagedir_nosave = NULL;
+ buffer = NULL;
}
@@ -437,7 +445,7 @@ static int enough_free_mem(unsigned int nr_pages)
(nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
}
-int alloc_data_pages(struct pbe *pblist, gfp_t gfp_mask, int safe_needed)
+static int alloc_data_pages(struct pbe *pblist, gfp_t gfp_mask, int safe_needed)
{
struct pbe *p;
@@ -504,7 +512,318 @@ asmlinkage int swsusp_save(void)
*/
nr_copy_pages = nr_pages;
+ nr_meta_pages = (nr_pages * sizeof(long) + PAGE_SIZE - 1) >> PAGE_SHIFT;
printk("swsusp: critical section/: done (%d pages copied)\n", nr_pages);
return 0;
}
+
+static void init_header(struct swsusp_info *info)
+{
+ memset(info, 0, sizeof(struct swsusp_info));
+ info->version_code = LINUX_VERSION_CODE;
+ info->num_physpages = num_physpages;
+ memcpy(&info->uts, &system_utsname, sizeof(system_utsname));
+ info->cpus = num_online_cpus();
+ info->image_pages = nr_copy_pages;
+ info->pages = nr_copy_pages + nr_meta_pages + 1;
+ info->size = info->pages;
+ info->size <<= PAGE_SHIFT;
+}
+
+/**
+ * pack_orig_addresses - the .orig_address fields of the PBEs from the
+ * list starting at @pbe are stored in the array @buf[] (1 page)
+ */
+
+static inline struct pbe *pack_orig_addresses(unsigned long *buf, struct pbe *pbe)
+{
+ int j;
+
+ for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
+ buf[j] = pbe->orig_address;
+ pbe = pbe->next;
+ }
+ if (!pbe)
+ for (; j < PAGE_SIZE / sizeof(long); j++)
+ buf[j] = 0;
+ return pbe;
+}
+
+/**
+ * snapshot_read_next - used for reading the system memory snapshot.
+ *
+ * On the first call to it @handle should point to a zeroed
+ * snapshot_handle structure. The structure gets updated and a pointer
+ * to it should be passed to this function every next time.
+ *
+ * The @count parameter should contain the number of bytes the caller
+ * wants to read from the snapshot. It must not be zero.
+ *
+ * On success the function returns a positive number. Then, the caller
+ * is allowed to read up to the returned number of bytes from the memory
+ * location computed by the data_of() macro. The number returned
+ * may be smaller than @count, but this only happens if the read would
+ * cross a page boundary otherwise.
+ *
+ * The function returns 0 to indicate the end of data stream condition,
+ * and a negative number is returned on error. In such cases the
+ * structure pointed to by @handle is not updated and should not be used
+ * any more.
+ */
+
+int snapshot_read_next(struct snapshot_handle *handle, size_t count)
+{
+ if (handle->page > nr_meta_pages + nr_copy_pages)
+ return 0;
+ if (!buffer) {
+ /* This makes the buffer be freed by swsusp_free() */
+ buffer = alloc_image_page(GFP_ATOMIC, 0);
+ if (!buffer)
+ return -ENOMEM;
+ }
+ if (!handle->offset) {
+ init_header((struct swsusp_info *)buffer);
+ handle->buffer = buffer;
+ handle->pbe = pagedir_nosave;
+ }
+ if (handle->prev < handle->page) {
+ if (handle->page <= nr_meta_pages) {
+ handle->pbe = pack_orig_addresses(buffer, handle->pbe);
+ if (!handle->pbe)
+ handle->pbe = pagedir_nosave;
+ } else {
+ handle->buffer = (void *)handle->pbe->address;
+ handle->pbe = handle->pbe->next;
+ }
+ handle->prev = handle->page;
+ }
+ handle->buf_offset = handle->page_offset;
+ if (handle->page_offset + count >= PAGE_SIZE) {
+ count = PAGE_SIZE - handle->page_offset;
+ handle->page_offset = 0;
+ handle->page++;
+ } else {
+ handle->page_offset += count;
+ }
+ handle->offset += count;
+ return count;
+}
+
+/**
+ * mark_unsafe_pages - mark the pages that cannot be used for storing
+ * the image during resume, because they conflict with the pages that
+ * had been used before suspend
+ */
+
+static int mark_unsafe_pages(struct pbe *pblist)
+{
+ struct zone *zone;
+ unsigned long zone_pfn;
+ struct pbe *p;
+
+ if (!pblist) /* a sanity check */
+ return -EINVAL;
+
+ /* Clear page flags */
+ for_each_zone (zone) {
+ for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
+ if (pfn_valid(zone_pfn + zone->zone_start_pfn))
+ ClearPageNosaveFree(pfn_to_page(zone_pfn +
+ zone->zone_start_pfn));
+ }
+
+ /* Mark orig addresses */
+ for_each_pbe (p, pblist) {
+ if (virt_addr_valid(p->orig_address))
+ SetPageNosaveFree(virt_to_page(p->orig_address));
+ else
+ return -EFAULT;
+ }
+
+ return 0;
+}
+
+static void copy_page_backup_list(struct pbe *dst, struct pbe *src)
+{
+ /* We assume both lists contain the same number of elements */
+ while (src) {
+ dst->orig_address = src->orig_address;
+ dst = dst->next;
+ src = src->next;
+ }
+}
+
+static int check_header(struct swsusp_info *info)
+{
+ char *reason = NULL;
+
+ if (info->version_code != LINUX_VERSION_CODE)
+ reason = "kernel version";
+ if (info->num_physpages != num_physpages)
+ reason = "memory size";
+ if (strcmp(info->uts.sysname,system_utsname.sysname))
+ reason = "system type";
+ if (strcmp(info->uts.release,system_utsname.release))
+ reason = "kernel release";
+ if (strcmp(info->uts.version,system_utsname.version))
+ reason = "version";
+ if (strcmp(info->uts.machine,system_utsname.machine))
+ reason = "machine";
+ if (reason) {
+ printk(KERN_ERR "swsusp: Resume mismatch: %s\n", reason);
+ return -EPERM;
+ }
+ return 0;
+}
+
+/**
+ * load header - check the image header and copy data from it
+ */
+
+static int load_header(struct snapshot_handle *handle,
+ struct swsusp_info *info)
+{
+ int error;
+ struct pbe *pblist;
+
+ error = check_header(info);
+ if (!error) {
+ pblist = alloc_pagedir(info->image_pages, GFP_ATOMIC, 0);
+ if (!pblist)
+ return -ENOMEM;
+ pagedir_nosave = pblist;
+ handle->pbe = pblist;
+ nr_copy_pages = info->image_pages;
+ nr_meta_pages = info->pages - info->image_pages - 1;
+ }
+ return error;
+}
+
+/**
+ * unpack_orig_addresses - copy the elements of @buf[] (1 page) to
+ * the PBEs in the list starting at @pbe
+ */
+
+static inline struct pbe *unpack_orig_addresses(unsigned long *buf,
+ struct pbe *pbe)
+{
+ int j;
+
+ for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
+ pbe->orig_address = buf[j];
+ pbe = pbe->next;
+ }
+ return pbe;
+}
+
+/**
+ * create_image - use metadata contained in the PBE list
+ * pointed to by pagedir_nosave to mark the pages that will
+ * be overwritten in the process of restoring the system
+ * memory state from the image and allocate memory for
+ * the image avoiding these pages
+ */
+
+static int create_image(struct snapshot_handle *handle)
+{
+ int error = 0;
+ struct pbe *p, *pblist;
+
+ p = pagedir_nosave;
+ error = mark_unsafe_pages(p);
+ if (!error) {
+ pblist = alloc_pagedir(nr_copy_pages, GFP_ATOMIC, 1);
+ if (pblist)
+ copy_page_backup_list(pblist, p);
+ free_pagedir(p);
+ if (!pblist)
+ error = -ENOMEM;
+ }
+ if (!error)
+ error = alloc_data_pages(pblist, GFP_ATOMIC, 1);
+ if (!error) {
+ release_eaten_pages();
+ pagedir_nosave = pblist;
+ } else {
+ pagedir_nosave = NULL;
+ handle->pbe = NULL;
+ nr_copy_pages = 0;
+ nr_meta_pages = 0;
+ }
+ return error;
+}
+
+/**
+ * snapshot_write_next - used for writing the system memory snapshot.
+ *
+ * On the first call to it @handle should point to a zeroed
+ * snapshot_handle structure. The structure gets updated and a pointer
+ * to it should be passed to this function every next time.
+ *
+ * The @count parameter should contain the number of bytes the caller
+ * wants to write to the image. It must not be zero.
+ *
+ * On success the function returns a positive number. Then, the caller
+ * is allowed to write up to the returned number of bytes to the memory
+ * location computed by the data_of() macro. The number returned
+ * may be smaller than @count, but this only happens if the write would
+ * cross a page boundary otherwise.
+ *
+ * The function returns 0 to indicate the "end of file" condition,
+ * and a negative number is returned on error. In such cases the
+ * structure pointed to by @handle is not updated and should not be used
+ * any more.
+ */
+
+int snapshot_write_next(struct snapshot_handle *handle, size_t count)
+{
+ int error = 0;
+
+ if (handle->prev && handle->page > nr_meta_pages + nr_copy_pages)
+ return 0;
+ if (!buffer) {
+ /* This makes the buffer be freed by swsusp_free() */
+ buffer = alloc_image_page(GFP_ATOMIC, 0);
+ if (!buffer)
+ return -ENOMEM;
+ }
+ if (!handle->offset)
+ handle->buffer = buffer;
+ if (handle->prev < handle->page) {
+ if (!handle->prev) {
+ error = load_header(handle, (struct swsusp_info *)buffer);
+ if (error)
+ return error;
+ } else if (handle->prev <= nr_meta_pages) {
+ handle->pbe = unpack_orig_addresses(buffer, handle->pbe);
+ if (!handle->pbe) {
+ error = create_image(handle);
+ if (error)
+ return error;
+ handle->pbe = pagedir_nosave;
+ handle->buffer = (void *)handle->pbe->address;
+ }
+ } else {
+ handle->pbe = handle->pbe->next;
+ handle->buffer = (void *)handle->pbe->address;
+ }
+ handle->prev = handle->page;
+ }
+ handle->buf_offset = handle->page_offset;
+ if (handle->page_offset + count >= PAGE_SIZE) {
+ count = PAGE_SIZE - handle->page_offset;
+ handle->page_offset = 0;
+ handle->page++;
+ } else {
+ handle->page_offset += count;
+ }
+ handle->offset += count;
+ return count;
+}
+
+int snapshot_image_loaded(struct snapshot_handle *handle)
+{
+ return !(!handle->pbe || handle->pbe->next || !nr_copy_pages ||
+ handle->page <= nr_meta_pages + nr_copy_pages);
+}
diff --git a/kernel/power/swap.c b/kernel/power/swap.c
new file mode 100644
index 000000000000..9177f3f73a6c
--- /dev/null
+++ b/kernel/power/swap.c
@@ -0,0 +1,544 @@
+/*
+ * linux/kernel/power/swap.c
+ *
+ * This file provides functions for reading the suspend image from
+ * and writing it to a swap partition.
+ *
+ * Copyright (C) 1998,2001-2005 Pavel Machek <pavel@suse.cz>
+ * Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
+ *
+ * This file is released under the GPLv2.
+ *
+ */
+
+#include <linux/module.h>
+#include <linux/smp_lock.h>
+#include <linux/file.h>
+#include <linux/utsname.h>
+#include <linux/version.h>
+#include <linux/delay.h>
+#include <linux/bitops.h>
+#include <linux/genhd.h>
+#include <linux/device.h>
+#include <linux/buffer_head.h>
+#include <linux/bio.h>
+#include <linux/swap.h>
+#include <linux/swapops.h>
+#include <linux/pm.h>
+
+#include "power.h"
+
+extern char resume_file[];
+
+#define SWSUSP_SIG "S1SUSPEND"
+
+static struct swsusp_header {
+ char reserved[PAGE_SIZE - 20 - sizeof(swp_entry_t)];
+ swp_entry_t image;
+ char orig_sig[10];
+ char sig[10];
+} __attribute__((packed, aligned(PAGE_SIZE))) swsusp_header;
+
+/*
+ * Saving part...
+ */
+
+static unsigned short root_swap = 0xffff;
+
+static int mark_swapfiles(swp_entry_t start)
+{
+ int error;
+
+ rw_swap_page_sync(READ,
+ swp_entry(root_swap, 0),
+ virt_to_page((unsigned long)&swsusp_header));
+ if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) ||
+ !memcmp("SWAPSPACE2",swsusp_header.sig, 10)) {
+ memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10);
+ memcpy(swsusp_header.sig,SWSUSP_SIG, 10);
+ swsusp_header.image = start;
+ error = rw_swap_page_sync(WRITE,
+ swp_entry(root_swap, 0),
+ virt_to_page((unsigned long)
+ &swsusp_header));
+ } else {
+ pr_debug("swsusp: Partition is not swap space.\n");
+ error = -ENODEV;
+ }
+ return error;
+}
+
+/**
+ * swsusp_swap_check - check if the resume device is a swap device
+ * and get its index (if so)
+ */
+
+static int swsusp_swap_check(void) /* This is called before saving image */
+{
+ int res = swap_type_of(swsusp_resume_device);
+
+ if (res >= 0) {
+ root_swap = res;
+ return 0;
+ }
+ return res;
+}
+
+/**
+ * write_page - Write one page to given swap location.
+ * @buf: Address we're writing.
+ * @offset: Offset of the swap page we're writing to.
+ */
+
+static int write_page(void *buf, unsigned long offset)
+{
+ swp_entry_t entry;
+ int error = -ENOSPC;
+
+ if (offset) {
+ entry = swp_entry(root_swap, offset);
+ error = rw_swap_page_sync(WRITE, entry, virt_to_page(buf));
+ }
+ return error;
+}
+
+/*
+ * The swap map is a data structure used for keeping track of each page
+ * written to a swap partition. It consists of many swap_map_page
+ * structures that contain each an array of MAP_PAGE_SIZE swap entries.
+ * These structures are stored on the swap and linked together with the
+ * help of the .next_swap member.
+ *
+ * The swap map is created during suspend. The swap map pages are
+ * allocated and populated one at a time, so we only need one memory
+ * page to set up the entire structure.
+ *
+ * During resume we also only need to use one swap_map_page structure
+ * at a time.
+ */
+
+#define MAP_PAGE_ENTRIES (PAGE_SIZE / sizeof(long) - 1)
+
+struct swap_map_page {
+ unsigned long entries[MAP_PAGE_ENTRIES];
+ unsigned long next_swap;
+};
+
+/**
+ * The swap_map_handle structure is used for handling swap in
+ * a file-alike way
+ */
+
+struct swap_map_handle {
+ struct swap_map_page *cur;
+ unsigned long cur_swap;
+ struct bitmap_page *bitmap;
+ unsigned int k;
+};
+
+static void release_swap_writer(struct swap_map_handle *handle)
+{
+ if (handle->cur)
+ free_page((unsigned long)handle->cur);
+ handle->cur = NULL;
+ if (handle->bitmap)
+ free_bitmap(handle->bitmap);
+ handle->bitmap = NULL;
+}
+
+static int get_swap_writer(struct swap_map_handle *handle)
+{
+ handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_KERNEL);
+ if (!handle->cur)
+ return -ENOMEM;
+ handle->bitmap = alloc_bitmap(count_swap_pages(root_swap, 0));
+ if (!handle->bitmap) {
+ release_swap_writer(handle);
+ return -ENOMEM;
+ }
+ handle->cur_swap = alloc_swap_page(root_swap, handle->bitmap);
+ if (!handle->cur_swap) {
+ release_swap_writer(handle);
+ return -ENOSPC;
+ }
+ handle->k = 0;
+ return 0;
+}
+
+static int swap_write_page(struct swap_map_handle *handle, void *buf)
+{
+ int error;
+ unsigned long offset;
+
+ if (!handle->cur)
+ return -EINVAL;
+ offset = alloc_swap_page(root_swap, handle->bitmap);
+ error = write_page(buf, offset);
+ if (error)
+ return error;
+ handle->cur->entries[handle->k++] = offset;
+ if (handle->k >= MAP_PAGE_ENTRIES) {
+ offset = alloc_swap_page(root_swap, handle->bitmap);
+ if (!offset)
+ return -ENOSPC;
+ handle->cur->next_swap = offset;
+ error = write_page(handle->cur, handle->cur_swap);
+ if (error)
+ return error;
+ memset(handle->cur, 0, PAGE_SIZE);
+ handle->cur_swap = offset;
+ handle->k = 0;
+ }
+ return 0;
+}
+
+static int flush_swap_writer(struct swap_map_handle *handle)
+{
+ if (handle->cur && handle->cur_swap)
+ return write_page(handle->cur, handle->cur_swap);
+ else
+ return -EINVAL;
+}
+
+/**
+ * save_image - save the suspend image data
+ */
+
+static int save_image(struct swap_map_handle *handle,
+ struct snapshot_handle *snapshot,
+ unsigned int nr_pages)
+{
+ unsigned int m;
+ int ret;
+ int error = 0;
+
+ printk("Saving image data pages (%u pages) ... ", nr_pages);
+ m = nr_pages / 100;
+ if (!m)
+ m = 1;
+ nr_pages = 0;
+ do {
+ ret = snapshot_read_next(snapshot, PAGE_SIZE);
+ if (ret > 0) {
+ error = swap_write_page(handle, data_of(*snapshot));
+ if (error)
+ break;
+ if (!(nr_pages % m))
+ printk("\b\b\b\b%3d%%", nr_pages / m);
+ nr_pages++;
+ }
+ } while (ret > 0);
+ if (!error)
+ printk("\b\b\b\bdone\n");
+ return error;
+}
+
+/**
+ * enough_swap - Make sure we have enough swap to save the image.
+ *
+ * Returns TRUE or FALSE after checking the total amount of swap
+ * space avaiable from the resume partition.
+ */
+
+static int enough_swap(unsigned int nr_pages)
+{
+ unsigned int free_swap = count_swap_pages(root_swap, 1);
+
+ pr_debug("swsusp: free swap pages: %u\n", free_swap);
+ return free_swap > (nr_pages + PAGES_FOR_IO +
+ (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
+}
+
+/**
+ * swsusp_write - Write entire image and metadata.
+ *
+ * It is important _NOT_ to umount filesystems at this point. We want
+ * them synced (in case something goes wrong) but we DO not want to mark
+ * filesystem clean: it is not. (And it does not matter, if we resume
+ * correctly, we'll mark system clean, anyway.)
+ */
+
+int swsusp_write(void)
+{
+ struct swap_map_handle handle;
+ struct snapshot_handle snapshot;
+ struct swsusp_info *header;
+ unsigned long start;
+ int error;
+
+ if ((error = swsusp_swap_check())) {
+ printk(KERN_ERR "swsusp: Cannot find swap device, try swapon -a.\n");
+ return error;
+ }
+ memset(&snapshot, 0, sizeof(struct snapshot_handle));
+ error = snapshot_read_next(&snapshot, PAGE_SIZE);
+ if (error < PAGE_SIZE)
+ return error < 0 ? error : -EFAULT;
+ header = (struct swsusp_info *)data_of(snapshot);
+ if (!enough_swap(header->pages)) {
+ printk(KERN_ERR "swsusp: Not enough free swap\n");
+ return -ENOSPC;
+ }
+ error = get_swap_writer(&handle);
+ if (!error) {
+ start = handle.cur_swap;
+ error = swap_write_page(&handle, header);
+ }
+ if (!error)
+ error = save_image(&handle, &snapshot, header->pages - 1);
+ if (!error) {
+ flush_swap_writer(&handle);
+ printk("S");
+ error = mark_swapfiles(swp_entry(root_swap, start));
+ printk("|\n");
+ }
+ if (error)
+ free_all_swap_pages(root_swap, handle.bitmap);
+ release_swap_writer(&handle);
+ return error;
+}
+
+/*
+ * Using bio to read from swap.
+ * This code requires a bit more work than just using buffer heads
+ * but, it is the recommended way for 2.5/2.6.
+ * The following are to signal the beginning and end of I/O. Bios
+ * finish asynchronously, while we want them to happen synchronously.
+ * A simple atomic_t, and a wait loop take care of this problem.
+ */
+
+static atomic_t io_done = ATOMIC_INIT(0);
+
+static int end_io(struct bio *bio, unsigned int num, int err)
+{
+ if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
+ panic("I/O error reading memory image");
+ atomic_set(&io_done, 0);
+ return 0;
+}
+
+static struct block_device *resume_bdev;
+
+/**
+ * submit - submit BIO request.
+ * @rw: READ or WRITE.
+ * @off physical offset of page.
+ * @page: page we're reading or writing.
+ *
+ * Straight from the textbook - allocate and initialize the bio.
+ * If we're writing, make sure the page is marked as dirty.
+ * Then submit it and wait.
+ */
+
+static int submit(int rw, pgoff_t page_off, void *page)
+{
+ int error = 0;
+ struct bio *bio;
+
+ bio = bio_alloc(GFP_ATOMIC, 1);
+ if (!bio)
+ return -ENOMEM;
+ bio->bi_sector = page_off * (PAGE_SIZE >> 9);
+ bio->bi_bdev = resume_bdev;
+ bio->bi_end_io = end_io;
+
+ if (bio_add_page(bio, virt_to_page(page), PAGE_SIZE, 0) < PAGE_SIZE) {
+ printk("swsusp: ERROR: adding page to bio at %ld\n",page_off);
+ error = -EFAULT;
+ goto Done;
+ }
+
+ atomic_set(&io_done, 1);
+ submit_bio(rw | (1 << BIO_RW_SYNC), bio);
+ while (atomic_read(&io_done))
+ yield();
+ if (rw == READ)
+ bio_set_pages_dirty(bio);
+ Done:
+ bio_put(bio);
+ return error;
+}
+
+static int bio_read_page(pgoff_t page_off, void *page)
+{
+ return submit(READ, page_off, page);
+}
+
+static int bio_write_page(pgoff_t page_off, void *page)
+{
+ return submit(WRITE, page_off, page);
+}
+
+/**
+ * The following functions allow us to read data using a swap map
+ * in a file-alike way
+ */
+
+static void release_swap_reader(struct swap_map_handle *handle)
+{
+ if (handle->cur)
+ free_page((unsigned long)handle->cur);
+ handle->cur = NULL;
+}
+
+static int get_swap_reader(struct swap_map_handle *handle,
+ swp_entry_t start)
+{
+ int error;
+
+ if (!swp_offset(start))
+ return -EINVAL;
+ handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
+ if (!handle->cur)
+ return -ENOMEM;
+ error = bio_read_page(swp_offset(start), handle->cur);
+ if (error) {
+ release_swap_reader(handle);
+ return error;
+ }
+ handle->k = 0;
+ return 0;
+}
+
+static int swap_read_page(struct swap_map_handle *handle, void *buf)
+{
+ unsigned long offset;
+ int error;
+
+ if (!handle->cur)
+ return -EINVAL;
+ offset = handle->cur->entries[handle->k];
+ if (!offset)
+ return -EFAULT;
+ error = bio_read_page(offset, buf);
+ if (error)
+ return error;
+ if (++handle->k >= MAP_PAGE_ENTRIES) {
+ handle->k = 0;
+ offset = handle->cur->next_swap;
+ if (!offset)
+ release_swap_reader(handle);
+ else
+ error = bio_read_page(offset, handle->cur);
+ }
+ return error;
+}
+
+/**
+ * load_image - load the image using the swap map handle
+ * @handle and the snapshot handle @snapshot
+ * (assume there are @nr_pages pages to load)
+ */
+
+static int load_image(struct swap_map_handle *handle,
+ struct snapshot_handle *snapshot,
+ unsigned int nr_pages)
+{
+ unsigned int m;
+ int ret;
+ int error = 0;
+
+ printk("Loading image data pages (%u pages) ... ", nr_pages);
+ m = nr_pages / 100;
+ if (!m)
+ m = 1;
+ nr_pages = 0;
+ do {
+ ret = snapshot_write_next(snapshot, PAGE_SIZE);
+ if (ret > 0) {
+ error = swap_read_page(handle, data_of(*snapshot));
+ if (error)
+ break;
+ if (!(nr_pages % m))
+ printk("\b\b\b\b%3d%%", nr_pages / m);
+ nr_pages++;
+ }
+ } while (ret > 0);
+ if (!error)
+ printk("\b\b\b\bdone\n");
+ if (!snapshot_image_loaded(snapshot))
+ error = -ENODATA;
+ return error;
+}
+
+int swsusp_read(void)
+{
+ int error;
+ struct swap_map_handle handle;
+ struct snapshot_handle snapshot;
+ struct swsusp_info *header;
+
+ if (IS_ERR(resume_bdev)) {
+ pr_debug("swsusp: block device not initialised\n");
+ return PTR_ERR(resume_bdev);
+ }
+
+ memset(&snapshot, 0, sizeof(struct snapshot_handle));
+ error = snapshot_write_next(&snapshot, PAGE_SIZE);
+ if (error < PAGE_SIZE)
+ return error < 0 ? error : -EFAULT;
+ header = (struct swsusp_info *)data_of(snapshot);
+ error = get_swap_reader(&handle, swsusp_header.image);
+ if (!error)
+ error = swap_read_page(&handle, header);
+ if (!error)
+ error = load_image(&handle, &snapshot, header->pages - 1);
+ release_swap_reader(&handle);
+
+ blkdev_put(resume_bdev);
+
+ if (!error)
+ pr_debug("swsusp: Reading resume file was successful\n");
+ else
+ pr_debug("swsusp: Error %d resuming\n", error);
+ return error;
+}
+
+/**
+ * swsusp_check - Check for swsusp signature in the resume device
+ */
+
+int swsusp_check(void)
+{
+ int error;
+
+ resume_bdev = open_by_devnum(swsusp_resume_device, FMODE_READ);
+ if (!IS_ERR(resume_bdev)) {
+ set_blocksize(resume_bdev, PAGE_SIZE);
+ memset(&swsusp_header, 0, sizeof(swsusp_header));
+ if ((error = bio_read_page(0, &swsusp_header)))
+ return error;
+ if (!memcmp(SWSUSP_SIG, swsusp_header.sig, 10)) {
+ memcpy(swsusp_header.sig, swsusp_header.orig_sig, 10);
+ /* Reset swap signature now */
+ error = bio_write_page(0, &swsusp_header);
+ } else {
+ return -EINVAL;
+ }
+ if (error)
+ blkdev_put(resume_bdev);
+ else
+ pr_debug("swsusp: Signature found, resuming\n");
+ } else {
+ error = PTR_ERR(resume_bdev);
+ }
+
+ if (error)
+ pr_debug("swsusp: Error %d check for resume file\n", error);
+
+ return error;
+}
+
+/**
+ * swsusp_close - close swap device.
+ */
+
+void swsusp_close(void)
+{
+ if (IS_ERR(resume_bdev)) {
+ pr_debug("swsusp: block device not initialised\n");
+ return;
+ }
+
+ blkdev_put(resume_bdev);
+}
diff --git a/kernel/power/swsusp.c b/kernel/power/swsusp.c
index 2d9d08f72f76..c4016cbbd3e0 100644
--- a/kernel/power/swsusp.c
+++ b/kernel/power/swsusp.c
@@ -31,41 +31,24 @@
* Fixed runaway init
*
* Rafael J. Wysocki <rjw@sisk.pl>
- * Added the swap map data structure and reworked the handling of swap
+ * Reworked the freeing of memory and the handling of swap
*
* More state savers are welcome. Especially for the scsi layer...
*
* For TODOs,FIXMEs also look in Documentation/power/swsusp.txt
*/
-#include <linux/module.h>
#include <linux/mm.h>
#include <linux/suspend.h>
-#include <linux/smp_lock.h>
-#include <linux/file.h>
-#include <linux/utsname.h>
-#include <linux/version.h>
-#include <linux/delay.h>
-#include <linux/bitops.h>
#include <linux/spinlock.h>
-#include <linux/genhd.h>
#include <linux/kernel.h>
#include <linux/major.h>
#include <linux/swap.h>
#include <linux/pm.h>
-#include <linux/device.h>
-#include <linux/buffer_head.h>
#include <linux/swapops.h>
#include <linux/bootmem.h>
#include <linux/syscalls.h>
#include <linux/highmem.h>
-#include <linux/bio.h>
-
-#include <asm/uaccess.h>
-#include <asm/mmu_context.h>
-#include <asm/pgtable.h>
-#include <asm/tlbflush.h>
-#include <asm/io.h>
#include "power.h"
@@ -77,6 +60,8 @@
*/
unsigned long image_size = 500 * 1024 * 1024;
+int in_suspend __nosavedata = 0;
+
#ifdef CONFIG_HIGHMEM
unsigned int count_highmem_pages(void);
int save_highmem(void);
@@ -87,471 +72,97 @@ static int restore_highmem(void) { return 0; }
static unsigned int count_highmem_pages(void) { return 0; }
#endif
-extern char resume_file[];
-
-#define SWSUSP_SIG "S1SUSPEND"
-
-static struct swsusp_header {
- char reserved[PAGE_SIZE - 20 - sizeof(swp_entry_t)];
- swp_entry_t image;
- char orig_sig[10];
- char sig[10];
-} __attribute__((packed, aligned(PAGE_SIZE))) swsusp_header;
-
-static struct swsusp_info swsusp_info;
-
-/*
- * Saving part...
- */
-
-static unsigned short root_swap = 0xffff;
-
-static int mark_swapfiles(swp_entry_t start)
-{
- int error;
-
- rw_swap_page_sync(READ,
- swp_entry(root_swap, 0),
- virt_to_page((unsigned long)&swsusp_header));
- if (!memcmp("SWAP-SPACE",swsusp_header.sig, 10) ||
- !memcmp("SWAPSPACE2",swsusp_header.sig, 10)) {
- memcpy(swsusp_header.orig_sig,swsusp_header.sig, 10);
- memcpy(swsusp_header.sig,SWSUSP_SIG, 10);
- swsusp_header.image = start;
- error = rw_swap_page_sync(WRITE,
- swp_entry(root_swap, 0),
- virt_to_page((unsigned long)
- &swsusp_header));
- } else {
- pr_debug("swsusp: Partition is not swap space.\n");
- error = -ENODEV;
- }
- return error;
-}
-
-/*
- * Check whether the swap device is the specified resume
- * device, irrespective of whether they are specified by
- * identical names.
- *
- * (Thus, device inode aliasing is allowed. You can say /dev/hda4
- * instead of /dev/ide/host0/bus0/target0/lun0/part4 [if using devfs]
- * and they'll be considered the same device. This is *necessary* for
- * devfs, since the resume code can only recognize the form /dev/hda4,
- * but the suspend code would see the long name.)
- */
-static inline int is_resume_device(const struct swap_info_struct *swap_info)
-{
- struct file *file = swap_info->swap_file;
- struct inode *inode = file->f_dentry->d_inode;
-
- return S_ISBLK(inode->i_mode) &&
- swsusp_resume_device == MKDEV(imajor(inode), iminor(inode));
-}
-
-static int swsusp_swap_check(void) /* This is called before saving image */
-{
- int i;
-
- spin_lock(&swap_lock);
- for (i = 0; i < MAX_SWAPFILES; i++) {
- if (!(swap_info[i].flags & SWP_WRITEOK))
- continue;
- if (!swsusp_resume_device || is_resume_device(swap_info + i)) {
- spin_unlock(&swap_lock);
- root_swap = i;
- return 0;
- }
- }
- spin_unlock(&swap_lock);
- return -ENODEV;
-}
-
-/**
- * write_page - Write one page to a fresh swap location.
- * @addr: Address we're writing.
- * @loc: Place to store the entry we used.
- *
- * Allocate a new swap entry and 'sync' it. Note we discard -EIO
- * errors. That is an artifact left over from swsusp. It did not
- * check the return of rw_swap_page_sync() at all, since most pages
- * written back to swap would return -EIO.
- * This is a partial improvement, since we will at least return other
- * errors, though we need to eventually fix the damn code.
- */
-static int write_page(unsigned long addr, swp_entry_t *loc)
-{
- swp_entry_t entry;
- int error = -ENOSPC;
-
- entry = get_swap_page_of_type(root_swap);
- if (swp_offset(entry)) {
- error = rw_swap_page_sync(WRITE, entry, virt_to_page(addr));
- if (!error || error == -EIO)
- *loc = entry;
- }
- return error;
-}
-
/**
- * Swap map-handling functions
- *
- * The swap map is a data structure used for keeping track of each page
- * written to the swap. It consists of many swap_map_page structures
- * that contain each an array of MAP_PAGE_SIZE swap entries.
- * These structures are linked together with the help of either the
- * .next (in memory) or the .next_swap (in swap) member.
+ * The following functions are used for tracing the allocated
+ * swap pages, so that they can be freed in case of an error.
*
- * The swap map is created during suspend. At that time we need to keep
- * it in memory, because we have to free all of the allocated swap
- * entries if an error occurs. The memory needed is preallocated
- * so that we know in advance if there's enough of it.
- *
- * The first swap_map_page structure is filled with the swap entries that
- * correspond to the first MAP_PAGE_SIZE data pages written to swap and
- * so on. After the all of the data pages have been written, the order
- * of the swap_map_page structures in the map is reversed so that they
- * can be read from swap in the original order. This causes the data
- * pages to be loaded in exactly the same order in which they have been
- * saved.
- *
- * During resume we only need to use one swap_map_page structure
- * at a time, which means that we only need to use two memory pages for
- * reading the image - one for reading the swap_map_page structures
- * and the second for reading the data pages from swap.
+ * The functions operate on a linked bitmap structure defined
+ * in power.h
*/
-#define MAP_PAGE_SIZE ((PAGE_SIZE - sizeof(swp_entry_t) - sizeof(void *)) \
- / sizeof(swp_entry_t))
-
-struct swap_map_page {
- swp_entry_t entries[MAP_PAGE_SIZE];
- swp_entry_t next_swap;
- struct swap_map_page *next;
-};
-
-static inline void free_swap_map(struct swap_map_page *swap_map)
+void free_bitmap(struct bitmap_page *bitmap)
{
- struct swap_map_page *swp;
+ struct bitmap_page *bp;
- while (swap_map) {
- swp = swap_map->next;
- free_page((unsigned long)swap_map);
- swap_map = swp;
+ while (bitmap) {
+ bp = bitmap->next;
+ free_page((unsigned long)bitmap);
+ bitmap = bp;
}
}
-static struct swap_map_page *alloc_swap_map(unsigned int nr_pages)
+struct bitmap_page *alloc_bitmap(unsigned int nr_bits)
{
- struct swap_map_page *swap_map, *swp;
- unsigned n = 0;
+ struct bitmap_page *bitmap, *bp;
+ unsigned int n;
- if (!nr_pages)
+ if (!nr_bits)
return NULL;
- pr_debug("alloc_swap_map(): nr_pages = %d\n", nr_pages);
- swap_map = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
- swp = swap_map;
- for (n = MAP_PAGE_SIZE; n < nr_pages; n += MAP_PAGE_SIZE) {
- swp->next = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
- swp = swp->next;
- if (!swp) {
- free_swap_map(swap_map);
+ bitmap = (struct bitmap_page *)get_zeroed_page(GFP_KERNEL);
+ bp = bitmap;
+ for (n = BITMAP_PAGE_BITS; n < nr_bits; n += BITMAP_PAGE_BITS) {
+ bp->next = (struct bitmap_page *)get_zeroed_page(GFP_KERNEL);
+ bp = bp->next;
+ if (!bp) {
+ free_bitmap(bitmap);
return NULL;
}
}
- return swap_map;
+ return bitmap;
}
-/**
- * reverse_swap_map - reverse the order of pages in the swap map
- * @swap_map
- */
-
-static inline struct swap_map_page *reverse_swap_map(struct swap_map_page *swap_map)
-{
- struct swap_map_page *prev, *next;
-
- prev = NULL;
- while (swap_map) {
- next = swap_map->next;
- swap_map->next = prev;
- prev = swap_map;
- swap_map = next;
- }
- return prev;
-}
-
-/**
- * free_swap_map_entries - free the swap entries allocated to store
- * the swap map @swap_map (this is only called in case of an error)
- */
-static inline void free_swap_map_entries(struct swap_map_page *swap_map)
-{
- while (swap_map) {
- if (swap_map->next_swap.val)
- swap_free(swap_map->next_swap);
- swap_map = swap_map->next;
- }
-}
-
-/**
- * save_swap_map - save the swap map used for tracing the data pages
- * stored in the swap
- */
-
-static int save_swap_map(struct swap_map_page *swap_map, swp_entry_t *start)
-{
- swp_entry_t entry = (swp_entry_t){0};
- int error;
-
- while (swap_map) {
- swap_map->next_swap = entry;
- if ((error = write_page((unsigned long)swap_map, &entry)))
- return error;
- swap_map = swap_map->next;
- }
- *start = entry;
- return 0;
-}
-
-/**
- * free_image_entries - free the swap entries allocated to store
- * the image data pages (this is only called in case of an error)
- */
-
-static inline void free_image_entries(struct swap_map_page *swp)
+static int bitmap_set(struct bitmap_page *bitmap, unsigned long bit)
{
- unsigned k;
+ unsigned int n;
- while (swp) {
- for (k = 0; k < MAP_PAGE_SIZE; k++)
- if (swp->entries[k].val)
- swap_free(swp->entries[k]);
- swp = swp->next;
+ n = BITMAP_PAGE_BITS;
+ while (bitmap && n <= bit) {
+ n += BITMAP_PAGE_BITS;
+ bitmap = bitmap->next;
}
-}
-
-/**
- * The swap_map_handle structure is used for handling the swap map in
- * a file-alike way
- */
-
-struct swap_map_handle {
- struct swap_map_page *cur;
- unsigned int k;
-};
-
-static inline void init_swap_map_handle(struct swap_map_handle *handle,
- struct swap_map_page *map)
-{
- handle->cur = map;
- handle->k = 0;
-}
-
-static inline int swap_map_write_page(struct swap_map_handle *handle,
- unsigned long addr)
-{
- int error;
-
- error = write_page(addr, handle->cur->entries + handle->k);
- if (error)
- return error;
- if (++handle->k >= MAP_PAGE_SIZE) {
- handle->cur = handle->cur->next;
- handle->k = 0;
+ if (!bitmap)
+ return -EINVAL;
+ n -= BITMAP_PAGE_BITS;
+ bit -= n;
+ n = 0;
+ while (bit >= BITS_PER_CHUNK) {
+ bit -= BITS_PER_CHUNK;
+ n++;
}
+ bitmap->chunks[n] |= (1UL << bit);
return 0;
}
-/**
- * save_image_data - save the data pages pointed to by the PBEs
- * from the list @pblist using the swap map handle @handle
- * (assume there are @nr_pages data pages to save)
- */
-
-static int save_image_data(struct pbe *pblist,
- struct swap_map_handle *handle,
- unsigned int nr_pages)
-{
- unsigned int m;
- struct pbe *p;
- int error = 0;
-
- printk("Saving image data pages (%u pages) ... ", nr_pages);
- m = nr_pages / 100;
- if (!m)
- m = 1;
- nr_pages = 0;
- for_each_pbe (p, pblist) {
- error = swap_map_write_page(handle, p->address);
- if (error)
- break;
- if (!(nr_pages % m))
- printk("\b\b\b\b%3d%%", nr_pages / m);
- nr_pages++;
- }
- if (!error)
- printk("\b\b\b\bdone\n");
- return error;
-}
-
-static void dump_info(void)
-{
- pr_debug(" swsusp: Version: %u\n",swsusp_info.version_code);
- pr_debug(" swsusp: Num Pages: %ld\n",swsusp_info.num_physpages);
- pr_debug(" swsusp: UTS Sys: %s\n",swsusp_info.uts.sysname);
- pr_debug(" swsusp: UTS Node: %s\n",swsusp_info.uts.nodename);
- pr_debug(" swsusp: UTS Release: %s\n",swsusp_info.uts.release);
- pr_debug(" swsusp: UTS Version: %s\n",swsusp_info.uts.version);
- pr_debug(" swsusp: UTS Machine: %s\n",swsusp_info.uts.machine);
- pr_debug(" swsusp: UTS Domain: %s\n",swsusp_info.uts.domainname);
- pr_debug(" swsusp: CPUs: %d\n",swsusp_info.cpus);
- pr_debug(" swsusp: Image: %ld Pages\n",swsusp_info.image_pages);
- pr_debug(" swsusp: Total: %ld Pages\n", swsusp_info.pages);
-}
-
-static void init_header(unsigned int nr_pages)
-{
- memset(&swsusp_info, 0, sizeof(swsusp_info));
- swsusp_info.version_code = LINUX_VERSION_CODE;
- swsusp_info.num_physpages = num_physpages;
- memcpy(&swsusp_info.uts, &system_utsname, sizeof(system_utsname));
-
- swsusp_info.cpus = num_online_cpus();
- swsusp_info.image_pages = nr_pages;
- swsusp_info.pages = nr_pages +
- ((nr_pages * sizeof(long) + PAGE_SIZE - 1) >> PAGE_SHIFT) + 1;
-}
-
-/**
- * pack_orig_addresses - the .orig_address fields of the PBEs from the
- * list starting at @pbe are stored in the array @buf[] (1 page)
- */
-
-static inline struct pbe *pack_orig_addresses(unsigned long *buf,
- struct pbe *pbe)
-{
- int j;
-
- for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
- buf[j] = pbe->orig_address;
- pbe = pbe->next;
- }
- if (!pbe)
- for (; j < PAGE_SIZE / sizeof(long); j++)
- buf[j] = 0;
- return pbe;
-}
-
-/**
- * save_image_metadata - save the .orig_address fields of the PBEs
- * from the list @pblist using the swap map handle @handle
- */
-
-static int save_image_metadata(struct pbe *pblist,
- struct swap_map_handle *handle)
+unsigned long alloc_swap_page(int swap, struct bitmap_page *bitmap)
{
- unsigned long *buf;
- unsigned int n = 0;
- struct pbe *p;
- int error = 0;
+ unsigned long offset;
- printk("Saving image metadata ... ");
- buf = (unsigned long *)get_zeroed_page(GFP_ATOMIC);
- if (!buf)
- return -ENOMEM;
- p = pblist;
- while (p) {
- p = pack_orig_addresses(buf, p);
- error = swap_map_write_page(handle, (unsigned long)buf);
- if (error)
- break;
- n++;
+ offset = swp_offset(get_swap_page_of_type(swap));
+ if (offset) {
+ if (bitmap_set(bitmap, offset)) {
+ swap_free(swp_entry(swap, offset));
+ offset = 0;
+ }
}
- free_page((unsigned long)buf);
- if (!error)
- printk("done (%u pages saved)\n", n);
- return error;
+ return offset;
}
-/**
- * enough_swap - Make sure we have enough swap to save the image.
- *
- * Returns TRUE or FALSE after checking the total amount of swap
- * space avaiable from the resume partition.
- */
-
-static int enough_swap(unsigned int nr_pages)
+void free_all_swap_pages(int swap, struct bitmap_page *bitmap)
{
- unsigned int free_swap = swap_info[root_swap].pages -
- swap_info[root_swap].inuse_pages;
-
- pr_debug("swsusp: free swap pages: %u\n", free_swap);
- return free_swap > (nr_pages + PAGES_FOR_IO +
- (nr_pages + PBES_PER_PAGE - 1) / PBES_PER_PAGE);
-}
+ unsigned int bit, n;
+ unsigned long test;
-/**
- * swsusp_write - Write entire image and metadata.
- *
- * It is important _NOT_ to umount filesystems at this point. We want
- * them synced (in case something goes wrong) but we DO not want to mark
- * filesystem clean: it is not. (And it does not matter, if we resume
- * correctly, we'll mark system clean, anyway.)
- */
-
-int swsusp_write(struct pbe *pblist, unsigned int nr_pages)
-{
- struct swap_map_page *swap_map;
- struct swap_map_handle handle;
- swp_entry_t start;
- int error;
-
- if ((error = swsusp_swap_check())) {
- printk(KERN_ERR "swsusp: Cannot find swap device, try swapon -a.\n");
- return error;
- }
- if (!enough_swap(nr_pages)) {
- printk(KERN_ERR "swsusp: Not enough free swap\n");
- return -ENOSPC;
+ bit = 0;
+ while (bitmap) {
+ for (n = 0; n < BITMAP_PAGE_CHUNKS; n++)
+ for (test = 1UL; test; test <<= 1) {
+ if (bitmap->chunks[n] & test)
+ swap_free(swp_entry(swap, bit));
+ bit++;
+ }
+ bitmap = bitmap->next;
}
-
- init_header(nr_pages);
- swap_map = alloc_swap_map(swsusp_info.pages);
- if (!swap_map)
- return -ENOMEM;
- init_swap_map_handle(&handle, swap_map);
-
- error = swap_map_write_page(&handle, (unsigned long)&swsusp_info);
- if (!error)
- error = save_image_metadata(pblist, &handle);
- if (!error)
- error = save_image_data(pblist, &handle, nr_pages);
- if (error)
- goto Free_image_entries;
-
- swap_map = reverse_swap_map(swap_map);
- error = save_swap_map(swap_map, &start);
- if (error)
- goto Free_map_entries;
-
- dump_info();
- printk( "S" );
- error = mark_swapfiles(start);
- printk( "|\n" );
- if (error)
- goto Free_map_entries;
-
-Free_swap_map:
- free_swap_map(swap_map);
- return error;
-
-Free_map_entries:
- free_swap_map_entries(swap_map);
-Free_image_entries:
- free_image_entries(swap_map);
- goto Free_swap_map;
}
/**
@@ -660,379 +271,3 @@ int swsusp_resume(void)
local_irq_enable();
return error;
}
-
-/**
- * mark_unsafe_pages - mark the pages that cannot be used for storing
- * the image during resume, because they conflict with the pages that
- * had been used before suspend
- */
-
-static void mark_unsafe_pages(struct pbe *pblist)
-{
- struct zone *zone;
- unsigned long zone_pfn;
- struct pbe *p;
-
- if (!pblist) /* a sanity check */
- return;
-
- /* Clear page flags */
- for_each_zone (zone) {
- for (zone_pfn = 0; zone_pfn < zone->spanned_pages; ++zone_pfn)
- if (pfn_valid(zone_pfn + zone->zone_start_pfn))
- ClearPageNosaveFree(pfn_to_page(zone_pfn +
- zone->zone_start_pfn));
- }
-
- /* Mark orig addresses */
- for_each_pbe (p, pblist)
- SetPageNosaveFree(virt_to_page(p->orig_address));
-
-}
-
-static void copy_page_backup_list(struct pbe *dst, struct pbe *src)
-{
- /* We assume both lists contain the same number of elements */
- while (src) {
- dst->orig_address = src->orig_address;
- dst = dst->next;
- src = src->next;
- }
-}
-
-/*
- * Using bio to read from swap.
- * This code requires a bit more work than just using buffer heads
- * but, it is the recommended way for 2.5/2.6.
- * The following are to signal the beginning and end of I/O. Bios
- * finish asynchronously, while we want them to happen synchronously.
- * A simple atomic_t, and a wait loop take care of this problem.
- */
-
-static atomic_t io_done = ATOMIC_INIT(0);
-
-static int end_io(struct bio *bio, unsigned int num, int err)
-{
- if (!test_bit(BIO_UPTODATE, &bio->bi_flags))
- panic("I/O error reading memory image");
- atomic_set(&io_done, 0);
- return 0;
-}
-
-static struct block_device *resume_bdev;
-
-/**
- * submit - submit BIO request.
- * @rw: READ or WRITE.
- * @off physical offset of page.
- * @page: page we're reading or writing.
- *
- * Straight from the textbook - allocate and initialize the bio.
- * If we're writing, make sure the page is marked as dirty.
- * Then submit it and wait.
- */
-
-static int submit(int rw, pgoff_t page_off, void *page)
-{
- int error = 0;
- struct bio *bio;
-
- bio = bio_alloc(GFP_ATOMIC, 1);
- if (!bio)
- return -ENOMEM;
- bio->bi_sector = page_off * (PAGE_SIZE >> 9);
- bio->bi_bdev = resume_bdev;
- bio->bi_end_io = end_io;
-
- if (bio_add_page(bio, virt_to_page(page), PAGE_SIZE, 0) < PAGE_SIZE) {
- printk("swsusp: ERROR: adding page to bio at %ld\n",page_off);
- error = -EFAULT;
- goto Done;
- }
-
-
- atomic_set(&io_done, 1);
- submit_bio(rw | (1 << BIO_RW_SYNC), bio);
- while (atomic_read(&io_done))
- yield();
- if (rw == READ)
- bio_set_pages_dirty(bio);
- Done:
- bio_put(bio);
- return error;
-}
-
-static int bio_read_page(pgoff_t page_off, void *page)
-{
- return submit(READ, page_off, page);
-}
-
-static int bio_write_page(pgoff_t page_off, void *page)
-{
- return submit(WRITE, page_off, page);
-}
-
-/**
- * The following functions allow us to read data using a swap map
- * in a file-alike way
- */
-
-static inline void release_swap_map_reader(struct swap_map_handle *handle)
-{
- if (handle->cur)
- free_page((unsigned long)handle->cur);
- handle->cur = NULL;
-}
-
-static inline int get_swap_map_reader(struct swap_map_handle *handle,
- swp_entry_t start)
-{
- int error;
-
- if (!swp_offset(start))
- return -EINVAL;
- handle->cur = (struct swap_map_page *)get_zeroed_page(GFP_ATOMIC);
- if (!handle->cur)
- return -ENOMEM;
- error = bio_read_page(swp_offset(start), handle->cur);
- if (error) {
- release_swap_map_reader(handle);
- return error;
- }
- handle->k = 0;
- return 0;
-}
-
-static inline int swap_map_read_page(struct swap_map_handle *handle, void *buf)
-{
- unsigned long offset;
- int error;
-
- if (!handle->cur)
- return -EINVAL;
- offset = swp_offset(handle->cur->entries[handle->k]);
- if (!offset)
- return -EINVAL;
- error = bio_read_page(offset, buf);
- if (error)
- return error;
- if (++handle->k >= MAP_PAGE_SIZE) {
- handle->k = 0;
- offset = swp_offset(handle->cur->next_swap);
- if (!offset)
- release_swap_map_reader(handle);
- else
- error = bio_read_page(offset, handle->cur);
- }
- return error;
-}
-
-static int check_header(void)
-{
- char *reason = NULL;
-
- dump_info();
- if (swsusp_info.version_code != LINUX_VERSION_CODE)
- reason = "kernel version";
- if (swsusp_info.num_physpages != num_physpages)
- reason = "memory size";
- if (strcmp(swsusp_info.uts.sysname,system_utsname.sysname))
- reason = "system type";
- if (strcmp(swsusp_info.uts.release,system_utsname.release))
- reason = "kernel release";
- if (strcmp(swsusp_info.uts.version,system_utsname.version))
- reason = "version";
- if (strcmp(swsusp_info.uts.machine,system_utsname.machine))
- reason = "machine";
- if (reason) {
- printk(KERN_ERR "swsusp: Resume mismatch: %s\n", reason);
- return -EPERM;
- }
- return 0;
-}
-
-/**
- * load_image_data - load the image data using the swap map handle
- * @handle and store them using the page backup list @pblist
- * (assume there are @nr_pages pages to load)
- */
-
-static int load_image_data(struct pbe *pblist,
- struct swap_map_handle *handle,
- unsigned int nr_pages)
-{
- int error;
- unsigned int m;
- struct pbe *p;
-
- if (!pblist)
- return -EINVAL;
- printk("Loading image data pages (%u pages) ... ", nr_pages);
- m = nr_pages / 100;
- if (!m)
- m = 1;
- nr_pages = 0;
- p = pblist;
- while (p) {
- error = swap_map_read_page(handle, (void *)p->address);
- if (error)
- break;
- p = p->next;
- if (!(nr_pages % m))
- printk("\b\b\b\b%3d%%", nr_pages / m);
- nr_pages++;
- }
- if (!error)
- printk("\b\b\b\bdone\n");
- return error;
-}
-
-/**
- * unpack_orig_addresses - copy the elements of @buf[] (1 page) to
- * the PBEs in the list starting at @pbe
- */
-
-static inline struct pbe *unpack_orig_addresses(unsigned long *buf,
- struct pbe *pbe)
-{
- int j;
-
- for (j = 0; j < PAGE_SIZE / sizeof(long) && pbe; j++) {
- pbe->orig_address = buf[j];
- pbe = pbe->next;
- }
- return pbe;
-}
-
-/**
- * load_image_metadata - load the image metadata using the swap map
- * handle @handle and put them into the PBEs in the list @pblist
- */
-
-static int load_image_metadata(struct pbe *pblist, struct swap_map_handle *handle)
-{
- struct pbe *p;
- unsigned long *buf;
- unsigned int n = 0;
- int error = 0;
-
- printk("Loading image metadata ... ");
- buf = (unsigned long *)get_zeroed_page(GFP_ATOMIC);
- if (!buf)
- return -ENOMEM;
- p = pblist;
- while (p) {
- error = swap_map_read_page(handle, buf);
- if (error)
- break;
- p = unpack_orig_addresses(buf, p);
- n++;
- }
- free_page((unsigned long)buf);
- if (!error)
- printk("done (%u pages loaded)\n", n);
- return error;
-}
-
-int swsusp_read(struct pbe **pblist_ptr)
-{
- int error;
- struct pbe *p, *pblist;
- struct swap_map_handle handle;
- unsigned int nr_pages;
-
- if (IS_ERR(resume_bdev)) {
- pr_debug("swsusp: block device not initialised\n");
- return PTR_ERR(resume_bdev);
- }
-
- error = get_swap_map_reader(&handle, swsusp_header.image);
- if (!error)
- error = swap_map_read_page(&handle, &swsusp_info);
- if (!error)
- error = check_header();
- if (error)
- return error;
- nr_pages = swsusp_info.image_pages;
- p = alloc_pagedir(nr_pages, GFP_ATOMIC, 0);
- if (!p)
- return -ENOMEM;
- error = load_image_metadata(p, &handle);
- if (!error) {
- mark_unsafe_pages(p);
- pblist = alloc_pagedir(nr_pages, GFP_ATOMIC, 1);
- if (pblist)
- copy_page_backup_list(pblist, p);
- free_pagedir(p);
- if (!pblist)
- error = -ENOMEM;
-
- /* Allocate memory for the image and read the data from swap */
- if (!error)
- error = alloc_data_pages(pblist, GFP_ATOMIC, 1);
- if (!error) {
- release_eaten_pages();
- error = load_image_data(pblist, &handle, nr_pages);
- }
- if (!error)
- *pblist_ptr = pblist;
- }
- release_swap_map_reader(&handle);
-
- blkdev_put(resume_bdev);
-
- if (!error)
- pr_debug("swsusp: Reading resume file was successful\n");
- else
- pr_debug("swsusp: Error %d resuming\n", error);
- return error;
-}
-
-/**
- * swsusp_check - Check for swsusp signature in the resume device
- */
-
-int swsusp_check(void)
-{
- int error;
-
- resume_bdev = open_by_devnum(swsusp_resume_device, FMODE_READ);
- if (!IS_ERR(resume_bdev)) {
- set_blocksize(resume_bdev, PAGE_SIZE);
- memset(&swsusp_header, 0, sizeof(swsusp_header));
- if ((error = bio_read_page(0, &swsusp_header)))
- return error;
- if (!memcmp(SWSUSP_SIG, swsusp_header.sig, 10)) {
- memcpy(swsusp_header.sig, swsusp_header.orig_sig, 10);
- /* Reset swap signature now */
- error = bio_write_page(0, &swsusp_header);
- } else {
- return -EINVAL;
- }
- if (error)
- blkdev_put(resume_bdev);
- else
- pr_debug("swsusp: Signature found, resuming\n");
- } else {
- error = PTR_ERR(resume_bdev);
- }
-
- if (error)
- pr_debug("swsusp: Error %d check for resume file\n", error);
-
- return error;
-}
-
-/**
- * swsusp_close - close swap device.
- */
-
-void swsusp_close(void)
-{
- if (IS_ERR(resume_bdev)) {
- pr_debug("swsusp: block device not initialised\n");
- return;
- }
-
- blkdev_put(resume_bdev);
-}
diff --git a/kernel/power/user.c b/kernel/power/user.c
new file mode 100644
index 000000000000..3f1539fbe48a
--- /dev/null
+++ b/kernel/power/user.c
@@ -0,0 +1,333 @@
+/*
+ * linux/kernel/power/user.c
+ *
+ * This file provides the user space interface for software suspend/resume.
+ *
+ * Copyright (C) 2006 Rafael J. Wysocki <rjw@sisk.pl>
+ *
+ * This file is released under the GPLv2.
+ *
+ */
+
+#include <linux/suspend.h>
+#include <linux/syscalls.h>
+#include <linux/string.h>
+#include <linux/device.h>
+#include <linux/miscdevice.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/swapops.h>
+#include <linux/pm.h>
+#include <linux/fs.h>
+
+#include <asm/uaccess.h>
+
+#include "power.h"
+
+#define SNAPSHOT_MINOR 231
+
+static struct snapshot_data {
+ struct snapshot_handle handle;
+ int swap;
+ struct bitmap_page *bitmap;
+ int mode;
+ char frozen;
+ char ready;
+} snapshot_state;
+
+static atomic_t device_available = ATOMIC_INIT(1);
+
+static int snapshot_open(struct inode *inode, struct file *filp)
+{
+ struct snapshot_data *data;
+
+ if (!atomic_add_unless(&device_available, -1, 0))
+ return -EBUSY;
+
+ if ((filp->f_flags & O_ACCMODE) == O_RDWR)
+ return -ENOSYS;
+
+ nonseekable_open(inode, filp);
+ data = &snapshot_state;
+ filp->private_data = data;
+ memset(&data->handle, 0, sizeof(struct snapshot_handle));
+ if ((filp->f_flags & O_ACCMODE) == O_RDONLY) {
+ data->swap = swsusp_resume_device ? swap_type_of(swsusp_resume_device) : -1;
+ data->mode = O_RDONLY;
+ } else {
+ data->swap = -1;
+ data->mode = O_WRONLY;
+ }
+ data->bitmap = NULL;
+ data->frozen = 0;
+ data->ready = 0;
+
+ return 0;
+}
+
+static int snapshot_release(struct inode *inode, struct file *filp)
+{
+ struct snapshot_data *data;
+
+ swsusp_free();
+ data = filp->private_data;
+ free_all_swap_pages(data->swap, data->bitmap);
+ free_bitmap(data->bitmap);
+ if (data->frozen) {
+ down(&pm_sem);
+ thaw_processes();
+ enable_nonboot_cpus();
+ up(&pm_sem);
+ }
+ atomic_inc(&device_available);
+ return 0;
+}
+
+static ssize_t snapshot_read(struct file *filp, char __user *buf,
+ size_t count, loff_t *offp)
+{
+ struct snapshot_data *data;
+ ssize_t res;
+
+ data = filp->private_data;
+ res = snapshot_read_next(&data->handle, count);
+ if (res > 0) {
+ if (copy_to_user(buf, data_of(data->handle), res))
+ res = -EFAULT;
+ else
+ *offp = data->handle.offset;
+ }
+ return res;
+}
+
+static ssize_t snapshot_write(struct file *filp, const char __user *buf,
+ size_t count, loff_t *offp)
+{
+ struct snapshot_data *data;
+ ssize_t res;
+
+ data = filp->private_data;
+ res = snapshot_write_next(&data->handle, count);
+ if (res > 0) {
+ if (copy_from_user(data_of(data->handle), buf, res))
+ res = -EFAULT;
+ else
+ *offp = data->handle.offset;
+ }
+ return res;
+}
+
+static int snapshot_ioctl(struct inode *inode, struct file *filp,
+ unsigned int cmd, unsigned long arg)
+{
+ int error = 0;
+ struct snapshot_data *data;
+ loff_t offset, avail;
+
+ if (_IOC_TYPE(cmd) != SNAPSHOT_IOC_MAGIC)
+ return -ENOTTY;
+ if (_IOC_NR(cmd) > SNAPSHOT_IOC_MAXNR)
+ return -ENOTTY;
+ if (!capable(CAP_SYS_ADMIN))
+ return -EPERM;
+
+ data = filp->private_data;
+
+ switch (cmd) {
+
+ case SNAPSHOT_FREEZE:
+ if (data->frozen)
+ break;
+ down(&pm_sem);
+ disable_nonboot_cpus();
+ if (freeze_processes()) {
+ thaw_processes();
+ enable_nonboot_cpus();
+ error = -EBUSY;
+ }
+ up(&pm_sem);
+ if (!error)
+ data->frozen = 1;
+ break;
+
+ case SNAPSHOT_UNFREEZE:
+ if (!data->frozen)
+ break;
+ down(&pm_sem);
+ thaw_processes();
+ enable_nonboot_cpus();
+ up(&pm_sem);
+ data->frozen = 0;
+ break;
+
+ case SNAPSHOT_ATOMIC_SNAPSHOT:
+ if (data->mode != O_RDONLY || !data->frozen || data->ready) {
+ error = -EPERM;
+ break;
+ }
+ down(&pm_sem);
+ /* Free memory before shutting down devices. */
+ error = swsusp_shrink_memory();
+ if (!error) {
+ error = device_suspend(PMSG_FREEZE);
+ if (!error) {
+ in_suspend = 1;
+ error = swsusp_suspend();
+ device_resume();
+ }
+ }
+ up(&pm_sem);
+ if (!error)
+ error = put_user(in_suspend, (unsigned int __user *)arg);
+ if (!error)
+ data->ready = 1;
+ break;
+
+ case SNAPSHOT_ATOMIC_RESTORE:
+ if (data->mode != O_WRONLY || !data->frozen ||
+ !snapshot_image_loaded(&data->handle)) {
+ error = -EPERM;
+ break;
+ }
+ down(&pm_sem);
+ pm_prepare_console();
+ error = device_suspend(PMSG_FREEZE);
+ if (!error) {
+ error = swsusp_resume();
+ device_resume();
+ }
+ pm_restore_console();
+ up(&pm_sem);
+ break;
+
+ case SNAPSHOT_FREE:
+ swsusp_free();
+ memset(&data->handle, 0, sizeof(struct snapshot_handle));
+ data->ready = 0;
+ break;
+
+ case SNAPSHOT_SET_IMAGE_SIZE:
+ image_size = arg;
+ break;
+
+ case SNAPSHOT_AVAIL_SWAP:
+ avail = count_swap_pages(data->swap, 1);
+ avail <<= PAGE_SHIFT;
+ error = put_user(avail, (loff_t __user *)arg);
+ break;
+
+ case SNAPSHOT_GET_SWAP_PAGE:
+ if (data->swap < 0 || data->swap >= MAX_SWAPFILES) {
+ error = -ENODEV;
+ break;
+ }
+ if (!data->bitmap) {
+ data->bitmap = alloc_bitmap(count_swap_pages(data->swap, 0));
+ if (!data->bitmap) {
+ error = -ENOMEM;
+ break;
+ }
+ }
+ offset = alloc_swap_page(data->swap, data->bitmap);
+ if (offset) {
+ offset <<= PAGE_SHIFT;
+ error = put_user(offset, (loff_t __user *)arg);
+ } else {
+ error = -ENOSPC;
+ }
+ break;
+
+ case SNAPSHOT_FREE_SWAP_PAGES:
+ if (data->swap < 0 || data->swap >= MAX_SWAPFILES) {
+ error = -ENODEV;
+ break;
+ }
+ free_all_swap_pages(data->swap, data->bitmap);
+ free_bitmap(data->bitmap);
+ data->bitmap = NULL;
+ break;
+
+ case SNAPSHOT_SET_SWAP_FILE:
+ if (!data->bitmap) {
+ /*
+ * User space encodes device types as two-byte values,
+ * so we need to recode them
+ */
+ if (old_decode_dev(arg)) {
+ data->swap = swap_type_of(old_decode_dev(arg));
+ if (data->swap < 0)
+ error = -ENODEV;
+ } else {
+ data->swap = -1;
+ error = -EINVAL;
+ }
+ } else {
+ error = -EPERM;
+ }
+ break;
+
+ case SNAPSHOT_S2RAM:
+ if (!data->frozen) {
+ error = -EPERM;
+ break;
+ }
+
+ if (down_trylock(&pm_sem)) {
+ error = -EBUSY;
+ break;
+ }
+
+ if (pm_ops->prepare) {
+ error = pm_ops->prepare(PM_SUSPEND_MEM);
+ if (error)
+ goto OutS3;
+ }
+
+ /* Put devices to sleep */
+ error = device_suspend(PMSG_SUSPEND);
+ if (error) {
+ printk(KERN_ERR "Failed to suspend some devices.\n");
+ } else {
+ /* Enter S3, system is already frozen */
+ suspend_enter(PM_SUSPEND_MEM);
+
+ /* Wake up devices */
+ device_resume();
+ }
+
+ if (pm_ops->finish)
+ pm_ops->finish(PM_SUSPEND_MEM);
+
+OutS3:
+ up(&pm_sem);
+ break;
+
+ default:
+ error = -ENOTTY;
+
+ }
+
+ return error;
+}
+
+static struct file_operations snapshot_fops = {
+ .open = snapshot_open,
+ .release = snapshot_release,
+ .read = snapshot_read,
+ .write = snapshot_write,
+ .llseek = no_llseek,
+ .ioctl = snapshot_ioctl,
+};
+
+static struct miscdevice snapshot_device = {
+ .minor = SNAPSHOT_MINOR,
+ .name = "snapshot",
+ .fops = &snapshot_fops,
+};
+
+static int __init snapshot_device_init(void)
+{
+ return misc_register(&snapshot_device);
+};
+
+device_initcall(snapshot_device_init);
diff --git a/kernel/profile.c b/kernel/profile.c
index f89248e6d704..ad81f799a9b4 100644
--- a/kernel/profile.c
+++ b/kernel/profile.c
@@ -23,6 +23,7 @@
#include <linux/cpu.h>
#include <linux/profile.h>
#include <linux/highmem.h>
+#include <linux/mutex.h>
#include <asm/sections.h>
#include <asm/semaphore.h>
@@ -44,7 +45,7 @@ static cpumask_t prof_cpu_mask = CPU_MASK_ALL;
#ifdef CONFIG_SMP
static DEFINE_PER_CPU(struct profile_hit *[2], cpu_profile_hits);
static DEFINE_PER_CPU(int, cpu_profile_flip);
-static DECLARE_MUTEX(profile_flip_mutex);
+static DEFINE_MUTEX(profile_flip_mutex);
#endif /* CONFIG_SMP */
static int __init profile_setup(char * str)
@@ -243,7 +244,7 @@ static void profile_flip_buffers(void)
{
int i, j, cpu;
- down(&profile_flip_mutex);
+ mutex_lock(&profile_flip_mutex);
j = per_cpu(cpu_profile_flip, get_cpu());
put_cpu();
on_each_cpu(__profile_flip_buffers, NULL, 0, 1);
@@ -259,14 +260,14 @@ static void profile_flip_buffers(void)
hits[i].hits = hits[i].pc = 0;
}
}
- up(&profile_flip_mutex);
+ mutex_unlock(&profile_flip_mutex);
}
static void profile_discard_flip_buffers(void)
{
int i, cpu;
- down(&profile_flip_mutex);
+ mutex_lock(&profile_flip_mutex);
i = per_cpu(cpu_profile_flip, get_cpu());
put_cpu();
on_each_cpu(__profile_flip_buffers, NULL, 0, 1);
@@ -274,7 +275,7 @@ static void profile_discard_flip_buffers(void)
struct profile_hit *hits = per_cpu(cpu_profile_hits, cpu)[i];
memset(hits, 0, NR_PROFILE_HIT*sizeof(struct profile_hit));
}
- up(&profile_flip_mutex);
+ mutex_unlock(&profile_flip_mutex);
}
void profile_hit(int type, void *__pc)
diff --git a/kernel/rcupdate.c b/kernel/rcupdate.c
index fedf5e369755..6df1559b1c02 100644
--- a/kernel/rcupdate.c
+++ b/kernel/rcupdate.c
@@ -47,15 +47,16 @@
#include <linux/notifier.h>
#include <linux/rcupdate.h>
#include <linux/cpu.h>
+#include <linux/mutex.h>
/* Definition for rcupdate control block. */
-struct rcu_ctrlblk rcu_ctrlblk = {
+static struct rcu_ctrlblk rcu_ctrlblk = {
.cur = -300,
.completed = -300,
.lock = SPIN_LOCK_UNLOCKED,
.cpumask = CPU_MASK_NONE,
};
-struct rcu_ctrlblk rcu_bh_ctrlblk = {
+static struct rcu_ctrlblk rcu_bh_ctrlblk = {
.cur = -300,
.completed = -300,
.lock = SPIN_LOCK_UNLOCKED,
@@ -75,7 +76,7 @@ static int rsinterval = 1000;
#endif
static atomic_t rcu_barrier_cpu_count;
-static struct semaphore rcu_barrier_sema;
+static DEFINE_MUTEX(rcu_barrier_mutex);
static struct completion rcu_barrier_completion;
#ifdef CONFIG_SMP
@@ -207,13 +208,13 @@ static void rcu_barrier_func(void *notused)
void rcu_barrier(void)
{
BUG_ON(in_interrupt());
- /* Take cpucontrol semaphore to protect against CPU hotplug */
- down(&rcu_barrier_sema);
+ /* Take cpucontrol mutex to protect against CPU hotplug */
+ mutex_lock(&rcu_barrier_mutex);
init_completion(&rcu_barrier_completion);
atomic_set(&rcu_barrier_cpu_count, 0);
on_each_cpu(rcu_barrier_func, NULL, 0, 1);
wait_for_completion(&rcu_barrier_completion);
- up(&rcu_barrier_sema);
+ mutex_unlock(&rcu_barrier_mutex);
}
EXPORT_SYMBOL_GPL(rcu_barrier);
@@ -549,7 +550,6 @@ static struct notifier_block __devinitdata rcu_nb = {
*/
void __init rcu_init(void)
{
- sema_init(&rcu_barrier_sema, 1);
rcu_cpu_notify(&rcu_nb, CPU_UP_PREPARE,
(void *)(long)smp_processor_id());
/* Register notifier for non-boot CPUs */
diff --git a/kernel/sched.c b/kernel/sched.c
index 6b6e0d70eb30..7ffaabd64f89 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -237,6 +237,7 @@ struct runqueue {
task_t *migration_thread;
struct list_head migration_queue;
+ int cpu;
#endif
#ifdef CONFIG_SCHEDSTATS
@@ -1654,6 +1655,9 @@ unsigned long nr_iowait(void)
/*
* double_rq_lock - safely lock two runqueues
*
+ * We must take them in cpu order to match code in
+ * dependent_sleeper and wake_dependent_sleeper.
+ *
* Note this does not disable interrupts like task_rq_lock,
* you need to do so manually before calling.
*/
@@ -1665,7 +1669,7 @@ static void double_rq_lock(runqueue_t *rq1, runqueue_t *rq2)
spin_lock(&rq1->lock);
__acquire(rq2->lock); /* Fake it out ;) */
} else {
- if (rq1 < rq2) {
+ if (rq1->cpu < rq2->cpu) {
spin_lock(&rq1->lock);
spin_lock(&rq2->lock);
} else {
@@ -1701,7 +1705,7 @@ static void double_lock_balance(runqueue_t *this_rq, runqueue_t *busiest)
__acquires(this_rq->lock)
{
if (unlikely(!spin_trylock(&busiest->lock))) {
- if (busiest < this_rq) {
+ if (busiest->cpu < this_rq->cpu) {
spin_unlock(&this_rq->lock);
spin_lock(&busiest->lock);
spin_lock(&this_rq->lock);
@@ -2869,7 +2873,7 @@ asmlinkage void __sched schedule(void)
*/
if (likely(!current->exit_state)) {
if (unlikely(in_atomic())) {
- printk(KERN_ERR "scheduling while atomic: "
+ printk(KERN_ERR "BUG: scheduling while atomic: "
"%s/0x%08x/%d\n",
current->comm, preempt_count(), current->pid);
dump_stack();
@@ -6029,6 +6033,7 @@ void __init sched_init(void)
rq->push_cpu = 0;
rq->migration_thread = NULL;
INIT_LIST_HEAD(&rq->migration_queue);
+ rq->cpu = i;
#endif
atomic_set(&rq->nr_iowait, 0);
@@ -6069,7 +6074,7 @@ void __might_sleep(char *file, int line)
if (time_before(jiffies, prev_jiffy + HZ) && prev_jiffy)
return;
prev_jiffy = jiffies;
- printk(KERN_ERR "Debug: sleeping function called from invalid"
+ printk(KERN_ERR "BUG: sleeping function called from invalid"
" context at %s:%d\n", file, line);
printk("in_atomic():%d, irqs_disabled():%d\n",
in_atomic(), irqs_disabled());
diff --git a/kernel/signal.c b/kernel/signal.c
index ea154104a00b..75f7341b0c39 100644
--- a/kernel/signal.c
+++ b/kernel/signal.c
@@ -1922,6 +1922,8 @@ int get_signal_to_deliver(siginfo_t *info, struct k_sigaction *return_ka,
sigset_t *mask = &current->blocked;
int signr = 0;
+ try_to_freeze();
+
relock:
spin_lock_irq(&current->sighand->siglock);
for (;;) {
@@ -2099,10 +2101,11 @@ long do_no_restart_syscall(struct restart_block *param)
int sigprocmask(int how, sigset_t *set, sigset_t *oldset)
{
int error;
- sigset_t old_block;
spin_lock_irq(&current->sighand->siglock);
- old_block = current->blocked;
+ if (oldset)
+ *oldset = current->blocked;
+
error = 0;
switch (how) {
case SIG_BLOCK:
@@ -2119,8 +2122,7 @@ int sigprocmask(int how, sigset_t *set, sigset_t *oldset)
}
recalc_sigpending();
spin_unlock_irq(&current->sighand->siglock);
- if (oldset)
- *oldset = old_block;
+
return error;
}
@@ -2307,7 +2309,6 @@ sys_rt_sigtimedwait(const sigset_t __user *uthese,
timeout = schedule_timeout_interruptible(timeout);
- try_to_freeze();
spin_lock_irq(&current->sighand->siglock);
sig = dequeue_signal(current, &these, &info);
current->blocked = current->real_blocked;
diff --git a/kernel/spinlock.c b/kernel/spinlock.c
index 0375fcd5921d..d1b810782bc4 100644
--- a/kernel/spinlock.c
+++ b/kernel/spinlock.c
@@ -179,16 +179,16 @@ EXPORT_SYMBOL(_write_lock);
#define BUILD_LOCK_OPS(op, locktype) \
void __lockfunc _##op##_lock(locktype##_t *lock) \
{ \
- preempt_disable(); \
for (;;) { \
+ preempt_disable(); \
if (likely(_raw_##op##_trylock(lock))) \
break; \
preempt_enable(); \
+ \
if (!(lock)->break_lock) \
(lock)->break_lock = 1; \
while (!op##_can_lock(lock) && (lock)->break_lock) \
cpu_relax(); \
- preempt_disable(); \
} \
(lock)->break_lock = 0; \
} \
@@ -199,19 +199,18 @@ unsigned long __lockfunc _##op##_lock_irqsave(locktype##_t *lock) \
{ \
unsigned long flags; \
\
- preempt_disable(); \
for (;;) { \
+ preempt_disable(); \
local_irq_save(flags); \
if (likely(_raw_##op##_trylock(lock))) \
break; \
local_irq_restore(flags); \
- \
preempt_enable(); \
+ \
if (!(lock)->break_lock) \
(lock)->break_lock = 1; \
while (!op##_can_lock(lock) && (lock)->break_lock) \
cpu_relax(); \
- preempt_disable(); \
} \
(lock)->break_lock = 0; \
return flags; \
diff --git a/kernel/sys.c b/kernel/sys.c
index f91218a5463e..c0fcad9f826c 100644
--- a/kernel/sys.c
+++ b/kernel/sys.c
@@ -1227,7 +1227,7 @@ asmlinkage long sys_setsid(void)
struct pid *pid;
int err = -EPERM;
- down(&tty_sem);
+ mutex_lock(&tty_mutex);
write_lock_irq(&tasklist_lock);
pid = find_pid(PIDTYPE_PGID, group_leader->pid);
@@ -1241,7 +1241,7 @@ asmlinkage long sys_setsid(void)
err = process_group(group_leader);
out:
write_unlock_irq(&tasklist_lock);
- up(&tty_sem);
+ mutex_unlock(&tty_mutex);
return err;
}
@@ -1677,9 +1677,6 @@ asmlinkage long sys_setrlimit(unsigned int resource, struct rlimit __user *rlim)
* a lot simpler! (Which we're not doing right now because we're not
* measuring them yet).
*
- * This expects to be called with tasklist_lock read-locked or better,
- * and the siglock not locked. It may momentarily take the siglock.
- *
* When sampling multiple threads for RUSAGE_SELF, under SMP we might have
* races with threads incrementing their own counters. But since word
* reads are atomic, we either get new values or old values and we don't
@@ -1687,6 +1684,25 @@ asmlinkage long sys_setrlimit(unsigned int resource, struct rlimit __user *rlim)
* the c* fields from p->signal from races with exit.c updating those
* fields when reaping, so a sample either gets all the additions of a
* given child after it's reaped, or none so this sample is before reaping.
+ *
+ * tasklist_lock locking optimisation:
+ * If we are current and single threaded, we do not need to take the tasklist
+ * lock or the siglock. No one else can take our signal_struct away,
+ * no one else can reap the children to update signal->c* counters, and
+ * no one else can race with the signal-> fields.
+ * If we do not take the tasklist_lock, the signal-> fields could be read
+ * out of order while another thread was just exiting. So we place a
+ * read memory barrier when we avoid the lock. On the writer side,
+ * write memory barrier is implied in __exit_signal as __exit_signal releases
+ * the siglock spinlock after updating the signal-> fields.
+ *
+ * We don't really need the siglock when we access the non c* fields
+ * of the signal_struct (for RUSAGE_SELF) even in multithreaded
+ * case, since we take the tasklist lock for read and the non c* signal->
+ * fields are updated only in __exit_signal, which is called with
+ * tasklist_lock taken for write, hence these two threads cannot execute
+ * concurrently.
+ *
*/
static void k_getrusage(struct task_struct *p, int who, struct rusage *r)
@@ -1694,13 +1710,23 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r)
struct task_struct *t;
unsigned long flags;
cputime_t utime, stime;
+ int need_lock = 0;
memset((char *) r, 0, sizeof *r);
+ utime = stime = cputime_zero;
- if (unlikely(!p->signal))
- return;
+ if (p != current || !thread_group_empty(p))
+ need_lock = 1;
- utime = stime = cputime_zero;
+ if (need_lock) {
+ read_lock(&tasklist_lock);
+ if (unlikely(!p->signal)) {
+ read_unlock(&tasklist_lock);
+ return;
+ }
+ } else
+ /* See locking comments above */
+ smp_rmb();
switch (who) {
case RUSAGE_BOTH:
@@ -1740,6 +1766,8 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r)
BUG();
}
+ if (need_lock)
+ read_unlock(&tasklist_lock);
cputime_to_timeval(utime, &r->ru_utime);
cputime_to_timeval(stime, &r->ru_stime);
}
@@ -1747,9 +1775,7 @@ static void k_getrusage(struct task_struct *p, int who, struct rusage *r)
int getrusage(struct task_struct *p, int who, struct rusage __user *ru)
{
struct rusage r;
- read_lock(&tasklist_lock);
k_getrusage(p, who, &r);
- read_unlock(&tasklist_lock);
return copy_to_user(ru, &r, sizeof(r)) ? -EFAULT : 0;
}
diff --git a/lib/reed_solomon/reed_solomon.c b/lib/reed_solomon/reed_solomon.c
index f5fef948a415..f8ac9fa95de1 100644
--- a/lib/reed_solomon/reed_solomon.c
+++ b/lib/reed_solomon/reed_solomon.c
@@ -44,12 +44,13 @@
#include <linux/module.h>
#include <linux/rslib.h>
#include <linux/slab.h>
+#include <linux/mutex.h>
#include <asm/semaphore.h>
/* This list holds all currently allocated rs control structures */
static LIST_HEAD (rslist);
/* Protection for the list */
-static DECLARE_MUTEX(rslistlock);
+static DEFINE_MUTEX(rslistlock);
/**
* rs_init - Initialize a Reed-Solomon codec
@@ -161,7 +162,7 @@ errrs:
*/
void free_rs(struct rs_control *rs)
{
- down(&rslistlock);
+ mutex_lock(&rslistlock);
rs->users--;
if(!rs->users) {
list_del(&rs->list);
@@ -170,7 +171,7 @@ void free_rs(struct rs_control *rs)
kfree(rs->genpoly);
kfree(rs);
}
- up(&rslistlock);
+ mutex_unlock(&rslistlock);
}
/**
@@ -201,7 +202,7 @@ struct rs_control *init_rs(int symsize, int gfpoly, int fcr, int prim,
if (nroots < 0 || nroots >= (1<<symsize))
return NULL;
- down(&rslistlock);
+ mutex_lock(&rslistlock);
/* Walk through the list and look for a matching entry */
list_for_each(tmp, &rslist) {
@@ -228,7 +229,7 @@ struct rs_control *init_rs(int symsize, int gfpoly, int fcr, int prim,
list_add(&rs->list, &rslist);
}
out:
- up(&rslistlock);
+ mutex_unlock(&rslistlock);
return rs;
}
diff --git a/mm/readahead.c b/mm/readahead.c
index 301b36c4a0ce..0f142a40984b 100644
--- a/mm/readahead.c
+++ b/mm/readahead.c
@@ -555,6 +555,7 @@ recheck:
out:
return ra->prev_page + 1;
}
+EXPORT_SYMBOL_GPL(page_cache_readahead);
/*
* handle_ra_miss() is called when it is known that a page which should have
diff --git a/mm/swapfile.c b/mm/swapfile.c
index 365ed6ff182d..39aa9d129612 100644
--- a/mm/swapfile.c
+++ b/mm/swapfile.c
@@ -45,7 +45,7 @@ static const char Unused_offset[] = "Unused swap offset entry ";
struct swap_list_t swap_list = {-1, -1};
-struct swap_info_struct swap_info[MAX_SWAPFILES];
+static struct swap_info_struct swap_info[MAX_SWAPFILES];
static DEFINE_MUTEX(swapon_mutex);
@@ -417,6 +417,61 @@ void free_swap_and_cache(swp_entry_t entry)
}
}
+#ifdef CONFIG_SOFTWARE_SUSPEND
+/*
+ * Find the swap type that corresponds to given device (if any)
+ *
+ * This is needed for software suspend and is done in such a way that inode
+ * aliasing is allowed.
+ */
+int swap_type_of(dev_t device)
+{
+ int i;
+
+ spin_lock(&swap_lock);
+ for (i = 0; i < nr_swapfiles; i++) {
+ struct inode *inode;
+
+ if (!(swap_info[i].flags & SWP_WRITEOK))
+ continue;
+ if (!device) {
+ spin_unlock(&swap_lock);
+ return i;
+ }
+ inode = swap_info->swap_file->f_dentry->d_inode;
+ if (S_ISBLK(inode->i_mode) &&
+ device == MKDEV(imajor(inode), iminor(inode))) {
+ spin_unlock(&swap_lock);
+ return i;
+ }
+ }
+ spin_unlock(&swap_lock);
+ return -ENODEV;
+}
+
+/*
+ * Return either the total number of swap pages of given type, or the number
+ * of free pages of that type (depending on @free)
+ *
+ * This is needed for software suspend
+ */
+unsigned int count_swap_pages(int type, int free)
+{
+ unsigned int n = 0;
+
+ if (type < nr_swapfiles) {
+ spin_lock(&swap_lock);
+ if (swap_info[type].flags & SWP_WRITEOK) {
+ n = swap_info[type].pages;
+ if (free)
+ n -= swap_info[type].inuse_pages;
+ }
+ spin_unlock(&swap_lock);
+ }
+ return n;
+}
+#endif
+
/*
* No need to decide whether this PTE shares the swap entry with others,
* just let do_wp_page work it out if a write is requested later - to
diff --git a/security/seclvl.c b/security/seclvl.c
index 8529ea6f7aa8..441beaf1bbc1 100644
--- a/security/seclvl.c
+++ b/security/seclvl.c
@@ -8,6 +8,7 @@
* Copyright (c) 2001 WireX Communications, Inc <chris@wirex.com>
* Copyright (c) 2001 Greg Kroah-Hartman <greg@kroah.com>
* Copyright (c) 2002 International Business Machines <robb@austin.ibm.com>
+ * Copyright (c) 2006 Davi E. M. Arnaut <davi.arnaut@gmail.com>
*
* This program is free software; you can redistribute it and/or modify
* it under the terms of the GNU General Public License as published by
@@ -31,6 +32,7 @@
#include <linux/kobject.h>
#include <linux/crypto.h>
#include <asm/scatterlist.h>
+#include <linux/scatterlist.h>
#include <linux/gfp.h>
#include <linux/sysfs.h>
@@ -194,35 +196,27 @@ static unsigned char hashedPassword[SHA1_DIGEST_SIZE];
* people...
*/
static int
-plaintext_to_sha1(unsigned char *hash, const char *plaintext, int len)
+plaintext_to_sha1(unsigned char *hash, const char *plaintext, unsigned int len)
{
- char *pgVirtAddr;
struct crypto_tfm *tfm;
- struct scatterlist sg[1];
+ struct scatterlist sg;
if (len > PAGE_SIZE) {
seclvl_printk(0, KERN_ERR, "Plaintext password too large (%d "
"characters). Largest possible is %lu "
"bytes.\n", len, PAGE_SIZE);
- return -ENOMEM;
+ return -EINVAL;
}
tfm = crypto_alloc_tfm("sha1", CRYPTO_TFM_REQ_MAY_SLEEP);
if (tfm == NULL) {
seclvl_printk(0, KERN_ERR,
"Failed to load transform for SHA1\n");
- return -ENOSYS;
+ return -EINVAL;
}
- // Just get a new page; don't play around with page boundaries
- // and scatterlists.
- pgVirtAddr = (char *)__get_free_page(GFP_KERNEL);
- sg[0].page = virt_to_page(pgVirtAddr);
- sg[0].offset = 0;
- sg[0].length = len;
- strncpy(pgVirtAddr, plaintext, len);
+ sg_init_one(&sg, (u8 *)plaintext, len);
crypto_digest_init(tfm);
- crypto_digest_update(tfm, sg, 1);
+ crypto_digest_update(tfm, &sg, 1);
crypto_digest_final(tfm, hash);
crypto_free_tfm(tfm);
- free_page((unsigned long)pgVirtAddr);
return 0;
}
@@ -234,11 +228,9 @@ static ssize_t
passwd_write_file(struct file * file, const char __user * buf,
size_t count, loff_t *ppos)
{
- int i;
- unsigned char tmp[SHA1_DIGEST_SIZE];
- char *page;
- int rc;
+ char *p;
int len;
+ unsigned char tmp[SHA1_DIGEST_SIZE];
if (!*passwd && !*sha1_passwd) {
seclvl_printk(0, KERN_ERR, "Attempt to password-unlock the "
@@ -251,38 +243,39 @@ passwd_write_file(struct file * file, const char __user * buf,
return -EINVAL;
}
- if (count < 0 || count >= PAGE_SIZE)
+ if (count >= PAGE_SIZE)
return -EINVAL;
if (*ppos != 0)
return -EINVAL;
- page = (char *)get_zeroed_page(GFP_KERNEL);
- if (!page)
+ p = kmalloc(count, GFP_KERNEL);
+ if (!p)
return -ENOMEM;
len = -EFAULT;
- if (copy_from_user(page, buf, count))
+ if (copy_from_user(p, buf, count))
goto out;
- len = strlen(page);
+ len = count;
/* ``echo "secret" > seclvl/passwd'' includes a newline */
- if (page[len - 1] == '\n')
+ if (p[len - 1] == '\n')
len--;
/* Hash the password, then compare the hashed values */
- if ((rc = plaintext_to_sha1(tmp, page, len))) {
+ if ((len = plaintext_to_sha1(tmp, p, len))) {
seclvl_printk(0, KERN_ERR, "Error hashing password: rc = "
- "[%d]\n", rc);
- return rc;
- }
- for (i = 0; i < SHA1_DIGEST_SIZE; i++) {
- if (hashedPassword[i] != tmp[i])
- return -EPERM;
+ "[%d]\n", len);
+ goto out;
}
+
+ len = -EPERM;
+ if (memcmp(hashedPassword, tmp, SHA1_DIGEST_SIZE))
+ goto out;
+
seclvl_printk(0, KERN_INFO,
"Password accepted; seclvl reduced to 0.\n");
seclvl = 0;
len = count;
out:
- free_page((unsigned long)page);
+ kfree (p);
return len;
}
@@ -295,13 +288,11 @@ static struct file_operations passwd_file_ops = {
*/
static int seclvl_ptrace(struct task_struct *parent, struct task_struct *child)
{
- if (seclvl >= 0) {
- if (child->pid == 1) {
- seclvl_printk(1, KERN_WARNING, "Attempt to ptrace "
- "the init process dissallowed in "
- "secure level %d\n", seclvl);
- return -EPERM;
- }
+ if (seclvl >= 0 && child->pid == 1) {
+ seclvl_printk(1, KERN_WARNING, "Attempt to ptrace "
+ "the init process dissallowed in "
+ "secure level %d\n", seclvl);
+ return -EPERM;
}
return 0;
}
@@ -312,55 +303,54 @@ static int seclvl_ptrace(struct task_struct *parent, struct task_struct *child)
*/
static int seclvl_capable(struct task_struct *tsk, int cap)
{
+ int rc = 0;
+
/* init can do anything it wants */
if (tsk->pid == 1)
return 0;
- switch (seclvl) {
- case 2:
- /* fall through */
- case 1:
- if (cap == CAP_LINUX_IMMUTABLE) {
+ if (seclvl > 0) {
+ rc = -EPERM;
+
+ if (cap == CAP_LINUX_IMMUTABLE)
seclvl_printk(1, KERN_WARNING, "Attempt to modify "
"the IMMUTABLE and/or APPEND extended "
"attribute on a file with the IMMUTABLE "
"and/or APPEND extended attribute set "
"denied in seclvl [%d]\n", seclvl);
- return -EPERM;
- } else if (cap == CAP_SYS_RAWIO) { // Somewhat broad...
+ else if (cap == CAP_SYS_RAWIO)
seclvl_printk(1, KERN_WARNING, "Attempt to perform "
"raw I/O while in secure level [%d] "
"denied\n", seclvl);
- return -EPERM;
- } else if (cap == CAP_NET_ADMIN) {
+ else if (cap == CAP_NET_ADMIN)
seclvl_printk(1, KERN_WARNING, "Attempt to perform "
"network administrative task while "
"in secure level [%d] denied\n", seclvl);
- return -EPERM;
- } else if (cap == CAP_SETUID) {
+ else if (cap == CAP_SETUID)
seclvl_printk(1, KERN_WARNING, "Attempt to setuid "
"while in secure level [%d] denied\n",
seclvl);
- return -EPERM;
- } else if (cap == CAP_SETGID) {
+ else if (cap == CAP_SETGID)
seclvl_printk(1, KERN_WARNING, "Attempt to setgid "
"while in secure level [%d] denied\n",
seclvl);
- } else if (cap == CAP_SYS_MODULE) {
+ else if (cap == CAP_SYS_MODULE)
seclvl_printk(1, KERN_WARNING, "Attempt to perform "
"a module operation while in secure "
"level [%d] denied\n", seclvl);
- return -EPERM;
- }
- break;
- default:
- break;
+ else
+ rc = 0;
}
- /* from dummy.c */
- if (cap_is_fs_cap(cap) ? tsk->fsuid == 0 : tsk->euid == 0)
- return 0; /* capability granted */
- seclvl_printk(1, KERN_WARNING, "Capability denied\n");
- return -EPERM; /* capability denied */
+
+ if (!rc) {
+ if (!(cap_is_fs_cap(cap) ? tsk->fsuid == 0 : tsk->euid == 0))
+ rc = -EPERM;
+ }
+
+ if (rc)
+ seclvl_printk(1, KERN_WARNING, "Capability denied\n");
+
+ return rc;
}
/**
@@ -466,12 +456,9 @@ static int seclvl_inode_setattr(struct dentry *dentry, struct iattr *iattr)
static void seclvl_file_free_security(struct file *filp)
{
struct dentry *dentry = filp->f_dentry;
- struct inode *inode = NULL;
- if (dentry) {
- inode = dentry->d_inode;
- seclvl_bd_release(inode);
- }
+ if (dentry)
+ seclvl_bd_release(dentry->d_inode);
}
/**
@@ -479,9 +466,7 @@ static void seclvl_file_free_security(struct file *filp)
*/
static int seclvl_umount(struct vfsmount *mnt, int flags)
{
- if (current->pid == 1)
- return 0;
- if (seclvl == 2) {
+ if (current->pid != 1 && seclvl == 2) {
seclvl_printk(1, KERN_WARNING, "Attempt to unmount in secure "
"level %d\n", seclvl);
return -EPERM;
@@ -505,8 +490,9 @@ static struct security_operations seclvl_ops = {
static int processPassword(void)
{
int rc = 0;
- hashedPassword[0] = '\0';
if (*passwd) {
+ char *p;
+
if (*sha1_passwd) {
seclvl_printk(0, KERN_ERR, "Error: Both "
"passwd and sha1_passwd "
@@ -514,12 +500,16 @@ static int processPassword(void)
"exclusive.\n");
return -EINVAL;
}
- if ((rc = plaintext_to_sha1(hashedPassword, passwd,
- strlen(passwd)))) {
+
+ p = kstrdup(passwd, GFP_KERNEL);
+ if (p == NULL)
+ return -ENOMEM;
+
+ if ((rc = plaintext_to_sha1(hashedPassword, p, strlen(p))))
seclvl_printk(0, KERN_ERR, "Error: SHA1 support not "
"in kernel\n");
- return rc;
- }
+
+ kfree (p);
/* All static data goes to the BSS, which zero's the
* plaintext password out for us. */
} else if (*sha1_passwd) { // Base 16
@@ -542,7 +532,7 @@ static int processPassword(void)
sha1_passwd[i + 2] = tmp;
}
}
- return 0;
+ return rc;
}
/**
@@ -552,28 +542,46 @@ struct dentry *dir_ino, *seclvl_ino, *passwd_ino;
static int seclvlfs_register(void)
{
+ int rc = 0;
+
dir_ino = securityfs_create_dir("seclvl", NULL);
- if (!dir_ino)
- return -EFAULT;
+
+ if (IS_ERR(dir_ino))
+ return PTR_ERR(dir_ino);
seclvl_ino = securityfs_create_file("seclvl", S_IRUGO | S_IWUSR,
dir_ino, &seclvl, &seclvl_file_ops);
- if (!seclvl_ino)
+ if (IS_ERR(seclvl_ino)) {
+ rc = PTR_ERR(seclvl_ino);
goto out_deldir;
+ }
if (*passwd || *sha1_passwd) {
passwd_ino = securityfs_create_file("passwd", S_IRUGO | S_IWUSR,
dir_ino, NULL, &passwd_file_ops);
- if (!passwd_ino)
+ if (IS_ERR(passwd_ino)) {
+ rc = PTR_ERR(passwd_ino);
goto out_delf;
+ }
}
- return 0;
+ return rc;
+
+out_delf:
+ securityfs_remove(seclvl_ino);
out_deldir:
securityfs_remove(dir_ino);
-out_delf:
+
+ return rc;
+}
+
+static void seclvlfs_unregister(void)
+{
securityfs_remove(seclvl_ino);
- return -EFAULT;
+ if (*passwd || *sha1_passwd)
+ securityfs_remove(passwd_ino);
+
+ securityfs_remove(dir_ino);
}
/**
@@ -582,6 +590,8 @@ out_delf:
static int __init seclvl_init(void)
{
int rc = 0;
+ static char once;
+
if (verbosity < 0 || verbosity > 1) {
printk(KERN_ERR "Error: bad verbosity [%d]; only 0 or 1 "
"are valid values\n", verbosity);
@@ -600,6 +610,11 @@ static int __init seclvl_init(void)
"module parameter(s): rc = [%d]\n", rc);
goto exit;
}
+
+ if ((rc = seclvlfs_register())) {
+ seclvl_printk(0, KERN_ERR, "Error registering with sysfs\n");
+ goto exit;
+ }
/* register ourselves with the security framework */
if (register_security(&seclvl_ops)) {
seclvl_printk(0, KERN_ERR,
@@ -611,20 +626,24 @@ static int __init seclvl_init(void)
seclvl_printk(0, KERN_ERR, "seclvl: Failure "
"registering with primary security "
"module.\n");
+ seclvlfs_unregister();
goto exit;
} /* if primary module registered */
secondary = 1;
} /* if we registered ourselves with the security framework */
- if ((rc = seclvlfs_register())) {
- seclvl_printk(0, KERN_ERR, "Error registering with sysfs\n");
- goto exit;
- }
+
seclvl_printk(0, KERN_INFO, "seclvl: Successfully initialized.\n");
+
+ if (once) {
+ once = 1;
+ seclvl_printk(0, KERN_INFO, "seclvl is going away. It has been "
+ "buggy for ages. Also, be warned that "
+ "Securelevels are useless.");
+ }
exit:
- if (rc) {
+ if (rc)
printk(KERN_ERR "seclvl: Error during initialization: rc = "
"[%d]\n", rc);
- }
return rc;
}
@@ -633,17 +652,14 @@ static int __init seclvl_init(void)
*/
static void __exit seclvl_exit(void)
{
- securityfs_remove(seclvl_ino);
- if (*passwd || *sha1_passwd)
- securityfs_remove(passwd_ino);
- securityfs_remove(dir_ino);
- if (secondary == 1) {
+ seclvlfs_unregister();
+
+ if (secondary)
mod_unreg_security(MY_NAME, &seclvl_ops);
- } else if (unregister_security(&seclvl_ops)) {
+ else if (unregister_security(&seclvl_ops))
seclvl_printk(0, KERN_INFO,
"seclvl: Failure unregistering with the "
"kernel\n");
- }
}
module_init(seclvl_init);
diff --git a/sound/oss/ac97_codec.c b/sound/oss/ac97_codec.c
index fd25aca25120..972327c97644 100644
--- a/sound/oss/ac97_codec.c
+++ b/sound/oss/ac97_codec.c
@@ -55,7 +55,7 @@
#include <linux/pci.h>
#include <linux/ac97_codec.h>
#include <asm/uaccess.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
#define CODEC_ID_BUFSZ 14
@@ -304,7 +304,7 @@ static const unsigned int ac97_oss_rm[] = {
static LIST_HEAD(codecs);
static LIST_HEAD(codec_drivers);
-static DECLARE_MUTEX(codec_sem);
+static DEFINE_MUTEX(codec_mutex);
/* reads the given OSS mixer from the ac97 the caller must have insured that the ac97 knows
about that given mixer, and should be holding a spinlock for the card */
@@ -769,9 +769,9 @@ void ac97_release_codec(struct ac97_codec *codec)
{
/* Remove from the list first, we don't want to be
"rediscovered" */
- down(&codec_sem);
+ mutex_lock(&codec_mutex);
list_del(&codec->list);
- up(&codec_sem);
+ mutex_unlock(&codec_mutex);
/*
* The driver needs to deal with internal
* locking to avoid accidents here.
@@ -889,7 +889,7 @@ int ac97_probe_codec(struct ac97_codec *codec)
* callbacks.
*/
- down(&codec_sem);
+ mutex_lock(&codec_mutex);
list_add(&codec->list, &codecs);
list_for_each(l, &codec_drivers) {
@@ -903,7 +903,7 @@ int ac97_probe_codec(struct ac97_codec *codec)
}
}
- up(&codec_sem);
+ mutex_unlock(&codec_mutex);
return 1;
}
@@ -1439,7 +1439,7 @@ int ac97_register_driver(struct ac97_driver *driver)
struct list_head *l;
struct ac97_codec *c;
- down(&codec_sem);
+ mutex_lock(&codec_mutex);
INIT_LIST_HEAD(&driver->list);
list_add(&driver->list, &codec_drivers);
@@ -1452,7 +1452,7 @@ int ac97_register_driver(struct ac97_driver *driver)
continue;
c->driver = driver;
}
- up(&codec_sem);
+ mutex_unlock(&codec_mutex);
return 0;
}
@@ -1471,7 +1471,7 @@ void ac97_unregister_driver(struct ac97_driver *driver)
struct list_head *l;
struct ac97_codec *c;
- down(&codec_sem);
+ mutex_lock(&codec_mutex);
list_del_init(&driver->list);
list_for_each(l, &codecs)
@@ -1483,7 +1483,7 @@ void ac97_unregister_driver(struct ac97_driver *driver)
}
}
- up(&codec_sem);
+ mutex_unlock(&codec_mutex);
}
EXPORT_SYMBOL_GPL(ac97_unregister_driver);
@@ -1494,14 +1494,14 @@ static int swap_headphone(int remove_master)
struct ac97_codec *c;
if (remove_master) {
- down(&codec_sem);
+ mutex_lock(&codec_mutex);
list_for_each(l, &codecs)
{
c = list_entry(l, struct ac97_codec, list);
if (supported_mixer(c, SOUND_MIXER_PHONEOUT))
c->supported_mixers &= ~SOUND_MASK_PHONEOUT;
}
- up(&codec_sem);
+ mutex_unlock(&codec_mutex);
} else
ac97_hw[SOUND_MIXER_PHONEOUT].offset = AC97_MASTER_VOL_STEREO;
diff --git a/sound/oss/aci.c b/sound/oss/aci.c
index 3928c2802cc4..3bfac375dbdb 100644
--- a/sound/oss/aci.c
+++ b/sound/oss/aci.c
@@ -56,7 +56,8 @@
#include <linux/module.h>
#include <linux/proc_fs.h>
#include <linux/slab.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
+
#include <asm/io.h>
#include <asm/uaccess.h>
#include "sound_config.h"
@@ -79,7 +80,7 @@ static int aci_micpreamp=3; /* microphone preamp-level that can't be *
* checked with ACI versions prior to 0xb0 */
static int mixer_device;
-static struct semaphore aci_sem;
+static struct mutex aci_mutex;
#ifdef MODULE
static int reset;
@@ -212,7 +213,7 @@ int aci_rw_cmd(int write1, int write2, int write3)
int write[] = {write1, write2, write3};
int read = -EINTR, i;
- if (down_interruptible(&aci_sem))
+ if (mutex_lock_interruptible(&aci_mutex))
goto out;
for (i=0; i<3; i++) {
@@ -227,7 +228,7 @@ int aci_rw_cmd(int write1, int write2, int write3)
}
read = aci_rawread();
-out_up: up(&aci_sem);
+out_up: mutex_unlock(&aci_mutex);
out: return read;
}
@@ -603,7 +604,7 @@ static int __init attach_aci(void)
char *boardname;
int i, rc = -EBUSY;
- init_MUTEX(&aci_sem);
+ mutex_init(&aci_mutex);
outb(0xE3, 0xf8f); /* Write MAD16 password */
aci_port = (inb(0xf90) & 0x10) ?
diff --git a/sound/oss/ad1889.c b/sound/oss/ad1889.c
index a0d73f343100..54dabf862802 100644
--- a/sound/oss/ad1889.c
+++ b/sound/oss/ad1889.c
@@ -38,6 +38,7 @@
#include <linux/ac97_codec.h>
#include <linux/sound.h>
#include <linux/interrupt.h>
+#include <linux/mutex.h>
#include <asm/delay.h>
#include <asm/io.h>
@@ -238,7 +239,7 @@ static ad1889_dev_t *ad1889_alloc_dev(struct pci_dev *pci)
for (i = 0; i < AD_MAX_STATES; i++) {
dev->state[i].card = dev;
- init_MUTEX(&dev->state[i].sem);
+ mutex_init(&dev->state[i].mutex);
init_waitqueue_head(&dev->state[i].dmabuf.wait);
}
@@ -461,7 +462,7 @@ static ssize_t ad1889_write(struct file *file, const char __user *buffer, size_t
ssize_t ret = 0;
DECLARE_WAITQUEUE(wait, current);
- down(&state->sem);
+ mutex_lock(&state->mutex);
#if 0
if (dmabuf->mapped) {
ret = -ENXIO;
@@ -546,7 +547,7 @@ static ssize_t ad1889_write(struct file *file, const char __user *buffer, size_t
err2:
remove_wait_queue(&state->dmabuf.wait, &wait);
err1:
- up(&state->sem);
+ mutex_unlock(&state->mutex);
return ret;
}
diff --git a/sound/oss/ad1889.h b/sound/oss/ad1889.h
index e04affce1dd1..861b3213f30b 100644
--- a/sound/oss/ad1889.h
+++ b/sound/oss/ad1889.h
@@ -100,7 +100,7 @@ typedef struct ad1889_state {
unsigned int subdivision;
} dmabuf;
- struct semaphore sem;
+ struct mutex mutex;
} ad1889_state_t;
typedef struct ad1889_dev {
diff --git a/sound/oss/ali5455.c b/sound/oss/ali5455.c
index 9c9e6c0410f2..62bb936b1f3d 100644
--- a/sound/oss/ali5455.c
+++ b/sound/oss/ali5455.c
@@ -64,6 +64,8 @@
#include <linux/smp_lock.h>
#include <linux/ac97_codec.h>
#include <linux/interrupt.h>
+#include <linux/mutex.h>
+
#include <asm/uaccess.h>
#ifndef PCI_DEVICE_ID_ALI_5455
@@ -234,7 +236,7 @@ struct ali_state {
struct ali_card *card; /* Card info */
/* single open lock mechanism, only used for recording */
- struct semaphore open_sem;
+ struct mutex open_mutex;
wait_queue_head_t open_wait;
/* file mode */
@@ -2807,7 +2809,7 @@ found_virt:
state->card = card;
state->magic = ALI5455_STATE_MAGIC;
init_waitqueue_head(&dmabuf->wait);
- init_MUTEX(&state->open_sem);
+ mutex_init(&state->open_mutex);
file->private_data = state;
dmabuf->trigger = 0;
/* allocate hardware channels */
@@ -3359,7 +3361,7 @@ static void __devinit ali_configure_clocking(void)
state->card = card;
state->magic = ALI5455_STATE_MAGIC;
init_waitqueue_head(&dmabuf->wait);
- init_MUTEX(&state->open_sem);
+ mutex_init(&state->open_mutex);
dmabuf->fmt = ALI5455_FMT_STEREO | ALI5455_FMT_16BIT;
dmabuf->trigger = PCM_ENABLE_OUTPUT;
ali_set_dac_rate(state, 48000);
diff --git a/sound/oss/au1000.c b/sound/oss/au1000.c
index c407de86cbb6..fe54de25aafc 100644
--- a/sound/oss/au1000.c
+++ b/sound/oss/au1000.c
@@ -68,6 +68,8 @@
#include <linux/smp_lock.h>
#include <linux/ac97_codec.h>
#include <linux/interrupt.h>
+#include <linux/mutex.h>
+
#include <asm/io.h>
#include <asm/uaccess.h>
#include <asm/mach-au1x00/au1000.h>
@@ -120,8 +122,8 @@ struct au1000_state {
int no_vra; // do not use VRA
spinlock_t lock;
- struct semaphore open_sem;
- struct semaphore sem;
+ struct mutex open_mutex;
+ struct mutex sem;
mode_t open_mode;
wait_queue_head_t open_wait;
@@ -1106,7 +1108,7 @@ static ssize_t au1000_read(struct file *file, char *buffer,
count *= db->cnt_factor;
- down(&s->sem);
+ mutex_lock(&s->sem);
add_wait_queue(&db->wait, &wait);
while (count > 0) {
@@ -1125,14 +1127,14 @@ static ssize_t au1000_read(struct file *file, char *buffer,
ret = -EAGAIN;
goto out;
}
- up(&s->sem);
+ mutex_unlock(&s->sem);
schedule();
if (signal_pending(current)) {
if (!ret)
ret = -ERESTARTSYS;
goto out2;
}
- down(&s->sem);
+ mutex_lock(&s->sem);
}
} while (avail <= 0);
@@ -1159,7 +1161,7 @@ static ssize_t au1000_read(struct file *file, char *buffer,
} // while (count > 0)
out:
- up(&s->sem);
+ mutex_unlock(&s->sem);
out2:
remove_wait_queue(&db->wait, &wait);
set_current_state(TASK_RUNNING);
@@ -1187,7 +1189,7 @@ static ssize_t au1000_write(struct file *file, const char *buffer,
count *= db->cnt_factor;
- down(&s->sem);
+ mutex_lock(&s->sem);
add_wait_queue(&db->wait, &wait);
while (count > 0) {
@@ -1204,14 +1206,14 @@ static ssize_t au1000_write(struct file *file, const char *buffer,
ret = -EAGAIN;
goto out;
}
- up(&s->sem);
+ mutex_unlock(&s->sem);
schedule();
if (signal_pending(current)) {
if (!ret)
ret = -ERESTARTSYS;
goto out2;
}
- down(&s->sem);
+ mutex_lock(&s->sem);
}
} while (avail <= 0);
@@ -1240,7 +1242,7 @@ static ssize_t au1000_write(struct file *file, const char *buffer,
} // while (count > 0)
out:
- up(&s->sem);
+ mutex_unlock(&s->sem);
out2:
remove_wait_queue(&db->wait, &wait);
set_current_state(TASK_RUNNING);
@@ -1298,7 +1300,7 @@ static int au1000_mmap(struct file *file, struct vm_area_struct *vma)
dbg("%s", __FUNCTION__);
lock_kernel();
- down(&s->sem);
+ mutex_lock(&s->sem);
if (vma->vm_flags & VM_WRITE)
db = &s->dma_dac;
else if (vma->vm_flags & VM_READ)
@@ -1324,7 +1326,7 @@ static int au1000_mmap(struct file *file, struct vm_area_struct *vma)
vma->vm_flags &= ~VM_IO;
db->mapped = 1;
out:
- up(&s->sem);
+ mutex_unlock(&s->sem);
unlock_kernel();
return ret;
}
@@ -1829,21 +1831,21 @@ static int au1000_open(struct inode *inode, struct file *file)
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
stop_dac(s);
@@ -1879,8 +1881,8 @@ static int au1000_open(struct inode *inode, struct file *file)
}
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&s->open_sem);
- init_MUTEX(&s->sem);
+ mutex_unlock(&s->open_mutex);
+ mutex_init(&s->sem);
return nonseekable_open(inode, file);
}
@@ -1896,7 +1898,7 @@ static int au1000_release(struct inode *inode, struct file *file)
lock_kernel();
}
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac(s);
dealloc_dmabuf(s, &s->dma_dac);
@@ -1906,7 +1908,7 @@ static int au1000_release(struct inode *inode, struct file *file)
dealloc_dmabuf(s, &s->dma_adc);
}
s->open_mode &= ((~file->f_mode) & (FMODE_READ|FMODE_WRITE));
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
wake_up(&s->open_wait);
unlock_kernel();
return 0;
@@ -1996,7 +1998,7 @@ static int __devinit au1000_probe(void)
init_waitqueue_head(&s->dma_adc.wait);
init_waitqueue_head(&s->dma_dac.wait);
init_waitqueue_head(&s->open_wait);
- init_MUTEX(&s->open_sem);
+ mutex_init(&s->open_mutex);
spin_lock_init(&s->lock);
s->codec.private_data = s;
s->codec.id = 0;
diff --git a/sound/oss/au1550_ac97.c b/sound/oss/au1550_ac97.c
index bdee0502f3e2..6a4956b8025d 100644
--- a/sound/oss/au1550_ac97.c
+++ b/sound/oss/au1550_ac97.c
@@ -52,6 +52,8 @@
#include <linux/spinlock.h>
#include <linux/smp_lock.h>
#include <linux/ac97_codec.h>
+#include <linux/mutex.h>
+
#include <asm/io.h>
#include <asm/uaccess.h>
#include <asm/hardirq.h>
@@ -90,8 +92,8 @@ static struct au1550_state {
int no_vra; /* do not use VRA */
spinlock_t lock;
- struct semaphore open_sem;
- struct semaphore sem;
+ struct mutex open_mutex;
+ struct mutex sem;
mode_t open_mode;
wait_queue_head_t open_wait;
@@ -1044,7 +1046,7 @@ au1550_read(struct file *file, char *buffer, size_t count, loff_t *ppos)
count *= db->cnt_factor;
- down(&s->sem);
+ mutex_lock(&s->sem);
add_wait_queue(&db->wait, &wait);
while (count > 0) {
@@ -1064,14 +1066,14 @@ au1550_read(struct file *file, char *buffer, size_t count, loff_t *ppos)
ret = -EAGAIN;
goto out;
}
- up(&s->sem);
+ mutex_unlock(&s->sem);
schedule();
if (signal_pending(current)) {
if (!ret)
ret = -ERESTARTSYS;
goto out2;
}
- down(&s->sem);
+ mutex_lock(&s->sem);
}
} while (avail <= 0);
@@ -1099,7 +1101,7 @@ au1550_read(struct file *file, char *buffer, size_t count, loff_t *ppos)
} /* while (count > 0) */
out:
- up(&s->sem);
+ mutex_unlock(&s->sem);
out2:
remove_wait_queue(&db->wait, &wait);
set_current_state(TASK_RUNNING);
@@ -1125,7 +1127,7 @@ au1550_write(struct file *file, const char *buffer, size_t count, loff_t * ppos)
count *= db->cnt_factor;
- down(&s->sem);
+ mutex_lock(&s->sem);
add_wait_queue(&db->wait, &wait);
while (count > 0) {
@@ -1143,14 +1145,14 @@ au1550_write(struct file *file, const char *buffer, size_t count, loff_t * ppos)
ret = -EAGAIN;
goto out;
}
- up(&s->sem);
+ mutex_unlock(&s->sem);
schedule();
if (signal_pending(current)) {
if (!ret)
ret = -ERESTARTSYS;
goto out2;
}
- down(&s->sem);
+ mutex_lock(&s->sem);
}
} while (avail <= 0);
@@ -1196,7 +1198,7 @@ au1550_write(struct file *file, const char *buffer, size_t count, loff_t * ppos)
} /* while (count > 0) */
out:
- up(&s->sem);
+ mutex_unlock(&s->sem);
out2:
remove_wait_queue(&db->wait, &wait);
set_current_state(TASK_RUNNING);
@@ -1253,7 +1255,7 @@ au1550_mmap(struct file *file, struct vm_area_struct *vma)
int ret = 0;
lock_kernel();
- down(&s->sem);
+ mutex_lock(&s->sem);
if (vma->vm_flags & VM_WRITE)
db = &s->dma_dac;
else if (vma->vm_flags & VM_READ)
@@ -1279,7 +1281,7 @@ au1550_mmap(struct file *file, struct vm_area_struct *vma)
vma->vm_flags &= ~VM_IO;
db->mapped = 1;
out:
- up(&s->sem);
+ mutex_unlock(&s->sem);
unlock_kernel();
return ret;
}
@@ -1790,21 +1792,21 @@ au1550_open(struct inode *inode, struct file *file)
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
stop_dac(s);
@@ -1840,8 +1842,8 @@ au1550_open(struct inode *inode, struct file *file)
}
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&s->open_sem);
- init_MUTEX(&s->sem);
+ mutex_unlock(&s->open_mutex);
+ mutex_init(&s->sem);
return 0;
}
@@ -1858,7 +1860,7 @@ au1550_release(struct inode *inode, struct file *file)
lock_kernel();
}
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac(s);
kfree(s->dma_dac.rawbuf);
@@ -1870,7 +1872,7 @@ au1550_release(struct inode *inode, struct file *file)
s->dma_adc.rawbuf = NULL;
}
s->open_mode &= ((~file->f_mode) & (FMODE_READ|FMODE_WRITE));
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
wake_up(&s->open_wait);
unlock_kernel();
return 0;
@@ -1902,7 +1904,7 @@ au1550_probe(void)
init_waitqueue_head(&s->dma_adc.wait);
init_waitqueue_head(&s->dma_dac.wait);
init_waitqueue_head(&s->open_wait);
- init_MUTEX(&s->open_sem);
+ mutex_init(&s->open_mutex);
spin_lock_init(&s->lock);
s->codec = ac97_alloc_codec();
diff --git a/sound/oss/btaudio.c b/sound/oss/btaudio.c
index 4007a5680acb..bfe3b534ef30 100644
--- a/sound/oss/btaudio.c
+++ b/sound/oss/btaudio.c
@@ -32,6 +32,8 @@
#include <linux/soundcard.h>
#include <linux/slab.h>
#include <linux/kdev_t.h>
+#include <linux/mutex.h>
+
#include <asm/uaccess.h>
#include <asm/io.h>
@@ -108,7 +110,7 @@ struct btaudio {
/* locking */
int users;
- struct semaphore lock;
+ struct mutex lock;
/* risc instructions */
unsigned int risc_size;
@@ -440,7 +442,7 @@ static struct file_operations btaudio_mixer_fops = {
static int btaudio_dsp_open(struct inode *inode, struct file *file,
struct btaudio *bta, int analog)
{
- down(&bta->lock);
+ mutex_lock(&bta->lock);
if (bta->users)
goto busy;
bta->users++;
@@ -452,11 +454,11 @@ static int btaudio_dsp_open(struct inode *inode, struct file *file,
bta->read_count = 0;
bta->sampleshift = 0;
- up(&bta->lock);
+ mutex_unlock(&bta->lock);
return 0;
busy:
- up(&bta->lock);
+ mutex_unlock(&bta->lock);
return -EBUSY;
}
@@ -496,11 +498,11 @@ static int btaudio_dsp_release(struct inode *inode, struct file *file)
{
struct btaudio *bta = file->private_data;
- down(&bta->lock);
+ mutex_lock(&bta->lock);
if (bta->recording)
stop_recording(bta);
bta->users--;
- up(&bta->lock);
+ mutex_unlock(&bta->lock);
return 0;
}
@@ -513,7 +515,7 @@ static ssize_t btaudio_dsp_read(struct file *file, char __user *buffer,
DECLARE_WAITQUEUE(wait, current);
add_wait_queue(&bta->readq, &wait);
- down(&bta->lock);
+ mutex_lock(&bta->lock);
while (swcount > 0) {
if (0 == bta->read_count) {
if (!bta->recording) {
@@ -528,10 +530,10 @@ static ssize_t btaudio_dsp_read(struct file *file, char __user *buffer,
ret = -EAGAIN;
break;
}
- up(&bta->lock);
+ mutex_unlock(&bta->lock);
current->state = TASK_INTERRUPTIBLE;
schedule();
- down(&bta->lock);
+ mutex_lock(&bta->lock);
if(signal_pending(current)) {
if (0 == ret)
ret = -EINTR;
@@ -604,7 +606,7 @@ static ssize_t btaudio_dsp_read(struct file *file, char __user *buffer,
if (bta->read_offset == bta->buf_size)
bta->read_offset = 0;
}
- up(&bta->lock);
+ mutex_unlock(&bta->lock);
remove_wait_queue(&bta->readq, &wait);
current->state = TASK_RUNNING;
return ret;
@@ -651,10 +653,10 @@ static int btaudio_dsp_ioctl(struct inode *inode, struct file *file,
bta->decimation = 0;
}
if (bta->recording) {
- down(&bta->lock);
+ mutex_lock(&bta->lock);
stop_recording(bta);
start_recording(bta);
- up(&bta->lock);
+ mutex_unlock(&bta->lock);
}
/* fall through */
case SOUND_PCM_READ_RATE:
@@ -716,10 +718,10 @@ static int btaudio_dsp_ioctl(struct inode *inode, struct file *file,
else
bta->bits = 16;
if (bta->recording) {
- down(&bta->lock);
+ mutex_lock(&bta->lock);
stop_recording(bta);
start_recording(bta);
- up(&bta->lock);
+ mutex_unlock(&bta->lock);
}
}
if (debug)
@@ -736,9 +738,9 @@ static int btaudio_dsp_ioctl(struct inode *inode, struct file *file,
case SNDCTL_DSP_RESET:
if (bta->recording) {
- down(&bta->lock);
+ mutex_lock(&bta->lock);
stop_recording(bta);
- up(&bta->lock);
+ mutex_unlock(&bta->lock);
}
return 0;
case SNDCTL_DSP_GETBLKSIZE:
@@ -941,7 +943,7 @@ static int __devinit btaudio_probe(struct pci_dev *pci_dev,
if (rate)
bta->rate = rate;
- init_MUTEX(&bta->lock);
+ mutex_init(&bta->lock);
init_waitqueue_head(&bta->readq);
if (-1 != latency) {
diff --git a/sound/oss/cmpci.c b/sound/oss/cmpci.c
index 7cfbb08db537..1fbd5137f6d7 100644
--- a/sound/oss/cmpci.c
+++ b/sound/oss/cmpci.c
@@ -138,6 +138,8 @@
#endif
#ifdef CONFIG_SOUND_CMPCI_JOYSTICK
#include <linux/gameport.h>
+#include <linux/mutex.h>
+
#endif
/* --------------------------------------------------------------------- */
@@ -392,7 +394,7 @@ struct cm_state {
unsigned char fmt, enable;
spinlock_t lock;
- struct semaphore open_sem;
+ struct mutex open_mutex;
mode_t open_mode;
wait_queue_head_t open_wait;
@@ -2825,21 +2827,21 @@ static int cm_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
if (file->f_mode & FMODE_READ) {
s->status &= ~DO_BIGENDIAN_R;
@@ -2867,7 +2869,7 @@ static int cm_open(struct inode *inode, struct file *file)
}
set_fmt(s, fmtm, fmts);
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -2879,7 +2881,7 @@ static int cm_release(struct inode *inode, struct file *file)
lock_kernel();
if (file->f_mode & FMODE_WRITE)
drain_dac(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac(s);
@@ -2903,7 +2905,7 @@ static int cm_release(struct inode *inode, struct file *file)
s->status &= ~DO_BIGENDIAN_R;
}
s->open_mode &= ~(file->f_mode & (FMODE_READ|FMODE_WRITE));
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
wake_up(&s->open_wait);
unlock_kernel();
return 0;
@@ -3080,7 +3082,7 @@ static int __devinit cm_probe(struct pci_dev *pcidev, const struct pci_device_id
init_waitqueue_head(&s->dma_adc.wait);
init_waitqueue_head(&s->dma_dac.wait);
init_waitqueue_head(&s->open_wait);
- init_MUTEX(&s->open_sem);
+ mutex_init(&s->open_mutex);
spin_lock_init(&s->lock);
s->magic = CM_MAGIC;
s->dev = pcidev;
diff --git a/sound/oss/cs4281/cs4281m.c b/sound/oss/cs4281/cs4281m.c
index 0720365f6438..0004442f9b7e 100644
--- a/sound/oss/cs4281/cs4281m.c
+++ b/sound/oss/cs4281/cs4281m.c
@@ -245,9 +245,9 @@ struct cs4281_state {
void *tmpbuff; // tmp buffer for sample conversions
unsigned ena;
spinlock_t lock;
- struct semaphore open_sem;
- struct semaphore open_sem_adc;
- struct semaphore open_sem_dac;
+ struct mutex open_sem;
+ struct mutex open_sem_adc;
+ struct mutex open_sem_dac;
mode_t open_mode;
wait_queue_head_t open_wait;
wait_queue_head_t open_wait_adc;
@@ -3598,20 +3598,20 @@ static int cs4281_release(struct inode *inode, struct file *file)
if (file->f_mode & FMODE_WRITE) {
drain_dac(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem_dac);
+ mutex_lock(&s->open_sem_dac);
stop_dac(s);
dealloc_dmabuf(s, &s->dma_dac);
s->open_mode &= ~FMODE_WRITE;
- up(&s->open_sem_dac);
+ mutex_unlock(&s->open_sem_dac);
wake_up(&s->open_wait_dac);
}
if (file->f_mode & FMODE_READ) {
drain_adc(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem_adc);
+ mutex_lock(&s->open_sem_adc);
stop_adc(s);
dealloc_dmabuf(s, &s->dma_adc);
s->open_mode &= ~FMODE_READ;
- up(&s->open_sem_adc);
+ mutex_unlock(&s->open_sem_adc);
wake_up(&s->open_wait_adc);
}
return 0;
@@ -3651,33 +3651,33 @@ static int cs4281_open(struct inode *inode, struct file *file)
return -ENODEV;
}
if (file->f_mode & FMODE_WRITE) {
- down(&s->open_sem_dac);
+ mutex_lock(&s->open_sem_dac);
while (s->open_mode & FMODE_WRITE) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem_dac);
+ mutex_unlock(&s->open_sem_dac);
return -EBUSY;
}
- up(&s->open_sem_dac);
+ mutex_unlock(&s->open_sem_dac);
interruptible_sleep_on(&s->open_wait_dac);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem_dac);
+ mutex_lock(&s->open_sem_dac);
}
}
if (file->f_mode & FMODE_READ) {
- down(&s->open_sem_adc);
+ mutex_lock(&s->open_sem_adc);
while (s->open_mode & FMODE_READ) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem_adc);
+ mutex_unlock(&s->open_sem_adc);
return -EBUSY;
}
- up(&s->open_sem_adc);
+ mutex_unlock(&s->open_sem_adc);
interruptible_sleep_on(&s->open_wait_adc);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem_adc);
+ mutex_lock(&s->open_sem_adc);
}
}
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
@@ -3691,7 +3691,7 @@ static int cs4281_open(struct inode *inode, struct file *file)
s->ena &= ~FMODE_READ;
s->dma_adc.ossfragshift = s->dma_adc.ossmaxfrags =
s->dma_adc.subdivision = 0;
- up(&s->open_sem_adc);
+ mutex_unlock(&s->open_sem_adc);
if (prog_dmabuf_adc(s)) {
CS_DBGOUT(CS_OPEN | CS_ERROR, 2, printk(KERN_ERR
@@ -3711,7 +3711,7 @@ static int cs4281_open(struct inode *inode, struct file *file)
s->ena &= ~FMODE_WRITE;
s->dma_dac.ossfragshift = s->dma_dac.ossmaxfrags =
s->dma_dac.subdivision = 0;
- up(&s->open_sem_dac);
+ mutex_unlock(&s->open_sem_dac);
if (prog_dmabuf_dac(s)) {
CS_DBGOUT(CS_OPEN | CS_ERROR, 2, printk(KERN_ERR
@@ -3978,17 +3978,17 @@ static int cs4281_midi_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
// wait for device to become free
- down(&s->open_sem);
+ mutex_lock(&s->open_sem);
while (s->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_sem);
return -EBUSY;
}
- up(&s->open_sem);
+ mutex_unlock(&s->open_sem);
interruptible_sleep_on(&s->open_wait);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_sem);
}
spin_lock_irqsave(&s->lock, flags);
if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -4018,7 +4018,7 @@ static int cs4281_midi_open(struct inode *inode, struct file *file)
(file->
f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ |
FMODE_MIDI_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_sem);
return nonseekable_open(inode, file);
}
@@ -4057,7 +4057,7 @@ static int cs4281_midi_release(struct inode *inode, struct file *file)
remove_wait_queue(&s->midi.owait, &wait);
current->state = TASK_RUNNING;
}
- down(&s->open_sem);
+ mutex_lock(&s->open_sem);
s->open_mode &=
(~(file->f_mode << FMODE_MIDI_SHIFT)) & (FMODE_MIDI_READ |
FMODE_MIDI_WRITE);
@@ -4067,7 +4067,7 @@ static int cs4281_midi_release(struct inode *inode, struct file *file)
del_timer(&s->midi.timer);
}
spin_unlock_irqrestore(&s->lock, flags);
- up(&s->open_sem);
+ mutex_unlock(&s->open_sem);
wake_up(&s->open_wait);
return 0;
}
@@ -4300,9 +4300,9 @@ static int __devinit cs4281_probe(struct pci_dev *pcidev,
init_waitqueue_head(&s->open_wait_dac);
init_waitqueue_head(&s->midi.iwait);
init_waitqueue_head(&s->midi.owait);
- init_MUTEX(&s->open_sem);
- init_MUTEX(&s->open_sem_adc);
- init_MUTEX(&s->open_sem_dac);
+ mutex_init(&s->open_sem);
+ mutex_init(&s->open_sem_adc);
+ mutex_init(&s->open_sem_dac);
spin_lock_init(&s->lock);
s->pBA0phys = pci_resource_start(pcidev, 0);
s->pBA1phys = pci_resource_start(pcidev, 1);
diff --git a/sound/oss/cs46xx.c b/sound/oss/cs46xx.c
index 58e25c82eaf2..53881bc91bba 100644
--- a/sound/oss/cs46xx.c
+++ b/sound/oss/cs46xx.c
@@ -90,6 +90,7 @@
#include <linux/init.h>
#include <linux/poll.h>
#include <linux/ac97_codec.h>
+#include <linux/mutex.h>
#include <asm/io.h>
#include <asm/dma.h>
@@ -238,7 +239,7 @@ struct cs_state {
struct cs_card *card; /* Card info */
/* single open lock mechanism, only used for recording */
- struct semaphore open_sem;
+ struct mutex open_mutex;
wait_queue_head_t open_wait;
/* file mode */
@@ -297,7 +298,7 @@ struct cs_state {
unsigned subdivision;
} dmabuf;
/* Guard against mmap/write/read races */
- struct semaphore sem;
+ struct mutex sem;
};
struct cs_card {
@@ -375,7 +376,7 @@ struct cs_card {
unsigned char ibuf[CS_MIDIINBUF];
unsigned char obuf[CS_MIDIOUTBUF];
mode_t open_mode;
- struct semaphore open_sem;
+ struct mutex open_mutex;
} midi;
struct cs46xx_pm pm;
};
@@ -1428,9 +1429,9 @@ static int prog_dmabuf(struct cs_state *state)
{
int ret;
- down(&state->sem);
+ mutex_lock(&state->sem);
ret = __prog_dmabuf(state);
- up(&state->sem);
+ mutex_unlock(&state->sem);
return ret;
}
@@ -1831,17 +1832,17 @@ static int cs_midi_open(struct inode *inode, struct file *file)
file->private_data = card;
/* wait for device to become free */
- down(&card->midi.open_sem);
+ mutex_lock(&card->midi.open_mutex);
while (card->midi.open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&card->midi.open_sem);
+ mutex_unlock(&card->midi.open_mutex);
return -EBUSY;
}
- up(&card->midi.open_sem);
+ mutex_unlock(&card->midi.open_mutex);
interruptible_sleep_on(&card->midi.open_wait);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&card->midi.open_sem);
+ mutex_lock(&card->midi.open_mutex);
}
spin_lock_irqsave(&card->midi.lock, flags);
if (!(card->midi.open_mode & (FMODE_READ | FMODE_WRITE))) {
@@ -1859,7 +1860,7 @@ static int cs_midi_open(struct inode *inode, struct file *file)
}
spin_unlock_irqrestore(&card->midi.lock, flags);
card->midi.open_mode |= (file->f_mode & (FMODE_READ | FMODE_WRITE));
- up(&card->midi.open_sem);
+ mutex_unlock(&card->midi.open_mutex);
return 0;
}
@@ -1891,9 +1892,9 @@ static int cs_midi_release(struct inode *inode, struct file *file)
remove_wait_queue(&card->midi.owait, &wait);
current->state = TASK_RUNNING;
}
- down(&card->midi.open_sem);
+ mutex_lock(&card->midi.open_mutex);
card->midi.open_mode &= (~(file->f_mode & (FMODE_READ | FMODE_WRITE)));
- up(&card->midi.open_sem);
+ mutex_unlock(&card->midi.open_mutex);
wake_up(&card->midi.open_wait);
return 0;
}
@@ -2081,7 +2082,7 @@ static ssize_t cs_read(struct file *file, char __user *buffer, size_t count, lof
if (!access_ok(VERIFY_WRITE, buffer, count))
return -EFAULT;
- down(&state->sem);
+ mutex_lock(&state->sem);
if (!dmabuf->ready && (ret = __prog_dmabuf(state)))
goto out2;
@@ -2114,13 +2115,13 @@ static ssize_t cs_read(struct file *file, char __user *buffer, size_t count, lof
if (!ret) ret = -EAGAIN;
goto out;
}
- up(&state->sem);
+ mutex_unlock(&state->sem);
schedule();
if (signal_pending(current)) {
if(!ret) ret = -ERESTARTSYS;
goto out;
}
- down(&state->sem);
+ mutex_lock(&state->sem);
if (dmabuf->mapped)
{
if(!ret)
@@ -2155,7 +2156,7 @@ static ssize_t cs_read(struct file *file, char __user *buffer, size_t count, lof
out:
remove_wait_queue(&state->dmabuf.wait, &wait);
out2:
- up(&state->sem);
+ mutex_unlock(&state->sem);
set_current_state(TASK_RUNNING);
CS_DBGOUT(CS_WAVE_READ | CS_FUNCTION, 4,
printk("cs46xx: cs_read()- %zd\n",ret) );
@@ -2184,7 +2185,7 @@ static ssize_t cs_write(struct file *file, const char __user *buffer, size_t cou
return -EFAULT;
dmabuf = &state->dmabuf;
- down(&state->sem);
+ mutex_lock(&state->sem);
if (dmabuf->mapped)
{
ret = -ENXIO;
@@ -2240,13 +2241,13 @@ static ssize_t cs_write(struct file *file, const char __user *buffer, size_t cou
if (!ret) ret = -EAGAIN;
goto out;
}
- up(&state->sem);
+ mutex_unlock(&state->sem);
schedule();
if (signal_pending(current)) {
if(!ret) ret = -ERESTARTSYS;
goto out;
}
- down(&state->sem);
+ mutex_lock(&state->sem);
if (dmabuf->mapped)
{
if(!ret)
@@ -2278,7 +2279,7 @@ static ssize_t cs_write(struct file *file, const char __user *buffer, size_t cou
start_dac(state);
}
out:
- up(&state->sem);
+ mutex_unlock(&state->sem);
remove_wait_queue(&state->dmabuf.wait, &wait);
set_current_state(TASK_RUNNING);
@@ -2411,7 +2412,7 @@ static int cs_mmap(struct file *file, struct vm_area_struct *vma)
goto out;
}
- down(&state->sem);
+ mutex_lock(&state->sem);
dmabuf = &state->dmabuf;
if (cs4x_pgoff(vma) != 0)
{
@@ -2438,7 +2439,7 @@ static int cs_mmap(struct file *file, struct vm_area_struct *vma)
CS_DBGOUT(CS_FUNCTION, 2, printk("cs46xx: cs_mmap()-\n") );
out:
- up(&state->sem);
+ mutex_unlock(&state->sem);
return ret;
}
@@ -3200,7 +3201,7 @@ static int cs_open(struct inode *inode, struct file *file)
if (state == NULL)
return -ENOMEM;
memset(state, 0, sizeof(struct cs_state));
- init_MUTEX(&state->sem);
+ mutex_init(&state->sem);
dmabuf = &state->dmabuf;
dmabuf->pbuf = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
if(dmabuf->pbuf==NULL)
@@ -3241,10 +3242,10 @@ static int cs_open(struct inode *inode, struct file *file)
state->virt = 0;
state->magic = CS_STATE_MAGIC;
init_waitqueue_head(&dmabuf->wait);
- init_MUTEX(&state->open_sem);
+ mutex_init(&state->open_mutex);
file->private_data = card;
- down(&state->open_sem);
+ mutex_lock(&state->open_mutex);
/* set default sample format. According to OSS Programmer's Guide /dev/dsp
should be default to unsigned 8-bits, mono, with sample rate 8kHz and
@@ -3260,7 +3261,7 @@ static int cs_open(struct inode *inode, struct file *file)
cs_set_divisor(dmabuf);
state->open_mode |= FMODE_READ;
- up(&state->open_sem);
+ mutex_unlock(&state->open_mutex);
}
if(file->f_mode & FMODE_WRITE)
{
@@ -3271,7 +3272,7 @@ static int cs_open(struct inode *inode, struct file *file)
if (state == NULL)
return -ENOMEM;
memset(state, 0, sizeof(struct cs_state));
- init_MUTEX(&state->sem);
+ mutex_init(&state->sem);
dmabuf = &state->dmabuf;
dmabuf->pbuf = (void *)get_zeroed_page(GFP_KERNEL | GFP_DMA);
if(dmabuf->pbuf==NULL)
@@ -3312,10 +3313,10 @@ static int cs_open(struct inode *inode, struct file *file)
state->virt = 1;
state->magic = CS_STATE_MAGIC;
init_waitqueue_head(&dmabuf->wait);
- init_MUTEX(&state->open_sem);
+ mutex_init(&state->open_mutex);
file->private_data = card;
- down(&state->open_sem);
+ mutex_lock(&state->open_mutex);
/* set default sample format. According to OSS Programmer's Guide /dev/dsp
should be default to unsigned 8-bits, mono, with sample rate 8kHz and
@@ -3331,7 +3332,7 @@ static int cs_open(struct inode *inode, struct file *file)
cs_set_divisor(dmabuf);
state->open_mode |= FMODE_WRITE;
- up(&state->open_sem);
+ mutex_unlock(&state->open_mutex);
if((ret = prog_dmabuf(state)))
return ret;
}
@@ -3363,14 +3364,14 @@ static int cs_release(struct inode *inode, struct file *file)
cs_clear_tail(state);
drain_dac(state, file->f_flags & O_NONBLOCK);
/* stop DMA state machine and free DMA buffers/channels */
- down(&state->open_sem);
+ mutex_lock(&state->open_mutex);
stop_dac(state);
dealloc_dmabuf(state);
state->card->free_pcm_channel(state->card, dmabuf->channel->num);
free_page((unsigned long)state->dmabuf.pbuf);
- /* we're covered by the open_sem */
- up(&state->open_sem);
+ /* we're covered by the open_mutex */
+ mutex_unlock(&state->open_mutex);
state->card->states[state->virt] = NULL;
state->open_mode &= (~file->f_mode) & (FMODE_READ|FMODE_WRITE);
@@ -3395,14 +3396,14 @@ static int cs_release(struct inode *inode, struct file *file)
{
CS_DBGOUT(CS_RELEASE, 2, printk("cs46xx: cs_release() FMODE_READ\n") );
dmabuf = &state->dmabuf;
- down(&state->open_sem);
+ mutex_lock(&state->open_mutex);
stop_adc(state);
dealloc_dmabuf(state);
state->card->free_pcm_channel(state->card, dmabuf->channel->num);
free_page((unsigned long)state->dmabuf.pbuf);
- /* we're covered by the open_sem */
- up(&state->open_sem);
+ /* we're covered by the open_mutex */
+ mutex_unlock(&state->open_mutex);
state->card->states[state->virt] = NULL;
state->open_mode &= (~file->f_mode) & (FMODE_READ|FMODE_WRITE);
@@ -5507,7 +5508,7 @@ static int __devinit cs46xx_probe(struct pci_dev *pci_dev,
}
init_waitqueue_head(&card->midi.open_wait);
- init_MUTEX(&card->midi.open_sem);
+ mutex_init(&card->midi.open_mutex);
init_waitqueue_head(&card->midi.iwait);
init_waitqueue_head(&card->midi.owait);
cs461x_pokeBA0(card, BA0_MIDCR, MIDCR_MRST);
diff --git a/sound/oss/dmasound/dmasound_awacs.c b/sound/oss/dmasound/dmasound_awacs.c
index 74f975676ccb..a17375141c3a 100644
--- a/sound/oss/dmasound/dmasound_awacs.c
+++ b/sound/oss/dmasound/dmasound_awacs.c
@@ -80,7 +80,7 @@
#include <linux/kmod.h>
#include <linux/interrupt.h>
#include <linux/input.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
#ifdef CONFIG_ADB_CUDA
#include <linux/cuda.h>
#endif
@@ -130,7 +130,7 @@ static struct resource awacs_rsrc[3];
static char awacs_name[64];
static int awacs_revision;
static int awacs_sleeping;
-static DECLARE_MUTEX(dmasound_sem);
+static DEFINE_MUTEX(dmasound_mutex);
static int sound_device_id; /* exists after iMac revA */
static int hw_can_byteswap = 1 ; /* most pmac sound h/w can */
@@ -312,11 +312,11 @@ extern int daca_enter_sleep(void);
extern int daca_leave_sleep(void);
#define TRY_LOCK() \
- if ((rc = down_interruptible(&dmasound_sem)) != 0) \
+ if ((rc = mutex_lock_interruptible(&dmasound_mutex)) != 0) \
return rc;
-#define LOCK() down(&dmasound_sem);
+#define LOCK() mutex_lock(&dmasound_mutex);
-#define UNLOCK() up(&dmasound_sem);
+#define UNLOCK() mutex_unlock(&dmasound_mutex);
/* We use different versions that the ones provided in dmasound.h
*
diff --git a/sound/oss/emu10k1/hwaccess.h b/sound/oss/emu10k1/hwaccess.h
index 104223a192aa..85e27bda694b 100644
--- a/sound/oss/emu10k1/hwaccess.h
+++ b/sound/oss/emu10k1/hwaccess.h
@@ -181,7 +181,7 @@ struct emu10k1_card
struct emu10k1_mpuout *mpuout;
struct emu10k1_mpuin *mpuin;
- struct semaphore open_sem;
+ struct mutex open_sem;
mode_t open_mode;
wait_queue_head_t open_wait;
diff --git a/sound/oss/emu10k1/main.c b/sound/oss/emu10k1/main.c
index 23241cbdd90f..0cd44a6f7ac0 100644
--- a/sound/oss/emu10k1/main.c
+++ b/sound/oss/emu10k1/main.c
@@ -1320,7 +1320,7 @@ static int __devinit emu10k1_probe(struct pci_dev *pci_dev, const struct pci_dev
card->is_aps = (subsysvid == EMU_APS_SUBID);
spin_lock_init(&card->lock);
- init_MUTEX(&card->open_sem);
+ mutex_init(&card->open_sem);
card->open_mode = 0;
init_waitqueue_head(&card->open_wait);
diff --git a/sound/oss/emu10k1/midi.c b/sound/oss/emu10k1/midi.c
index b40b5f97aace..959a96794dba 100644
--- a/sound/oss/emu10k1/midi.c
+++ b/sound/oss/emu10k1/midi.c
@@ -110,21 +110,21 @@ match:
#endif
/* Wait for device to become free */
- down(&card->open_sem);
+ mutex_lock(&card->open_sem);
while (card->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
if (file->f_flags & O_NONBLOCK) {
- up(&card->open_sem);
+ mutex_unlock(&card->open_sem);
return -EBUSY;
}
- up(&card->open_sem);
+ mutex_unlock(&card->open_sem);
interruptible_sleep_on(&card->open_wait);
if (signal_pending(current)) {
return -ERESTARTSYS;
}
- down(&card->open_sem);
+ mutex_lock(&card->open_sem);
}
if ((midi_dev = (struct emu10k1_mididevice *) kmalloc(sizeof(*midi_dev), GFP_KERNEL)) == NULL)
@@ -183,7 +183,7 @@ match:
card->open_mode |= (file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE);
- up(&card->open_sem);
+ mutex_unlock(&card->open_sem);
return nonseekable_open(inode, file);
}
@@ -234,9 +234,9 @@ static int emu10k1_midi_release(struct inode *inode, struct file *file)
kfree(midi_dev);
- down(&card->open_sem);
+ mutex_lock(&card->open_sem);
card->open_mode &= ~((file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE));
- up(&card->open_sem);
+ mutex_unlock(&card->open_sem);
wake_up_interruptible(&card->open_wait);
unlock_kernel();
diff --git a/sound/oss/es1370.c b/sound/oss/es1370.c
index ae55c536613a..094f569cc6e0 100644
--- a/sound/oss/es1370.c
+++ b/sound/oss/es1370.c
@@ -157,6 +157,7 @@
#include <linux/gameport.h>
#include <linux/wait.h>
#include <linux/dma-mapping.h>
+#include <linux/mutex.h>
#include <asm/io.h>
#include <asm/page.h>
@@ -346,7 +347,7 @@ struct es1370_state {
unsigned sctrl;
spinlock_t lock;
- struct semaphore open_sem;
+ struct mutex open_mutex;
mode_t open_mode;
wait_queue_head_t open_wait;
@@ -393,7 +394,7 @@ struct es1370_state {
struct gameport *gameport;
#endif
- struct semaphore sem;
+ struct mutex mutex;
};
/* --------------------------------------------------------------------- */
@@ -1159,7 +1160,7 @@ static ssize_t es1370_read(struct file *file, char __user *buffer, size_t count,
return -ENXIO;
if (!access_ok(VERIFY_WRITE, buffer, count))
return -EFAULT;
- down(&s->sem);
+ mutex_lock(&s->mutex);
if (!s->dma_adc.ready && (ret = prog_dmabuf_adc(s)))
goto out;
@@ -1183,14 +1184,14 @@ static ssize_t es1370_read(struct file *file, char __user *buffer, size_t count,
ret = -EAGAIN;
goto out;
}
- up(&s->sem);
+ mutex_unlock(&s->mutex);
schedule();
if (signal_pending(current)) {
if (!ret)
ret = -ERESTARTSYS;
goto out;
}
- down(&s->sem);
+ mutex_lock(&s->mutex);
if (s->dma_adc.mapped)
{
ret = -ENXIO;
@@ -1215,7 +1216,7 @@ static ssize_t es1370_read(struct file *file, char __user *buffer, size_t count,
start_adc(s);
}
out:
- up(&s->sem);
+ mutex_unlock(&s->mutex);
remove_wait_queue(&s->dma_adc.wait, &wait);
set_current_state(TASK_RUNNING);
return ret;
@@ -1235,7 +1236,7 @@ static ssize_t es1370_write(struct file *file, const char __user *buffer, size_t
return -ENXIO;
if (!access_ok(VERIFY_READ, buffer, count))
return -EFAULT;
- down(&s->sem);
+ mutex_lock(&s->mutex);
if (!s->dma_dac2.ready && (ret = prog_dmabuf_dac2(s)))
goto out;
ret = 0;
@@ -1263,14 +1264,14 @@ static ssize_t es1370_write(struct file *file, const char __user *buffer, size_t
ret = -EAGAIN;
goto out;
}
- up(&s->sem);
+ mutex_unlock(&s->mutex);
schedule();
if (signal_pending(current)) {
if (!ret)
ret = -ERESTARTSYS;
goto out;
}
- down(&s->sem);
+ mutex_lock(&s->mutex);
if (s->dma_dac2.mapped)
{
ret = -ENXIO;
@@ -1296,7 +1297,7 @@ static ssize_t es1370_write(struct file *file, const char __user *buffer, size_t
start_dac2(s);
}
out:
- up(&s->sem);
+ mutex_unlock(&s->mutex);
remove_wait_queue(&s->dma_dac2.wait, &wait);
set_current_state(TASK_RUNNING);
return ret;
@@ -1348,7 +1349,7 @@ static int es1370_mmap(struct file *file, struct vm_area_struct *vma)
VALIDATE_STATE(s);
lock_kernel();
- down(&s->sem);
+ mutex_lock(&s->mutex);
if (vma->vm_flags & VM_WRITE) {
if ((ret = prog_dmabuf_dac2(s)) != 0) {
goto out;
@@ -1380,7 +1381,7 @@ static int es1370_mmap(struct file *file, struct vm_area_struct *vma)
}
db->mapped = 1;
out:
- up(&s->sem);
+ mutex_unlock(&s->mutex);
unlock_kernel();
return ret;
}
@@ -1752,21 +1753,21 @@ static int es1370_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
spin_lock_irqsave(&s->lock, flags);
if (!(s->open_mode & (FMODE_READ|FMODE_WRITE)))
@@ -1793,8 +1794,8 @@ static int es1370_open(struct inode *inode, struct file *file)
outl(s->ctrl, s->io+ES1370_REG_CONTROL);
spin_unlock_irqrestore(&s->lock, flags);
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&s->open_sem);
- init_MUTEX(&s->sem);
+ mutex_unlock(&s->open_mutex);
+ mutex_init(&s->mutex);
return nonseekable_open(inode, file);
}
@@ -1806,7 +1807,7 @@ static int es1370_release(struct inode *inode, struct file *file)
lock_kernel();
if (file->f_mode & FMODE_WRITE)
drain_dac2(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac2(s);
synchronize_irq(s->irq);
@@ -1818,7 +1819,7 @@ static int es1370_release(struct inode *inode, struct file *file)
}
s->open_mode &= ~(file->f_mode & (FMODE_READ|FMODE_WRITE));
wake_up(&s->open_wait);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
unlock_kernel();
return 0;
}
@@ -2198,21 +2199,21 @@ static int es1370_open_dac(struct inode *inode, struct file *file)
return -EINVAL;
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & FMODE_DAC) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
s->dma_dac1.ossfragshift = s->dma_dac1.ossmaxfrags = s->dma_dac1.subdivision = 0;
s->dma_dac1.enabled = 1;
@@ -2227,7 +2228,7 @@ static int es1370_open_dac(struct inode *inode, struct file *file)
outl(s->ctrl, s->io+ES1370_REG_CONTROL);
spin_unlock_irqrestore(&s->lock, flags);
s->open_mode |= FMODE_DAC;
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -2238,12 +2239,12 @@ static int es1370_release_dac(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
lock_kernel();
drain_dac1(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
stop_dac1(s);
dealloc_dmabuf(s, &s->dma_dac1);
s->open_mode &= ~FMODE_DAC;
wake_up(&s->open_wait);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
unlock_kernel();
return 0;
}
@@ -2430,21 +2431,21 @@ static int es1370_midi_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
spin_lock_irqsave(&s->lock, flags);
if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2465,7 +2466,7 @@ static int es1370_midi_open(struct inode *inode, struct file *file)
es1370_handle_midi(s);
spin_unlock_irqrestore(&s->lock, flags);
s->open_mode |= (file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -2499,7 +2500,7 @@ static int es1370_midi_release(struct inode *inode, struct file *file)
remove_wait_queue(&s->midi.owait, &wait);
set_current_state(TASK_RUNNING);
}
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
s->open_mode &= ~((file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ|FMODE_MIDI_WRITE));
spin_lock_irqsave(&s->lock, flags);
if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2508,7 +2509,7 @@ static int es1370_midi_release(struct inode *inode, struct file *file)
}
spin_unlock_irqrestore(&s->lock, flags);
wake_up(&s->open_wait);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
unlock_kernel();
return 0;
}
@@ -2638,7 +2639,7 @@ static int __devinit es1370_probe(struct pci_dev *pcidev, const struct pci_devic
init_waitqueue_head(&s->open_wait);
init_waitqueue_head(&s->midi.iwait);
init_waitqueue_head(&s->midi.owait);
- init_MUTEX(&s->open_sem);
+ mutex_init(&s->open_mutex);
spin_lock_init(&s->lock);
s->magic = ES1370_MAGIC;
s->dev = pcidev;
diff --git a/sound/oss/es1371.c b/sound/oss/es1371.c
index 5c697f162579..4400c8538686 100644
--- a/sound/oss/es1371.c
+++ b/sound/oss/es1371.c
@@ -129,6 +129,7 @@
#include <linux/gameport.h>
#include <linux/wait.h>
#include <linux/dma-mapping.h>
+#include <linux/mutex.h>
#include <asm/io.h>
#include <asm/page.h>
@@ -419,7 +420,7 @@ struct es1371_state {
unsigned dac1rate, dac2rate, adcrate;
spinlock_t lock;
- struct semaphore open_sem;
+ struct mutex open_mutex;
mode_t open_mode;
wait_queue_head_t open_wait;
@@ -462,7 +463,7 @@ struct es1371_state {
struct gameport *gameport;
#endif
- struct semaphore sem;
+ struct mutex sem;
};
/* --------------------------------------------------------------------- */
@@ -1346,7 +1347,7 @@ static ssize_t es1371_read(struct file *file, char __user *buffer, size_t count,
return -ENXIO;
if (!access_ok(VERIFY_WRITE, buffer, count))
return -EFAULT;
- down(&s->sem);
+ mutex_lock(&s->sem);
if (!s->dma_adc.ready && (ret = prog_dmabuf_adc(s)))
goto out2;
@@ -1370,14 +1371,14 @@ static ssize_t es1371_read(struct file *file, char __user *buffer, size_t count,
ret = -EAGAIN;
goto out;
}
- up(&s->sem);
+ mutex_unlock(&s->sem);
schedule();
if (signal_pending(current)) {
if (!ret)
ret = -ERESTARTSYS;
goto out2;
}
- down(&s->sem);
+ mutex_lock(&s->sem);
if (s->dma_adc.mapped)
{
ret = -ENXIO;
@@ -1402,7 +1403,7 @@ static ssize_t es1371_read(struct file *file, char __user *buffer, size_t count,
start_adc(s);
}
out:
- up(&s->sem);
+ mutex_unlock(&s->sem);
out2:
remove_wait_queue(&s->dma_adc.wait, &wait);
set_current_state(TASK_RUNNING);
@@ -1423,7 +1424,7 @@ static ssize_t es1371_write(struct file *file, const char __user *buffer, size_t
return -ENXIO;
if (!access_ok(VERIFY_READ, buffer, count))
return -EFAULT;
- down(&s->sem);
+ mutex_lock(&s->sem);
if (!s->dma_dac2.ready && (ret = prog_dmabuf_dac2(s)))
goto out3;
ret = 0;
@@ -1451,14 +1452,14 @@ static ssize_t es1371_write(struct file *file, const char __user *buffer, size_t
ret = -EAGAIN;
goto out;
}
- up(&s->sem);
+ mutex_unlock(&s->sem);
schedule();
if (signal_pending(current)) {
if (!ret)
ret = -ERESTARTSYS;
goto out2;
}
- down(&s->sem);
+ mutex_lock(&s->sem);
if (s->dma_dac2.mapped)
{
ret = -ENXIO;
@@ -1484,7 +1485,7 @@ static ssize_t es1371_write(struct file *file, const char __user *buffer, size_t
start_dac2(s);
}
out:
- up(&s->sem);
+ mutex_unlock(&s->sem);
out2:
remove_wait_queue(&s->dma_dac2.wait, &wait);
out3:
@@ -1538,7 +1539,7 @@ static int es1371_mmap(struct file *file, struct vm_area_struct *vma)
VALIDATE_STATE(s);
lock_kernel();
- down(&s->sem);
+ mutex_lock(&s->sem);
if (vma->vm_flags & VM_WRITE) {
if ((ret = prog_dmabuf_dac2(s)) != 0) {
@@ -1571,7 +1572,7 @@ static int es1371_mmap(struct file *file, struct vm_area_struct *vma)
}
db->mapped = 1;
out:
- up(&s->sem);
+ mutex_unlock(&s->sem);
unlock_kernel();
return ret;
}
@@ -1938,21 +1939,21 @@ static int es1371_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
if (file->f_mode & FMODE_READ) {
s->dma_adc.ossfragshift = s->dma_adc.ossmaxfrags = s->dma_adc.subdivision = 0;
@@ -1982,8 +1983,8 @@ static int es1371_open(struct inode *inode, struct file *file)
outl(s->sctrl, s->io+ES1371_REG_SERIAL_CONTROL);
spin_unlock_irqrestore(&s->lock, flags);
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&s->open_sem);
- init_MUTEX(&s->sem);
+ mutex_unlock(&s->open_mutex);
+ mutex_init(&s->sem);
return nonseekable_open(inode, file);
}
@@ -1995,7 +1996,7 @@ static int es1371_release(struct inode *inode, struct file *file)
lock_kernel();
if (file->f_mode & FMODE_WRITE)
drain_dac2(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac2(s);
dealloc_dmabuf(s, &s->dma_dac2);
@@ -2005,7 +2006,7 @@ static int es1371_release(struct inode *inode, struct file *file)
dealloc_dmabuf(s, &s->dma_adc);
}
s->open_mode &= ~(file->f_mode & (FMODE_READ|FMODE_WRITE));
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
wake_up(&s->open_wait);
unlock_kernel();
return 0;
@@ -2377,21 +2378,21 @@ static int es1371_open_dac(struct inode *inode, struct file *file)
return -EINVAL;
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & FMODE_DAC) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
s->dma_dac1.ossfragshift = s->dma_dac1.ossmaxfrags = s->dma_dac1.subdivision = 0;
s->dma_dac1.enabled = 1;
@@ -2405,7 +2406,7 @@ static int es1371_open_dac(struct inode *inode, struct file *file)
outl(s->sctrl, s->io+ES1371_REG_SERIAL_CONTROL);
spin_unlock_irqrestore(&s->lock, flags);
s->open_mode |= FMODE_DAC;
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -2416,11 +2417,11 @@ static int es1371_release_dac(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
lock_kernel();
drain_dac1(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
stop_dac1(s);
dealloc_dmabuf(s, &s->dma_dac1);
s->open_mode &= ~FMODE_DAC;
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
wake_up(&s->open_wait);
unlock_kernel();
return 0;
@@ -2608,21 +2609,21 @@ static int es1371_midi_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
spin_lock_irqsave(&s->lock, flags);
if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2643,7 +2644,7 @@ static int es1371_midi_open(struct inode *inode, struct file *file)
es1371_handle_midi(s);
spin_unlock_irqrestore(&s->lock, flags);
s->open_mode |= (file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -2676,7 +2677,7 @@ static int es1371_midi_release(struct inode *inode, struct file *file)
remove_wait_queue(&s->midi.owait, &wait);
set_current_state(TASK_RUNNING);
}
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
s->open_mode &= ~((file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ|FMODE_MIDI_WRITE));
spin_lock_irqsave(&s->lock, flags);
if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2684,7 +2685,7 @@ static int es1371_midi_release(struct inode *inode, struct file *file)
outl(s->ctrl, s->io+ES1371_REG_CONTROL);
}
spin_unlock_irqrestore(&s->lock, flags);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
wake_up(&s->open_wait);
unlock_kernel();
return 0;
@@ -2884,7 +2885,7 @@ static int __devinit es1371_probe(struct pci_dev *pcidev, const struct pci_devic
init_waitqueue_head(&s->open_wait);
init_waitqueue_head(&s->midi.iwait);
init_waitqueue_head(&s->midi.owait);
- init_MUTEX(&s->open_sem);
+ mutex_init(&s->open_mutex);
spin_lock_init(&s->lock);
s->magic = ES1371_MAGIC;
s->dev = pcidev;
diff --git a/sound/oss/esssolo1.c b/sound/oss/esssolo1.c
index 849b59f67ef5..78d3e29ce968 100644
--- a/sound/oss/esssolo1.c
+++ b/sound/oss/esssolo1.c
@@ -105,6 +105,8 @@
#include <linux/gameport.h>
#include <linux/wait.h>
#include <linux/dma-mapping.h>
+#include <linux/mutex.h>
+
#include <asm/io.h>
#include <asm/page.h>
@@ -191,7 +193,7 @@ struct solo1_state {
unsigned ena;
spinlock_t lock;
- struct semaphore open_sem;
+ struct mutex open_mutex;
mode_t open_mode;
wait_queue_head_t open_wait;
@@ -1581,7 +1583,7 @@ static int solo1_release(struct inode *inode, struct file *file)
lock_kernel();
if (file->f_mode & FMODE_WRITE)
drain_dac(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac(s);
outb(0, s->iobase+6); /* disable DMA */
@@ -1595,7 +1597,7 @@ static int solo1_release(struct inode *inode, struct file *file)
}
s->open_mode &= ~(FMODE_READ | FMODE_WRITE);
wake_up(&s->open_wait);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
unlock_kernel();
return 0;
}
@@ -1624,21 +1626,21 @@ static int solo1_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & (FMODE_READ | FMODE_WRITE)) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
s->fmt = AFMT_U8;
s->channels = 1;
@@ -1650,7 +1652,7 @@ static int solo1_open(struct inode *inode, struct file *file)
s->dma_dac.ossfragshift = s->dma_dac.ossmaxfrags = s->dma_dac.subdivision = 0;
s->dma_dac.enabled = 1;
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
prog_codec(s);
return nonseekable_open(inode, file);
}
@@ -1911,21 +1913,21 @@ static int solo1_midi_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
spin_lock_irqsave(&s->lock, flags);
if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -1951,7 +1953,7 @@ static int solo1_midi_open(struct inode *inode, struct file *file)
}
spin_unlock_irqrestore(&s->lock, flags);
s->open_mode |= (file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -1985,7 +1987,7 @@ static int solo1_midi_release(struct inode *inode, struct file *file)
remove_wait_queue(&s->midi.owait, &wait);
set_current_state(TASK_RUNNING);
}
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
s->open_mode &= ~((file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ|FMODE_MIDI_WRITE));
spin_lock_irqsave(&s->lock, flags);
if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -1994,7 +1996,7 @@ static int solo1_midi_release(struct inode *inode, struct file *file)
}
spin_unlock_irqrestore(&s->lock, flags);
wake_up(&s->open_wait);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
unlock_kernel();
return 0;
}
@@ -2132,24 +2134,24 @@ static int solo1_dmfm_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & FMODE_DMFM) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
if (!request_region(s->sbbase, FMSYNTH_EXTENT, "ESS Solo1")) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
printk(KERN_ERR "solo1: FM synth io ports in use, opl3 loaded?\n");
return -EBUSY;
}
@@ -2161,7 +2163,7 @@ static int solo1_dmfm_open(struct inode *inode, struct file *file)
outb(5, s->sbbase+2);
outb(1, s->sbbase+3); /* enable OPL3 */
s->open_mode |= FMODE_DMFM;
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -2172,7 +2174,7 @@ static int solo1_dmfm_release(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
lock_kernel();
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
s->open_mode &= ~FMODE_DMFM;
for (regb = 0xb0; regb < 0xb9; regb++) {
outb(regb, s->sbbase);
@@ -2182,7 +2184,7 @@ static int solo1_dmfm_release(struct inode *inode, struct file *file)
}
release_region(s->sbbase, FMSYNTH_EXTENT);
wake_up(&s->open_wait);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
unlock_kernel();
return 0;
}
@@ -2362,7 +2364,7 @@ static int __devinit solo1_probe(struct pci_dev *pcidev, const struct pci_device
init_waitqueue_head(&s->open_wait);
init_waitqueue_head(&s->midi.iwait);
init_waitqueue_head(&s->midi.owait);
- init_MUTEX(&s->open_sem);
+ mutex_init(&s->open_mutex);
spin_lock_init(&s->lock);
s->magic = SOLO1_MAGIC;
s->dev = pcidev;
diff --git a/sound/oss/forte.c b/sound/oss/forte.c
index 8406bc90c4ff..0294eec8ad90 100644
--- a/sound/oss/forte.c
+++ b/sound/oss/forte.c
@@ -43,6 +43,7 @@
#include <linux/interrupt.h>
#include <linux/proc_fs.h>
+#include <linux/mutex.h>
#include <asm/uaccess.h>
#include <asm/io.h>
@@ -185,7 +186,7 @@ struct forte_chip {
unsigned long iobase;
int irq;
- struct semaphore open_sem; /* Device access */
+ struct mutex open_mutex; /* Device access */
spinlock_t lock; /* State */
spinlock_t ac97_lock;
@@ -1242,13 +1243,13 @@ forte_dsp_open (struct inode *inode, struct file *file)
struct forte_chip *chip = forte; /* FIXME: HACK FROM HELL! */
if (file->f_flags & O_NONBLOCK) {
- if (down_trylock (&chip->open_sem)) {
+ if (!mutex_trylock(&chip->open_mutex)) {
DPRINTK ("%s: returning -EAGAIN\n", __FUNCTION__);
return -EAGAIN;
}
}
else {
- if (down_interruptible (&chip->open_sem)) {
+ if (mutex_lock_interruptible(&chip->open_mutex)) {
DPRINTK ("%s: returning -ERESTARTSYS\n", __FUNCTION__);
return -ERESTARTSYS;
}
@@ -1302,7 +1303,7 @@ forte_dsp_release (struct inode *inode, struct file *file)
spin_unlock_irq (&chip->lock);
}
- up (&chip->open_sem);
+ mutex_unlock(&chip->open_mutex);
return ret;
}
@@ -2011,7 +2012,7 @@ forte_probe (struct pci_dev *pci_dev, const struct pci_device_id *pci_id)
memset (chip, 0, sizeof (struct forte_chip));
chip->pci_dev = pci_dev;
- init_MUTEX(&chip->open_sem);
+ mutex_init(&chip->open_mutex);
spin_lock_init (&chip->lock);
spin_lock_init (&chip->ac97_lock);
diff --git a/sound/oss/hal2.c b/sound/oss/hal2.c
index afe97c4ce069..dd4f59d30a3a 100644
--- a/sound/oss/hal2.c
+++ b/sound/oss/hal2.c
@@ -32,6 +32,8 @@
#include <linux/dma-mapping.h>
#include <linux/sound.h>
#include <linux/soundcard.h>
+#include <linux/mutex.h>
+
#include <asm/io.h>
#include <asm/sgi/hpc3.h>
@@ -92,7 +94,7 @@ struct hal2_codec {
wait_queue_head_t dma_wait;
spinlock_t lock;
- struct semaphore sem;
+ struct mutex sem;
int usecount; /* recording and playback are
* independent */
@@ -1178,7 +1180,7 @@ static ssize_t hal2_read(struct file *file, char *buffer,
if (!count)
return 0;
- if (down_interruptible(&adc->sem))
+ if (mutex_lock_interruptible(&adc->sem))
return -EINTR;
if (file->f_flags & O_NONBLOCK) {
err = hal2_get_buffer(hal2, buffer, count);
@@ -1217,7 +1219,7 @@ static ssize_t hal2_read(struct file *file, char *buffer,
}
} while (count > 0 && err >= 0);
}
- up(&adc->sem);
+ mutex_unlock(&adc->sem);
return err;
}
@@ -1232,7 +1234,7 @@ static ssize_t hal2_write(struct file *file, const char *buffer,
if (!count)
return 0;
- if (down_interruptible(&dac->sem))
+ if (mutex_lock_interruptible(&dac->sem))
return -EINTR;
if (file->f_flags & O_NONBLOCK) {
err = hal2_add_buffer(hal2, buf, count);
@@ -1271,7 +1273,7 @@ static ssize_t hal2_write(struct file *file, const char *buffer,
}
} while (count > 0 && err >= 0);
}
- up(&dac->sem);
+ mutex_unlock(&dac->sem);
return err;
}
@@ -1356,20 +1358,20 @@ static int hal2_release(struct inode *inode, struct file *file)
if (file->f_mode & FMODE_READ) {
struct hal2_codec *adc = &hal2->adc;
- down(&adc->sem);
+ mutex_lock(&adc->sem);
hal2_stop_adc(hal2);
hal2_free_adc_dmabuf(adc);
adc->usecount--;
- up(&adc->sem);
+ mutex_unlock(&adc->sem);
}
if (file->f_mode & FMODE_WRITE) {
struct hal2_codec *dac = &hal2->dac;
- down(&dac->sem);
+ mutex_lock(&dac->sem);
hal2_sync_dac(hal2);
hal2_free_dac_dmabuf(dac);
dac->usecount--;
- up(&dac->sem);
+ mutex_unlock(&dac->sem);
}
return 0;
@@ -1400,7 +1402,7 @@ static void hal2_init_codec(struct hal2_codec *codec, struct hpc3_regs *hpc3,
codec->pbus.pbusnr = index;
codec->pbus.pbus = &hpc3->pbdma[index];
init_waitqueue_head(&codec->dma_wait);
- init_MUTEX(&codec->sem);
+ mutex_init(&codec->sem);
spin_lock_init(&codec->lock);
}
diff --git a/sound/oss/i810_audio.c b/sound/oss/i810_audio.c
index abc242abd5b1..dd2b871cdac5 100644
--- a/sound/oss/i810_audio.c
+++ b/sound/oss/i810_audio.c
@@ -100,6 +100,8 @@
#include <linux/smp_lock.h>
#include <linux/ac97_codec.h>
#include <linux/bitops.h>
+#include <linux/mutex.h>
+
#include <asm/uaccess.h>
#define DRIVER_VERSION "1.01"
@@ -331,7 +333,7 @@ struct i810_state {
struct i810_card *card; /* Card info */
/* single open lock mechanism, only used for recording */
- struct semaphore open_sem;
+ struct mutex open_mutex;
wait_queue_head_t open_wait;
/* file mode */
@@ -2597,7 +2599,7 @@ found_virt:
state->card = card;
state->magic = I810_STATE_MAGIC;
init_waitqueue_head(&dmabuf->wait);
- init_MUTEX(&state->open_sem);
+ mutex_init(&state->open_mutex);
file->private_data = state;
dmabuf->trigger = 0;
@@ -3213,7 +3215,7 @@ static void __devinit i810_configure_clocking (void)
state->card = card;
state->magic = I810_STATE_MAGIC;
init_waitqueue_head(&dmabuf->wait);
- init_MUTEX(&state->open_sem);
+ mutex_init(&state->open_mutex);
dmabuf->fmt = I810_FMT_STEREO | I810_FMT_16BIT;
dmabuf->trigger = PCM_ENABLE_OUTPUT;
i810_set_spdif_output(state, -1, 0);
diff --git a/sound/oss/ite8172.c b/sound/oss/ite8172.c
index 8fd2f9a9e668..ffcb910f5c3e 100644
--- a/sound/oss/ite8172.c
+++ b/sound/oss/ite8172.c
@@ -71,6 +71,8 @@
#include <linux/smp_lock.h>
#include <linux/ac97_codec.h>
#include <linux/interrupt.h>
+#include <linux/mutex.h>
+
#include <asm/io.h>
#include <asm/dma.h>
#include <asm/uaccess.h>
@@ -304,7 +306,7 @@ struct it8172_state {
unsigned dacrate, adcrate;
spinlock_t lock;
- struct semaphore open_sem;
+ struct mutex open_mutex;
mode_t open_mode;
wait_queue_head_t open_wait;
@@ -1801,21 +1803,21 @@ static int it8172_open(struct inode *inode, struct file *file)
}
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
spin_lock_irqsave(&s->lock, flags);
@@ -1850,7 +1852,7 @@ static int it8172_open(struct inode *inode, struct file *file)
spin_unlock_irqrestore(&s->lock, flags);
s->open_mode |= (file->f_mode & (FMODE_READ | FMODE_WRITE));
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -1864,7 +1866,7 @@ static int it8172_release(struct inode *inode, struct file *file)
lock_kernel();
if (file->f_mode & FMODE_WRITE)
drain_dac(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac(s);
dealloc_dmabuf(s, &s->dma_dac);
@@ -1874,7 +1876,7 @@ static int it8172_release(struct inode *inode, struct file *file)
dealloc_dmabuf(s, &s->dma_adc);
}
s->open_mode &= ((~file->f_mode) & (FMODE_READ|FMODE_WRITE));
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
wake_up(&s->open_wait);
unlock_kernel();
return 0;
@@ -1997,7 +1999,7 @@ static int __devinit it8172_probe(struct pci_dev *pcidev,
init_waitqueue_head(&s->dma_adc.wait);
init_waitqueue_head(&s->dma_dac.wait);
init_waitqueue_head(&s->open_wait);
- init_MUTEX(&s->open_sem);
+ mutex_init(&s->open_mutex);
spin_lock_init(&s->lock);
s->dev = pcidev;
s->io = pci_resource_start(pcidev, 0);
diff --git a/sound/oss/maestro.c b/sound/oss/maestro.c
index d4b569acf764..e647f2f86279 100644
--- a/sound/oss/maestro.c
+++ b/sound/oss/maestro.c
@@ -223,6 +223,8 @@
#include <linux/reboot.h>
#include <linux/bitops.h>
#include <linux/wait.h>
+#include <linux/mutex.h>
+
#include <asm/current.h>
#include <asm/dma.h>
@@ -397,7 +399,7 @@ struct ess_state {
/* this locks around the oss state in the driver */
spinlock_t lock;
/* only let 1 be opening at a time */
- struct semaphore open_sem;
+ struct mutex open_mutex;
wait_queue_head_t open_wait;
mode_t open_mode;
@@ -3020,26 +3022,26 @@ ess_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EWOULDBLOCK;
}
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
interruptible_sleep_on(&s->open_wait);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
/* under semaphore.. */
if ((s->card->dmapages==NULL) && allocate_buffers(s)) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -ENOMEM;
}
- /* we're covered by the open_sem */
+ /* we're covered by the open_mutex */
if( ! s->card->dsps_open ) {
maestro_power(s->card,ACPI_D0);
start_bob(s);
@@ -3076,7 +3078,7 @@ ess_open(struct inode *inode, struct file *file)
set_fmt(s, fmtm, fmts);
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -3089,7 +3091,7 @@ ess_release(struct inode *inode, struct file *file)
lock_kernel();
if (file->f_mode & FMODE_WRITE)
drain_dac(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac(s);
}
@@ -3098,7 +3100,7 @@ ess_release(struct inode *inode, struct file *file)
}
s->open_mode &= (~file->f_mode) & (FMODE_READ|FMODE_WRITE);
- /* we're covered by the open_sem */
+ /* we're covered by the open_mutex */
M_printk("maestro: %d dsps now alive\n",s->card->dsps_open-1);
if( --s->card->dsps_open <= 0) {
s->card->dsps_open = 0;
@@ -3106,7 +3108,7 @@ ess_release(struct inode *inode, struct file *file)
free_buffers(s);
maestro_power(s->card,ACPI_D2);
}
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
wake_up(&s->open_wait);
unlock_kernel();
return 0;
@@ -3466,7 +3468,7 @@ maestro_probe(struct pci_dev *pcidev,const struct pci_device_id *pdid)
init_waitqueue_head(&s->dma_dac.wait);
init_waitqueue_head(&s->open_wait);
spin_lock_init(&s->lock);
- init_MUTEX(&s->open_sem);
+ mutex_init(&s->open_mutex);
s->magic = ESS_STATE_MAGIC;
s->apu[0] = 6*i;
diff --git a/sound/oss/maestro3.c b/sound/oss/maestro3.c
index f3dec70fcb9b..66044aff2586 100644
--- a/sound/oss/maestro3.c
+++ b/sound/oss/maestro3.c
@@ -144,6 +144,8 @@
#include <linux/spinlock.h>
#include <linux/ac97_codec.h>
#include <linux/wait.h>
+#include <linux/mutex.h>
+
#include <asm/io.h>
#include <asm/dma.h>
@@ -205,7 +207,7 @@ struct m3_state {
when irqhandler uses s->lock
and m3_assp_read uses card->lock ?
*/
- struct semaphore open_sem;
+ struct mutex open_mutex;
wait_queue_head_t open_wait;
mode_t open_mode;
@@ -2013,17 +2015,17 @@ static int m3_open(struct inode *inode, struct file *file)
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EWOULDBLOCK;
}
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
interruptible_sleep_on(&s->open_wait);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
spin_lock_irqsave(&c->lock, flags);
@@ -2047,7 +2049,7 @@ static int m3_open(struct inode *inode, struct file *file)
set_fmt(s, fmtm, fmts);
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
spin_unlock_irqrestore(&c->lock, flags);
return nonseekable_open(inode, file);
}
@@ -2062,7 +2064,7 @@ static int m3_release(struct inode *inode, struct file *file)
if (file->f_mode & FMODE_WRITE)
drain_dac(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
spin_lock_irqsave(&card->lock, flags);
if (file->f_mode & FMODE_WRITE) {
@@ -2083,7 +2085,7 @@ static int m3_release(struct inode *inode, struct file *file)
s->open_mode &= (~file->f_mode) & (FMODE_READ|FMODE_WRITE);
spin_unlock_irqrestore(&card->lock, flags);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
wake_up(&s->open_wait);
return 0;
@@ -2679,7 +2681,7 @@ static int __devinit m3_probe(struct pci_dev *pci_dev, const struct pci_device_i
init_waitqueue_head(&s->dma_adc.wait);
init_waitqueue_head(&s->dma_dac.wait);
init_waitqueue_head(&s->open_wait);
- init_MUTEX(&(s->open_sem));
+ mutex_init(&(s->open_mutex));
s->magic = M3_STATE_MAGIC;
m3_assp_client_init(s);
diff --git a/sound/oss/nec_vrc5477.c b/sound/oss/nec_vrc5477.c
index fbb9170e8e0a..21c1954d9108 100644
--- a/sound/oss/nec_vrc5477.c
+++ b/sound/oss/nec_vrc5477.c
@@ -78,6 +78,8 @@
#include <linux/spinlock.h>
#include <linux/smp_lock.h>
#include <linux/ac97_codec.h>
+#include <linux/mutex.h>
+
#include <asm/io.h>
#include <asm/dma.h>
#include <asm/uaccess.h>
@@ -198,7 +200,7 @@ struct vrc5477_ac97_state {
unsigned short extended_status;
spinlock_t lock;
- struct semaphore open_sem;
+ struct mutex open_mutex;
mode_t open_mode;
wait_queue_head_t open_wait;
@@ -1617,22 +1619,22 @@ static int vrc5477_ac97_open(struct inode *inode, struct file *file)
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
spin_lock_irqsave(&s->lock, flags);
@@ -1659,7 +1661,7 @@ static int vrc5477_ac97_open(struct inode *inode, struct file *file)
bailout:
spin_unlock_irqrestore(&s->lock, flags);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return ret;
}
@@ -1671,7 +1673,7 @@ static int vrc5477_ac97_release(struct inode *inode, struct file *file)
lock_kernel();
if (file->f_mode & FMODE_WRITE)
drain_dac(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac(s);
dealloc_dmabuf(s, &s->dma_dac);
@@ -1681,7 +1683,7 @@ static int vrc5477_ac97_release(struct inode *inode, struct file *file)
dealloc_dmabuf(s, &s->dma_adc);
}
s->open_mode &= (~file->f_mode) & (FMODE_READ|FMODE_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
wake_up(&s->open_wait);
unlock_kernel();
return 0;
@@ -1867,7 +1869,7 @@ static int __devinit vrc5477_ac97_probe(struct pci_dev *pcidev,
init_waitqueue_head(&s->dma_adc.wait);
init_waitqueue_head(&s->dma_dac.wait);
init_waitqueue_head(&s->open_wait);
- init_MUTEX(&s->open_sem);
+ mutex_init(&s->open_mutex);
spin_lock_init(&s->lock);
s->dev = pcidev;
diff --git a/sound/oss/rme96xx.c b/sound/oss/rme96xx.c
index faa0b7919b65..a1ec9d131ab3 100644
--- a/sound/oss/rme96xx.c
+++ b/sound/oss/rme96xx.c
@@ -58,6 +58,7 @@ TODO:
#include <linux/interrupt.h>
#include <linux/poll.h>
#include <linux/wait.h>
+#include <linux/mutex.h>
#include <asm/dma.h>
#include <asm/page.h>
@@ -326,7 +327,7 @@ typedef struct _rme96xx_info {
/* waiting and locking */
wait_queue_head_t wait;
- struct semaphore open_sem;
+ struct mutex open_mutex;
wait_queue_head_t open_wait;
} dma[RME96xx_MAX_DEVS];
@@ -842,7 +843,7 @@ static void busmaster_free(void* ptr,int size) {
static int rme96xx_dmabuf_init(rme96xx_info * s,struct dmabuf* dma,int ioffset,int ooffset) {
- init_MUTEX(&dma->open_sem);
+ mutex_init(&dma->open_mutex);
init_waitqueue_head(&dma->open_wait);
init_waitqueue_head(&dma->wait);
dma->s = s;
@@ -1469,21 +1470,21 @@ static int rme96xx_open(struct inode *in, struct file *f)
dma = &s->dma[devnum];
f->private_data = dma;
/* wait for device to become free */
- down(&dma->open_sem);
+ mutex_lock(&dma->open_mutex);
while (dma->open_mode & f->f_mode) {
if (f->f_flags & O_NONBLOCK) {
- up(&dma->open_sem);
+ mutex_unlock(&dma->open_mutex);
return -EBUSY;
}
add_wait_queue(&dma->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&dma->open_sem);
+ mutex_unlock(&dma->open_mutex);
schedule();
remove_wait_queue(&dma->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&dma->open_sem);
+ mutex_lock(&dma->open_mutex);
}
COMM ("hardware open")
@@ -1492,7 +1493,7 @@ static int rme96xx_open(struct inode *in, struct file *f)
dma->open_mode |= (f->f_mode & (FMODE_READ | FMODE_WRITE));
dma->opened = 1;
- up(&dma->open_sem);
+ mutex_unlock(&dma->open_mutex);
DBG(printk("device num %d open finished\n",devnum));
return 0;
@@ -1524,7 +1525,7 @@ static int rme96xx_release(struct inode *in, struct file *file)
}
wake_up(&dma->open_wait);
- up(&dma->open_sem);
+ mutex_unlock(&dma->open_mutex);
return 0;
}
diff --git a/sound/oss/sonicvibes.c b/sound/oss/sonicvibes.c
index 71b05e2f6977..69a4b8778b51 100644
--- a/sound/oss/sonicvibes.c
+++ b/sound/oss/sonicvibes.c
@@ -116,6 +116,8 @@
#include <linux/spinlock.h>
#include <linux/smp_lock.h>
#include <linux/gameport.h>
+#include <linux/mutex.h>
+
#include <asm/io.h>
#include <asm/uaccess.h>
@@ -328,7 +330,7 @@ struct sv_state {
unsigned char fmt, enable;
spinlock_t lock;
- struct semaphore open_sem;
+ struct mutex open_mutex;
mode_t open_mode;
wait_queue_head_t open_wait;
@@ -1922,21 +1924,21 @@ static int sv_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & file->f_mode) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
if (file->f_mode & FMODE_READ) {
fmtm &= ~((SV_CFMT_STEREO | SV_CFMT_16BIT) << SV_CFMT_CSHIFT);
@@ -1956,7 +1958,7 @@ static int sv_open(struct inode *inode, struct file *file)
}
set_fmt(s, fmtm, fmts);
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -1968,7 +1970,7 @@ static int sv_release(struct inode *inode, struct file *file)
lock_kernel();
if (file->f_mode & FMODE_WRITE)
drain_dac(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac(s);
dealloc_dmabuf(s, &s->dma_dac);
@@ -1979,7 +1981,7 @@ static int sv_release(struct inode *inode, struct file *file)
}
s->open_mode &= ~(file->f_mode & (FMODE_READ|FMODE_WRITE));
wake_up(&s->open_wait);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
unlock_kernel();
return 0;
}
@@ -2167,21 +2169,21 @@ static int sv_midi_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & (file->f_mode << FMODE_MIDI_SHIFT)) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
spin_lock_irqsave(&s->lock, flags);
if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2210,7 +2212,7 @@ static int sv_midi_open(struct inode *inode, struct file *file)
}
spin_unlock_irqrestore(&s->lock, flags);
s->open_mode |= (file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ | FMODE_MIDI_WRITE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -2248,7 +2250,7 @@ static int sv_midi_release(struct inode *inode, struct file *file)
remove_wait_queue(&s->midi.owait, &wait);
set_current_state(TASK_RUNNING);
}
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
s->open_mode &= ~((file->f_mode << FMODE_MIDI_SHIFT) & (FMODE_MIDI_READ|FMODE_MIDI_WRITE));
spin_lock_irqsave(&s->lock, flags);
if (!(s->open_mode & (FMODE_MIDI_READ | FMODE_MIDI_WRITE))) {
@@ -2257,7 +2259,7 @@ static int sv_midi_release(struct inode *inode, struct file *file)
}
spin_unlock_irqrestore(&s->lock, flags);
wake_up(&s->open_wait);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
unlock_kernel();
return 0;
}
@@ -2388,21 +2390,21 @@ static int sv_dmfm_open(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
file->private_data = s;
/* wait for device to become free */
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
while (s->open_mode & FMODE_DMFM) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return -EBUSY;
}
add_wait_queue(&s->open_wait, &wait);
__set_current_state(TASK_INTERRUPTIBLE);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
schedule();
remove_wait_queue(&s->open_wait, &wait);
set_current_state(TASK_RUNNING);
if (signal_pending(current))
return -ERESTARTSYS;
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
}
/* init the stuff */
outb(1, s->iosynth);
@@ -2412,7 +2414,7 @@ static int sv_dmfm_open(struct inode *inode, struct file *file)
outb(5, s->iosynth+2);
outb(1, s->iosynth+3); /* enable OPL3 */
s->open_mode |= FMODE_DMFM;
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
return nonseekable_open(inode, file);
}
@@ -2423,7 +2425,7 @@ static int sv_dmfm_release(struct inode *inode, struct file *file)
VALIDATE_STATE(s);
lock_kernel();
- down(&s->open_sem);
+ mutex_lock(&s->open_mutex);
s->open_mode &= ~FMODE_DMFM;
for (regb = 0xb0; regb < 0xb9; regb++) {
outb(regb, s->iosynth);
@@ -2432,7 +2434,7 @@ static int sv_dmfm_release(struct inode *inode, struct file *file)
outb(0, s->iosynth+3);
}
wake_up(&s->open_wait);
- up(&s->open_sem);
+ mutex_unlock(&s->open_mutex);
unlock_kernel();
return 0;
}
@@ -2582,7 +2584,7 @@ static int __devinit sv_probe(struct pci_dev *pcidev, const struct pci_device_id
init_waitqueue_head(&s->open_wait);
init_waitqueue_head(&s->midi.iwait);
init_waitqueue_head(&s->midi.owait);
- init_MUTEX(&s->open_sem);
+ mutex_init(&s->open_mutex);
spin_lock_init(&s->lock);
s->magic = SV_MAGIC;
s->dev = pcidev;
diff --git a/sound/oss/swarm_cs4297a.c b/sound/oss/swarm_cs4297a.c
index df4d3771fa84..dce9016cbcfd 100644
--- a/sound/oss/swarm_cs4297a.c
+++ b/sound/oss/swarm_cs4297a.c
@@ -76,6 +76,7 @@
#include <linux/init.h>
#include <linux/poll.h>
#include <linux/smp_lock.h>
+#include <linux/mutex.h>
#include <asm/byteorder.h>
#include <asm/dma.h>
@@ -291,9 +292,9 @@ struct cs4297a_state {
unsigned conversion:1; // conversion from 16 to 8 bit in progress
unsigned ena;
spinlock_t lock;
- struct semaphore open_sem;
- struct semaphore open_sem_adc;
- struct semaphore open_sem_dac;
+ struct mutex open_mutex;
+ struct mutex open_sem_adc;
+ struct mutex open_sem_dac;
mode_t open_mode;
wait_queue_head_t open_wait;
wait_queue_head_t open_wait_adc;
@@ -2352,20 +2353,20 @@ static int cs4297a_release(struct inode *inode, struct file *file)
if (file->f_mode & FMODE_WRITE) {
drain_dac(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem_dac);
+ mutex_lock(&s->open_sem_dac);
stop_dac(s);
dealloc_dmabuf(s, &s->dma_dac);
s->open_mode &= ~FMODE_WRITE;
- up(&s->open_sem_dac);
+ mutex_unlock(&s->open_sem_dac);
wake_up(&s->open_wait_dac);
}
if (file->f_mode & FMODE_READ) {
drain_adc(s, file->f_flags & O_NONBLOCK);
- down(&s->open_sem_adc);
+ mutex_lock(&s->open_sem_adc);
stop_adc(s);
dealloc_dmabuf(s, &s->dma_adc);
s->open_mode &= ~FMODE_READ;
- up(&s->open_sem_adc);
+ mutex_unlock(&s->open_sem_adc);
wake_up(&s->open_wait_adc);
}
return 0;
@@ -2413,37 +2414,37 @@ static int cs4297a_open(struct inode *inode, struct file *file)
;
}
- down(&s->open_sem_dac);
+ mutex_lock(&s->open_sem_dac);
while (s->open_mode & FMODE_WRITE) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem_dac);
+ mutex_unlock(&s->open_sem_dac);
return -EBUSY;
}
- up(&s->open_sem_dac);
+ mutex_unlock(&s->open_sem_dac);
interruptible_sleep_on(&s->open_wait_dac);
if (signal_pending(current)) {
printk("open - sig pending\n");
return -ERESTARTSYS;
}
- down(&s->open_sem_dac);
+ mutex_lock(&s->open_sem_dac);
}
}
if (file->f_mode & FMODE_READ) {
- down(&s->open_sem_adc);
+ mutex_lock(&s->open_sem_adc);
while (s->open_mode & FMODE_READ) {
if (file->f_flags & O_NONBLOCK) {
- up(&s->open_sem_adc);
+ mutex_unlock(&s->open_sem_adc);
return -EBUSY;
}
- up(&s->open_sem_adc);
+ mutex_unlock(&s->open_sem_adc);
interruptible_sleep_on(&s->open_wait_adc);
if (signal_pending(current)) {
printk("open - sig pending\n");
return -ERESTARTSYS;
}
- down(&s->open_sem_adc);
+ mutex_lock(&s->open_sem_adc);
}
}
s->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
@@ -2456,7 +2457,7 @@ static int cs4297a_open(struct inode *inode, struct file *file)
s->ena &= ~FMODE_READ;
s->dma_adc.ossfragshift = s->dma_adc.ossmaxfrags =
s->dma_adc.subdivision = 0;
- up(&s->open_sem_adc);
+ mutex_unlock(&s->open_sem_adc);
if (prog_dmabuf_adc(s)) {
CS_DBGOUT(CS_OPEN | CS_ERROR, 2, printk(KERN_ERR
@@ -2474,7 +2475,7 @@ static int cs4297a_open(struct inode *inode, struct file *file)
s->ena &= ~FMODE_WRITE;
s->dma_dac.ossfragshift = s->dma_dac.ossmaxfrags =
s->dma_dac.subdivision = 0;
- up(&s->open_sem_dac);
+ mutex_unlock(&s->open_sem_dac);
if (prog_dmabuf_dac(s)) {
CS_DBGOUT(CS_OPEN | CS_ERROR, 2, printk(KERN_ERR
@@ -2631,8 +2632,8 @@ static int __init cs4297a_init(void)
init_waitqueue_head(&s->open_wait);
init_waitqueue_head(&s->open_wait_adc);
init_waitqueue_head(&s->open_wait_dac);
- init_MUTEX(&s->open_sem_adc);
- init_MUTEX(&s->open_sem_dac);
+ mutex_init(&s->open_sem_adc);
+ mutex_init(&s->open_sem_dac);
spin_lock_init(&s->lock);
s->irq = K_INT_SER_1;
diff --git a/sound/oss/trident.c b/sound/oss/trident.c
index a21c663e7e12..e61a454a8150 100644
--- a/sound/oss/trident.c
+++ b/sound/oss/trident.c
@@ -190,7 +190,7 @@
*
* Lock order (high->low)
* lock - hardware lock
- * open_sem - guard opens
+ * open_mutex - guard opens
* sem - guard dmabuf, write re-entry etc
*/
@@ -216,6 +216,8 @@
#include <linux/pm.h>
#include <linux/gameport.h>
#include <linux/kernel.h>
+#include <linux/mutex.h>
+
#include <asm/uaccess.h>
#include <asm/io.h>
#include <asm/dma.h>
@@ -349,7 +351,7 @@ struct trident_state {
unsigned chans_num;
unsigned long fmt_flag;
/* Guard against mmap/write/read races */
- struct semaphore sem;
+ struct mutex sem;
};
@@ -402,7 +404,7 @@ struct trident_card {
struct trident_card *next;
/* single open lock mechanism, only used for recording */
- struct semaphore open_sem;
+ struct mutex open_mutex;
/* The trident has a certain amount of cross channel interaction
so we use a single per card lock */
@@ -1881,7 +1883,7 @@ trident_read(struct file *file, char __user *buffer, size_t count, loff_t * ppos
if (!access_ok(VERIFY_WRITE, buffer, count))
return -EFAULT;
- down(&state->sem);
+ mutex_lock(&state->sem);
if (!dmabuf->ready && (ret = prog_dmabuf_record(state)))
goto out;
@@ -1913,7 +1915,7 @@ trident_read(struct file *file, char __user *buffer, size_t count, loff_t * ppos
goto out;
}
- up(&state->sem);
+ mutex_unlock(&state->sem);
/* No matter how much space left in the buffer, */
/* we have to wait until CSO == ESO/2 or CSO == ESO */
/* when address engine interrupts */
@@ -1940,7 +1942,7 @@ trident_read(struct file *file, char __user *buffer, size_t count, loff_t * ppos
ret = -ERESTARTSYS;
goto out;
}
- down(&state->sem);
+ mutex_lock(&state->sem);
if (dmabuf->mapped) {
if (!ret)
ret = -ENXIO;
@@ -1968,7 +1970,7 @@ trident_read(struct file *file, char __user *buffer, size_t count, loff_t * ppos
start_adc(state);
}
out:
- up(&state->sem);
+ mutex_unlock(&state->sem);
return ret;
}
@@ -1996,7 +1998,7 @@ trident_write(struct file *file, const char __user *buffer, size_t count, loff_t
* Guard against an mmap or ioctl while writing
*/
- down(&state->sem);
+ mutex_lock(&state->sem);
if (dmabuf->mapped) {
ret = -ENXIO;
@@ -2045,7 +2047,7 @@ trident_write(struct file *file, const char __user *buffer, size_t count, loff_t
tmo = (dmabuf->dmasize * HZ) / (dmabuf->rate * 2);
tmo >>= sample_shift[dmabuf->fmt];
unlock_set_fmt(state);
- up(&state->sem);
+ mutex_unlock(&state->sem);
/* There are two situations when sleep_on_timeout */
/* returns, one is when the interrupt is serviced */
@@ -2073,7 +2075,7 @@ trident_write(struct file *file, const char __user *buffer, size_t count, loff_t
ret = -ERESTARTSYS;
goto out_nolock;
}
- down(&state->sem);
+ mutex_lock(&state->sem);
if (dmabuf->mapped) {
if (!ret)
ret = -ENXIO;
@@ -2131,7 +2133,7 @@ trident_write(struct file *file, const char __user *buffer, size_t count, loff_t
start_dac(state);
}
out:
- up(&state->sem);
+ mutex_unlock(&state->sem);
out_nolock:
return ret;
}
@@ -2152,24 +2154,24 @@ trident_poll(struct file *file, struct poll_table_struct *wait)
* prog_dmabuf events
*/
- down(&state->sem);
+ mutex_lock(&state->sem);
if (file->f_mode & FMODE_WRITE) {
if (!dmabuf->ready && prog_dmabuf_playback(state)) {
- up(&state->sem);
+ mutex_unlock(&state->sem);
return 0;
}
poll_wait(file, &dmabuf->wait, wait);
}
if (file->f_mode & FMODE_READ) {
if (!dmabuf->ready && prog_dmabuf_record(state)) {
- up(&state->sem);
+ mutex_unlock(&state->sem);
return 0;
}
poll_wait(file, &dmabuf->wait, wait);
}
- up(&state->sem);
+ mutex_unlock(&state->sem);
spin_lock_irqsave(&state->card->lock, flags);
trident_update_ptr(state);
@@ -2207,7 +2209,7 @@ trident_mmap(struct file *file, struct vm_area_struct *vma)
* a read or write against an mmap.
*/
- down(&state->sem);
+ mutex_lock(&state->sem);
if (vma->vm_flags & VM_WRITE) {
if ((ret = prog_dmabuf_playback(state)) != 0)
@@ -2232,7 +2234,7 @@ trident_mmap(struct file *file, struct vm_area_struct *vma)
dmabuf->mapped = 1;
ret = 0;
out:
- up(&state->sem);
+ mutex_unlock(&state->sem);
return ret;
}
@@ -2429,15 +2431,15 @@ trident_ioctl(struct inode *inode, struct file *file,
unlock_set_fmt(state);
break;
}
- down(&state->card->open_sem);
+ mutex_lock(&state->card->open_mutex);
ret = ali_allocate_other_states_resources(state, 6);
if (ret < 0) {
- up(&state->card->open_sem);
+ mutex_unlock(&state->card->open_mutex);
unlock_set_fmt(state);
break;
}
state->card->multi_channel_use_count++;
- up(&state->card->open_sem);
+ mutex_unlock(&state->card->open_mutex);
} else
val = 2; /*yield to 2-channels */
} else
@@ -2727,11 +2729,11 @@ trident_open(struct inode *inode, struct file *file)
/* find an available virtual channel (instance of /dev/dsp) */
while (card != NULL) {
- down(&card->open_sem);
+ mutex_lock(&card->open_mutex);
if (file->f_mode & FMODE_READ) {
/* Skip opens on cards that are in 6 channel mode */
if (card->multi_channel_use_count > 0) {
- up(&card->open_sem);
+ mutex_unlock(&card->open_mutex);
card = card->next;
continue;
}
@@ -2740,16 +2742,16 @@ trident_open(struct inode *inode, struct file *file)
if (card->states[i] == NULL) {
state = card->states[i] = kmalloc(sizeof(*state), GFP_KERNEL);
if (state == NULL) {
- up(&card->open_sem);
+ mutex_unlock(&card->open_mutex);
return -ENOMEM;
}
memset(state, 0, sizeof(*state));
- init_MUTEX(&state->sem);
+ mutex_init(&state->sem);
dmabuf = &state->dmabuf;
goto found_virt;
}
}
- up(&card->open_sem);
+ mutex_unlock(&card->open_mutex);
card = card->next;
}
/* no more virtual channel avaiable */
@@ -2816,7 +2818,7 @@ trident_open(struct inode *inode, struct file *file)
}
state->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&card->open_sem);
+ mutex_unlock(&card->open_mutex);
pr_debug("trident: open virtual channel %d, hard channel %d\n",
state->virt, dmabuf->channel->num);
@@ -2845,7 +2847,7 @@ trident_release(struct inode *inode, struct file *file)
state->virt, dmabuf->channel->num);
/* stop DMA state machine and free DMA buffers/channels */
- down(&card->open_sem);
+ mutex_lock(&card->open_mutex);
if (file->f_mode & FMODE_WRITE) {
stop_dac(state);
@@ -2878,8 +2880,8 @@ trident_release(struct inode *inode, struct file *file)
card->states[state->virt] = NULL;
kfree(state);
- /* we're covered by the open_sem */
- up(&card->open_sem);
+ /* we're covered by the open_mutex */
+ mutex_unlock(&card->open_mutex);
return 0;
}
@@ -4405,7 +4407,7 @@ trident_probe(struct pci_dev *pci_dev, const struct pci_device_id *pci_id)
card->banks[BANK_B].addresses = &bank_b_addrs;
card->banks[BANK_B].bitmap = 0UL;
- init_MUTEX(&card->open_sem);
+ mutex_init(&card->open_mutex);
spin_lock_init(&card->lock);
init_timer(&card->timer);
diff --git a/sound/oss/via82cxxx_audio.c b/sound/oss/via82cxxx_audio.c
index 83edda93f0b4..1a921ee71aba 100644
--- a/sound/oss/via82cxxx_audio.c
+++ b/sound/oss/via82cxxx_audio.c
@@ -38,7 +38,8 @@
#include <linux/dma-mapping.h>
#include <asm/io.h>
#include <asm/uaccess.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
+
#include "sound_config.h"
#include "dev_table.h"
#include "mpu401.h"
@@ -311,8 +312,8 @@ struct via_info {
int mixer_vol; /* 8233/35 volume - not yet implemented */
- struct semaphore syscall_sem;
- struct semaphore open_sem;
+ struct mutex syscall_mutex;
+ struct mutex open_mutex;
/* The 8233/8235 have 4 DX audio channels, two record and
one six channel out. We bind ch_in to DX 1, ch_out to multichannel
@@ -505,10 +506,10 @@ static inline int via_syscall_down (struct via_info *card, int nonblock)
nonblock = 0;
if (nonblock) {
- if (down_trylock (&card->syscall_sem))
+ if (!mutex_trylock(&card->syscall_mutex))
return -EAGAIN;
} else {
- if (down_interruptible (&card->syscall_sem))
+ if (mutex_lock_interruptible(&card->syscall_mutex))
return -ERESTARTSYS;
}
@@ -1609,7 +1610,7 @@ static int via_mixer_ioctl (struct inode *inode, struct file *file, unsigned int
#endif
rc = codec->mixer_ioctl(codec, cmd, arg);
- up (&card->syscall_sem);
+ mutex_unlock(&card->syscall_mutex);
out:
DPRINTK ("EXIT, returning %d\n", rc);
@@ -2228,7 +2229,7 @@ static int via_dsp_mmap(struct file *file, struct vm_area_struct *vma)
if (wr)
card->ch_out.is_mapped = 1;
- up (&card->syscall_sem);
+ mutex_unlock(&card->syscall_mutex);
rc = 0;
out:
@@ -2256,7 +2257,7 @@ handle_one_block:
/* Thomas Sailer:
* But also to ourselves, release semaphore if we do so */
if (need_resched()) {
- up(&card->syscall_sem);
+ mutex_unlock(&card->syscall_mutex);
schedule ();
ret = via_syscall_down (card, nonblock);
if (ret)
@@ -2286,7 +2287,7 @@ handle_one_block:
break;
}
- up(&card->syscall_sem);
+ mutex_unlock(&card->syscall_mutex);
DPRINTK ("Sleeping on block %d\n", n);
schedule();
@@ -2402,7 +2403,7 @@ static ssize_t via_dsp_read(struct file *file, char __user *buffer, size_t count
rc = via_dsp_do_read (card, buffer, count, nonblock);
out_up:
- up (&card->syscall_sem);
+ mutex_unlock(&card->syscall_mutex);
out:
DPRINTK ("EXIT, returning %ld\n",(long) rc);
return rc;
@@ -2426,7 +2427,7 @@ handle_one_block:
/* Thomas Sailer:
* But also to ourselves, release semaphore if we do so */
if (need_resched()) {
- up(&card->syscall_sem);
+ mutex_unlock(&card->syscall_mutex);
schedule ();
ret = via_syscall_down (card, nonblock);
if (ret)
@@ -2456,7 +2457,7 @@ handle_one_block:
break;
}
- up(&card->syscall_sem);
+ mutex_unlock(&card->syscall_mutex);
DPRINTK ("Sleeping on page %d, tmp==%d, ir==%d\n", n, tmp, chan->is_record);
schedule();
@@ -2585,7 +2586,7 @@ static ssize_t via_dsp_write(struct file *file, const char __user *buffer, size_
rc = via_dsp_do_write (card, buffer, count, nonblock);
out_up:
- up (&card->syscall_sem);
+ mutex_unlock(&card->syscall_mutex);
out:
DPRINTK ("EXIT, returning %ld\n",(long) rc);
return rc;
@@ -2634,7 +2635,7 @@ static unsigned int via_dsp_poll(struct file *file, struct poll_table_struct *wa
* Sleeps until all playback has been flushed to the audio
* hardware.
*
- * Locking: inside card->syscall_sem
+ * Locking: inside card->syscall_mutex
*/
static int via_dsp_drain_playback (struct via_info *card,
@@ -2692,7 +2693,7 @@ static int via_dsp_drain_playback (struct via_info *card,
printk (KERN_ERR "sleeping but not active\n");
#endif
- up(&card->syscall_sem);
+ mutex_unlock(&card->syscall_mutex);
DPRINTK ("sleeping, nbufs=%d\n", atomic_read (&chan->n_frags));
schedule();
@@ -2748,7 +2749,7 @@ out:
*
* Handles SNDCTL_DSP_GETISPACE and SNDCTL_DSP_GETOSPACE.
*
- * Locking: inside card->syscall_sem
+ * Locking: inside card->syscall_mutex
*/
static int via_dsp_ioctl_space (struct via_info *card,
@@ -2793,7 +2794,7 @@ static int via_dsp_ioctl_space (struct via_info *card,
*
* Handles SNDCTL_DSP_GETIPTR and SNDCTL_DSP_GETOPTR.
*
- * Locking: inside card->syscall_sem
+ * Locking: inside card->syscall_mutex
*/
static int via_dsp_ioctl_ptr (struct via_info *card,
@@ -3221,7 +3222,7 @@ static int via_dsp_ioctl (struct inode *inode, struct file *file,
break;
}
- up (&card->syscall_sem);
+ mutex_unlock(&card->syscall_mutex);
DPRINTK ("EXIT, returning %d\n", rc);
return rc;
}
@@ -3264,12 +3265,12 @@ static int via_dsp_open (struct inode *inode, struct file *file)
match:
if (nonblock) {
- if (down_trylock (&card->open_sem)) {
+ if (!mutex_trylock(&card->open_mutex)) {
DPRINTK ("EXIT, returning -EAGAIN\n");
return -EAGAIN;
}
} else {
- if (down_interruptible (&card->open_sem)) {
+ if (mutex_lock_interruptible(&card->open_mutex)) {
DPRINTK ("EXIT, returning -ERESTARTSYS\n");
return -ERESTARTSYS;
}
@@ -3355,8 +3356,8 @@ static int via_dsp_release(struct inode *inode, struct file *file)
via_chan_buffer_free (card, &card->ch_in);
}
- up (&card->syscall_sem);
- up (&card->open_sem);
+ mutex_unlock(&card->syscall_mutex);
+ mutex_unlock(&card->open_mutex);
DPRINTK ("EXIT, returning 0\n");
return 0;
@@ -3414,8 +3415,8 @@ static int __devinit via_init_one (struct pci_dev *pdev, const struct pci_device
card->card_num = via_num_cards++;
spin_lock_init (&card->lock);
spin_lock_init (&card->ac97_lock);
- init_MUTEX (&card->syscall_sem);
- init_MUTEX (&card->open_sem);
+ mutex_init(&card->syscall_mutex);
+ mutex_init(&card->open_mutex);
/* we must init these now, in case the intr handler needs them */
via_chan_init_defaults (card, &card->ch_out);
diff --git a/sound/oss/vwsnd.c b/sound/oss/vwsnd.c
index 265423054caf..b372e88e857f 100644
--- a/sound/oss/vwsnd.c
+++ b/sound/oss/vwsnd.c
@@ -94,7 +94,7 @@
* Open will block until the previous client has closed the
* device, unless O_NONBLOCK is specified.
*
- * The semaphore devc->io_sema serializes PCM I/O syscalls. This
+ * The semaphore devc->io_mutex serializes PCM I/O syscalls. This
* is unnecessary in Linux 2.2, because the kernel lock
* serializes read, write, and ioctl globally, but it's there,
* ready for the brave, new post-kernel-lock world.
@@ -105,7 +105,7 @@
* area it owns and update its pointers. See pcm_output() and
* pcm_input() for most of the gory stuff.
*
- * devc->mix_sema serializes all mixer ioctls. This is also
+ * devc->mix_mutex serializes all mixer ioctls. This is also
* redundant because of the kernel lock.
*
* The lowest level lock is lith->lithium_lock. It is a
@@ -148,7 +148,8 @@
#include <linux/smp_lock.h>
#include <linux/wait.h>
#include <linux/interrupt.h>
-#include <asm/semaphore.h>
+#include <linux/mutex.h>
+
#include <asm/mach-visws/cobalt.h>
#include "sound_config.h"
@@ -1447,11 +1448,11 @@ typedef enum vwsnd_port_flags {
*
* port->lock protects: hwstate, flags, swb_[iu]_avail.
*
- * devc->io_sema protects: swstate, sw_*, swb_[iu]_idx.
+ * devc->io_mutex protects: swstate, sw_*, swb_[iu]_idx.
*
* everything else is only written by open/release or
* pcm_{setup,shutdown}(), which are serialized by a
- * combination of devc->open_sema and devc->io_sema.
+ * combination of devc->open_mutex and devc->io_mutex.
*/
typedef struct vwsnd_port {
@@ -1507,9 +1508,9 @@ typedef struct vwsnd_dev {
int audio_minor; /* minor number of audio device */
int mixer_minor; /* minor number of mixer device */
- struct semaphore open_sema;
- struct semaphore io_sema;
- struct semaphore mix_sema;
+ struct mutex open_mutex;
+ struct mutex io_mutex;
+ struct mutex mix_mutex;
mode_t open_mode;
wait_queue_head_t open_wait;
@@ -1633,7 +1634,7 @@ static __inline__ unsigned int swb_inc_i(vwsnd_port_t *port, int inc)
* mode-setting ioctls have been done, but before the first I/O is
* done.
*
- * Locking: called with devc->io_sema held.
+ * Locking: called with devc->io_mutex held.
*
* Returns 0 on success, -errno on failure.
*/
@@ -2319,9 +2320,9 @@ static ssize_t vwsnd_audio_read(struct file *file,
vwsnd_dev_t *devc = file->private_data;
ssize_t ret;
- down(&devc->io_sema);
+ mutex_lock(&devc->io_mutex);
ret = vwsnd_audio_do_read(file, buffer, count, ppos);
- up(&devc->io_sema);
+ mutex_unlock(&devc->io_mutex);
return ret;
}
@@ -2394,9 +2395,9 @@ static ssize_t vwsnd_audio_write(struct file *file,
vwsnd_dev_t *devc = file->private_data;
ssize_t ret;
- down(&devc->io_sema);
+ mutex_lock(&devc->io_mutex);
ret = vwsnd_audio_do_write(file, buffer, count, ppos);
- up(&devc->io_sema);
+ mutex_unlock(&devc->io_mutex);
return ret;
}
@@ -2891,9 +2892,9 @@ static int vwsnd_audio_ioctl(struct inode *inode,
vwsnd_dev_t *devc = (vwsnd_dev_t *) file->private_data;
int ret;
- down(&devc->io_sema);
+ mutex_lock(&devc->io_mutex);
ret = vwsnd_audio_do_ioctl(inode, file, cmd, arg);
- up(&devc->io_sema);
+ mutex_unlock(&devc->io_mutex);
return ret;
}
@@ -2929,9 +2930,9 @@ static int vwsnd_audio_open(struct inode *inode, struct file *file)
return -ENODEV;
}
- down(&devc->open_sema);
+ mutex_lock(&devc->open_mutex);
while (devc->open_mode & file->f_mode) {
- up(&devc->open_sema);
+ mutex_unlock(&devc->open_mutex);
if (file->f_flags & O_NONBLOCK) {
DEC_USE_COUNT;
return -EBUSY;
@@ -2941,10 +2942,10 @@ static int vwsnd_audio_open(struct inode *inode, struct file *file)
DEC_USE_COUNT;
return -ERESTARTSYS;
}
- down(&devc->open_sema);
+ mutex_lock(&devc->open_mutex);
}
devc->open_mode |= file->f_mode & (FMODE_READ | FMODE_WRITE);
- up(&devc->open_sema);
+ mutex_unlock(&devc->open_mutex);
/* get default sample format from minor number. */
@@ -2960,7 +2961,7 @@ static int vwsnd_audio_open(struct inode *inode, struct file *file)
/* Initialize vwsnd_ports. */
- down(&devc->io_sema);
+ mutex_lock(&devc->io_mutex);
{
if (file->f_mode & FMODE_READ) {
devc->rport.swstate = SW_INITIAL;
@@ -2987,7 +2988,7 @@ static int vwsnd_audio_open(struct inode *inode, struct file *file)
devc->wport.frag_count = 0;
}
}
- up(&devc->io_sema);
+ mutex_unlock(&devc->io_mutex);
file->private_data = devc;
DBGRV();
@@ -3005,7 +3006,7 @@ static int vwsnd_audio_release(struct inode *inode, struct file *file)
int err = 0;
lock_kernel();
- down(&devc->io_sema);
+ mutex_lock(&devc->io_mutex);
{
DBGEV("(inode=0x%p, file=0x%p)\n", inode, file);
@@ -3022,13 +3023,13 @@ static int vwsnd_audio_release(struct inode *inode, struct file *file)
if (wport)
wport->swstate = SW_OFF;
}
- up(&devc->io_sema);
+ mutex_unlock(&devc->io_mutex);
- down(&devc->open_sema);
+ mutex_lock(&devc->open_mutex);
{
devc->open_mode &= ~file->f_mode;
}
- up(&devc->open_sema);
+ mutex_unlock(&devc->open_mutex);
wake_up(&devc->open_wait);
DEC_USE_COUNT;
DBGR();
@@ -3213,7 +3214,7 @@ static int vwsnd_mixer_ioctl(struct inode *ioctl,
DBGEV("(devc=0x%p, cmd=0x%x, arg=0x%lx)\n", devc, cmd, arg);
- down(&devc->mix_sema);
+ mutex_lock(&devc->mix_mutex);
{
if ((cmd & ~nrmask) == MIXER_READ(0))
retval = mixer_read_ioctl(devc, nr, (void __user *) arg);
@@ -3222,7 +3223,7 @@ static int vwsnd_mixer_ioctl(struct inode *ioctl,
else
retval = -EINVAL;
}
- up(&devc->mix_sema);
+ mutex_unlock(&devc->mix_mutex);
return retval;
}
@@ -3376,9 +3377,9 @@ static int __init attach_vwsnd(struct address_info *hw_config)
/* Initialize as much of *devc as possible */
- init_MUTEX(&devc->open_sema);
- init_MUTEX(&devc->io_sema);
- init_MUTEX(&devc->mix_sema);
+ mutex_init(&devc->open_mutex);
+ mutex_init(&devc->io_mutex);
+ mutex_init(&devc->mix_mutex);
devc->open_mode = 0;
spin_lock_init(&devc->rport.lock);
init_waitqueue_head(&devc->rport.queue);
diff --git a/sound/oss/ymfpci.c b/sound/oss/ymfpci.c
index f8bd72e46f57..bf90c124a7e6 100644
--- a/sound/oss/ymfpci.c
+++ b/sound/oss/ymfpci.c
@@ -1918,10 +1918,10 @@ static int ymf_open(struct inode *inode, struct file *file)
if (unit == NULL)
return -ENODEV;
- down(&unit->open_sem);
+ mutex_lock(&unit->open_mutex);
if ((state = ymf_state_alloc(unit)) == NULL) {
- up(&unit->open_sem);
+ mutex_unlock(&unit->open_mutex);
return -ENOMEM;
}
list_add_tail(&state->chain, &unit->states);
@@ -1956,7 +1956,7 @@ static int ymf_open(struct inode *inode, struct file *file)
ymfpci_writeb(unit, YDSXGR_TIMERCTRL,
(YDSXGR_TIMERCTRL_TEN|YDSXGR_TIMERCTRL_TIEN));
#endif
- up(&unit->open_sem);
+ mutex_unlock(&unit->open_mutex);
return nonseekable_open(inode, file);
@@ -1974,7 +1974,7 @@ out_nodma:
list_del(&state->chain);
kfree(state);
- up(&unit->open_sem);
+ mutex_unlock(&unit->open_mutex);
return err;
}
@@ -1987,7 +1987,7 @@ static int ymf_release(struct inode *inode, struct file *file)
ymfpci_writeb(unit, YDSXGR_TIMERCTRL, 0);
#endif
- down(&unit->open_sem);
+ mutex_lock(&unit->open_mutex);
/*
* XXX Solve the case of O_NONBLOCK close - don't deallocate here.
@@ -2004,7 +2004,7 @@ static int ymf_release(struct inode *inode, struct file *file)
file->private_data = NULL; /* Can you tell I programmed Solaris */
kfree(state);
- up(&unit->open_sem);
+ mutex_unlock(&unit->open_mutex);
return 0;
}
@@ -2532,7 +2532,7 @@ static int __devinit ymf_probe_one(struct pci_dev *pcidev, const struct pci_devi
spin_lock_init(&codec->reg_lock);
spin_lock_init(&codec->voice_lock);
spin_lock_init(&codec->ac97_lock);
- init_MUTEX(&codec->open_sem);
+ mutex_init(&codec->open_mutex);
INIT_LIST_HEAD(&codec->states);
codec->pci = pcidev;
diff --git a/sound/oss/ymfpci.h b/sound/oss/ymfpci.h
index f810a100c641..ac1785f2b7e7 100644
--- a/sound/oss/ymfpci.h
+++ b/sound/oss/ymfpci.h
@@ -22,6 +22,7 @@
*
*/
#include <linux/config.h>
+#include <linux/mutex.h>
/*
* Direct registers
@@ -279,7 +280,7 @@ struct ymf_unit {
/* soundcore stuff */
int dev_audio;
- struct semaphore open_sem;
+ struct mutex open_mutex;
struct list_head ymf_devs;
struct list_head states; /* List of states for this unit */