diff options
Diffstat (limited to 'fs')
41 files changed, 446 insertions, 398 deletions
diff --git a/fs/btrfs/backref.c b/fs/btrfs/backref.c index b9a843226de8..633c701a287d 100644 --- a/fs/btrfs/backref.c +++ b/fs/btrfs/backref.c @@ -297,7 +297,7 @@ static int __add_delayed_refs(struct btrfs_delayed_ref_head *head, u64 seq, struct btrfs_delayed_extent_op *extent_op = head->extent_op; struct rb_node *n = &head->node.rb_node; int sgn; - int ret; + int ret = 0; if (extent_op && extent_op->update_key) btrfs_disk_key_to_cpu(info_key, &extent_op->key); @@ -392,7 +392,7 @@ static int __add_inline_refs(struct btrfs_fs_info *fs_info, struct btrfs_key *info_key, int *info_level, struct list_head *prefs) { - int ret; + int ret = 0; int slot; struct extent_buffer *leaf; struct btrfs_key key; diff --git a/fs/btrfs/check-integrity.c b/fs/btrfs/check-integrity.c index ad0b3ba735b7..b669a7d8e499 100644 --- a/fs/btrfs/check-integrity.c +++ b/fs/btrfs/check-integrity.c @@ -1662,7 +1662,7 @@ static void btrfsic_process_written_block(struct btrfsic_dev_state *dev_state, block = btrfsic_block_hashtable_lookup(bdev, dev_bytenr, &state->block_hashtable); if (NULL != block) { - u64 bytenr; + u64 bytenr = 0; struct list_head *elem_ref_to; struct list_head *tmp_ref_to; @@ -2777,9 +2777,10 @@ int btrfsic_submit_bh(int rw, struct buffer_head *bh) printk(KERN_INFO "submit_bh(rw=0x%x, blocknr=%lu (bytenr %llu)," " size=%lu, data=%p, bdev=%p)\n", - rw, bh->b_blocknr, - (unsigned long long)dev_bytenr, bh->b_size, - bh->b_data, bh->b_bdev); + rw, (unsigned long)bh->b_blocknr, + (unsigned long long)dev_bytenr, + (unsigned long)bh->b_size, bh->b_data, + bh->b_bdev); btrfsic_process_written_block(dev_state, dev_bytenr, bh->b_data, bh->b_size, NULL, NULL, bh, rw); @@ -2844,7 +2845,7 @@ void btrfsic_submit_bio(int rw, struct bio *bio) printk(KERN_INFO "submit_bio(rw=0x%x, bi_vcnt=%u," " bi_sector=%lu (bytenr %llu), bi_bdev=%p)\n", - rw, bio->bi_vcnt, bio->bi_sector, + rw, bio->bi_vcnt, (unsigned long)bio->bi_sector, (unsigned long long)dev_bytenr, bio->bi_bdev); diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c index 7aa9cd36bf1b..811d9f918b1c 100644 --- a/fs/btrfs/disk-io.c +++ b/fs/btrfs/disk-io.c @@ -962,6 +962,13 @@ static int btree_releasepage(struct page *page, gfp_t gfp_flags) tree = &BTRFS_I(page->mapping->host)->io_tree; map = &BTRFS_I(page->mapping->host)->extent_tree; + /* + * We need to mask out eg. __GFP_HIGHMEM and __GFP_DMA32 as we're doing + * slab allocation from alloc_extent_state down the callchain where + * it'd hit a BUG_ON as those flags are not allowed. + */ + gfp_flags &= ~GFP_SLAB_BUG_MASK; + ret = try_release_extent_state(map, tree, page, gfp_flags); if (!ret) return 0; diff --git a/fs/btrfs/extent-tree.c b/fs/btrfs/extent-tree.c index 700879ed64cf..283af7a676a3 100644 --- a/fs/btrfs/extent-tree.c +++ b/fs/btrfs/extent-tree.c @@ -34,23 +34,24 @@ #include "locking.h" #include "free-space-cache.h" -/* control flags for do_chunk_alloc's force field +/* + * control flags for do_chunk_alloc's force field * CHUNK_ALLOC_NO_FORCE means to only allocate a chunk * if we really need one. * - * CHUNK_ALLOC_FORCE means it must try to allocate one - * * CHUNK_ALLOC_LIMITED means to only try and allocate one * if we have very few chunks already allocated. This is * used as part of the clustering code to help make sure * we have a good pool of storage to cluster in, without * filling the FS with empty chunks * + * CHUNK_ALLOC_FORCE means it must try to allocate one + * */ enum { CHUNK_ALLOC_NO_FORCE = 0, - CHUNK_ALLOC_FORCE = 1, - CHUNK_ALLOC_LIMITED = 2, + CHUNK_ALLOC_LIMITED = 1, + CHUNK_ALLOC_FORCE = 2, }; /* @@ -3414,7 +3415,7 @@ static int do_chunk_alloc(struct btrfs_trans_handle *trans, again: spin_lock(&space_info->lock); - if (space_info->force_alloc) + if (force < space_info->force_alloc) force = space_info->force_alloc; if (space_info->full) { spin_unlock(&space_info->lock); @@ -5794,6 +5795,7 @@ int btrfs_reserve_extent(struct btrfs_trans_handle *trans, u64 search_end, struct btrfs_key *ins, u64 data) { + bool final_tried = false; int ret; u64 search_start = 0; @@ -5813,22 +5815,25 @@ again: search_start, search_end, hint_byte, ins, data); - if (ret == -ENOSPC && num_bytes > min_alloc_size) { - num_bytes = num_bytes >> 1; - num_bytes = num_bytes & ~(root->sectorsize - 1); - num_bytes = max(num_bytes, min_alloc_size); - do_chunk_alloc(trans, root->fs_info->extent_root, - num_bytes, data, CHUNK_ALLOC_FORCE); - goto again; - } - if (ret == -ENOSPC && btrfs_test_opt(root, ENOSPC_DEBUG)) { - struct btrfs_space_info *sinfo; - - sinfo = __find_space_info(root->fs_info, data); - printk(KERN_ERR "btrfs allocation failed flags %llu, " - "wanted %llu\n", (unsigned long long)data, - (unsigned long long)num_bytes); - dump_space_info(sinfo, num_bytes, 1); + if (ret == -ENOSPC) { + if (!final_tried) { + num_bytes = num_bytes >> 1; + num_bytes = num_bytes & ~(root->sectorsize - 1); + num_bytes = max(num_bytes, min_alloc_size); + do_chunk_alloc(trans, root->fs_info->extent_root, + num_bytes, data, CHUNK_ALLOC_FORCE); + if (num_bytes == min_alloc_size) + final_tried = true; + goto again; + } else if (btrfs_test_opt(root, ENOSPC_DEBUG)) { + struct btrfs_space_info *sinfo; + + sinfo = __find_space_info(root->fs_info, data); + printk(KERN_ERR "btrfs allocation failed flags %llu, " + "wanted %llu\n", (unsigned long long)data, + (unsigned long long)num_bytes); + dump_space_info(sinfo, num_bytes, 1); + } } trace_btrfs_reserved_extent_alloc(root, ins->objectid, ins->offset); diff --git a/fs/btrfs/extent_io.c b/fs/btrfs/extent_io.c index 9d09a4f81875..fcf77e1ded40 100644 --- a/fs/btrfs/extent_io.c +++ b/fs/btrfs/extent_io.c @@ -3909,6 +3909,8 @@ int extent_range_uptodate(struct extent_io_tree *tree, while (start <= end) { index = start >> PAGE_CACHE_SHIFT; page = find_get_page(tree->mapping, index); + if (!page) + return 1; uptodate = PageUptodate(page); page_cache_release(page); if (!uptodate) { diff --git a/fs/btrfs/free-space-cache.c b/fs/btrfs/free-space-cache.c index d20ff87ca603..c2f20594c9f7 100644 --- a/fs/btrfs/free-space-cache.c +++ b/fs/btrfs/free-space-cache.c @@ -2242,7 +2242,7 @@ u64 btrfs_alloc_from_cluster(struct btrfs_block_group_cache *block_group, if (entry->bitmap) { ret = btrfs_alloc_from_bitmap(block_group, cluster, entry, bytes, - min_start); + cluster->window_start); if (ret == 0) { node = rb_next(&entry->offset_index); if (!node) @@ -2251,6 +2251,7 @@ u64 btrfs_alloc_from_cluster(struct btrfs_block_group_cache *block_group, offset_index); continue; } + cluster->window_start += bytes; } else { ret = entry->offset; @@ -2475,7 +2476,7 @@ setup_cluster_bitmap(struct btrfs_block_group_cache *block_group, } list_for_each_entry(entry, bitmaps, list) { - if (entry->bytes < min_bytes) + if (entry->bytes < bytes) continue; ret = btrfs_bitmap_cluster(block_group, entry, cluster, offset, bytes, cont1_bytes, min_bytes); diff --git a/fs/btrfs/inode.c b/fs/btrfs/inode.c index 0da19a0ea00d..32214fe0f7e3 100644 --- a/fs/btrfs/inode.c +++ b/fs/btrfs/inode.c @@ -6401,18 +6401,23 @@ int btrfs_page_mkwrite(struct vm_area_struct *vma, struct vm_fault *vmf) unsigned long zero_start; loff_t size; int ret; + int reserved = 0; u64 page_start; u64 page_end; ret = btrfs_delalloc_reserve_space(inode, PAGE_CACHE_SIZE); - if (!ret) + if (!ret) { ret = btrfs_update_time(vma->vm_file); + reserved = 1; + } if (ret) { if (ret == -ENOMEM) ret = VM_FAULT_OOM; else /* -ENOSPC, -EIO, etc */ ret = VM_FAULT_SIGBUS; - goto out; + if (reserved) + goto out; + goto out_noreserve; } ret = VM_FAULT_NOPAGE; /* make the VM retry the fault */ @@ -6495,6 +6500,7 @@ out_unlock: unlock_page(page); out: btrfs_delalloc_release_space(inode, PAGE_CACHE_SIZE); +out_noreserve: return ret; } diff --git a/fs/btrfs/ioctl.c b/fs/btrfs/ioctl.c index ab620014bcc3..03bb62a9ee24 100644 --- a/fs/btrfs/ioctl.c +++ b/fs/btrfs/ioctl.c @@ -1065,7 +1065,7 @@ int btrfs_defrag_file(struct inode *inode, struct file *file, i = range->start >> PAGE_CACHE_SHIFT; } if (!max_to_defrag) - max_to_defrag = last_index; + max_to_defrag = last_index + 1; /* * make writeback starts from i, so the defrag range can be diff --git a/fs/btrfs/tree-log.c b/fs/btrfs/tree-log.c index cb877e0886a7..966cc74f5d6c 100644 --- a/fs/btrfs/tree-log.c +++ b/fs/btrfs/tree-log.c @@ -1957,7 +1957,8 @@ static int wait_log_commit(struct btrfs_trans_handle *trans, finish_wait(&root->log_commit_wait[index], &wait); mutex_lock(&root->log_mutex); - } while (root->log_transid < transid + 2 && + } while (root->fs_info->last_trans_log_full_commit != + trans->transid && root->log_transid < transid + 2 && atomic_read(&root->log_commit[index])); return 0; } @@ -1966,7 +1967,8 @@ static int wait_for_writer(struct btrfs_trans_handle *trans, struct btrfs_root *root) { DEFINE_WAIT(wait); - while (atomic_read(&root->log_writers)) { + while (root->fs_info->last_trans_log_full_commit != + trans->transid && atomic_read(&root->log_writers)) { prepare_to_wait(&root->log_writer_wait, &wait, TASK_UNINTERRUPTIBLE); mutex_unlock(&root->log_mutex); diff --git a/fs/ceph/caps.c b/fs/ceph/caps.c index b60fc8bfb3e9..620daad201db 100644 --- a/fs/ceph/caps.c +++ b/fs/ceph/caps.c @@ -641,10 +641,10 @@ static int __cap_is_valid(struct ceph_cap *cap) unsigned long ttl; u32 gen; - spin_lock(&cap->session->s_cap_lock); + spin_lock(&cap->session->s_gen_ttl_lock); gen = cap->session->s_cap_gen; ttl = cap->session->s_cap_ttl; - spin_unlock(&cap->session->s_cap_lock); + spin_unlock(&cap->session->s_gen_ttl_lock); if (cap->cap_gen < gen || time_after_eq(jiffies, ttl)) { dout("__cap_is_valid %p cap %p issued %s " diff --git a/fs/ceph/dir.c b/fs/ceph/dir.c index 618246bc2196..3e8094be4604 100644 --- a/fs/ceph/dir.c +++ b/fs/ceph/dir.c @@ -975,10 +975,10 @@ static int dentry_lease_is_valid(struct dentry *dentry) di = ceph_dentry(dentry); if (di->lease_session) { s = di->lease_session; - spin_lock(&s->s_cap_lock); + spin_lock(&s->s_gen_ttl_lock); gen = s->s_cap_gen; ttl = s->s_cap_ttl; - spin_unlock(&s->s_cap_lock); + spin_unlock(&s->s_gen_ttl_lock); if (di->lease_gen == gen && time_before(jiffies, dentry->d_time) && diff --git a/fs/ceph/mds_client.c b/fs/ceph/mds_client.c index 23ab6a3f1825..866e8d7ca37d 100644 --- a/fs/ceph/mds_client.c +++ b/fs/ceph/mds_client.c @@ -262,6 +262,7 @@ static int parse_reply_info(struct ceph_msg *msg, /* trace */ ceph_decode_32_safe(&p, end, len, bad); if (len > 0) { + ceph_decode_need(&p, end, len, bad); err = parse_reply_info_trace(&p, p+len, info, features); if (err < 0) goto out_bad; @@ -270,6 +271,7 @@ static int parse_reply_info(struct ceph_msg *msg, /* extra */ ceph_decode_32_safe(&p, end, len, bad); if (len > 0) { + ceph_decode_need(&p, end, len, bad); err = parse_reply_info_extra(&p, p+len, info, features); if (err < 0) goto out_bad; @@ -398,9 +400,11 @@ static struct ceph_mds_session *register_session(struct ceph_mds_client *mdsc, s->s_con.peer_name.type = CEPH_ENTITY_TYPE_MDS; s->s_con.peer_name.num = cpu_to_le64(mds); - spin_lock_init(&s->s_cap_lock); + spin_lock_init(&s->s_gen_ttl_lock); s->s_cap_gen = 0; s->s_cap_ttl = 0; + + spin_lock_init(&s->s_cap_lock); s->s_renew_requested = 0; s->s_renew_seq = 0; INIT_LIST_HEAD(&s->s_caps); @@ -2326,10 +2330,10 @@ static void handle_session(struct ceph_mds_session *session, case CEPH_SESSION_STALE: pr_info("mds%d caps went stale, renewing\n", session->s_mds); - spin_lock(&session->s_cap_lock); + spin_lock(&session->s_gen_ttl_lock); session->s_cap_gen++; session->s_cap_ttl = 0; - spin_unlock(&session->s_cap_lock); + spin_unlock(&session->s_gen_ttl_lock); send_renew_caps(mdsc, session); break; diff --git a/fs/ceph/mds_client.h b/fs/ceph/mds_client.h index a50ca0e39475..8c7c04ebb595 100644 --- a/fs/ceph/mds_client.h +++ b/fs/ceph/mds_client.h @@ -117,10 +117,13 @@ struct ceph_mds_session { void *s_authorizer_buf, *s_authorizer_reply_buf; size_t s_authorizer_buf_len, s_authorizer_reply_buf_len; - /* protected by s_cap_lock */ - spinlock_t s_cap_lock; + /* protected by s_gen_ttl_lock */ + spinlock_t s_gen_ttl_lock; u32 s_cap_gen; /* inc each time we get mds stale msg */ unsigned long s_cap_ttl; /* when session caps expire */ + + /* protected by s_cap_lock */ + spinlock_t s_cap_lock; struct list_head s_caps; /* all caps issued by this session */ int s_nr_caps, s_trim_caps; int s_num_cap_releases; diff --git a/fs/ceph/xattr.c b/fs/ceph/xattr.c index 857214ae8c08..a76f697303d9 100644 --- a/fs/ceph/xattr.c +++ b/fs/ceph/xattr.c @@ -111,8 +111,10 @@ static size_t ceph_vxattrcb_layout(struct ceph_inode_info *ci, char *val, } static struct ceph_vxattr_cb ceph_file_vxattrs[] = { + { true, "ceph.file.layout", ceph_vxattrcb_layout}, + /* The following extended attribute name is deprecated */ { true, "ceph.layout", ceph_vxattrcb_layout}, - { NULL, NULL } + { true, NULL, NULL } }; static struct ceph_vxattr_cb *ceph_inode_vxattrs(struct inode *inode) diff --git a/fs/ecryptfs/crypto.c b/fs/ecryptfs/crypto.c index 2a834255c75d..63ab24510649 100644 --- a/fs/ecryptfs/crypto.c +++ b/fs/ecryptfs/crypto.c @@ -417,17 +417,6 @@ static int ecryptfs_encrypt_extent(struct page *enc_extent_page, (unsigned long long)(extent_base + extent_offset), rc); goto out; } - if (unlikely(ecryptfs_verbosity > 0)) { - ecryptfs_printk(KERN_DEBUG, "Encrypting extent " - "with iv:\n"); - ecryptfs_dump_hex(extent_iv, crypt_stat->iv_bytes); - ecryptfs_printk(KERN_DEBUG, "First 8 bytes before " - "encryption:\n"); - ecryptfs_dump_hex((char *) - (page_address(page) - + (extent_offset * crypt_stat->extent_size)), - 8); - } rc = ecryptfs_encrypt_page_offset(crypt_stat, enc_extent_page, 0, page, (extent_offset * crypt_stat->extent_size), @@ -440,14 +429,6 @@ static int ecryptfs_encrypt_extent(struct page *enc_extent_page, goto out; } rc = 0; - if (unlikely(ecryptfs_verbosity > 0)) { - ecryptfs_printk(KERN_DEBUG, "Encrypt extent [0x%.16llx]; " - "rc = [%d]\n", - (unsigned long long)(extent_base + extent_offset), rc); - ecryptfs_printk(KERN_DEBUG, "First 8 bytes after " - "encryption:\n"); - ecryptfs_dump_hex((char *)(page_address(enc_extent_page)), 8); - } out: return rc; } @@ -543,17 +524,6 @@ static int ecryptfs_decrypt_extent(struct page *page, (unsigned long long)(extent_base + extent_offset), rc); goto out; } - if (unlikely(ecryptfs_verbosity > 0)) { - ecryptfs_printk(KERN_DEBUG, "Decrypting extent " - "with iv:\n"); - ecryptfs_dump_hex(extent_iv, crypt_stat->iv_bytes); - ecryptfs_printk(KERN_DEBUG, "First 8 bytes before " - "decryption:\n"); - ecryptfs_dump_hex((char *) - (page_address(enc_extent_page) - + (extent_offset * crypt_stat->extent_size)), - 8); - } rc = ecryptfs_decrypt_page_offset(crypt_stat, page, (extent_offset * crypt_stat->extent_size), @@ -567,16 +537,6 @@ static int ecryptfs_decrypt_extent(struct page *page, goto out; } rc = 0; - if (unlikely(ecryptfs_verbosity > 0)) { - ecryptfs_printk(KERN_DEBUG, "Decrypt extent [0x%.16llx]; " - "rc = [%d]\n", - (unsigned long long)(extent_base + extent_offset), rc); - ecryptfs_printk(KERN_DEBUG, "First 8 bytes after " - "decryption:\n"); - ecryptfs_dump_hex((char *)(page_address(page) - + (extent_offset - * crypt_stat->extent_size)), 8); - } out: return rc; } @@ -1590,8 +1550,8 @@ int ecryptfs_read_and_validate_xattr_region(struct dentry *dentry, */ int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry) { - int rc = 0; - char *page_virt = NULL; + int rc; + char *page_virt; struct inode *ecryptfs_inode = ecryptfs_dentry->d_inode; struct ecryptfs_crypt_stat *crypt_stat = &ecryptfs_inode_to_private(ecryptfs_inode)->crypt_stat; @@ -1616,11 +1576,13 @@ int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry) ecryptfs_dentry, ECRYPTFS_VALIDATE_HEADER_SIZE); if (rc) { + /* metadata is not in the file header, so try xattrs */ memset(page_virt, 0, PAGE_CACHE_SIZE); rc = ecryptfs_read_xattr_region(page_virt, ecryptfs_inode); if (rc) { printk(KERN_DEBUG "Valid eCryptfs headers not found in " - "file header region or xattr region\n"); + "file header region or xattr region, inode %lu\n", + ecryptfs_inode->i_ino); rc = -EINVAL; goto out; } @@ -1629,7 +1591,8 @@ int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry) ECRYPTFS_DONT_VALIDATE_HEADER_SIZE); if (rc) { printk(KERN_DEBUG "Valid eCryptfs headers not found in " - "file xattr region either\n"); + "file xattr region either, inode %lu\n", + ecryptfs_inode->i_ino); rc = -EINVAL; } if (crypt_stat->mount_crypt_stat->flags @@ -1640,7 +1603,8 @@ int ecryptfs_read_metadata(struct dentry *ecryptfs_dentry) "crypto metadata only in the extended attribute " "region, but eCryptfs was mounted without " "xattr support enabled. eCryptfs will not treat " - "this like an encrypted file.\n"); + "this like an encrypted file, inode %lu\n", + ecryptfs_inode->i_ino); rc = -EINVAL; } } diff --git a/fs/ecryptfs/ecryptfs_kernel.h b/fs/ecryptfs/ecryptfs_kernel.h index a9f29b12fbf2..a2362df58ae8 100644 --- a/fs/ecryptfs/ecryptfs_kernel.h +++ b/fs/ecryptfs/ecryptfs_kernel.h @@ -151,6 +151,11 @@ ecryptfs_get_key_payload_data(struct key *key) * dentry name */ #define ECRYPTFS_TAG_73_PACKET_TYPE 0x49 /* FEK-encrypted filename as * metadata */ +#define ECRYPTFS_MIN_PKT_LEN_SIZE 1 /* Min size to specify packet length */ +#define ECRYPTFS_MAX_PKT_LEN_SIZE 2 /* Pass at least this many bytes to + * ecryptfs_parse_packet_length() and + * ecryptfs_write_packet_length() + */ /* Constraint: ECRYPTFS_FILENAME_MIN_RANDOM_PREPEND_BYTES >= * ECRYPTFS_MAX_IV_BYTES */ #define ECRYPTFS_FILENAME_MIN_RANDOM_PREPEND_BYTES 16 diff --git a/fs/ecryptfs/inode.c b/fs/ecryptfs/inode.c index 19a8ca4ab1dd..19892d7d2ed1 100644 --- a/fs/ecryptfs/inode.c +++ b/fs/ecryptfs/inode.c @@ -822,18 +822,6 @@ static int truncate_upper(struct dentry *dentry, struct iattr *ia, size_t num_zeros = (PAGE_CACHE_SIZE - (ia->ia_size & ~PAGE_CACHE_MASK)); - - /* - * XXX(truncate) this should really happen at the begginning - * of ->setattr. But the code is too messy to that as part - * of a larger patch. ecryptfs is also totally missing out - * on the inode_change_ok check at the beginning of - * ->setattr while would include this. - */ - rc = inode_newsize_ok(inode, ia->ia_size); - if (rc) - goto out; - if (!(crypt_stat->flags & ECRYPTFS_ENCRYPTED)) { truncate_setsize(inode, ia->ia_size); lower_ia->ia_size = ia->ia_size; @@ -883,6 +871,28 @@ out: return rc; } +static int ecryptfs_inode_newsize_ok(struct inode *inode, loff_t offset) +{ + struct ecryptfs_crypt_stat *crypt_stat; + loff_t lower_oldsize, lower_newsize; + + crypt_stat = &ecryptfs_inode_to_private(inode)->crypt_stat; + lower_oldsize = upper_size_to_lower_size(crypt_stat, + i_size_read(inode)); + lower_newsize = upper_size_to_lower_size(crypt_stat, offset); + if (lower_newsize > lower_oldsize) { + /* + * The eCryptfs inode and the new *lower* size are mixed here + * because we may not have the lower i_mutex held and/or it may + * not be appropriate to call inode_newsize_ok() with inodes + * from other filesystems. + */ + return inode_newsize_ok(inode, lower_newsize); + } + + return 0; +} + /** * ecryptfs_truncate * @dentry: The ecryptfs layer dentry @@ -899,6 +909,10 @@ int ecryptfs_truncate(struct dentry *dentry, loff_t new_length) struct iattr lower_ia = { .ia_valid = 0 }; int rc; + rc = ecryptfs_inode_newsize_ok(dentry->d_inode, new_length); + if (rc) + return rc; + rc = truncate_upper(dentry, &ia, &lower_ia); if (!rc && lower_ia.ia_valid & ATTR_SIZE) { struct dentry *lower_dentry = ecryptfs_dentry_to_lower(dentry); @@ -978,6 +992,16 @@ static int ecryptfs_setattr(struct dentry *dentry, struct iattr *ia) } } mutex_unlock(&crypt_stat->cs_mutex); + + rc = inode_change_ok(inode, ia); + if (rc) + goto out; + if (ia->ia_valid & ATTR_SIZE) { + rc = ecryptfs_inode_newsize_ok(inode, ia->ia_size); + if (rc) + goto out; + } + if (S_ISREG(inode->i_mode)) { rc = filemap_write_and_wait(inode->i_mapping); if (rc) diff --git a/fs/ecryptfs/keystore.c b/fs/ecryptfs/keystore.c index ac1ad48c2376..8e3b943e330f 100644 --- a/fs/ecryptfs/keystore.c +++ b/fs/ecryptfs/keystore.c @@ -109,7 +109,7 @@ int ecryptfs_parse_packet_length(unsigned char *data, size_t *size, (*size) += ((unsigned char)(data[1]) + 192); (*length_size) = 2; } else if (data[0] == 255) { - /* Five-byte length; we're not supposed to see this */ + /* If support is added, adjust ECRYPTFS_MAX_PKT_LEN_SIZE */ ecryptfs_printk(KERN_ERR, "Five-byte packet length not " "supported\n"); rc = -EINVAL; @@ -126,7 +126,7 @@ out: /** * ecryptfs_write_packet_length * @dest: The byte array target into which to write the length. Must - * have at least 5 bytes allocated. + * have at least ECRYPTFS_MAX_PKT_LEN_SIZE bytes allocated. * @size: The length to write. * @packet_size_length: The number of bytes used to encode the packet * length is written to this address. @@ -146,6 +146,7 @@ int ecryptfs_write_packet_length(char *dest, size_t size, dest[1] = ((size - 192) % 256); (*packet_size_length) = 2; } else { + /* If support is added, adjust ECRYPTFS_MAX_PKT_LEN_SIZE */ rc = -EINVAL; ecryptfs_printk(KERN_WARNING, "Unsupported packet size: [%zd]\n", size); diff --git a/fs/ecryptfs/miscdev.c b/fs/ecryptfs/miscdev.c index 940a82e63dc3..349209dc6a91 100644 --- a/fs/ecryptfs/miscdev.c +++ b/fs/ecryptfs/miscdev.c @@ -218,6 +218,29 @@ out_unlock: return rc; } +/* + * miscdevfs packet format: + * Octet 0: Type + * Octets 1-4: network byte order msg_ctx->counter + * Octets 5-N0: Size of struct ecryptfs_message to follow + * Octets N0-N1: struct ecryptfs_message (including data) + * + * Octets 5-N1 not written if the packet type does not include a message + */ +#define PKT_TYPE_SIZE 1 +#define PKT_CTR_SIZE 4 +#define MIN_NON_MSG_PKT_SIZE (PKT_TYPE_SIZE + PKT_CTR_SIZE) +#define MIN_MSG_PKT_SIZE (PKT_TYPE_SIZE + PKT_CTR_SIZE \ + + ECRYPTFS_MIN_PKT_LEN_SIZE) +/* 4 + ECRYPTFS_MAX_ENCRYPTED_KEY_BYTES comes from tag 65 packet format */ +#define MAX_MSG_PKT_SIZE (PKT_TYPE_SIZE + PKT_CTR_SIZE \ + + ECRYPTFS_MAX_PKT_LEN_SIZE \ + + sizeof(struct ecryptfs_message) \ + + 4 + ECRYPTFS_MAX_ENCRYPTED_KEY_BYTES) +#define PKT_TYPE_OFFSET 0 +#define PKT_CTR_OFFSET PKT_TYPE_SIZE +#define PKT_LEN_OFFSET (PKT_TYPE_SIZE + PKT_CTR_SIZE) + /** * ecryptfs_miscdev_read - format and send message from queue * @file: fs/ecryptfs/euid miscdevfs handle (ignored) @@ -237,7 +260,7 @@ ecryptfs_miscdev_read(struct file *file, char __user *buf, size_t count, struct ecryptfs_daemon *daemon; struct ecryptfs_msg_ctx *msg_ctx; size_t packet_length_size; - char packet_length[3]; + char packet_length[ECRYPTFS_MAX_PKT_LEN_SIZE]; size_t i; size_t total_length; uid_t euid = current_euid(); @@ -305,15 +328,8 @@ check_list: packet_length_size = 0; msg_ctx->msg_size = 0; } - /* miscdevfs packet format: - * Octet 0: Type - * Octets 1-4: network byte order msg_ctx->counter - * Octets 5-N0: Size of struct ecryptfs_message to follow - * Octets N0-N1: struct ecryptfs_message (including data) - * - * Octets 5-N1 not written if the packet type does not - * include a message */ - total_length = (1 + 4 + packet_length_size + msg_ctx->msg_size); + total_length = (PKT_TYPE_SIZE + PKT_CTR_SIZE + packet_length_size + + msg_ctx->msg_size); if (count < total_length) { rc = 0; printk(KERN_WARNING "%s: Only given user buffer of " @@ -324,9 +340,10 @@ check_list: rc = -EFAULT; if (put_user(msg_ctx->type, buf)) goto out_unlock_msg_ctx; - if (put_user(cpu_to_be32(msg_ctx->counter), (__be32 __user *)(buf + 1))) + if (put_user(cpu_to_be32(msg_ctx->counter), + (__be32 __user *)(&buf[PKT_CTR_OFFSET]))) goto out_unlock_msg_ctx; - i = 5; + i = PKT_TYPE_SIZE + PKT_CTR_SIZE; if (msg_ctx->msg) { if (copy_to_user(&buf[i], packet_length, packet_length_size)) goto out_unlock_msg_ctx; @@ -391,12 +408,6 @@ out: * @count: Amount of data in @buf * @ppos: Pointer to offset in file (ignored) * - * miscdevfs packet format: - * Octet 0: Type - * Octets 1-4: network byte order msg_ctx->counter (0's for non-response) - * Octets 5-N0: Size of struct ecryptfs_message to follow - * Octets N0-N1: struct ecryptfs_message (including data) - * * Returns the number of bytes read from @buf */ static ssize_t @@ -405,60 +416,78 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf, { __be32 counter_nbo; u32 seq; - size_t packet_size, packet_size_length, i; - ssize_t sz = 0; + size_t packet_size, packet_size_length; char *data; uid_t euid = current_euid(); - int rc; + unsigned char packet_size_peek[ECRYPTFS_MAX_PKT_LEN_SIZE]; + ssize_t rc; - if (count == 0) - goto out; + if (count == 0) { + return 0; + } else if (count == MIN_NON_MSG_PKT_SIZE) { + /* Likely a harmless MSG_HELO or MSG_QUIT - no packet length */ + goto memdup; + } else if (count < MIN_MSG_PKT_SIZE || count > MAX_MSG_PKT_SIZE) { + printk(KERN_WARNING "%s: Acceptable packet size range is " + "[%d-%lu], but amount of data written is [%zu].", + __func__, MIN_MSG_PKT_SIZE, MAX_MSG_PKT_SIZE, count); + return -EINVAL; + } + + if (copy_from_user(packet_size_peek, &buf[PKT_LEN_OFFSET], + sizeof(packet_size_peek))) { + printk(KERN_WARNING "%s: Error while inspecting packet size\n", + __func__); + return -EFAULT; + } + rc = ecryptfs_parse_packet_length(packet_size_peek, &packet_size, + &packet_size_length); + if (rc) { + printk(KERN_WARNING "%s: Error parsing packet length; " + "rc = [%zd]\n", __func__, rc); + return rc; + } + + if ((PKT_TYPE_SIZE + PKT_CTR_SIZE + packet_size_length + packet_size) + != count) { + printk(KERN_WARNING "%s: Invalid packet size [%zu]\n", __func__, + packet_size); + return -EINVAL; + } + +memdup: data = memdup_user(buf, count); if (IS_ERR(data)) { printk(KERN_ERR "%s: memdup_user returned error [%ld]\n", __func__, PTR_ERR(data)); - goto out; + return PTR_ERR(data); } - sz = count; - i = 0; - switch (data[i++]) { + switch (data[PKT_TYPE_OFFSET]) { case ECRYPTFS_MSG_RESPONSE: - if (count < (1 + 4 + 1 + sizeof(struct ecryptfs_message))) { + if (count < (MIN_MSG_PKT_SIZE + + sizeof(struct ecryptfs_message))) { printk(KERN_WARNING "%s: Minimum acceptable packet " "size is [%zd], but amount of data written is " "only [%zd]. Discarding response packet.\n", __func__, - (1 + 4 + 1 + sizeof(struct ecryptfs_message)), - count); + (MIN_MSG_PKT_SIZE + + sizeof(struct ecryptfs_message)), count); + rc = -EINVAL; goto out_free; } - memcpy(&counter_nbo, &data[i], 4); + memcpy(&counter_nbo, &data[PKT_CTR_OFFSET], PKT_CTR_SIZE); seq = be32_to_cpu(counter_nbo); - i += 4; - rc = ecryptfs_parse_packet_length(&data[i], &packet_size, - &packet_size_length); + rc = ecryptfs_miscdev_response( + &data[PKT_LEN_OFFSET + packet_size_length], + packet_size, euid, current_user_ns(), + task_pid(current), seq); if (rc) { - printk(KERN_WARNING "%s: Error parsing packet length; " - "rc = [%d]\n", __func__, rc); - goto out_free; - } - i += packet_size_length; - if ((1 + 4 + packet_size_length + packet_size) != count) { - printk(KERN_WARNING "%s: (1 + packet_size_length([%zd])" - " + packet_size([%zd]))([%zd]) != " - "count([%zd]). Invalid packet format.\n", - __func__, packet_size_length, packet_size, - (1 + packet_size_length + packet_size), count); - goto out_free; - } - rc = ecryptfs_miscdev_response(&data[i], packet_size, - euid, current_user_ns(), - task_pid(current), seq); - if (rc) printk(KERN_WARNING "%s: Failed to deliver miscdev " - "response to requesting operation; rc = [%d]\n", + "response to requesting operation; rc = [%zd]\n", __func__, rc); + goto out_free; + } break; case ECRYPTFS_MSG_HELO: case ECRYPTFS_MSG_QUIT: @@ -467,12 +496,13 @@ ecryptfs_miscdev_write(struct file *file, const char __user *buf, ecryptfs_printk(KERN_WARNING, "Dropping miscdev " "message of unrecognized type [%d]\n", data[0]); - break; + rc = -EINVAL; + goto out_free; } + rc = count; out_free: kfree(data); -out: - return sz; + return rc; } diff --git a/fs/ecryptfs/mmap.c b/fs/ecryptfs/mmap.c index 6a44148c5fb9..10ec695ccd68 100644 --- a/fs/ecryptfs/mmap.c +++ b/fs/ecryptfs/mmap.c @@ -57,6 +57,10 @@ struct page *ecryptfs_get_locked_page(struct inode *inode, loff_t index) * @page: Page that is locked before this call is made * * Returns zero on success; non-zero otherwise + * + * This is where we encrypt the data and pass the encrypted data to + * the lower filesystem. In OpenPGP-compatible mode, we operate on + * entire underlying packets. */ static int ecryptfs_writepage(struct page *page, struct writeback_control *wbc) { @@ -481,10 +485,6 @@ int ecryptfs_write_inode_size_to_metadata(struct inode *ecryptfs_inode) * @copied: The amount of data copied * @page: The eCryptfs page * @fsdata: The fsdata (unused) - * - * This is where we encrypt the data and pass the encrypted data to - * the lower filesystem. In OpenPGP-compatible mode, we operate on - * entire underlying packets. */ static int ecryptfs_write_end(struct file *file, struct address_space *mapping, diff --git a/fs/ecryptfs/read_write.c b/fs/ecryptfs/read_write.c index 3745f7c2b9c2..5c0106f75775 100644 --- a/fs/ecryptfs/read_write.c +++ b/fs/ecryptfs/read_write.c @@ -130,13 +130,18 @@ int ecryptfs_write(struct inode *ecryptfs_inode, char *data, loff_t offset, pgoff_t ecryptfs_page_idx = (pos >> PAGE_CACHE_SHIFT); size_t start_offset_in_page = (pos & ~PAGE_CACHE_MASK); size_t num_bytes = (PAGE_CACHE_SIZE - start_offset_in_page); - size_t total_remaining_bytes = ((offset + size) - pos); + loff_t total_remaining_bytes = ((offset + size) - pos); + + if (fatal_signal_pending(current)) { + rc = -EINTR; + break; + } if (num_bytes > total_remaining_bytes) num_bytes = total_remaining_bytes; if (pos < offset) { /* remaining zeros to write, up to destination offset */ - size_t total_remaining_zeros = (offset - pos); + loff_t total_remaining_zeros = (offset - pos); if (num_bytes > total_remaining_zeros) num_bytes = total_remaining_zeros; @@ -193,15 +198,19 @@ int ecryptfs_write(struct inode *ecryptfs_inode, char *data, loff_t offset, } pos += num_bytes; } - if ((offset + size) > ecryptfs_file_size) { - i_size_write(ecryptfs_inode, (offset + size)); + if (pos > ecryptfs_file_size) { + i_size_write(ecryptfs_inode, pos); if (crypt_stat->flags & ECRYPTFS_ENCRYPTED) { - rc = ecryptfs_write_inode_size_to_metadata( + int rc2; + + rc2 = ecryptfs_write_inode_size_to_metadata( ecryptfs_inode); - if (rc) { + if (rc2) { printk(KERN_ERR "Problem with " "ecryptfs_write_inode_size_to_metadata; " - "rc = [%d]\n", rc); + "rc = [%d]\n", rc2); + if (!rc) + rc = rc2; goto out; } } @@ -273,76 +282,3 @@ int ecryptfs_read_lower_page_segment(struct page *page_for_ecryptfs, flush_dcache_page(page_for_ecryptfs); return rc; } - -#if 0 -/** - * ecryptfs_read - * @data: The virtual address into which to write the data read (and - * possibly decrypted) from the lower file - * @offset: The offset in the decrypted view of the file from which to - * read into @data - * @size: The number of bytes to read into @data - * @ecryptfs_file: The eCryptfs file from which to read - * - * Read an arbitrary amount of data from an arbitrary location in the - * eCryptfs page cache. This is done on an extent-by-extent basis; - * individual extents are decrypted and read from the lower page - * cache (via VFS reads). This function takes care of all the - * address translation to locations in the lower filesystem. - * - * Returns zero on success; non-zero otherwise - */ -int ecryptfs_read(char *data, loff_t offset, size_t size, - struct file *ecryptfs_file) -{ - struct inode *ecryptfs_inode = ecryptfs_file->f_dentry->d_inode; - struct page *ecryptfs_page; - char *ecryptfs_page_virt; - loff_t ecryptfs_file_size = i_size_read(ecryptfs_inode); - loff_t data_offset = 0; - loff_t pos; - int rc = 0; - - if ((offset + size) > ecryptfs_file_size) { - rc = -EINVAL; - printk(KERN_ERR "%s: Attempt to read data past the end of the " - "file; offset = [%lld]; size = [%td]; " - "ecryptfs_file_size = [%lld]\n", - __func__, offset, size, ecryptfs_file_size); - goto out; - } - pos = offset; - while (pos < (offset + size)) { - pgoff_t ecryptfs_page_idx = (pos >> PAGE_CACHE_SHIFT); - size_t start_offset_in_page = (pos & ~PAGE_CACHE_MASK); - size_t num_bytes = (PAGE_CACHE_SIZE - start_offset_in_page); - size_t total_remaining_bytes = ((offset + size) - pos); - - if (num_bytes > total_remaining_bytes) - num_bytes = total_remaining_bytes; - ecryptfs_page = ecryptfs_get_locked_page(ecryptfs_inode, - ecryptfs_page_idx); - if (IS_ERR(ecryptfs_page)) { - rc = PTR_ERR(ecryptfs_page); - printk(KERN_ERR "%s: Error getting page at " - "index [%ld] from eCryptfs inode " - "mapping; rc = [%d]\n", __func__, - ecryptfs_page_idx, rc); - goto out; - } - ecryptfs_page_virt = kmap_atomic(ecryptfs_page, KM_USER0); - memcpy((data + data_offset), - ((char *)ecryptfs_page_virt + start_offset_in_page), - num_bytes); - kunmap_atomic(ecryptfs_page_virt, KM_USER0); - flush_dcache_page(ecryptfs_page); - SetPageUptodate(ecryptfs_page); - unlock_page(ecryptfs_page); - page_cache_release(ecryptfs_page); - pos += num_bytes; - data_offset += num_bytes; - } -out: - return rc; -} -#endif /* 0 */ diff --git a/fs/exec.c b/fs/exec.c index aeb135c7ff5c..92ce83a11e90 100644 --- a/fs/exec.c +++ b/fs/exec.c @@ -1071,6 +1071,21 @@ void set_task_comm(struct task_struct *tsk, char *buf) perf_event_comm(tsk); } +static void filename_to_taskname(char *tcomm, const char *fn, unsigned int len) +{ + int i, ch; + + /* Copies the binary name from after last slash */ + for (i = 0; (ch = *(fn++)) != '\0';) { + if (ch == '/') + i = 0; /* overwrite what we wrote */ + else + if (i < len - 1) + tcomm[i++] = ch; + } + tcomm[i] = '\0'; +} + int flush_old_exec(struct linux_binprm * bprm) { int retval; @@ -1085,6 +1100,7 @@ int flush_old_exec(struct linux_binprm * bprm) set_mm_exe_file(bprm->mm, bprm->file); + filename_to_taskname(bprm->tcomm, bprm->filename, sizeof(bprm->tcomm)); /* * Release all of the old mmap stuff */ @@ -1116,10 +1132,6 @@ EXPORT_SYMBOL(would_dump); void setup_new_exec(struct linux_binprm * bprm) { - int i, ch; - const char *name; - char tcomm[sizeof(current->comm)]; - arch_pick_mmap_layout(current->mm); /* This is the point of no return */ @@ -1130,18 +1142,7 @@ void setup_new_exec(struct linux_binprm * bprm) else set_dumpable(current->mm, suid_dumpable); - name = bprm->filename; - - /* Copies the binary name from after last slash */ - for (i=0; (ch = *(name++)) != '\0';) { - if (ch == '/') - i = 0; /* overwrite what we wrote */ - else - if (i < (sizeof(tcomm) - 1)) - tcomm[i++] = ch; - } - tcomm[i] = '\0'; - set_task_comm(current, tcomm); + set_task_comm(current, bprm->tcomm); /* Set the new mm task size. We have to do that late because it may * depend on TIF_32BIT which is only updated in flush_thread() on diff --git a/fs/ext2/ioctl.c b/fs/ext2/ioctl.c index 1089f760c847..2de655f5d625 100644 --- a/fs/ext2/ioctl.c +++ b/fs/ext2/ioctl.c @@ -77,10 +77,11 @@ long ext2_ioctl(struct file *filp, unsigned int cmd, unsigned long arg) flags = flags & EXT2_FL_USER_MODIFIABLE; flags |= oldflags & ~EXT2_FL_USER_MODIFIABLE; ei->i_flags = flags; - mutex_unlock(&inode->i_mutex); ext2_set_inode_flags(inode); inode->i_ctime = CURRENT_TIME_SEC; + mutex_unlock(&inode->i_mutex); + mark_inode_dirty(inode); setflags_out: mnt_drop_write_file(filp); @@ -88,20 +89,29 @@ setflags_out: } case EXT2_IOC_GETVERSION: return put_user(inode->i_generation, (int __user *) arg); - case EXT2_IOC_SETVERSION: + case EXT2_IOC_SETVERSION: { + __u32 generation; + if (!inode_owner_or_capable(inode)) return -EPERM; ret = mnt_want_write_file(filp); if (ret) return ret; - if (get_user(inode->i_generation, (int __user *) arg)) { + if (get_user(generation, (int __user *) arg)) { ret = -EFAULT; - } else { - inode->i_ctime = CURRENT_TIME_SEC; - mark_inode_dirty(inode); + goto setversion_out; } + + mutex_lock(&inode->i_mutex); + inode->i_ctime = CURRENT_TIME_SEC; + inode->i_generation = generation; + mutex_unlock(&inode->i_mutex); + + mark_inode_dirty(inode); +setversion_out: mnt_drop_write_file(filp); return ret; + } case EXT2_IOC_GETRSVSZ: if (test_opt(inode->i_sb, RESERVATION) && S_ISREG(inode->i_mode) diff --git a/fs/jbd/checkpoint.c b/fs/jbd/checkpoint.c index 5d1a00a5041b..05f0754f2b46 100644 --- a/fs/jbd/checkpoint.c +++ b/fs/jbd/checkpoint.c @@ -453,8 +453,6 @@ out: * * Return <0 on error, 0 on success, 1 if there was nothing to clean up. * - * Called with the journal lock held. - * * This is the only part of the journaling code which really needs to be * aware of transaction aborts. Checkpointing involves writing to the * main filesystem area rather than to the journal, so it can proceed @@ -472,13 +470,14 @@ int cleanup_journal_tail(journal_t *journal) if (is_journal_aborted(journal)) return 1; - /* OK, work out the oldest transaction remaining in the log, and + /* + * OK, work out the oldest transaction remaining in the log, and * the log block it starts at. * * If the log is now empty, we need to work out which is the * next transaction ID we will write, and where it will - * start. */ - + * start. + */ spin_lock(&journal->j_state_lock); spin_lock(&journal->j_list_lock); transaction = journal->j_checkpoint_transactions; @@ -504,7 +503,25 @@ int cleanup_journal_tail(journal_t *journal) spin_unlock(&journal->j_state_lock); return 1; } + spin_unlock(&journal->j_state_lock); + + /* + * We need to make sure that any blocks that were recently written out + * --- perhaps by log_do_checkpoint() --- are flushed out before we + * drop the transactions from the journal. It's unlikely this will be + * necessary, especially with an appropriately sized journal, but we + * need this to guarantee correctness. Fortunately + * cleanup_journal_tail() doesn't get called all that often. + */ + if (journal->j_flags & JFS_BARRIER) + blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL); + spin_lock(&journal->j_state_lock); + if (!tid_gt(first_tid, journal->j_tail_sequence)) { + spin_unlock(&journal->j_state_lock); + /* Someone else cleaned up journal so return 0 */ + return 0; + } /* OK, update the superblock to recover the freed space. * Physical blocks come first: have we wrapped beyond the end of * the log? */ diff --git a/fs/jbd/recovery.c b/fs/jbd/recovery.c index 5b43e96788e6..008bf062fd26 100644 --- a/fs/jbd/recovery.c +++ b/fs/jbd/recovery.c @@ -20,6 +20,7 @@ #include <linux/fs.h> #include <linux/jbd.h> #include <linux/errno.h> +#include <linux/blkdev.h> #endif /* @@ -263,6 +264,9 @@ int journal_recover(journal_t *journal) err2 = sync_blockdev(journal->j_fs_dev); if (!err) err = err2; + /* Flush disk caches to get replayed data on the permanent storage */ + if (journal->j_flags & JFS_BARRIER) + blkdev_issue_flush(journal->j_fs_dev, GFP_KERNEL, NULL); return err; } diff --git a/fs/jffs2/erase.c b/fs/jffs2/erase.c index a01cdad6aad1..eafb8d37a6fb 100644 --- a/fs/jffs2/erase.c +++ b/fs/jffs2/erase.c @@ -335,7 +335,7 @@ static int jffs2_block_check_erase(struct jffs2_sb_info *c, struct jffs2_erasebl void *ebuf; uint32_t ofs; size_t retlen; - int ret = -EIO; + int ret; unsigned long *wordebuf; ret = mtd_point(c->mtd, jeb->offset, c->sector_size, &retlen, diff --git a/fs/logfs/dev_mtd.c b/fs/logfs/dev_mtd.c index e97404d611e0..9c501449450d 100644 --- a/fs/logfs/dev_mtd.c +++ b/fs/logfs/dev_mtd.c @@ -152,9 +152,6 @@ static struct page *logfs_mtd_find_first_sb(struct super_block *sb, u64 *ofs) filler_t *filler = logfs_mtd_readpage; struct mtd_info *mtd = super->s_mtd; - if (!mtd_can_have_bb(mtd)) - return NULL; - *ofs = 0; while (mtd_block_isbad(mtd, *ofs)) { *ofs += mtd->erasesize; @@ -172,9 +169,6 @@ static struct page *logfs_mtd_find_last_sb(struct super_block *sb, u64 *ofs) filler_t *filler = logfs_mtd_readpage; struct mtd_info *mtd = super->s_mtd; - if (!mtd_can_have_bb(mtd)) - return NULL; - *ofs = mtd->size - mtd->erasesize; while (mtd_block_isbad(mtd, *ofs)) { *ofs -= mtd->erasesize; diff --git a/fs/logfs/dir.c b/fs/logfs/dir.c index 501043e8966c..3de7a32cadbe 100644 --- a/fs/logfs/dir.c +++ b/fs/logfs/dir.c @@ -71,7 +71,7 @@ static int write_dir(struct inode *dir, struct logfs_disk_dentry *dd, static int write_inode(struct inode *inode) { - return __logfs_write_inode(inode, WF_LOCK); + return __logfs_write_inode(inode, NULL, WF_LOCK); } static s64 dir_seek_data(struct inode *inode, s64 pos) diff --git a/fs/logfs/file.c b/fs/logfs/file.c index b548c87a86f1..3886cded283c 100644 --- a/fs/logfs/file.c +++ b/fs/logfs/file.c @@ -230,7 +230,9 @@ int logfs_fsync(struct file *file, loff_t start, loff_t end, int datasync) return ret; mutex_lock(&inode->i_mutex); + logfs_get_wblocks(sb, NULL, WF_LOCK); logfs_write_anchor(sb); + logfs_put_wblocks(sb, NULL, WF_LOCK); mutex_unlock(&inode->i_mutex); return 0; diff --git a/fs/logfs/gc.c b/fs/logfs/gc.c index caa4419285dc..d4efb061bdc5 100644 --- a/fs/logfs/gc.c +++ b/fs/logfs/gc.c @@ -367,7 +367,7 @@ static struct gc_candidate *get_candidate(struct super_block *sb) int i, max_dist; struct gc_candidate *cand = NULL, *this; - max_dist = min(no_free_segments(sb), LOGFS_NO_AREAS); + max_dist = min(no_free_segments(sb), LOGFS_NO_AREAS - 1); for (i = max_dist; i >= 0; i--) { this = first_in_list(&super->s_low_list[i]); diff --git a/fs/logfs/inode.c b/fs/logfs/inode.c index 388df1aa35e5..a422f42238b2 100644 --- a/fs/logfs/inode.c +++ b/fs/logfs/inode.c @@ -286,7 +286,7 @@ static int logfs_write_inode(struct inode *inode, struct writeback_control *wbc) if (logfs_inode(inode)->li_flags & LOGFS_IF_STILLBORN) return 0; - ret = __logfs_write_inode(inode, flags); + ret = __logfs_write_inode(inode, NULL, flags); LOGFS_BUG_ON(ret, inode->i_sb); return ret; } @@ -363,7 +363,9 @@ static void logfs_init_once(void *_li) static int logfs_sync_fs(struct super_block *sb, int wait) { + logfs_get_wblocks(sb, NULL, WF_LOCK); logfs_write_anchor(sb); + logfs_put_wblocks(sb, NULL, WF_LOCK); return 0; } diff --git a/fs/logfs/journal.c b/fs/logfs/journal.c index 9da29706f91c..1e1c369df22b 100644 --- a/fs/logfs/journal.c +++ b/fs/logfs/journal.c @@ -612,7 +612,6 @@ static size_t __logfs_write_je(struct super_block *sb, void *buf, u16 type, if (len == 0) return logfs_write_header(super, header, 0, type); - BUG_ON(len > sb->s_blocksize); compr_len = logfs_compress(buf, data, len, sb->s_blocksize); if (compr_len < 0 || type == JE_ANCHOR) { memcpy(data, buf, len); diff --git a/fs/logfs/logfs.h b/fs/logfs/logfs.h index 926373866a55..5f0937609465 100644 --- a/fs/logfs/logfs.h +++ b/fs/logfs/logfs.h @@ -528,7 +528,7 @@ void logfs_destroy_inode_cache(void); void logfs_set_blocks(struct inode *inode, u64 no); /* these logically belong into inode.c but actually reside in readwrite.c */ int logfs_read_inode(struct inode *inode); -int __logfs_write_inode(struct inode *inode, long flags); +int __logfs_write_inode(struct inode *inode, struct page *, long flags); void logfs_evict_inode(struct inode *inode); /* journal.c */ @@ -577,6 +577,8 @@ void initialize_block_counters(struct page *page, struct logfs_block *block, __be64 *array, int page_is_empty); int logfs_exist_block(struct inode *inode, u64 bix); int get_page_reserve(struct inode *inode, struct page *page); +void logfs_get_wblocks(struct super_block *sb, struct page *page, int lock); +void logfs_put_wblocks(struct super_block *sb, struct page *page, int lock); extern struct logfs_block_ops indirect_block_ops; /* segment.c */ @@ -594,6 +596,7 @@ int logfs_init_mapping(struct super_block *sb); void logfs_sync_area(struct logfs_area *area); void logfs_sync_segments(struct super_block *sb); void freeseg(struct super_block *sb, u32 segno); +void free_areas(struct super_block *sb); /* area handling */ int logfs_init_areas(struct super_block *sb); diff --git a/fs/logfs/readwrite.c b/fs/logfs/readwrite.c index 2ac4217b7901..4153e65b0148 100644 --- a/fs/logfs/readwrite.c +++ b/fs/logfs/readwrite.c @@ -244,8 +244,7 @@ static void preunlock_page(struct super_block *sb, struct page *page, int lock) * is waiting for s_write_mutex. We annotate this fact by setting PG_pre_locked * in addition to PG_locked. */ -static void logfs_get_wblocks(struct super_block *sb, struct page *page, - int lock) +void logfs_get_wblocks(struct super_block *sb, struct page *page, int lock) { struct logfs_super *super = logfs_super(sb); @@ -260,8 +259,7 @@ static void logfs_get_wblocks(struct super_block *sb, struct page *page, } } -static void logfs_put_wblocks(struct super_block *sb, struct page *page, - int lock) +void logfs_put_wblocks(struct super_block *sb, struct page *page, int lock) { struct logfs_super *super = logfs_super(sb); @@ -424,7 +422,7 @@ static void inode_write_block(struct logfs_block *block) if (inode->i_ino == LOGFS_INO_MASTER) logfs_write_anchor(inode->i_sb); else { - ret = __logfs_write_inode(inode, 0); + ret = __logfs_write_inode(inode, NULL, 0); /* see indirect_write_block comment */ BUG_ON(ret); } @@ -560,8 +558,13 @@ static void inode_free_block(struct super_block *sb, struct logfs_block *block) static void indirect_free_block(struct super_block *sb, struct logfs_block *block) { - ClearPagePrivate(block->page); - block->page->private = 0; + struct page *page = block->page; + + if (PagePrivate(page)) { + ClearPagePrivate(page); + page_cache_release(page); + set_page_private(page, 0); + } __free_block(sb, block); } @@ -650,8 +653,11 @@ static void alloc_data_block(struct inode *inode, struct page *page) logfs_unpack_index(page->index, &bix, &level); block = __alloc_block(inode->i_sb, inode->i_ino, bix, level); block->page = page; + SetPagePrivate(page); - page->private = (unsigned long)block; + page_cache_get(page); + set_page_private(page, (unsigned long) block); + block->ops = &indirect_block_ops; } @@ -1570,11 +1576,15 @@ int logfs_write_buf(struct inode *inode, struct page *page, long flags) static int __logfs_delete(struct inode *inode, struct page *page) { long flags = WF_DELETE; + int err; inode->i_ctime = inode->i_mtime = CURRENT_TIME; if (page->index < I0_BLOCKS) return logfs_write_direct(inode, page, flags); + err = grow_inode(inode, page->index, 0); + if (err) + return err; return logfs_write_rec(inode, page, page->index, 0, flags); } @@ -1623,7 +1633,7 @@ int logfs_rewrite_block(struct inode *inode, u64 bix, u64 ofs, if (inode->i_ino == LOGFS_INO_MASTER) logfs_write_anchor(inode->i_sb); else { - err = __logfs_write_inode(inode, flags); + err = __logfs_write_inode(inode, page, flags); } } } @@ -1873,7 +1883,7 @@ int logfs_truncate(struct inode *inode, u64 target) logfs_get_wblocks(sb, NULL, 1); err = __logfs_truncate(inode, size); if (!err) - err = __logfs_write_inode(inode, 0); + err = __logfs_write_inode(inode, NULL, 0); logfs_put_wblocks(sb, NULL, 1); } @@ -1901,8 +1911,11 @@ static void move_page_to_inode(struct inode *inode, struct page *page) li->li_block = block; block->page = NULL; - page->private = 0; - ClearPagePrivate(page); + if (PagePrivate(page)) { + ClearPagePrivate(page); + page_cache_release(page); + set_page_private(page, 0); + } } static void move_inode_to_page(struct page *page, struct inode *inode) @@ -1918,8 +1931,12 @@ static void move_inode_to_page(struct page *page, struct inode *inode) BUG_ON(PagePrivate(page)); block->ops = &indirect_block_ops; block->page = page; - page->private = (unsigned long)block; - SetPagePrivate(page); + + if (!PagePrivate(page)) { + SetPagePrivate(page); + page_cache_get(page); + set_page_private(page, (unsigned long) block); + } block->inode = NULL; li->li_block = NULL; @@ -2106,14 +2123,14 @@ void logfs_set_segment_unreserved(struct super_block *sb, u32 segno, u32 ec) ec_level); } -int __logfs_write_inode(struct inode *inode, long flags) +int __logfs_write_inode(struct inode *inode, struct page *page, long flags) { struct super_block *sb = inode->i_sb; int ret; - logfs_get_wblocks(sb, NULL, flags & WF_LOCK); + logfs_get_wblocks(sb, page, flags & WF_LOCK); ret = do_write_inode(inode); - logfs_put_wblocks(sb, NULL, flags & WF_LOCK); + logfs_put_wblocks(sb, page, flags & WF_LOCK); return ret; } diff --git a/fs/logfs/segment.c b/fs/logfs/segment.c index 9d5187353255..ab798ed1cc88 100644 --- a/fs/logfs/segment.c +++ b/fs/logfs/segment.c @@ -86,7 +86,11 @@ int __logfs_buf_write(struct logfs_area *area, u64 ofs, void *buf, size_t len, BUG_ON(!page); /* FIXME: reserve a pool */ SetPageUptodate(page); memcpy(page_address(page) + offset, buf, copylen); - SetPagePrivate(page); + + if (!PagePrivate(page)) { + SetPagePrivate(page); + page_cache_get(page); + } page_cache_release(page); buf += copylen; @@ -110,7 +114,10 @@ static void pad_partial_page(struct logfs_area *area) page = get_mapping_page(sb, index, 0); BUG_ON(!page); /* FIXME: reserve a pool */ memset(page_address(page) + offset, 0xff, len); - SetPagePrivate(page); + if (!PagePrivate(page)) { + SetPagePrivate(page); + page_cache_get(page); + } page_cache_release(page); } } @@ -130,7 +137,10 @@ static void pad_full_pages(struct logfs_area *area) BUG_ON(!page); /* FIXME: reserve a pool */ SetPageUptodate(page); memset(page_address(page), 0xff, PAGE_CACHE_SIZE); - SetPagePrivate(page); + if (!PagePrivate(page)) { + SetPagePrivate(page); + page_cache_get(page); + } page_cache_release(page); index++; no_indizes--; @@ -485,8 +495,12 @@ static void move_btree_to_page(struct inode *inode, struct page *page, mempool_free(item, super->s_alias_pool); } block->page = page; - SetPagePrivate(page); - page->private = (unsigned long)block; + + if (!PagePrivate(page)) { + SetPagePrivate(page); + page_cache_get(page); + set_page_private(page, (unsigned long) block); + } block->ops = &indirect_block_ops; initialize_block_counters(page, block, data, 0); } @@ -536,8 +550,12 @@ void move_page_to_btree(struct page *page) list_add(&item->list, &block->item_list); } block->page = NULL; - ClearPagePrivate(page); - page->private = 0; + + if (PagePrivate(page)) { + ClearPagePrivate(page); + page_cache_release(page); + set_page_private(page, 0); + } block->ops = &btree_block_ops; err = alias_tree_insert(block->sb, block->ino, block->bix, block->level, block); @@ -702,7 +720,10 @@ void freeseg(struct super_block *sb, u32 segno) page = find_get_page(mapping, ofs >> PAGE_SHIFT); if (!page) continue; - ClearPagePrivate(page); + if (PagePrivate(page)) { + ClearPagePrivate(page); + page_cache_release(page); + } page_cache_release(page); } } @@ -841,6 +862,16 @@ static void free_area(struct logfs_area *area) kfree(area); } +void free_areas(struct super_block *sb) +{ + struct logfs_super *super = logfs_super(sb); + int i; + + for_each_area(i) + free_area(super->s_area[i]); + free_area(super->s_journal_area); +} + static struct logfs_area *alloc_area(struct super_block *sb) { struct logfs_area *area; @@ -923,10 +954,6 @@ err: void logfs_cleanup_areas(struct super_block *sb) { struct logfs_super *super = logfs_super(sb); - int i; btree_grim_visitor128(&super->s_object_alias_tree, 0, kill_alias); - for_each_area(i) - free_area(super->s_area[i]); - free_area(super->s_journal_area); } diff --git a/fs/logfs/super.c b/fs/logfs/super.c index e795c234ea33..c9ee7f5d1caf 100644 --- a/fs/logfs/super.c +++ b/fs/logfs/super.c @@ -486,14 +486,15 @@ static void logfs_kill_sb(struct super_block *sb) /* Alias entries slow down mount, so evict as many as possible */ sync_filesystem(sb); logfs_write_anchor(sb); + free_areas(sb); /* * From this point on alias entries are simply dropped - and any * writes to the object store are considered bugs. */ - super->s_flags |= LOGFS_SB_FLAG_SHUTDOWN; log_super("LogFS: Now in shutdown\n"); generic_shutdown_super(sb); + super->s_flags |= LOGFS_SB_FLAG_SHUTDOWN; BUG_ON(super->s_dirty_used_bytes || super->s_dirty_free_bytes); diff --git a/fs/proc/base.c b/fs/proc/base.c index 9cde9edf9c4d..d4548dd49b02 100644 --- a/fs/proc/base.c +++ b/fs/proc/base.c @@ -198,26 +198,6 @@ static int proc_root_link(struct dentry *dentry, struct path *path) return result; } -static struct mm_struct *mm_access(struct task_struct *task, unsigned int mode) -{ - struct mm_struct *mm; - int err; - - err = mutex_lock_killable(&task->signal->cred_guard_mutex); - if (err) - return ERR_PTR(err); - - mm = get_task_mm(task); - if (mm && mm != current->mm && - !ptrace_may_access(task, mode)) { - mmput(mm); - mm = ERR_PTR(-EACCES); - } - mutex_unlock(&task->signal->cred_guard_mutex); - - return mm; -} - struct mm_struct *mm_for_maps(struct task_struct *task) { return mm_access(task, PTRACE_MODE_READ); @@ -711,6 +691,13 @@ static int mem_open(struct inode* inode, struct file* file) if (IS_ERR(mm)) return PTR_ERR(mm); + if (mm) { + /* ensure this mm_struct can't be freed */ + atomic_inc(&mm->mm_count); + /* but do not pin its memory */ + mmput(mm); + } + /* OK to pass negative loff_t, we can catch out-of-range */ file->f_mode |= FMODE_UNSIGNED_OFFSET; file->private_data = mm; @@ -718,57 +705,13 @@ static int mem_open(struct inode* inode, struct file* file) return 0; } -static ssize_t mem_read(struct file * file, char __user * buf, - size_t count, loff_t *ppos) +static ssize_t mem_rw(struct file *file, char __user *buf, + size_t count, loff_t *ppos, int write) { - int ret; - char *page; - unsigned long src = *ppos; struct mm_struct *mm = file->private_data; - - if (!mm) - return 0; - - page = (char *)__get_free_page(GFP_TEMPORARY); - if (!page) - return -ENOMEM; - - ret = 0; - - while (count > 0) { - int this_len, retval; - - this_len = (count > PAGE_SIZE) ? PAGE_SIZE : count; - retval = access_remote_vm(mm, src, page, this_len, 0); - if (!retval) { - if (!ret) - ret = -EIO; - break; - } - - if (copy_to_user(buf, page, retval)) { - ret = -EFAULT; - break; - } - - ret += retval; - src += retval; - buf += retval; - count -= retval; - } - *ppos = src; - - free_page((unsigned long) page); - return ret; -} - -static ssize_t mem_write(struct file * file, const char __user *buf, - size_t count, loff_t *ppos) -{ - int copied; + unsigned long addr = *ppos; + ssize_t copied; char *page; - unsigned long dst = *ppos; - struct mm_struct *mm = file->private_data; if (!mm) return 0; @@ -778,31 +721,54 @@ static ssize_t mem_write(struct file * file, const char __user *buf, return -ENOMEM; copied = 0; + if (!atomic_inc_not_zero(&mm->mm_users)) + goto free; + while (count > 0) { - int this_len, retval; + int this_len = min_t(int, count, PAGE_SIZE); - this_len = (count > PAGE_SIZE) ? PAGE_SIZE : count; - if (copy_from_user(page, buf, this_len)) { + if (write && copy_from_user(page, buf, this_len)) { copied = -EFAULT; break; } - retval = access_remote_vm(mm, dst, page, this_len, 1); - if (!retval) { + + this_len = access_remote_vm(mm, addr, page, this_len, write); + if (!this_len) { if (!copied) copied = -EIO; break; } - copied += retval; - buf += retval; - dst += retval; - count -= retval; + + if (!write && copy_to_user(buf, page, this_len)) { + copied = -EFAULT; + break; + } + + buf += this_len; + addr += this_len; + copied += this_len; + count -= this_len; } - *ppos = dst; + *ppos = addr; + mmput(mm); +free: free_page((unsigned long) page); return copied; } +static ssize_t mem_read(struct file *file, char __user *buf, + size_t count, loff_t *ppos) +{ + return mem_rw(file, buf, count, ppos, 0); +} + +static ssize_t mem_write(struct file *file, const char __user *buf, + size_t count, loff_t *ppos) +{ + return mem_rw(file, (char __user*)buf, count, ppos, 1); +} + loff_t mem_lseek(struct file *file, loff_t offset, int orig) { switch (orig) { @@ -822,8 +788,8 @@ loff_t mem_lseek(struct file *file, loff_t offset, int orig) static int mem_release(struct inode *inode, struct file *file) { struct mm_struct *mm = file->private_data; - - mmput(mm); + if (mm) + mmdrop(mm); return 0; } diff --git a/fs/quota/dquot.c b/fs/quota/dquot.c index 5ec59b20cf76..46741970371b 100644 --- a/fs/quota/dquot.c +++ b/fs/quota/dquot.c @@ -2125,6 +2125,8 @@ static int vfs_load_quota_inode(struct inode *inode, int type, int format_id, mutex_unlock(&dqopt->dqio_mutex); goto out_file_init; } + if (dqopt->flags & DQUOT_QUOTA_SYS_FILE) + dqopt->info[type].dqi_flags |= DQF_SYS_FILE; mutex_unlock(&dqopt->dqio_mutex); spin_lock(&dq_state_lock); dqopt->flags |= dquot_state_flag(flags, type); @@ -2464,7 +2466,7 @@ int dquot_get_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii) spin_lock(&dq_data_lock); ii->dqi_bgrace = mi->dqi_bgrace; ii->dqi_igrace = mi->dqi_igrace; - ii->dqi_flags = mi->dqi_flags & DQF_MASK; + ii->dqi_flags = mi->dqi_flags & DQF_GETINFO_MASK; ii->dqi_valid = IIF_ALL; spin_unlock(&dq_data_lock); mutex_unlock(&sb_dqopt(sb)->dqonoff_mutex); @@ -2490,8 +2492,8 @@ int dquot_set_dqinfo(struct super_block *sb, int type, struct if_dqinfo *ii) if (ii->dqi_valid & IIF_IGRACE) mi->dqi_igrace = ii->dqi_igrace; if (ii->dqi_valid & IIF_FLAGS) - mi->dqi_flags = (mi->dqi_flags & ~DQF_MASK) | - (ii->dqi_flags & DQF_MASK); + mi->dqi_flags = (mi->dqi_flags & ~DQF_SETINFO_MASK) | + (ii->dqi_flags & DQF_SETINFO_MASK); spin_unlock(&dq_data_lock); mark_info_dirty(sb, type); /* Force write to disk */ diff --git a/fs/sysfs/file.c b/fs/sysfs/file.c index 62f4fb37789e..00012e31829d 100644 --- a/fs/sysfs/file.c +++ b/fs/sysfs/file.c @@ -493,6 +493,12 @@ int sysfs_attr_ns(struct kobject *kobj, const struct attribute *attr, const void *ns = NULL; int err; + if (!dir_sd) { + WARN(1, KERN_ERR "sysfs: kobject %s without dirent\n", + kobject_name(kobj)); + return -ENOENT; + } + err = 0; if (!sysfs_ns_type(dir_sd)) goto out; diff --git a/fs/sysfs/inode.c b/fs/sysfs/inode.c index 4a802b4a9056..85eb81683a29 100644 --- a/fs/sysfs/inode.c +++ b/fs/sysfs/inode.c @@ -318,8 +318,11 @@ int sysfs_hash_and_remove(struct sysfs_dirent *dir_sd, const void *ns, const cha struct sysfs_addrm_cxt acxt; struct sysfs_dirent *sd; - if (!dir_sd) + if (!dir_sd) { + WARN(1, KERN_WARNING "sysfs: can not remove '%s', no directory\n", + name); return -ENOENT; + } sysfs_addrm_start(&acxt, dir_sd); diff --git a/fs/xfs/xfs_vnodeops.c b/fs/xfs/xfs_vnodeops.c index 0cf52da9d246..ebdb88840a47 100644 --- a/fs/xfs/xfs_vnodeops.c +++ b/fs/xfs/xfs_vnodeops.c @@ -131,7 +131,8 @@ xfs_readlink( __func__, (unsigned long long) ip->i_ino, (long long) pathlen); ASSERT(0); - return XFS_ERROR(EFSCORRUPTED); + error = XFS_ERROR(EFSCORRUPTED); + goto out; } |