sh: Flush only the needed range when unmapping a VMA.

This follows the ARM change from Aaro Koskinen:

	When unmapping N pages (e.g. shared memory) the amount of TLB
	flushes done can be (N*PAGE_SIZE/ZAP_BLOCK_SIZE)*N although it
	should be N at maximum. With PREEMPT kernel ZAP_BLOCK_SIZE is 8
	pages, so there is a noticeable performance penalty when
	unmapping a large VMA and the system is spending its time in
	flush_tlb_range().

	The problem is that tlb_end_vma() is always flushing the full VMA
	range. The subrange that needs to be flushed can be calculated by
	tlb_remove_tlb_entry(). This approach was suggested by Hugh
	Dickins, and is also used by other arches.

	The speed increase is roughly 3x for 8M mappings and for larger
	mappings even more.

Bits and peices are taken from the ARM patch as well as the existing
arch/um implementation that is quite similar.

The end result is a significant reduction in both partial and full TLB
flushes initiated through flush_tlb_range().

At the same time, the nommu implementation was broken, had a superfluous
cache flush, and subsequently would have triggered a BUG_ON() if a
code-path had triggered it. Tidy this up for correctness and provide a
nopped-out implementation there.

More background on the initial discussion can be found at:

	http://marc.info/?t=123609820900002&r=1&w=2
	http://marc.info/?t=123660375800003&r=1&w=2

Signed-off-by: Paul Mundt <lethal@linux-sh.org>

1 files changed, 92 insertions, 8 deletions

diff --git a/arch/sh/include/asm/tlb.h b/arch/sh/include/asm/tlb.h
index 88ff1ae8a6b..9c16f737074 100644
--- a/arch/sh/include/asm/tlb.h
+++ b/arch/sh/include/asm/tlb.h
@@ -6,22 +6,106 @@
 #endif
 
 #ifndef __ASSEMBLY__
+#include <linux/pagemap.h>
+
+#ifdef CONFIG_MMU
+#include <asm/pgalloc.h>
+#include <asm/tlbflush.h>
+
+/*
+ * TLB handling.  This allows us to remove pages from the page
+ * tables, and efficiently handle the TLB issues.
+ */
+struct mmu_gather {
+	struct mm_struct	*mm;
+	unsigned int		fullmm;
+	unsigned long		start, end;
+};
 
-#define tlb_start_vma(tlb, vma) \
-	flush_cache_range(vma, vma->vm_start, vma->vm_end)
+DECLARE_PER_CPU(struct mmu_gather, mmu_gathers);
 
-#define tlb_end_vma(tlb, vma)	\
-	flush_tlb_range(vma, vma->vm_start, vma->vm_end)
+static inline void init_tlb_gather(struct mmu_gather *tlb)
+{
+	tlb->start = TASK_SIZE;
+	tlb->end = 0;
 
-#define __tlb_remove_tlb_entry(tlb, pte, address)	do { } while (0)
+	if (tlb->fullmm) {
+		tlb->start = 0;
+		tlb->end = TASK_SIZE;
+	}
+}
+
+static inline struct mmu_gather *
+tlb_gather_mmu(struct mm_struct *mm, unsigned int full_mm_flush)
+{
+	struct mmu_gather *tlb = &get_cpu_var(mmu_gathers);
+
+	tlb->mm = mm;
+	tlb->fullmm = full_mm_flush;
+
+	init_tlb_gather(tlb);
+
+	return tlb;
+}
+
+static inline void
+tlb_finish_mmu(struct mmu_gather *tlb, unsigned long start, unsigned long end)
+{
+	if (tlb->fullmm)
+		flush_tlb_mm(tlb->mm);
+
+	/* keep the page table cache within bounds */
+	check_pgt_cache();
+
+	put_cpu_var(mmu_gathers);
+}
+
+static inline void
+tlb_remove_tlb_entry(struct mmu_gather *tlb, pte_t *ptep, unsigned long address)
+{
+	if (tlb->start > address)
+		tlb->start = address;
+	if (tlb->end < address + PAGE_SIZE)
+		tlb->end = address + PAGE_SIZE;
+}
 
 /*
- * Flush whole TLBs for MM
+ * In the case of tlb vma handling, we can optimise these away in the
+ * case where we're doing a full MM flush.  When we're doing a munmap,
+ * the vmas are adjusted to only cover the region to be torn down.
  */
-#define tlb_flush(tlb)				flush_tlb_mm((tlb)->mm)
+static inline void
+tlb_start_vma(struct mmu_gather *tlb, struct vm_area_struct *vma)
+{
+	if (!tlb->fullmm)
+		flush_cache_range(vma, vma->vm_start, vma->vm_end);
+}
+
+static inline void
+tlb_end_vma(struct mmu_gather *tlb, struct vm_area_struct *vma)
+{
+	if (!tlb->fullmm && tlb->end) {
+		flush_tlb_range(vma, tlb->start, tlb->end);
+		init_tlb_gather(tlb);
+	}
+}
+
+#define tlb_remove_page(tlb,page)	free_page_and_swap_cache(page)
+#define pte_free_tlb(tlb, ptep)		pte_free((tlb)->mm, ptep)
+#define pmd_free_tlb(tlb, pmdp)		pmd_free((tlb)->mm, pmdp)
+#define pud_free_tlb(tlb, pudp)		pud_free((tlb)->mm, pudp)
+
+#define tlb_migrate_finish(mm)		do { } while (0)
+
+#else /* CONFIG_MMU */
+
+#define tlb_start_vma(tlb, vma)				do { } while (0)
+#define tlb_end_vma(tlb, vma)				do { } while (0)
+#define __tlb_remove_tlb_entry(tlb, pte, address)	do { } while (0)
+#define tlb_flush(tlb)					do { } while (0)
 
-#include <linux/pagemap.h>
 #include <asm-generic/tlb.h>
 
+#endif /* CONFIG_MMU */
 #endif /* __ASSEMBLY__ */
 #endif /* __ASM_SH_TLB_H */