On Wed, 22 Jul 2015, Catalin Marinas wrote:
> When the page table entry is a huge page (and not a table), there is no
> need to flush the TLB by range. This patch changes flush_tlb_range() to
> flush_tlb_page() in functions where we know the pmd entry is a huge
> page.
>
> Signed-off-by: Catalin Marinas <catalin.marinas@xxxxxxx>
> Cc: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
> Cc: Andrea Arcangeli <aarcange@xxxxxxxxxx>
> ---
>
> Hi,
>
> That's just a minor improvement but it saves iterating over each small
> page in a huge page when a single TLB entry is used (we already have a
> similar assumption in __tlb_adjust_range).
>
For x86 smp, this seems to mean the difference between unconditional
flush_tlb_page() and local_flush_tlb() due to
tlb_single_page_flush_ceiling, so I don't think this just removes the
iteration.
> Thanks.
>
> mm/pgtable-generic.c | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/mm/pgtable-generic.c b/mm/pgtable-generic.c
> index 6b674e00153c..ff17eca26211 100644
> --- a/mm/pgtable-generic.c
> +++ b/mm/pgtable-generic.c
> @@ -67,7 +67,7 @@ int pmdp_set_access_flags(struct vm_area_struct *vma,
> VM_BUG_ON(address & ~HPAGE_PMD_MASK);
> if (changed) {
> set_pmd_at(vma->vm_mm, address, pmdp, entry);
> - flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
> + flush_tlb_page(vma, address);
> }
> return changed;
> #else /* CONFIG_TRANSPARENT_HUGEPAGE */
> @@ -101,7 +101,7 @@ int pmdp_clear_flush_young(struct vm_area_struct *vma,
> #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
> young = pmdp_test_and_clear_young(vma, address, pmdp);
> if (young)
> - flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
> + flush_tlb_page(vma, address);
> return young;
> }
> #endif
> @@ -128,7 +128,7 @@ pmd_t pmdp_huge_clear_flush(struct vm_area_struct *vma, unsigned long address,
> VM_BUG_ON(address & ~HPAGE_PMD_MASK);
> VM_BUG_ON(!pmd_trans_huge(*pmdp));
> pmd = pmdp_huge_get_and_clear(vma->vm_mm, address, pmdp);
> - flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
> + flush_tlb_page(vma, address);
> return pmd;
> }
> #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
> @@ -143,7 +143,7 @@ void pmdp_splitting_flush(struct vm_area_struct *vma, unsigned long address,
> VM_BUG_ON(address & ~HPAGE_PMD_MASK);
> set_pmd_at(vma->vm_mm, address, pmdp, pmd);
> /* tlb flush only to serialize against gup-fast */
> - flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
> + flush_tlb_page(vma, address);
> }
> #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
> #endif
> @@ -195,7 +195,7 @@ void pmdp_invalidate(struct vm_area_struct *vma, unsigned long address,
> {
> pmd_t entry = *pmdp;
> set_pmd_at(vma->vm_mm, address, pmdp, pmd_mknotpresent(entry));
> - flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
> + flush_tlb_page(vma, address);
> }
> #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
> #endif
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@xxxxxxxxx. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>