diff options
| author | SeongJae Park <sj@kernel.org> | 2025-04-09 17:00:22 -0700 |
|---|---|---|
| committer | Andrew Morton <akpm@linux-foundation.org> | 2025-05-11 17:48:27 -0700 |
| commit | 43c4cfde7e37460482e41dbb6837d4bff290d2cc (patch) | |
| tree | 5e96c4d62cb115c0f3be8b2fc399e568cdf3c396 | |
| parent | de8efdf8cd273619121cbc2f0f70dddc74bc586e (diff) | |
mm/madvise: batch tlb flushes for MADV_DONTNEED[_LOCKED]
MADV_DONTNEED[_LOCKED] handling for [process_]madvise() flushes tlb for
each vma of each address range. Update the logic to do tlb flushes in a
batched way. Initialize an mmu_gather object from do_madvise() and
vector_madvise(), which are the entry level functions for
[process_]madvise(), respectively. And pass those objects to the function
for per-vma work, via madvise_behavior struct. Make the per-vma logic not
flushes tlb on their own but just saves the tlb entries to the received
mmu_gather object. For this internal logic change, make
zap_page_range_single_batched() non-static and use it directly from
madvise_dontneed_single_vma(). Finally, the entry level functions flush
the tlb entries that gathered for the entire user request, at once.
Link: https://lkml.kernel.org/r/20250410000022.1901-5-sj@kernel.org
Signed-off-by: SeongJae Park <sj@kernel.org>
Reviewed-by: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
| -rw-r--r-- | mm/internal.h | 3 | ||||
| -rw-r--r-- | mm/madvise.c | 11 | ||||
| -rw-r--r-- | mm/memory.c | 4 |
3 files changed, 13 insertions, 5 deletions
diff --git a/mm/internal.h b/mm/internal.h index 0cf1f534ee1a..780481a8be0e 100644 --- a/mm/internal.h +++ b/mm/internal.h @@ -430,6 +430,9 @@ void unmap_page_range(struct mmu_gather *tlb, struct vm_area_struct *vma, unsigned long addr, unsigned long end, struct zap_details *details); +void zap_page_range_single_batched(struct mmu_gather *tlb, + struct vm_area_struct *vma, unsigned long addr, + unsigned long size, struct zap_details *details); int folio_unmap_invalidate(struct address_space *mapping, struct folio *folio, gfp_t gfp); diff --git a/mm/madvise.c b/mm/madvise.c index 951038a9f36f..8433ac9b27e0 100644 --- a/mm/madvise.c +++ b/mm/madvise.c @@ -851,7 +851,8 @@ static int madvise_free_single_vma(struct madvise_behavior *madv_behavior, * An interface that causes the system to free clean pages and flush * dirty pages is already available as msync(MS_INVALIDATE). */ -static long madvise_dontneed_single_vma(struct vm_area_struct *vma, +static long madvise_dontneed_single_vma(struct madvise_behavior *madv_behavior, + struct vm_area_struct *vma, unsigned long start, unsigned long end) { struct zap_details details = { @@ -859,7 +860,8 @@ static long madvise_dontneed_single_vma(struct vm_area_struct *vma, .even_cows = true, }; - zap_page_range_single(vma, start, end - start, &details); + zap_page_range_single_batched( + madv_behavior->tlb, vma, start, end - start, &details); return 0; } @@ -950,7 +952,8 @@ static long madvise_dontneed_free(struct vm_area_struct *vma, } if (behavior == MADV_DONTNEED || behavior == MADV_DONTNEED_LOCKED) - return madvise_dontneed_single_vma(vma, start, end); + return madvise_dontneed_single_vma( + madv_behavior, vma, start, end); else if (behavior == MADV_FREE) return madvise_free_single_vma(madv_behavior, vma, start, end); else @@ -1628,6 +1631,8 @@ static void madvise_unlock(struct mm_struct *mm, int behavior) static bool madvise_batch_tlb_flush(int behavior) { switch (behavior) { + case MADV_DONTNEED: + case MADV_DONTNEED_LOCKED: case MADV_FREE: return true; default: diff --git a/mm/memory.c b/mm/memory.c index 33d34722e339..71c255f3fdcc 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -1998,7 +1998,7 @@ void unmap_vmas(struct mmu_gather *tlb, struct ma_state *mas, mmu_notifier_invalidate_range_end(&range); } -/* +/** * zap_page_range_single_batched - remove user pages in a given range * @tlb: pointer to the caller's struct mmu_gather * @vma: vm_area_struct holding the applicable pages @@ -2009,7 +2009,7 @@ void unmap_vmas(struct mmu_gather *tlb, struct ma_state *mas, * @tlb shouldn't be NULL. The range must fit into one VMA. If @vma is for * hugetlb, @tlb is flushed and re-initialized by this function. */ -static void zap_page_range_single_batched(struct mmu_gather *tlb, +void zap_page_range_single_batched(struct mmu_gather *tlb, struct vm_area_struct *vma, unsigned long address, unsigned long size, struct zap_details *details) { |