Skip to contentSkip to conversation

mm-hugetlb-considering-pmd-sharing-when-flushing-cache-tlbs.patch removed from -mm tree

merged mm-stable Andrew Morton 1 participant 1 message May 10, 2022

lore
Share & tools
lore

Andrew Morton

author tree notice Applied
#0 lore raw

The quilt patch titled
Subject: mm: hugetlb: considering PMD sharing when flushing cache/TLBs
has been removed from the -mm tree. Its filename was
mm-hugetlb-considering-pmd-sharing-when-flushing-cache-tlbs.patch

This patch was dropped because it was merged into mm-stable

------------------------------------------------------
From: Baolin Wang <[email protected]>
Subject: mm: hugetlb: considering PMD sharing when flushing cache/TLBs

This patchset fixes some cache flushing issues if PMD sharing is possible for hugetlb pages, which were found by code inspection. Meanwhile Mike found the flush_cache_page() can not cover the whole size of a hugetlb page on some architectures [1], so I added a new patch 3 to fix this issue, since I found only try_to_unmap_one() and try_to_migrate_one() need to fix after some investigation.

[1] https://lore.kernel.org/linux-mm/[email protected]/

This patch (of 3):

When moving hugetlb page tables, the cache flushing is called in move_page_tables() without considering the shared PMDs, which may be cause cache issues on some architectures.

Thus we should move the hugetlb cache flushing into move_hugetlb_page_tables() with considering the shared PMDs ranges, calculated by adjust_range_if_pmd_sharing_possible(). Meanwhile also expanding the TLBs flushing range in case of shared PMDs.

Note this is discovered via code inspection, and did not meet a real problem in practice so far.

---

 mm/hugetlb.c |   17 +++++++++++++++--
 mm/mremap.c  |    2 +-
 2 files changed, 16 insertions(+), 3 deletions(-)
mm/hugetlb.c renamed from mm/hugetlb.c~mm-hugetlb-considering-pmd-sharing-when-flushing-cache-tlbs +15 -2
@@ -4922,10 +4922,17 @@ int move_hugetlb_page_tables(struct vm_a
 	unsigned long old_addr_copy;
 	pte_t *src_pte, *dst_pte;
 	struct mmu_notifier_range range;
+	bool shared_pmd = false;
 
 	mmu_notifier_range_init(&range, MMU_NOTIFY_CLEAR, 0, vma, mm, old_addr,
 				old_end);
 	adjust_range_if_pmd_sharing_possible(vma, &range.start, &range.end);
+	/*
+	 * In case of shared PMDs, we should cover the maximum possible
+	 * range.
+	 */
+	flush_cache_range(vma, range.start, range.end);
+
 	mmu_notifier_invalidate_range_start(&range);
 	/* Prevent race with file truncation */
 	i_mmap_lock_write(mapping);
@@ -4942,8 +4949,10 @@ int move_hugetlb_page_tables(struct vm_a
 		 */
 		old_addr_copy = old_addr;
 
-		if (huge_pmd_unshare(mm, vma, &old_addr_copy, src_pte))
+		if (huge_pmd_unshare(mm, vma, &old_addr_copy, src_pte)) {
+			shared_pmd = true;
 			continue;
+		}
 
 		dst_pte = huge_pte_alloc(mm, new_vma, new_addr, sz);
 		if (!dst_pte)
@@ -4951,7 +4960,11 @@ int move_hugetlb_page_tables(struct vm_a
 
 		move_huge_pte(vma, old_addr, new_addr, src_pte, dst_pte);
 	}
-	flush_tlb_range(vma, old_end - len, old_end);
+
+	if (shared_pmd)
+		flush_tlb_range(vma, range.start, range.end);
+	else
+		flush_tlb_range(vma, old_end - len, old_end);
 	mmu_notifier_invalidate_range_end(&range);
 	i_mmap_unlock_write(mapping);
 
mm/mremap.c renamed from mm/mremap.c~mm-hugetlb-considering-pmd-sharing-when-flushing-cache-tlbs +1 -1
@@ -490,12 +490,12 @@ unsigned long move_page_tables(struct vm
 		return 0;
 
 	old_end = old_addr + len;
-	flush_cache_range(vma, old_addr, old_end);
 
 	if (is_vm_hugetlb_page(vma))
 		return move_hugetlb_page_tables(vma, new_vma, old_addr,
 						new_addr, len);
 
+	flush_cache_range(vma, old_addr, old_end);
 	mmu_notifier_range_init(&range, MMU_NOTIFY_UNMAP, 0, vma, vma->vm_mm,
 				old_addr, old_end);
 	mmu_notifier_invalidate_range_start(&range);

_

Patches currently in -mm which might be from [email protected] are