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/064da3bb-5b4b-7332-a722-c5a541128705@oracle.com/


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.

Link: https://lkml.kernel.org/r/cover.1651056365.git.baolin.wang@linux.alibaba.com
Link: https://lkml.kernel.org/r/0443c8cf20db554d3ff4b439b30e0ff26c0181dd.1651056365.git.baolin.wang@linux.alibaba.com
Fixes: 550a7d60bd5e ("mm, hugepages: add mremap() support for hugepage backed vma")
Signed-off-by: Baolin Wang <baolin.wang@linux.alibaba.com>
Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Reviewed-by: Muchun Song <songmuchun@bytedance.com>
Cc: Mina Almasry <almasrymina@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>

2 files changed, 16 insertions, 3 deletions

diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 4454c42a82f0..fc31a8f49067 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -4922,10 +4922,17 @@ int move_hugetlb_page_tables(struct vm_area_struct *vma,
 	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_area_struct *vma,
 		 */
 		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_area_struct *vma,
 
 		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);
 
diff --git a/mm/mremap.c b/mm/mremap.c
index 98f50e633009..097002506d2d 100644
--- a/mm/mremap.c
+++ b/mm/mremap.c
@@ -490,12 +490,12 @@ unsigned long move_page_tables(struct vm_area_struct *vma,
 		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);