Skip to contentSkip to conversation

mm/thp: fix __split_huge_pmd_locked() on shmem migration entry (fwd)

PATCH In review Hugh Dickins 3 participants 3 messages Jun 9, 2021

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH v2 1/10 mm/thp: fix __split_huge_pmd_locked() on shmem migration entry (fwd)

---------- Forwarded message ----------
Date: Tue, 8 Jun 2021 21:00:12 -0700 (PDT)
From: Hugh Dickins <[email protected]>
To: Andrew Morton <[email protected]>
Cc: Hugh Dickins <[email protected]>,
Kirill A. Shutemov <[email protected]>,
Yang Shi <[email protected]>, Wang Yugui <[email protected]>, Matthew Wilcox <[email protected]>,
Naoya Horiguchi <[email protected]>,
Alistair Popple <[email protected]>, Ralph Campbell <[email protected]>, Zi Yan <[email protected]>, Miaohe Lin <[email protected]>, Minchan Kim <[email protected]>, Jue Wang <[email protected]>, Peter Xu <[email protected]>, Jan Kara <[email protected]>,
Shakeel Butt <[email protected]>, Oscar Salvador <[email protected]>
Subject: [PATCH v2 01/10] mm/thp: fix __split_huge_pmd_locked() on shmem
migration entry

Stressing huge tmpfs page migration racing hole punch often crashed on the VM_BUG_ON(!pmd_present) in pmdp_huge_clear_flush(), with DEBUG_VM=y kernel; or shortly afterwards, on a bad dereference in __split_huge_pmd_locked() when DEBUG_VM=n. They forgot to allow for pmd migration entries in the non-anonymous case.

Full disclosure: those particular experiments were on a kernel with more relaxed mmap_lock and i_mmap_rwsem locking, and were not repeated on the vanilla kernel: it is conceivable that stricter locking happens to avoid those cases, or makes them less likely; but __split_huge_pmd_locked() already allowed for pmd migration entries when handling anonymous THPs, so this commit brings the shmem and file THP handling into line.

And while there: use old_pmd rather than _pmd, as in the following blocks; and make it clearer to the eye that the !vma_is_anonymous() block is self-contained, making an early return after accounting for unmapping.

Fixes e71769ae5260 ("mm: enable thp migration for shmem thp")

---
v2: omit is_huge_zero_pmd() mods (done differently in next), per Kirill

 mm/huge_memory.c     | 27 ++++++++++++++++++---------
 mm/pgtable-generic.c |  5 ++---
 2 files changed, 20 insertions(+), 12 deletions(-)
mm/huge_memory.c +18 -9
index 63ed6b25deaa..42cfefc6e66e 100644
@@ -2044,7 +2044,7 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 	count_vm_event(THP_SPLIT_PMD);
 
 	if (!vma_is_anonymous(vma)) {
-		_pmd = pmdp_huge_clear_flush_notify(vma, haddr, pmd);
+		old_pmd = pmdp_huge_clear_flush_notify(vma, haddr, pmd);
 		/*
 		 * We are going to unmap this huge page. So
 		 * just go ahead and zap it
@@ -2053,16 +2053,25 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 			zap_deposited_table(mm, pmd);
 		if (vma_is_special_huge(vma))
 			return;
-		page = pmd_page(_pmd);
-		if (!PageDirty(page) && pmd_dirty(_pmd))
-			set_page_dirty(page);
-		if (!PageReferenced(page) && pmd_young(_pmd))
-			SetPageReferenced(page);
-		page_remove_rmap(page, true);
-		put_page(page);
+		if (unlikely(is_pmd_migration_entry(old_pmd))) {
+			swp_entry_t entry;
+
+			entry = pmd_to_swp_entry(old_pmd);
+			page = migration_entry_to_page(entry);
+		} else {
+			page = pmd_page(old_pmd);
+			if (!PageDirty(page) && pmd_dirty(old_pmd))
+				set_page_dirty(page);
+			if (!PageReferenced(page) && pmd_young(old_pmd))
+				SetPageReferenced(page);
+			page_remove_rmap(page, true);
+			put_page(page);
+		}
 		add_mm_counter(mm, mm_counter_file(page), -HPAGE_PMD_NR);
 		return;
-	} else if (pmd_trans_huge(*pmd) && is_huge_zero_pmd(*pmd)) {
+	}
+
+	if (pmd_trans_huge(*pmd) && is_huge_zero_pmd(*pmd)) {
 		/*
 		 * FIXME: Do we want to invalidate secondary mmu by calling
 		 * mmu_notifier_invalidate_range() see comments below inside
mm/pgtable-generic.c +2 -3
index c2210e1cdb51..4e640baf9794 100644
@@ -135,9 +135,8 @@ pmd_t pmdp_huge_clear_flush(struct vm_area_struct *vma, unsigned long address,
 {
 	pmd_t pmd;
 	VM_BUG_ON(address & ~HPAGE_PMD_MASK);
-	VM_BUG_ON(!pmd_present(*pmdp));
-	/* Below assumes pmd_present() is true */
-	VM_BUG_ON(!pmd_trans_huge(*pmdp) && !pmd_devmap(*pmdp));
+	VM_BUG_ON(pmd_present(*pmdp) && !pmd_trans_huge(*pmdp) &&
+			   !pmd_devmap(*pmdp));
 	pmd = pmdp_huge_get_and_clear(vma->vm_mm, address, pmdp);
 	flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
 	return pmd;
Signature · 1 line
2.26.2
Acked-by on v2 1/10
#1 lore raw
Quoted 32 lines · Hugh Dickins

---------- Forwarded message ----------
Date: Tue, 8 Jun 2021 21:00:12 -0700 (PDT)
From: Hugh Dickins <[email protected]>
To: Andrew Morton <[email protected]>
Cc: Hugh Dickins <[email protected]>,
Kirill A. Shutemov <[email protected]>,
Yang Shi <[email protected]>, Wang Yugui <[email protected]>, Matthew Wilcox <[email protected]>,
Naoya Horiguchi <[email protected]>,
Alistair Popple <[email protected]>, Ralph Campbell <[email protected]>, Zi Yan <[email protected]>, Miaohe Lin <[email protected]>, Minchan Kim <[email protected]>, Jue Wang <[email protected]>, Peter Xu <[email protected]>, Jan Kara <[email protected]>,
Shakeel Butt <[email protected]>, Oscar Salvador <[email protected]>
Subject: [PATCH v2 01/10] mm/thp: fix __split_huge_pmd_locked() on shmem
migration entry

Stressing huge tmpfs page migration racing hole punch often crashed on the VM_BUG_ON(!pmd_present) in pmdp_huge_clear_flush(), with DEBUG_VM=y kernel; or shortly afterwards, on a bad dereference in __split_huge_pmd_locked() when DEBUG_VM=n. They forgot to allow for pmd migration entries in the non-anonymous case.

Full disclosure: those particular experiments were on a kernel with more relaxed mmap_lock and i_mmap_rwsem locking, and were not repeated on the vanilla kernel: it is conceivable that stricter locking happens to avoid those cases, or makes them less likely; but __split_huge_pmd_locked() already allowed for pmd migration entries when handling anonymous THPs, so this commit brings the shmem and file THP handling into line.

Quoted 7 lines

And while there: use old_pmd rather than _pmd, as in the following blocks; and make it clearer to the eye that the !vma_is_anonymous() block is self-contained, making an early return after accounting for unmapping.

Fixes e71769ae5260 ("mm: enable thp migration for shmem thp")
Signature · 1 line
 Kirill A. Shutemov
Reviewed-by on v2 1/10
#2 lore raw
In reply to Hugh Dickins
Quoted 33 lines · Hugh Dickins

---------- Forwarded message ----------
Date: Tue, 8 Jun 2021 21:00:12 -0700 (PDT)
From: Hugh Dickins <[email protected]>
To: Andrew Morton <[email protected]>
Cc: Hugh Dickins <[email protected]>,
Kirill A. Shutemov <[email protected]>,
Yang Shi <[email protected]>, Wang Yugui <[email protected]>, Matthew Wilcox <[email protected]>,
Naoya Horiguchi <[email protected]>,
Alistair Popple <[email protected]>, Ralph Campbell <[email protected]>, Zi Yan <[email protected]>, Miaohe Lin <[email protected]>, Minchan Kim <[email protected]>, Jue Wang <[email protected]>, Peter Xu <[email protected]>, Jan Kara <[email protected]>,
Shakeel Butt <[email protected]>, Oscar Salvador <[email protected]>
Subject: [PATCH v2 01/10] mm/thp: fix __split_huge_pmd_locked() on shmem
migration entry

Stressing huge tmpfs page migration racing hole punch often crashed on the VM_BUG_ON(!pmd_present) in pmdp_huge_clear_flush(), with DEBUG_VM=y kernel; or shortly afterwards, on a bad dereference in __split_huge_pmd_locked() when DEBUG_VM=n. They forgot to allow for pmd migration entries in the non-anonymous case.

Full disclosure: those particular experiments were on a kernel with more relaxed mmap_lock and i_mmap_rwsem locking, and were not repeated on the vanilla kernel: it is conceivable that stricter locking happens to avoid those cases, or makes them less likely; but __split_huge_pmd_locked() already allowed for pmd migration entries when handling anonymous THPs, so this commit brings the shmem and file THP handling into line.

Quoted 7 lines

And while there: use old_pmd rather than _pmd, as in the following blocks; and make it clearer to the eye that the !vma_is_anonymous() block is self-contained, making an early return after accounting for unmapping.

Fixes e71769ae5260 ("mm: enable thp migration for shmem thp")
Quoted 74 lines

---
v2: omit is_huge_zero_pmd() mods (done differently in next), per Kirill

 mm/huge_memory.c     | 27 ++++++++++++++++++---------
 mm/pgtable-generic.c |  5 ++---
 2 files changed, 20 insertions(+), 12 deletions(-)
mm/huge_memory.c +18 -9
index 63ed6b25deaa..42cfefc6e66e 100644
@@ -2044,7 +2044,7 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
        count_vm_event(THP_SPLIT_PMD);
 
        if (!vma_is_anonymous(vma)) {
-               _pmd = pmdp_huge_clear_flush_notify(vma, haddr, pmd);
+               old_pmd = pmdp_huge_clear_flush_notify(vma, haddr, pmd);
                /*
                 * We are going to unmap this huge page. So
                 * just go ahead and zap it
@@ -2053,16 +2053,25 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
                        zap_deposited_table(mm, pmd);
                if (vma_is_special_huge(vma))
                        return;
-               page = pmd_page(_pmd);
-               if (!PageDirty(page) && pmd_dirty(_pmd))
-                       set_page_dirty(page);
-               if (!PageReferenced(page) && pmd_young(_pmd))
-                       SetPageReferenced(page);
-               page_remove_rmap(page, true);
-               put_page(page);
+               if (unlikely(is_pmd_migration_entry(old_pmd))) {
+                       swp_entry_t entry;
+
+                       entry = pmd_to_swp_entry(old_pmd);
+                       page = migration_entry_to_page(entry);
+               } else {
+                       page = pmd_page(old_pmd);
+                       if (!PageDirty(page) && pmd_dirty(old_pmd))
+                               set_page_dirty(page);
+                       if (!PageReferenced(page) && pmd_young(old_pmd))
+                               SetPageReferenced(page);
+                       page_remove_rmap(page, true);
+                       put_page(page);
+               }
                add_mm_counter(mm, mm_counter_file(page), -HPAGE_PMD_NR);
                return;
-       } else if (pmd_trans_huge(*pmd) && is_huge_zero_pmd(*pmd)) {
+       }
+
+       if (pmd_trans_huge(*pmd) && is_huge_zero_pmd(*pmd)) {
                /*
                 * FIXME: Do we want to invalidate secondary mmu by calling
                 * mmu_notifier_invalidate_range() see comments below inside
mm/pgtable-generic.c +2 -4
index c2210e1cdb51..4e640baf9794 100644
@@ -135,9 +135,8 @@ pmd_t pmdp_huge_clear_flush(struct vm_area_struct *vma, unsigned long address,
 {
        pmd_t pmd;
        VM_BUG_ON(address & ~HPAGE_PMD_MASK);
-       VM_BUG_ON(!pmd_present(*pmdp));
-       /* Below assumes pmd_present() is true */
-       VM_BUG_ON(!pmd_trans_huge(*pmdp) && !pmd_devmap(*pmdp));
+       VM_BUG_ON(pmd_present(*pmdp) && !pmd_trans_huge(*pmdp) &&
+                          !pmd_devmap(*pmdp));
        pmd = pmdp_huge_get_and_clear(vma->vm_mm, address, pmdp);
        flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
        return pmd;
--

2.26.2