Skip to contentSkip to conversation

mm-gup-reliable-r-o-long-term-pinning-in-cow-mappings.patch removed from -mm tree

merged mm-stable Andrew Morton 1 participant 1 message Dec 1, 2022

lore
Share & tools
lore

Andrew Morton

author tree notice Applied
#0 lore raw

The quilt patch titled
Subject: mm/gup: reliable R/O long-term pinning in COW mappings
has been removed from the -mm tree. Its filename was
mm-gup-reliable-r-o-long-term-pinning-in-cow-mappings.patch

This patch was dropped because it was merged into the mm-stable branch of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

------------------------------------------------------
From: David Hildenbrand <[email protected]>
Subject: mm/gup: reliable R/O long-term pinning in COW mappings
Date: Wed, 16 Nov 2022 11:26:48 +0100

We already support reliable R/O pinning of anonymous memory. However, assume we end up pinning (R/O long-term) a pagecache page or the shared zeropage inside a writable private ("COW") mapping. The next write access will trigger a write-fault and replace the pinned page by an exclusive anonymous page in the process page tables to break COW: the pinned page no longer corresponds to the page mapped into the process' page table.

Now that FAULT_FLAG_UNSHARE can break COW on anything mapped into a COW mapping, let's properly break COW first before R/O long-term pinning something that's not an exclusive anon page inside a COW mapping. FAULT_FLAG_UNSHARE will break COW and map an exclusive anon page instead that can get pinned safely.

With this change, we can stop using FOLL_FORCE|FOLL_WRITE for reliable R/O long-term pinning in COW mappings.

With this change, the new R/O long-term pinning tests for non-anonymous memory succeed:
# [RUN] R/O longterm GUP pin ... with shared zeropage
ok 151 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP pin ... with memfd
ok 152 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP pin ... with tmpfile
ok 153 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP pin ... with huge zeropage
ok 154 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP pin ... with memfd hugetlb (2048 kB) ok 155 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP pin ... with memfd hugetlb (1048576 kB) ok 156 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP-fast pin ... with shared zeropage ok 157 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP-fast pin ... with memfd
ok 158 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP-fast pin ... with tmpfile
ok 159 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP-fast pin ... with huge zeropage ok 160 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP-fast pin ... with memfd hugetlb (2048 kB) ok 161 Longterm R/O pin is reliable
# [RUN] R/O longterm GUP-fast pin ... with memfd hugetlb (1048576 kB) ok 162 Longterm R/O pin is reliable

Note 1: We don't care about short-term R/O-pinning, because they have snapshot semantics: they are not supposed to observe modifications that happen after pinning.

As one example, assume we start direct I/O to read from a page and store page content into a file: modifications to page content after starting direct I/O are not guaranteed to end up in the file. So even if we'd pin the shared zeropage, the end result would be as expected -- getting zeroes stored to the file.

Note 2: For shared mappings we'll now always fallback to the slow path to lookup the VMA when R/O long-term pining. While that's the necessary price we have to pay right now, it's actually not that bad in practice: most FOLL_LONGTERM users already specify FOLL_WRITE, for example, along with FOLL_FORCE because they tried dealing with COW mappings correctly ...

Note 3: For users that use FOLL_LONGTERM right now without FOLL_WRITE, such as VFIO, we'd now no longer pin the shared zeropage. Instead, we'd populate exclusive anon pages that we can pin. There was a concern that this could affect the memlock limit of existing setups.

For example, a VM running with VFIO could run into the memlock limit and fail to run. However, we essentially had the same behavior already in commit 17839856fd58 ("gup: document and work around "COW can break either way" issue") which got merged into some enterprise distros, and there were not any such complaints. So most probably, we're fine.

---

 include/linux/mm.h |   27 ++++++++++++++++++++++++---
 mm/gup.c           |   10 +++++-----
 mm/huge_memory.c   |    2 +-
 mm/hugetlb.c       |    7 ++++---
 4 files changed, 34 insertions(+), 12 deletions(-)
include/linux/mm.h renamed from include/linux/mm.h~mm-gup-reliable-r-o-long-term-pinning-in-cow-mappings +24 -3
@@ -3149,8 +3149,12 @@ static inline int vm_fault_to_errno(vm_f
  * Must be called with the (sub)page that's actually referenced via the
  * page table entry, which might not necessarily be the head page for a
  * PTE-mapped THP.
+ *
+ * If the vma is NULL, we're coming from the GUP-fast path and might have
+ * to fallback to the slow path just to lookup the vma.
  */
-static inline bool gup_must_unshare(unsigned int flags, struct page *page)
+static inline bool gup_must_unshare(struct vm_area_struct *vma,
+				    unsigned int flags, struct page *page)
 {
 	/*
 	 * FOLL_WRITE is implicitly handled correctly as the page table entry
@@ -3163,8 +3167,25 @@ static inline bool gup_must_unshare(unsi
 	 * Note: PageAnon(page) is stable until the page is actually getting
 	 * freed.
 	 */
-	if (!PageAnon(page))
-		return false;
+	if (!PageAnon(page)) {
+		/*
+		 * We only care about R/O long-term pining: R/O short-term
+		 * pinning does not have the semantics to observe successive
+		 * changes through the process page tables.
+		 */
+		if (!(flags & FOLL_LONGTERM))
+			return false;
+
+		/* We really need the vma ... */
+		if (!vma)
+			return true;
+
+		/*
+		 * ... because we only care about writable private ("COW")
+		 * mappings where we have to break COW early.
+		 */
+		return is_cow_mapping(vma->vm_flags);
+	}
 
 	/* Paired with a memory barrier in page_try_share_anon_rmap(). */
 	if (IS_ENABLED(CONFIG_HAVE_FAST_GUP))
mm/gup.c renamed from mm/gup.c~mm-gup-reliable-r-o-long-term-pinning-in-cow-mappings +5 -5
@@ -603,7 +603,7 @@ retry:
 		}
 	}
 
-	if (!pte_write(pte) && gup_must_unshare(flags, page)) {
+	if (!pte_write(pte) && gup_must_unshare(vma, flags, page)) {
 		page = ERR_PTR(-EMLINK);
 		goto out;
 	}
@@ -2380,7 +2380,7 @@ static int gup_pte_range(pmd_t pmd, pmd_
 			goto pte_unmap;
 		}
 
-		if (!pte_write(pte) && gup_must_unshare(flags, page)) {
+		if (!pte_write(pte) && gup_must_unshare(NULL, flags, page)) {
 			gup_put_folio(folio, 1, flags);
 			goto pte_unmap;
 		}
@@ -2566,7 +2566,7 @@ static int gup_hugepte(pte_t *ptep, unsi
 		return 0;
 	}
 
-	if (!pte_write(pte) && gup_must_unshare(flags, &folio->page)) {
+	if (!pte_write(pte) && gup_must_unshare(NULL, flags, &folio->page)) {
 		gup_put_folio(folio, refs, flags);
 		return 0;
 	}
@@ -2632,7 +2632,7 @@ static int gup_huge_pmd(pmd_t orig, pmd_
 		return 0;
 	}
 
-	if (!pmd_write(orig) && gup_must_unshare(flags, &folio->page)) {
+	if (!pmd_write(orig) && gup_must_unshare(NULL, flags, &folio->page)) {
 		gup_put_folio(folio, refs, flags);
 		return 0;
 	}
@@ -2672,7 +2672,7 @@ static int gup_huge_pud(pud_t orig, pud_
 		return 0;
 	}
 
-	if (!pud_write(orig) && gup_must_unshare(flags, &folio->page)) {
+	if (!pud_write(orig) && gup_must_unshare(NULL, flags, &folio->page)) {
 		gup_put_folio(folio, refs, flags);
 		return 0;
 	}
mm/huge_memory.c renamed from mm/huge_memory.c~mm-gup-reliable-r-o-long-term-pinning-in-cow-mappings +1 -1
@@ -1480,7 +1480,7 @@ struct page *follow_trans_huge_pmd(struc
 	if (pmd_protnone(*pmd) && !gup_can_follow_protnone(flags))
 		return NULL;
 
-	if (!pmd_write(*pmd) && gup_must_unshare(flags, page))
+	if (!pmd_write(*pmd) && gup_must_unshare(vma, flags, page))
 		return ERR_PTR(-EMLINK);
 
 	VM_BUG_ON_PAGE((flags & FOLL_PIN) && PageAnon(page) &&
mm/hugetlb.c renamed from mm/hugetlb.c~mm-gup-reliable-r-o-long-term-pinning-in-cow-mappings +4 -3
@@ -6197,7 +6197,8 @@ static void record_subpages_vmas(struct
 	}
 }
 
-static inline bool __follow_hugetlb_must_fault(unsigned int flags, pte_t *pte,
+static inline bool __follow_hugetlb_must_fault(struct vm_area_struct *vma,
+					       unsigned int flags, pte_t *pte,
 					       bool *unshare)
 {
 	pte_t pteval = huge_ptep_get(pte);
@@ -6209,7 +6210,7 @@ static inline bool __follow_hugetlb_must
 		return false;
 	if (flags & FOLL_WRITE)
 		return true;
-	if (gup_must_unshare(flags, pte_page(pteval))) {
+	if (gup_must_unshare(vma, flags, pte_page(pteval))) {
 		*unshare = true;
 		return true;
 	}
@@ -6338,7 +6339,7 @@ long follow_hugetlb_page(struct mm_struc
 		 * directly from any kind of swap entries.
 		 */
 		if (absent ||
-		    __follow_hugetlb_must_fault(flags, pte, &unshare)) {
+		    __follow_hugetlb_must_fault(vma, flags, pte, &unshare)) {
 			vm_fault_t ret;
 			unsigned int fault_flags = 0;
 

_

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

selftests-vm-add-ksm-unmerge-tests.patch mm-pagewalk-dont-trigger-test_walk-in-walk_page_vma.patch selftests-vm-add-test-to-measure-madv_unmergeable-performance.patch mm-ksm-simplify-break_ksm-to-not-rely-on-vm_fault_write.patch mm-remove-vm_fault_write.patch mm-ksm-fix-ksm-cow-breaking-with-userfaultfd-wp-via-fault_flag_unshare.patch mm-pagewalk-add-walk_page_range_vma.patch mm-ksm-convert-break_ksm-to-use-walk_page_range_vma.patch mm-gup-remove-foll_migration.patch