mm/pagemap: add PAGE_IS_ACCESSED for RWP tracking

PAGEMAP_SCAN already reports PAGE_IS_WRITTEN from the inverted uffd PTE bit, targeting the UFFDIO_WRITEPROTECT workflow. UFFDIO_RWPROTECT reuses the same PTE bit as a marker for read-write protection, but "has been written" and "has been accessed" are distinct semantic signals — they happen to share one PTE bit today only because the two implementations share infrastructure. Give RWP its own pagemap category so the UAPI does not conflate them: PAGE_IS_WRITTEN reported on VM_UFFD_WP VMAs, !pte_uffd(pte) PAGE_IS_ACCESSED reported on VM_UFFD_RWP VMAs, !pte_uffd(pte) Both still read the same PTE bit today, but each is scoped to the VMA whose registered mode makes the bit meaningful. If a future implementation moves RWP to a separate PTE bit, only PAGE_IS_ACCESSED switches over. This is a UAPI narrowing. Outside VM_UFFD_WP VMAs the uffd bit is always clear, so PAGEMAP_SCAN used to flag PAGE_IS_WRITTEN on every present PTE there — a meaningless duplicate of PAGE_IS_PRESENT. Now PAGE_IS_WRITTEN fires only inside VM_UFFD_WP VMAs. pagemap_hugetlb_category() now takes the vma like its PTE/PMD peers.

Kiryl Shutsemau (Meta) committed 4 months ago

Commit 33896481 parent 2d427bb4 files · +67 −29Browse files

Unified · Split

4 files

Documentation/admin-guide/mm/pagemap.rst +10 −3
fs/proc/task_mmu.c +55 −26
include/uapi/linux/fs.h +1 −0
tools/include/uapi/linux/fs.h +1 −0

Documentation/admin-guide/mm/pagemap.rst+10−3

     * Bit  55    pte is soft-dirty (see
       Documentation/admin-guide/mm/soft-dirty.rst)
     * Bit  56    page exclusively mapped (since 4.2)
-    * Bit  57    pte is uffd-wp write-protected (since 5.13) (see
+    * Bit  57    pte is tracked by userfaultfd (since 5.13) — in a
-      Documentation/admin-guide/mm/userfaultfd.rst)
+      ``VM_UFFD_WP`` VMA this indicates a write-protected PTE; in a
+      ``VM_UFFD_RWP`` VMA it indicates an RWP-protected PTE. WP and
+      RWP are mutually exclusive per VMA, so the meaning is
+      unambiguous. See Documentation/admin-guide/mm/userfaultfd.rst.
     * Bit  58    pte is a guard region (since 6.15) (see madvise (2) man page)
     * Bits 59-60 zero
     * Bit  61    page is file-page or shared-anon (since 3.5)
 Following flags about pages are currently supported:
 - ``PAGE_IS_WPALLOWED`` - Page has async-write-protection enabled
-- ``PAGE_IS_WRITTEN`` - Page has been written to from the time it was write protected
+- ``PAGE_IS_WRITTEN`` - Page in a ``UFFDIO_REGISTER_MODE_WP`` VMA has been
+  written to since it was write-protected. Only reported inside such VMAs.
 - ``PAGE_IS_FILE`` - Page is file backed
 - ``PAGE_IS_PRESENT`` - Page is present in the memory
 - ``PAGE_IS_SWAPPED`` - Page is in swapped
 - ``PAGE_IS_HUGE`` - Page is PMD-mapped THP or Hugetlb backed
 - ``PAGE_IS_SOFT_DIRTY`` - Page is soft-dirty
 - ``PAGE_IS_GUARD`` - Page is a part of a guard region
+- ``PAGE_IS_ACCESSED`` - Page in a ``UFFDIO_REGISTER_MODE_RWP`` VMA has been
+  accessed since RWP was applied. Only reported inside such VMAs. See
+  Documentation/admin-guide/mm/userfaultfd.rst for the RWP workflow.
 The ``struct pm_scan_arg`` is used as the argument of the IOCTL.

fs/proc/task_mmu.c+55−26

  * Bits 5-54  swap offset if swapped
  * Bit  55    pte is soft-dirty (see Documentation/admin-guide/mm/soft-dirty.rst)
  * Bit  56    page exclusively mapped
- * Bit  57    pte is uffd-wp write-protected
+ * Bit  57    pte is tracked by userfaultfd (uffd-wp or RWP)
  * Bit  58    pte is a guard region
  * Bits 59-60 zero
  * Bit  61    page is file-page or shared-anon
 				 PAGE_IS_FILE |	PAGE_IS_PRESENT |	\
 				 PAGE_IS_SWAPPED | PAGE_IS_PFNZERO |	\
 				 PAGE_IS_HUGE | PAGE_IS_SOFT_DIRTY |	\
-				 PAGE_IS_GUARD)
+				 PAGE_IS_GUARD | PAGE_IS_ACCESSED)
 #define PM_SCAN_FLAGS		(PM_SCAN_WP_MATCHING | PM_SCAN_CHECK_WPASYNC)
 struct pagemap_scan_private {
 	if (pte_none(pte)) {
 		/*
-		 * An unpopulated pte carries no uffd-wp marker, i.e. it is not
+		 * An unpopulated pte carries no uffd bit, i.e. it is not
-		 * write-protected, the same condition under which the present
+		 * write-protected. The PAGE_IS_WRITTEN fast path in
-		 * and swap cases below report PAGE_IS_WRITTEN. Report it here
+		 * pagemap_scan_pmd_entry() is now gated on a VM_UFFD_WP VMA;
-		 * too so this generic path agrees with the PAGE_IS_WRITTEN fast
+		 * gate the pte_none report here the same way so the two paths
-		 * path in pagemap_scan_pmd_entry(), which reports pte_none as
+		 * still agree. RWP has no such fast path and an unpopulated
-		 * written and, under PM_SCAN_WP_MATCHING, arms a marker. The
+		 * page is not part of the RWP working set, so it is reported as
-		 * fast path applies no VMA test, so neither does this.
+		 * neither.
 		 */
-		return PAGE_IS_WRITTEN;
+		if (userfaultfd_wp(vma))
+			return PAGE_IS_WRITTEN;
+		return 0;
 	}
 	if (pte_present(pte)) {
 		categories = PAGE_IS_PRESENT;
-		if (!pte_uffd(pte))
+		if (!pte_uffd(pte)) {
-			categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_wp(vma))
+				categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_rwp(vma))
+				categories |= PAGE_IS_ACCESSED;
+		}
 		if (p->masks_of_interest & PAGE_IS_FILE) {
 			page = vm_normal_page(vma, addr, pte);
 		categories = PAGE_IS_SWAPPED;
-		if (!pte_swp_uffd_any(pte))
+		if (!pte_swp_uffd_any(pte)) {
-			categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_wp(vma))
+				categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_rwp(vma))
+				categories |= PAGE_IS_ACCESSED;
+		}
 		entry = softleaf_from_pte(pte);
 		if (softleaf_is_guard_marker(entry))
 		struct page *page;
 		categories |= PAGE_IS_PRESENT;
-		if (!pmd_uffd(pmd))
+		if (!pmd_uffd(pmd)) {
-			categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_wp(vma))
+				categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_rwp(vma))
+				categories |= PAGE_IS_ACCESSED;
+		}
 		if (p->masks_of_interest & PAGE_IS_FILE) {
 			page = vm_normal_page_pmd(vma, addr, pmd);
 			categories |= PAGE_IS_SOFT_DIRTY;
 	} else {
 		categories |= PAGE_IS_SWAPPED;
-		if (!pmd_swp_uffd(pmd))
+		if (!pmd_swp_uffd(pmd)) {
-			categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_wp(vma))
+				categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_rwp(vma))
+				categories |= PAGE_IS_ACCESSED;
+		}
 		if (pmd_swp_soft_dirty(pmd))
 			categories |= PAGE_IS_SOFT_DIRTY;
 #endif /* CONFIG_TRANSPARENT_HUGEPAGE */
 #ifdef CONFIG_HUGETLB_PAGE
-static unsigned long pagemap_hugetlb_category(pte_t pte)
+static unsigned long pagemap_hugetlb_category(struct vm_area_struct *vma,
+					      pte_t pte)
 {
 	unsigned long categories = PAGE_IS_HUGE;
 	if (pte_present(pte)) {
 		categories |= PAGE_IS_PRESENT;
-		if (!huge_pte_uffd(pte))
+		if (!huge_pte_uffd(pte)) {
-			categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_wp(vma))
+				categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_rwp(vma))
+				categories |= PAGE_IS_ACCESSED;
+		}
 		if (!PageAnon(pte_page(pte)))
 			categories |= PAGE_IS_FILE;
 		if (is_zero_pfn(pte_pfn(pte)))
 	} else {
 		categories |= PAGE_IS_SWAPPED;
-		if (!pte_swp_uffd_any(pte))
+		if (!pte_swp_uffd_any(pte)) {
-			categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_wp(vma))
+				categories |= PAGE_IS_WRITTEN;
+			if (userfaultfd_rwp(vma))
+				categories |= PAGE_IS_ACCESSED;
+		}
 		if (pte_swp_soft_dirty(pte))
 			categories |= PAGE_IS_SOFT_DIRTY;
 	}
 		goto flush_and_return;
 	}
-	if (!p->arg.category_anyof_mask && !p->arg.category_inverted &&
+	if (userfaultfd_wp(vma) && !p->arg.category_anyof_mask &&
+	    !p->arg.category_inverted &&
 	    p->arg.category_mask == PAGE_IS_WRITTEN &&
 	    p->arg.return_mask == PAGE_IS_WRITTEN) {
 		for (addr = start; addr < end; pte++, addr += PAGE_SIZE) {
 		/* Go the short route when not write-protecting pages. */
 		pte = huge_ptep_get(walk->mm, start, ptep);
-		categories = p->cur_vma_category | pagemap_hugetlb_category(pte);
+		categories = p->cur_vma_category |
+			     pagemap_hugetlb_category(vma, pte);
 		if (!pagemap_scan_is_interesting_page(categories, p))
 			return 0;
 	ptl = huge_pte_lock(hstate_vma(vma), vma->vm_mm, ptep);
 	pte = huge_ptep_get(walk->mm, start, ptep);
-	categories = p->cur_vma_category | pagemap_hugetlb_category(pte);
+	categories = p->cur_vma_category | pagemap_hugetlb_category(vma, pte);
 	if (!pagemap_scan_is_interesting_page(categories, p))
 		goto out_unlock;

include/uapi/linux/fs.h+1−0

@@ -471,6 +471,7 @@ typedef int __bitwise __kernel_rwf_t; · 470 unmodified lines
471	#define PAGE_IS_HUGE (1 << 6)	471	#define PAGE_IS_HUGE (1 << 6)
472	#define PAGE_IS_SOFT_DIRTY (1 << 7)	472	#define PAGE_IS_SOFT_DIRTY (1 << 7)
473	#define PAGE_IS_GUARD (1 << 8)	473	#define PAGE_IS_GUARD (1 << 8)
		474	#define PAGE_IS_ACCESSED (1 << 9)
474		475
475	/*	476	/*
476	* struct page_region - Page region with flags	477	* struct page_region - Page region with flags

tools/include/uapi/linux/fs.h+1−0

@@ -364,6 +364,7 @@ typedef int __bitwise __kernel_rwf_t; · 363 unmodified lines
364	#define PAGE_IS_HUGE (1 << 6)	364	#define PAGE_IS_HUGE (1 << 6)
365	#define PAGE_IS_SOFT_DIRTY (1 << 7)	365	#define PAGE_IS_SOFT_DIRTY (1 << 7)
366	#define PAGE_IS_GUARD (1 << 8)	366	#define PAGE_IS_GUARD (1 << 8)
		367	#define PAGE_IS_ACCESSED (1 << 9)
367		368
368	/*	369	/*
369	* struct page_region - Page region with flags	370	* struct page_region - Page region with flags