linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Peter Xu <peterx@redhat.com>
To: linux-kernel@vger.kernel.org, linux-mm@kvack.org
Cc: Mike Kravetz <mike.kravetz@oracle.com>,
	Nadav Amit <nadav.amit@gmail.com>,
	Matthew Wilcox <willy@infradead.org>,
	Mike Rapoport <rppt@linux.vnet.ibm.com>,
	David Hildenbrand <david@redhat.com>,
	Hugh Dickins <hughd@google.com>,
	Jerome Glisse <jglisse@redhat.com>,
	"Kirill A . Shutemov" <kirill@shutemov.name>,
	Andrea Arcangeli <aarcange@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Axel Rasmussen <axelrasmussen@google.com>,
	Alistair Popple <apopple@nvidia.com>,
	peterx@redhat.com
Subject: [PATCH v8 17/23] mm/hugetlb: Only drop uffd-wp special pte if required
Date: Mon,  4 Apr 2022 21:49:15 -0400	[thread overview]
Message-ID: <20220405014915.14873-1-peterx@redhat.com> (raw)
In-Reply-To: <20220405014646.13522-1-peterx@redhat.com>

As with shmem uffd-wp special ptes, only drop the uffd-wp special swap pte if
unmapping an entire vma or synchronized such that faults can not race with the
unmap operation.  This requires passing zap_flags all the way to the lowest
level hugetlb unmap routine: __unmap_hugepage_range.

In general, unmap calls originated in hugetlbfs code will pass the
ZAP_FLAG_DROP_MARKER flag as synchronization is in place to prevent faults.
The exception is hole punch which will first unmap without any synchronization.
Later when hole punch actually removes the page from the file, it will check to
see if there was a subsequent fault and if so take the hugetlb fault mutex
while unmapping again.  This second unmap will pass in ZAP_FLAG_DROP_MARKER.

The justification of "whether to apply ZAP_FLAG_DROP_MARKER flag when unmap a
hugetlb range" is (IMHO): we should never reach a state when a page fault could
errornously fault in a page-cache page that was wr-protected to be writable,
even in an extremely short period.  That could happen if e.g. we pass
ZAP_FLAG_DROP_MARKER when hugetlbfs_punch_hole() calls hugetlb_vmdelete_list(),
because if a page faults after that call and before remove_inode_hugepages() is
executed, the page cache can be mapped writable again in the small racy window,
that can cause unexpected data overwritten.

Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
---
 fs/hugetlbfs/inode.c    | 15 +++++++++------
 include/linux/hugetlb.h |  8 +++++---
 mm/hugetlb.c            | 33 +++++++++++++++++++++++++--------
 mm/memory.c             |  5 ++++-
 4 files changed, 43 insertions(+), 18 deletions(-)

diff --git a/fs/hugetlbfs/inode.c b/fs/hugetlbfs/inode.c
index 99c7477cee5c..8b5b9df2be7d 100644
--- a/fs/hugetlbfs/inode.c
+++ b/fs/hugetlbfs/inode.c
@@ -404,7 +404,8 @@ static void remove_huge_page(struct page *page)
 }
 
 static void
-hugetlb_vmdelete_list(struct rb_root_cached *root, pgoff_t start, pgoff_t end)
+hugetlb_vmdelete_list(struct rb_root_cached *root, pgoff_t start, pgoff_t end,
+		      unsigned long zap_flags)
 {
 	struct vm_area_struct *vma;
 
@@ -438,7 +439,7 @@ hugetlb_vmdelete_list(struct rb_root_cached *root, pgoff_t start, pgoff_t end)
 		}
 
 		unmap_hugepage_range(vma, vma->vm_start + v_offset, v_end,
-									NULL);
+				     NULL, zap_flags);
 	}
 }
 
@@ -516,7 +517,8 @@ static void remove_inode_hugepages(struct inode *inode, loff_t lstart,
 				mutex_lock(&hugetlb_fault_mutex_table[hash]);
 				hugetlb_vmdelete_list(&mapping->i_mmap,
 					index * pages_per_huge_page(h),
-					(index + 1) * pages_per_huge_page(h));
+					(index + 1) * pages_per_huge_page(h),
+					ZAP_FLAG_DROP_MARKER);
 				i_mmap_unlock_write(mapping);
 			}
 
@@ -582,7 +584,8 @@ static void hugetlb_vmtruncate(struct inode *inode, loff_t offset)
 	i_mmap_lock_write(mapping);
 	i_size_write(inode, offset);
 	if (!RB_EMPTY_ROOT(&mapping->i_mmap.rb_root))
-		hugetlb_vmdelete_list(&mapping->i_mmap, pgoff, 0);
+		hugetlb_vmdelete_list(&mapping->i_mmap, pgoff, 0,
+				      ZAP_FLAG_DROP_MARKER);
 	i_mmap_unlock_write(mapping);
 	remove_inode_hugepages(inode, offset, LLONG_MAX);
 }
@@ -615,8 +618,8 @@ static long hugetlbfs_punch_hole(struct inode *inode, loff_t offset, loff_t len)
 		i_mmap_lock_write(mapping);
 		if (!RB_EMPTY_ROOT(&mapping->i_mmap.rb_root))
 			hugetlb_vmdelete_list(&mapping->i_mmap,
-						hole_start >> PAGE_SHIFT,
-						hole_end  >> PAGE_SHIFT);
+					      hole_start >> PAGE_SHIFT,
+					      hole_end >> PAGE_SHIFT, 0);
 		i_mmap_unlock_write(mapping);
 		remove_inode_hugepages(inode, hole_start, hole_end);
 		inode_unlock(inode);
diff --git a/include/linux/hugetlb.h b/include/linux/hugetlb.h
index 38c5ac28b787..ab48b3bbb0e6 100644
--- a/include/linux/hugetlb.h
+++ b/include/linux/hugetlb.h
@@ -143,11 +143,12 @@ long follow_hugetlb_page(struct mm_struct *, struct vm_area_struct *,
 			 unsigned long *, unsigned long *, long, unsigned int,
 			 int *);
 void unmap_hugepage_range(struct vm_area_struct *,
-			  unsigned long, unsigned long, struct page *);
+			  unsigned long, unsigned long, struct page *,
+			  unsigned long);
 void __unmap_hugepage_range_final(struct mmu_gather *tlb,
 			  struct vm_area_struct *vma,
 			  unsigned long start, unsigned long end,
-			  struct page *ref_page);
+			  struct page *ref_page, unsigned long zap_flags);
 void hugetlb_report_meminfo(struct seq_file *);
 int hugetlb_report_node_meminfo(char *buf, int len, int nid);
 void hugetlb_show_meminfo(void);
@@ -400,7 +401,8 @@ static inline unsigned long hugetlb_change_protection(
 
 static inline void __unmap_hugepage_range_final(struct mmu_gather *tlb,
 			struct vm_area_struct *vma, unsigned long start,
-			unsigned long end, struct page *ref_page)
+			unsigned long end, struct page *ref_page,
+			unsigned long zap_flags)
 {
 	BUG();
 }
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 578c48ef931a..e4af8b357b90 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -4947,7 +4947,7 @@ int move_hugetlb_page_tables(struct vm_area_struct *vma,
 
 static void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct *vma,
 				   unsigned long start, unsigned long end,
-				   struct page *ref_page)
+				   struct page *ref_page, unsigned long zap_flags)
 {
 	struct mm_struct *mm = vma->vm_mm;
 	unsigned long address;
@@ -5003,7 +5003,18 @@ static void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct
 		 * unmapped and its refcount is dropped, so just clear pte here.
 		 */
 		if (unlikely(!pte_present(pte))) {
-			huge_pte_clear(mm, address, ptep, sz);
+			/*
+			 * If the pte was wr-protected by uffd-wp in any of the
+			 * swap forms, meanwhile the caller does not want to
+			 * drop the uffd-wp bit in this zap, then replace the
+			 * pte with a marker.
+			 */
+			if (pte_swp_uffd_wp_any(pte) &&
+			    !(zap_flags & ZAP_FLAG_DROP_MARKER))
+				set_huge_pte_at(mm, address, ptep,
+						make_pte_marker(PTE_MARKER_UFFD_WP));
+			else
+				huge_pte_clear(mm, address, ptep, sz);
 			spin_unlock(ptl);
 			continue;
 		}
@@ -5031,7 +5042,11 @@ static void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct
 		tlb_remove_huge_tlb_entry(h, tlb, ptep, address);
 		if (huge_pte_dirty(pte))
 			set_page_dirty(page);
-
+		/* Leave a uffd-wp pte marker if needed */
+		if (huge_pte_uffd_wp(pte) &&
+		    !(zap_flags & ZAP_FLAG_DROP_MARKER))
+			set_huge_pte_at(mm, address, ptep,
+					make_pte_marker(PTE_MARKER_UFFD_WP));
 		hugetlb_count_sub(pages_per_huge_page(h), mm);
 		page_remove_rmap(page, vma, true);
 
@@ -5065,9 +5080,10 @@ static void __unmap_hugepage_range(struct mmu_gather *tlb, struct vm_area_struct
 
 void __unmap_hugepage_range_final(struct mmu_gather *tlb,
 			  struct vm_area_struct *vma, unsigned long start,
-			  unsigned long end, struct page *ref_page)
+			  unsigned long end, struct page *ref_page,
+			  unsigned long zap_flags)
 {
-	__unmap_hugepage_range(tlb, vma, start, end, ref_page);
+	__unmap_hugepage_range(tlb, vma, start, end, ref_page, zap_flags);
 
 	/*
 	 * Clear this flag so that x86's huge_pmd_share page_table_shareable
@@ -5083,12 +5099,13 @@ void __unmap_hugepage_range_final(struct mmu_gather *tlb,
 }
 
 void unmap_hugepage_range(struct vm_area_struct *vma, unsigned long start,
-			  unsigned long end, struct page *ref_page)
+			  unsigned long end, struct page *ref_page,
+			  unsigned long zap_flags)
 {
 	struct mmu_gather tlb;
 
 	tlb_gather_mmu(&tlb, vma->vm_mm);
-	__unmap_hugepage_range(&tlb, vma, start, end, ref_page);
+	__unmap_hugepage_range(&tlb, vma, start, end, ref_page, zap_flags);
 	tlb_finish_mmu(&tlb);
 }
 
@@ -5143,7 +5160,7 @@ static void unmap_ref_private(struct mm_struct *mm, struct vm_area_struct *vma,
 		 */
 		if (!is_vma_resv_set(iter_vma, HPAGE_RESV_OWNER))
 			unmap_hugepage_range(iter_vma, address,
-					     address + huge_page_size(h), page);
+					     address + huge_page_size(h), page, 0);
 	}
 	i_mmap_unlock_write(mapping);
 }
diff --git a/mm/memory.c b/mm/memory.c
index 8ba1bb196095..9808edfe18d4 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1675,8 +1675,11 @@ static void unmap_single_vma(struct mmu_gather *tlb,
 			 * safe to do nothing in this case.
 			 */
 			if (vma->vm_file) {
+				unsigned long zap_flags = details ?
+				    details->zap_flags : 0;
 				i_mmap_lock_write(vma->vm_file->f_mapping);
-				__unmap_hugepage_range_final(tlb, vma, start, end, NULL);
+				__unmap_hugepage_range_final(tlb, vma, start, end,
+							     NULL, zap_flags);
 				i_mmap_unlock_write(vma->vm_file->f_mapping);
 			}
 		} else
-- 
2.32.0


  parent reply	other threads:[~2022-04-05  2:42 UTC|newest]

Thread overview: 63+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-04-05  1:46 [PATCH v8 00/23] userfaultfd-wp: Support shmem and hugetlbfs Peter Xu
2022-04-05  1:46 ` [PATCH v8 01/23] mm: Introduce PTE_MARKER swap entry Peter Xu
2022-04-12  1:07   ` Alistair Popple
2022-04-12 19:45     ` Peter Xu
2022-04-13  0:30       ` Alistair Popple
2022-04-13 13:44         ` Peter Xu
2022-04-19  8:25   ` Alistair Popple
2022-04-19 19:44     ` Peter Xu
2022-04-05  1:48 ` [PATCH v8 02/23] mm: Teach core mm about pte markers Peter Xu
2022-04-12  1:22   ` Alistair Popple
2022-04-12 19:53     ` Peter Xu
2022-04-05  1:48 ` [PATCH v8 03/23] mm: Check against orig_pte for finish_fault() Peter Xu
2022-04-12  2:05   ` Alistair Popple
2022-04-12 19:54     ` Peter Xu
     [not found]   ` <CGME20220413140330eucas1p167da41e079712b829ef8237dc27b049c@eucas1p1.samsung.com>
2022-04-13 14:03     ` Marek Szyprowski
2022-04-13 16:43       ` Peter Xu
2022-04-14  7:51         ` Marek Szyprowski
2022-04-14 16:30           ` Peter Xu
2022-04-14 20:57             ` Andrew Morton
2022-04-14 21:08               ` Peter Xu
2022-04-15 14:21   ` Guenter Roeck
2022-04-15 14:41     ` Peter Xu
2022-04-05  1:48 ` [PATCH v8 04/23] mm/uffd: PTE_MARKER_UFFD_WP Peter Xu
2022-04-06  1:41   ` kernel test robot
2022-04-05  1:48 ` [PATCH v8 05/23] mm/shmem: Take care of UFFDIO_COPY_MODE_WP Peter Xu
2022-04-05  1:48 ` [PATCH v8 06/23] mm/shmem: Handle uffd-wp special pte in page fault handler Peter Xu
2022-05-11 16:30   ` David Hildenbrand
2022-05-12 16:34     ` Peter Xu
2022-04-05  1:48 ` [PATCH v8 07/23] mm/shmem: Persist uffd-wp bit across zapping for file-backed Peter Xu
2022-04-05  1:48 ` [PATCH v8 08/23] mm/shmem: Allow uffd wr-protect none pte for file-backed mem Peter Xu
2022-04-05  1:48 ` [PATCH v8 09/23] mm/shmem: Allows file-back mem to be uffd wr-protected on thps Peter Xu
2022-04-05  1:48 ` [PATCH v8 10/23] mm/shmem: Handle uffd-wp during fork() Peter Xu
2022-04-06  6:16   ` kernel test robot
2022-04-06 12:18     ` Peter Xu
2022-04-05  1:48 ` [PATCH v8 11/23] mm/hugetlb: Introduce huge pte version of uffd-wp helpers Peter Xu
2022-04-05  1:49 ` [PATCH v8 12/23] mm/hugetlb: Hook page faults for uffd write protection Peter Xu
2022-04-05  1:49 ` [PATCH v8 13/23] mm/hugetlb: Take care of UFFDIO_COPY_MODE_WP Peter Xu
2022-04-05  1:49 ` [PATCH v8 14/23] mm/hugetlb: Handle UFFDIO_WRITEPROTECT Peter Xu
2022-04-05  1:49 ` [PATCH v8 15/23] mm/hugetlb: Handle pte markers in page faults Peter Xu
2022-04-06 13:37   ` kernel test robot
2022-04-06 15:02     ` Peter Xu
2022-04-05  1:49 ` [PATCH v8 16/23] mm/hugetlb: Allow uffd wr-protect none ptes Peter Xu
2022-04-05  1:49 ` Peter Xu [this message]
2022-04-05  1:49 ` [PATCH v8 18/23] mm/hugetlb: Handle uffd-wp during fork() Peter Xu
2022-04-05  1:49 ` [PATCH v8 19/23] mm/khugepaged: Don't recycle vma pgtable if uffd-wp registered Peter Xu
2022-04-05  1:49 ` [PATCH v8 20/23] mm/pagemap: Recognize uffd-wp bit for shmem/hugetlbfs Peter Xu
2022-04-05  1:49 ` [PATCH v8 21/23] mm/uffd: Enable write protection for shmem & hugetlbfs Peter Xu
2022-04-05  1:49 ` [PATCH v8 22/23] mm: Enable PTE markers by default Peter Xu
2022-04-19 15:13   ` Johannes Weiner
2022-04-19 19:59     ` Peter Xu
2022-04-19 20:14       ` Johannes Weiner
2022-04-19 20:28         ` Peter Xu
2022-04-19 21:24           ` Johannes Weiner
2022-04-19 22:01             ` Peter Xu
2022-04-20 13:46               ` Johannes Weiner
2022-04-20 14:25                 ` Peter Xu
2022-04-05  1:49 ` [PATCH v8 23/23] selftests/uffd: Enable uffd-wp for shmem/hugetlbfs Peter Xu
2022-04-05 22:16 ` [PATCH v8 00/23] userfaultfd-wp: Support shmem and hugetlbfs Andrew Morton
2022-04-05 22:42   ` Peter Xu
2022-04-05 22:49     ` Andrew Morton
2022-04-05 23:02       ` Peter Xu
2022-04-05 23:08         ` Andrew Morton
2022-05-10 19:05 ` Andrew Morton

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20220405014915.14873-1-peterx@redhat.com \
    --to=peterx@redhat.com \
    --cc=aarcange@redhat.com \
    --cc=akpm@linux-foundation.org \
    --cc=apopple@nvidia.com \
    --cc=axelrasmussen@google.com \
    --cc=david@redhat.com \
    --cc=hughd@google.com \
    --cc=jglisse@redhat.com \
    --cc=kirill@shutemov.name \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=mike.kravetz@oracle.com \
    --cc=nadav.amit@gmail.com \
    --cc=rppt@linux.vnet.ibm.com \
    --cc=willy@infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).