From: Qi Zheng <zhengqi.arch@bytedance.com>
To: david@redhat.com, jannh@google.com, hughd@google.com,
willy@infradead.org, muchun.song@linux.dev, vbabka@kernel.org,
akpm@linux-foundation.org, peterx@redhat.com
Cc: mgorman@suse.de, catalin.marinas@arm.com, will@kernel.org,
dave.hansen@linux.intel.com, luto@kernel.org,
peterz@infradead.org, x86@kernel.org, lorenzo.stoakes@oracle.com,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
zokeefe@google.com, rientjes@google.com,
Qi Zheng <zhengqi.arch@bytedance.com>
Subject: [PATCH v3 2/9] mm: userfaultfd: recheck dst_pmd entry in move_pages_pte()
Date: Thu, 14 Nov 2024 14:59:53 +0800 [thread overview]
Message-ID: <ce9830b7725bf2b85d55a57c6317ed1141e96acb.1731566457.git.zhengqi.arch@bytedance.com> (raw)
In-Reply-To: <cover.1731566457.git.zhengqi.arch@bytedance.com>
In move_pages_pte(), since dst_pte needs to be none, the subsequent
pte_same() check cannot prevent the dst_pte page from being freed
concurrently, so we also need to abtain dst_pmdval and recheck pmd_same().
Otherwise, once we support empty PTE page reclaimation for anonymous
pages, it may result in moving the src_pte page into the dts_pte page that
is about to be freed by RCU.
Signed-off-by: Qi Zheng <zhengqi.arch@bytedance.com>
---
mm/userfaultfd.c | 51 +++++++++++++++++++++++++++++++-----------------
1 file changed, 33 insertions(+), 18 deletions(-)
diff --git a/mm/userfaultfd.c b/mm/userfaultfd.c
index 60a0be33766ff..8e16dc290ddf1 100644
--- a/mm/userfaultfd.c
+++ b/mm/userfaultfd.c
@@ -1020,6 +1020,14 @@ void double_pt_unlock(spinlock_t *ptl1,
__release(ptl2);
}
+static inline bool is_pte_pages_stable(pte_t *dst_pte, pte_t *src_pte,
+ pte_t orig_dst_pte, pte_t orig_src_pte,
+ pmd_t *dst_pmd, pmd_t dst_pmdval)
+{
+ return pte_same(ptep_get(src_pte), orig_src_pte) &&
+ pte_same(ptep_get(dst_pte), orig_dst_pte) &&
+ pmd_same(dst_pmdval, pmdp_get_lockless(dst_pmd));
+}
static int move_present_pte(struct mm_struct *mm,
struct vm_area_struct *dst_vma,
@@ -1027,6 +1035,7 @@ static int move_present_pte(struct mm_struct *mm,
unsigned long dst_addr, unsigned long src_addr,
pte_t *dst_pte, pte_t *src_pte,
pte_t orig_dst_pte, pte_t orig_src_pte,
+ pmd_t *dst_pmd, pmd_t dst_pmdval,
spinlock_t *dst_ptl, spinlock_t *src_ptl,
struct folio *src_folio)
{
@@ -1034,8 +1043,8 @@ static int move_present_pte(struct mm_struct *mm,
double_pt_lock(dst_ptl, src_ptl);
- if (!pte_same(ptep_get(src_pte), orig_src_pte) ||
- !pte_same(ptep_get(dst_pte), orig_dst_pte)) {
+ if (!is_pte_pages_stable(dst_pte, src_pte, orig_dst_pte, orig_src_pte,
+ dst_pmd, dst_pmdval)) {
err = -EAGAIN;
goto out;
}
@@ -1071,6 +1080,7 @@ static int move_swap_pte(struct mm_struct *mm,
unsigned long dst_addr, unsigned long src_addr,
pte_t *dst_pte, pte_t *src_pte,
pte_t orig_dst_pte, pte_t orig_src_pte,
+ pmd_t *dst_pmd, pmd_t dst_pmdval,
spinlock_t *dst_ptl, spinlock_t *src_ptl)
{
if (!pte_swp_exclusive(orig_src_pte))
@@ -1078,8 +1088,8 @@ static int move_swap_pte(struct mm_struct *mm,
double_pt_lock(dst_ptl, src_ptl);
- if (!pte_same(ptep_get(src_pte), orig_src_pte) ||
- !pte_same(ptep_get(dst_pte), orig_dst_pte)) {
+ if (!is_pte_pages_stable(dst_pte, src_pte, orig_dst_pte, orig_src_pte,
+ dst_pmd, dst_pmdval)) {
double_pt_unlock(dst_ptl, src_ptl);
return -EAGAIN;
}
@@ -1097,13 +1107,14 @@ static int move_zeropage_pte(struct mm_struct *mm,
unsigned long dst_addr, unsigned long src_addr,
pte_t *dst_pte, pte_t *src_pte,
pte_t orig_dst_pte, pte_t orig_src_pte,
+ pmd_t *dst_pmd, pmd_t dst_pmdval,
spinlock_t *dst_ptl, spinlock_t *src_ptl)
{
pte_t zero_pte;
double_pt_lock(dst_ptl, src_ptl);
- if (!pte_same(ptep_get(src_pte), orig_src_pte) ||
- !pte_same(ptep_get(dst_pte), orig_dst_pte)) {
+ if (!is_pte_pages_stable(dst_pte, src_pte, orig_dst_pte, orig_src_pte,
+ dst_pmd, dst_pmdval)) {
double_pt_unlock(dst_ptl, src_ptl);
return -EAGAIN;
}
@@ -1136,6 +1147,7 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
pte_t *src_pte = NULL;
pte_t *dst_pte = NULL;
pmd_t dummy_pmdval;
+ pmd_t dst_pmdval;
struct folio *src_folio = NULL;
struct anon_vma *src_anon_vma = NULL;
struct mmu_notifier_range range;
@@ -1148,11 +1160,11 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
retry:
/*
* Use the maywrite version to indicate that dst_pte will be modified,
- * but since we will use pte_same() to detect the change of the pte
- * entry, there is no need to get pmdval, so just pass a dummy variable
- * to it.
+ * since dst_pte needs to be none, the subsequent pte_same() check
+ * cannot prevent the dst_pte page from being freed concurrently, so we
+ * also need to abtain dst_pmdval and recheck pmd_same() later.
*/
- dst_pte = pte_offset_map_rw_nolock(mm, dst_pmd, dst_addr, &dummy_pmdval,
+ dst_pte = pte_offset_map_rw_nolock(mm, dst_pmd, dst_addr, &dst_pmdval,
&dst_ptl);
/* Retry if a huge pmd materialized from under us */
@@ -1161,7 +1173,11 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
goto out;
}
- /* same as dst_pte */
+ /*
+ * Unlike dst_pte, the subsequent pte_same() check can ensure the
+ * stability of the src_pte page, so there is no need to get pmdval,
+ * just pass a dummy variable to it.
+ */
src_pte = pte_offset_map_rw_nolock(mm, src_pmd, src_addr, &dummy_pmdval,
&src_ptl);
@@ -1213,7 +1229,7 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
err = move_zeropage_pte(mm, dst_vma, src_vma,
dst_addr, src_addr, dst_pte, src_pte,
orig_dst_pte, orig_src_pte,
- dst_ptl, src_ptl);
+ dst_pmd, dst_pmdval, dst_ptl, src_ptl);
goto out;
}
@@ -1303,8 +1319,8 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
err = move_present_pte(mm, dst_vma, src_vma,
dst_addr, src_addr, dst_pte, src_pte,
- orig_dst_pte, orig_src_pte,
- dst_ptl, src_ptl, src_folio);
+ orig_dst_pte, orig_src_pte, dst_pmd,
+ dst_pmdval, dst_ptl, src_ptl, src_folio);
} else {
entry = pte_to_swp_entry(orig_src_pte);
if (non_swap_entry(entry)) {
@@ -1319,10 +1335,9 @@ static int move_pages_pte(struct mm_struct *mm, pmd_t *dst_pmd, pmd_t *src_pmd,
goto out;
}
- err = move_swap_pte(mm, dst_addr, src_addr,
- dst_pte, src_pte,
- orig_dst_pte, orig_src_pte,
- dst_ptl, src_ptl);
+ err = move_swap_pte(mm, dst_addr, src_addr, dst_pte, src_pte,
+ orig_dst_pte, orig_src_pte, dst_pmd,
+ dst_pmdval, dst_ptl, src_ptl);
}
out:
--
2.20.1
next prev parent reply other threads:[~2024-11-14 7:00 UTC|newest]
Thread overview: 28+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-11-14 6:59 [PATCH v3 0/9] synchronously scan and reclaim empty user PTE pages Qi Zheng
2024-11-14 6:59 ` [PATCH v3 1/9] mm: khugepaged: recheck pmd state in retract_page_tables() Qi Zheng
2024-11-14 6:59 ` Qi Zheng [this message]
2024-11-14 6:59 ` [PATCH v3 3/9] mm: introduce zap_nonpresent_ptes() Qi Zheng
2024-11-14 6:59 ` [PATCH v3 4/9] mm: introduce skip_none_ptes() Qi Zheng
2024-11-14 8:04 ` David Hildenbrand
2024-11-14 9:20 ` Qi Zheng
2024-11-14 12:32 ` David Hildenbrand
2024-11-14 12:51 ` Qi Zheng
2024-11-14 21:19 ` David Hildenbrand
2024-11-15 3:03 ` Qi Zheng
2024-11-15 10:22 ` David Hildenbrand
2024-11-15 14:41 ` Qi Zheng
2024-11-15 14:59 ` David Hildenbrand
2024-11-18 3:35 ` Qi Zheng
2024-11-18 9:29 ` David Hildenbrand
2024-11-18 10:34 ` Qi Zheng
2024-11-18 10:41 ` David Hildenbrand
2024-11-18 10:56 ` Qi Zheng
2024-11-18 10:59 ` David Hildenbrand
2024-11-18 11:13 ` Qi Zheng
2024-11-19 9:55 ` David Hildenbrand
2024-11-19 10:03 ` Qi Zheng
2024-11-14 6:59 ` [PATCH v3 5/9] mm: introduce do_zap_pte_range() Qi Zheng
2024-11-14 6:59 ` [PATCH v3 6/9] mm: make zap_pte_range() handle full within-PMD range Qi Zheng
2024-11-14 6:59 ` [PATCH v3 7/9] mm: pgtable: try to reclaim empty PTE page in madvise(MADV_DONTNEED) Qi Zheng
2024-11-14 6:59 ` [PATCH v3 8/9] x86: mm: free page table pages by RCU instead of semi RCU Qi Zheng
2024-11-14 7:00 ` [PATCH v3 9/9] x86: select ARCH_SUPPORTS_PT_RECLAIM if X86_64 Qi Zheng
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=ce9830b7725bf2b85d55a57c6317ed1141e96acb.1731566457.git.zhengqi.arch@bytedance.com \
--to=zhengqi.arch@bytedance.com \
--cc=akpm@linux-foundation.org \
--cc=catalin.marinas@arm.com \
--cc=dave.hansen@linux.intel.com \
--cc=david@redhat.com \
--cc=hughd@google.com \
--cc=jannh@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lorenzo.stoakes@oracle.com \
--cc=luto@kernel.org \
--cc=mgorman@suse.de \
--cc=muchun.song@linux.dev \
--cc=peterx@redhat.com \
--cc=peterz@infradead.org \
--cc=rientjes@google.com \
--cc=vbabka@kernel.org \
--cc=will@kernel.org \
--cc=willy@infradead.org \
--cc=x86@kernel.org \
--cc=zokeefe@google.com \
/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