From: Max Boone <mboone@akamai.com>
To: Andrew Morton <akpm@linux-foundation.org>,
David Hildenbrand <david@kernel.org>
Cc: Lorenzo Stoakes <lorenzo.stoakes@oracle.com>,
"Liam R . Howlett" <Liam.Howlett@oracle.com>,
Vlastimil Babka <vbabka@suse.cz>, Mike Rapoport <rppt@kernel.org>,
Suren Baghdasaryan <surenb@google.com>,
Michal Hocko <mhocko@suse.com>,
Alex Williamson <alex@shazbot.org>,
linux-mm@kvack.org, kvm@vger.kernel.org,
linux-kernel@vger.kernel.org, Max Tottenham <mtottenh@akamai.com>,
Josh Hunt <johunt@akamai.com>, Matt Pelland <mpelland@akamai.com>,
Max Boone <mboone@akamai.com>
Subject: [RFC 1/1] mm/pagewalk: don't split device-backed huge pfnmaps
Date: Mon, 9 Mar 2026 18:49:49 +0100 [thread overview]
Message-ID: <20260309174949.2514565-2-mboone@akamai.com> (raw)
In-Reply-To: <20260309174949.2514565-1-mboone@akamai.com>
Don't split and descend on special PMD/PUDs, which are generally
device-backed huge pfnmaps as used by vfio for BAR mapping. These
can be faulted back in after splitting and before descending, which
can race to an illegal read.
Signed-off-by: Max Boone <mboone@akamai.com>
Signed-off-by: Max Tottenham <mtottenh@akamai.com>
---
mm/pagewalk.c | 24 ++++++++++++++++++++----
1 file changed, 20 insertions(+), 4 deletions(-)
diff --git a/mm/pagewalk.c b/mm/pagewalk.c
index a94c401ab..d1460dd84 100644
--- a/mm/pagewalk.c
+++ b/mm/pagewalk.c
@@ -147,10 +147,18 @@ static int walk_pmd_range(pud_t *pud, unsigned long addr, unsigned long end,
continue;
}
- if (walk->vma)
+ if (walk->vma) {
+ /*
+ * Don't descend into device-backed pfnmaps,
+ * they might refault the PMD entry.
+ */
+ if (unlikely(pmd_special(*pmd)))
+ continue;
+
split_huge_pmd(walk->vma, pmd, addr);
- else if (pmd_leaf(*pmd) || !pmd_present(*pmd))
+ } else if (pmd_leaf(*pmd) || !pmd_present(*pmd)) {
continue; /* Nothing to do. */
+ }
err = walk_pte_range(pmd, addr, next, walk);
if (err)
@@ -213,10 +221,18 @@ static int walk_pud_range(p4d_t *p4d, unsigned long addr, unsigned long end,
continue;
}
- if (walk->vma)
+ if (walk->vma) {
+ /*
+ * Don't descend into device-backed pfnmaps,
+ * they might refault the PUD entry.
+ */
+ if (unlikely(pud_special(*pud)))
+ continue;
+
split_huge_pud(walk->vma, pud, addr);
- else if (pud_leaf(*pud) || !pud_present(*pud))
+ } else if (pud_leaf(*pud) || !pud_present(*pud)) {
continue; /* Nothing to do. */
+ }
if (pud_none(*pud))
goto again;
--
2.34.1
next prev parent reply other threads:[~2026-03-09 17:51 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-03-09 17:49 [RFC 0/1] Avoid pagewalk hugepage-split race with VFIO DMA set Max Boone
2026-03-09 17:49 ` Max Boone [this message]
2026-03-09 20:19 ` [RFC 1/1] mm/pagewalk: don't split device-backed huge pfnmaps David Hildenbrand (Arm)
2026-03-09 22:47 ` Boone, Max
2026-03-09 23:02 ` Boone, Max
2026-03-10 9:11 ` David Hildenbrand (Arm)
2026-03-10 11:38 ` Boone, Max
2026-03-10 15:19 ` David Hildenbrand (Arm)
2026-03-11 9:42 ` Boone, Max
2026-03-11 9:59 ` David Hildenbrand (Arm)
2026-03-11 10:34 ` Boone, Max
2026-03-11 10:45 ` David Hildenbrand (Arm)
2026-03-11 11:14 ` Boone, Max
2026-03-11 11:59 ` David Hildenbrand (Arm)
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=20260309174949.2514565-2-mboone@akamai.com \
--to=mboone@akamai.com \
--cc=Liam.Howlett@oracle.com \
--cc=akpm@linux-foundation.org \
--cc=alex@shazbot.org \
--cc=david@kernel.org \
--cc=johunt@akamai.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=lorenzo.stoakes@oracle.com \
--cc=mhocko@suse.com \
--cc=mpelland@akamai.com \
--cc=mtottenh@akamai.com \
--cc=rppt@kernel.org \
--cc=surenb@google.com \
--cc=vbabka@suse.cz \
/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