From: "Lorenzo Stoakes (ARM)" <ljs@kernel.org>
To: "David Hildenbrand (Arm)" <david@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>,
"Liam R. Howlett" <liam@infradead.org>,
Vlastimil Babka <vbabka@kernel.org>,
Mike Rapoport <rppt@kernel.org>,
Suren Baghdasaryan <surenb@google.com>,
Michal Hocko <mhocko@suse.com>,
Alex Williamson <alex@shazbot.org>, Peter Xu <peterx@redhat.com>,
Zi Yan <ziy@nvidia.com>,
Baolin Wang <baolin.wang@linux.alibaba.com>,
Nico Pache <nico.pache@linux.dev>,
Ryan Roberts <ryan.roberts@arm.com>, Dev Jain <dev.jain@arm.com>,
Barry Song <baohua@kernel.org>,
Lance Yang <lance.yang@linux.dev>,
Usama Arif <usama.arif@linux.dev>,
Kiryl Shutsemau <kas@kernel.org>, Wei Liu <wei.liu@kernel.org>,
Roman Kisel <romank@linux.microsoft.com>,
Saurabh Sengar <ssengar@linux.microsoft.com>,
Naman Jain <namjain@linux.microsoft.com>,
Michael Kelley <mhklinux@outlook.com>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
Eliav Farber <farbere@amazon.com>,
stable@vger.kernel.org
Subject: Re: [PATCH 1/3] mm/memory: remove racy huge PMD assertion from walk_to_pmd()
Date: Tue, 17 Nov 2026 09:35:30 +0000 [thread overview]
Message-ID: <avwbrON3_cCtaR20@gremlin> (raw)
In-Reply-To: <61d4f3ee-7a0b-4a8b-b5a2-35380dd46b4c@kernel.org>
On Mon, Nov 16, 2026 at 09:06:21PM +0100, David Hildenbrand (Arm) wrote:
> On 11/16/26 18:10, Lorenzo Stoakes (ARM) wrote:
> > On Mon, Nov 16, 2026 at 04:56:25PM +0100, David Hildenbrand (Arm) wrote:
> >> On 11/15/26 21:58, Lorenzo Stoakes (ARM) wrote:
> >>> From: Eliav Farber <farbere@amazon.com>
> >>>
> >>> walk_to_pmd() asserts that the PMD entry it returns is not huge, but does
> >>> so without holding the PMD lock, so the assertion is only valid if nothing
> >>> can concurrently install a huge PMD entry in the range.
> >>>
> >>> This is not the case for a VMA whose driver implements both ->fault and
> >>> ->huge_fault and which invokes vmf_insert_pfn() from the former.
> >>>
> >>> The existence of prctl(PR_SET_THP_DISABLE) is problematic here, as it only
> >>> takes the mmap write lock and faults taken under the VMA lock are
> >>> unaffected by this.
> >>>
> >>> This can result in two concurrent threads faulting the same thing
> >>> disagreeing as to whether to attempt a huge fault and thus ->fault and
> >>> ->huge_fault running at the same time and triggering this assertion.
> >>>
> >>> So just remove the assert.
> >>>
> >>> Target the commit which introduced the first driver able to do this for the
> >>> fix - vfio-pci. DAX also mixes fault orders within a VMA, but serialises
> >>> them using its own mapping entry lock.
> >>>
> >>> Link: https://lore.kernel.org/all/20261112051440.16140-2-farbere@amazon.com/
> >>> Fixes: f9e54c3a2f5b ("vfio/pci: implement huge_fault support")
> >>> Cc: stable@vger.kernel.org
> >>> Signed-off-by: Eliav Farber <farbere@amazon.com>
> >>> [ ljs: rewrote commit message ]
> >>> Signed-off-by: Lorenzo Stoakes (ARM) <ljs@kernel.org>
> >>> ---
> >>> mm/memory.c | 1 -
> >>> 1 file changed, 1 deletion(-)
> >>>
> >>> diff --git a/mm/memory.c b/mm/memory.c
> >>> index 3a20d64fd276..e03110906aa8 100644
> >>> --- a/mm/memory.c
> >>> +++ b/mm/memory.c
> >>> @@ -2384,7 +2384,6 @@ static pmd_t *walk_to_pmd(struct mm_struct *mm, unsigned long addr)
> >>> if (!pmd)
> >>> return NULL;
> >>>
> >>> - VM_BUG_ON(pmd_trans_huge(*pmd));
> >>> return pmd;
> >>> }
> >>>
> >>>
> >>
> >> Could this also stumble over a huge PUD somehow? Should we just prepare it for
> >> any leaves and bail out if we find some?
> >
> > No, walk_to_pmd() can only come from the order-0 path right now and
> > __handle_mm_fault() makes sure that can't happen in the fault path:
>
> That implies that get_locked_pte() and all insert_pages interfaces can only be
> called from that path in __handle_mm_fault().
>
> x86 an s390x use get_locked_pte() for some other paths, so I wouldn't want to
> trust on that :)
Well, I had claude go look through everything and there are no paths which are
problematic _now_ :) but future-proofing is a good idea because assumptions like
this bite you in the... assembly *ahem*
>
> (did I say that I hate walk_to_pmd and get_locked_pte? But you likely guessed it
> at this point lol)
Yeah they're bloody awful aren't they? :)
>
> >
> > vmf.pmd = pmd_alloc(mm, vmf.pud, address);
> > if (!vmf.pmd)
> > return VM_FAULT_OOM;
> >
> > /* Huge pud page fault raced with pmd_alloc? */
> > if (pud_trans_unstable(vmf.pud))
> > goto retry_pud;
> >
> > So by the time any ->fault path runs, the PUD is stable.
> >
> > We don't have to worry about the mmap paths as at that point nothing has
> > faulted in and it's all fresh.
> >
> > So it's all safe as things stand.
> >
> > But could update for future-proofing like:
> >
> > /* PUD leaf might have been installed. */
> > if (pud_trans_unstable(pud))
> > return NULL;
>
> Yeah, that makes sense. I was just wondering whether we'd actually want to
> return a proper error then instead, because this would not be an -ENOMEM case.
Let me have a fiddle and see what I can come up with, maybe we can use this as
an excuse to improve things a bit... :)
>
> --
> Cheers,
>
> David
--
Cheers, Lorenzo
next prev parent reply other threads:[~2026-11-17 9:35 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-11-15 20:58 [PATCH 0/3] mm: fix races between huge and non-huge faults in PFN mappings Lorenzo Stoakes (ARM)
2026-11-15 20:58 ` [PATCH 1/3] mm/memory: remove racy huge PMD assertion from walk_to_pmd() Lorenzo Stoakes (ARM)
2026-11-16 10:35 ` Jose A. Perez de Azpillaga
2026-11-16 15:56 ` David Hildenbrand (Arm)
2026-11-16 17:10 ` Lorenzo Stoakes (ARM)
2026-11-16 20:06 ` David Hildenbrand (Arm)
2026-11-17 9:35 ` Lorenzo Stoakes (ARM) [this message]
2026-11-16 17:40 ` Gregory Price (Meta)
2026-11-15 20:58 ` [PATCH 2/3] mm/memory: do not report OOM if insert_pfn() races a huge fault Lorenzo Stoakes (ARM)
2026-11-16 10:36 ` Jose A. Perez de Azpillaga
2026-11-16 15:59 ` David Hildenbrand (Arm)
2026-11-16 17:21 ` Lorenzo Stoakes (ARM)
2026-11-16 20:10 ` David Hildenbrand (Arm)
2026-11-16 18:13 ` Gregory Price (Meta)
2026-11-15 20:58 ` [PATCH 3/3] mm/huge_memory: only compare PFNs of huge entries on huge fault insert Lorenzo Stoakes (ARM)
2026-11-16 10:42 ` Jose A. Perez de Azpillaga
2026-11-16 16:01 ` David Hildenbrand (Arm)
2026-11-16 16:55 ` Lorenzo Stoakes (ARM)
2026-11-16 20:09 ` David Hildenbrand (Arm)
2026-11-16 18:22 ` Gregory Price (Meta)
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=avwbrON3_cCtaR20@gremlin \
--to=ljs@kernel.org \
--cc=akpm@linux-foundation.org \
--cc=alex@shazbot.org \
--cc=baohua@kernel.org \
--cc=baolin.wang@linux.alibaba.com \
--cc=david@kernel.org \
--cc=dev.jain@arm.com \
--cc=farbere@amazon.com \
--cc=kas@kernel.org \
--cc=lance.yang@linux.dev \
--cc=liam@infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=mhklinux@outlook.com \
--cc=mhocko@suse.com \
--cc=namjain@linux.microsoft.com \
--cc=nico.pache@linux.dev \
--cc=peterx@redhat.com \
--cc=romank@linux.microsoft.com \
--cc=rppt@kernel.org \
--cc=ryan.roberts@arm.com \
--cc=ssengar@linux.microsoft.com \
--cc=stable@vger.kernel.org \
--cc=surenb@google.com \
--cc=usama.arif@linux.dev \
--cc=vbabka@kernel.org \
--cc=wei.liu@kernel.org \
--cc=ziy@nvidia.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