From: Will Deacon <will@kernel.org>
To: Yu Zhao <yuzhao@google.com>
Cc: linux-kernel@vger.kernel.org, kernel-team@android.com,
Catalin Marinas <catalin.marinas@arm.com>,
Minchan Kim <minchan@kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Linus Torvalds <torvalds@linux-foundation.org>,
Anshuman Khandual <anshuman.khandual@arm.com>,
linux-mm@kvack.org, linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH 6/6] mm: proc: Avoid fullmm flush for young/dirty bit toggling
Date: Tue, 24 Nov 2020 14:31:16 +0000 [thread overview]
Message-ID: <20201124143116.GA13352@willie-the-truck> (raw)
In-Reply-To: <20201124011334.GA140483@google.com>
On Mon, Nov 23, 2020 at 06:13:34PM -0700, Yu Zhao wrote:
> On Mon, Nov 23, 2020 at 09:17:51PM +0000, Will Deacon wrote:
> > On Mon, Nov 23, 2020 at 01:04:03PM -0700, Yu Zhao wrote:
> > > On Mon, Nov 23, 2020 at 06:35:55PM +0000, Will Deacon wrote:
> > > > On Fri, Nov 20, 2020 at 01:40:05PM -0700, Yu Zhao wrote:
> > > > > On Fri, Nov 20, 2020 at 02:35:57PM +0000, Will Deacon wrote:
> > > > > > clear_refs_write() uses the 'fullmm' API for invalidating TLBs after
> > > > > > updating the page-tables for the current mm. However, since the mm is not
> > > > > > being freed, this can result in stale TLB entries on architectures which
> > > > > > elide 'fullmm' invalidation.
> > > > > >
> > > > > > Ensure that TLB invalidation is performed after updating soft-dirty
> > > > > > entries via clear_refs_write() by using the non-fullmm API to MMU gather.
> > > > > >
> > > > > > Signed-off-by: Will Deacon <will@kernel.org>
> > > > > > ---
> > > > > > fs/proc/task_mmu.c | 2 +-
> > > > > > 1 file changed, 1 insertion(+), 1 deletion(-)
> > > > > >
> > > > > > diff --git a/fs/proc/task_mmu.c b/fs/proc/task_mmu.c
> > > > > > index a76d339b5754..316af047f1aa 100644
> > > > > > --- a/fs/proc/task_mmu.c
> > > > > > +++ b/fs/proc/task_mmu.c
> > > > > > @@ -1238,7 +1238,7 @@ static ssize_t clear_refs_write(struct file *file, const char __user *buf,
> > > > > > count = -EINTR;
> > > > > > goto out_mm;
> > > > > > }
> > > > > > - tlb_gather_mmu_fullmm(&tlb, mm);
> > > > > > + tlb_gather_mmu(&tlb, mm, 0, TASK_SIZE);
> > > > >
> > > > > Let's assume my reply to patch 4 is wrong, and therefore we still need
> > > > > tlb_gather/finish_mmu() here. But then wouldn't this change deprive
> > > > > architectures other than ARM the opportunity to optimize based on the
> > > > > fact it's a full-mm flush?
> > >
> > > I double checked my conclusion on patch 4, and aside from a couple
> > > of typos, it still seems correct after the weekend.
> >
> > I still need to digest that, but I would prefer that we restore the
> > invalidation first, and then have a subsequent commit to relax it. I find
> > it hard to believe that the behaviour in mainline at the moment is deliberate.
> >
> > That is, I'm not against optimising this, but I'd rather get it "obviously
> > correct" first and the current code is definitely not that.
>
> I wouldn't mind having this patch and patch 4 if the invalidation they
> restore were in a correct state -- b3a81d0841a9 ("mm: fix KSM data
> corruption") isn't correct to start with.
>
> It is complicated, so please bear with me. Let's study this by looking
> at examples this time.
Thanks for putting these together. If you're right, then it looks like it's
even worse than I thought :(
> > > > Only for the soft-dirty case, but I think TLB invalidation is required
> > > > there because we are write-protecting the entries and I don't see any
> > > > mechanism to handle lazy invalidation for that (compared with the aging
> > > > case, which is handled via pte_accessible()).
> > >
> > > The lazy invalidation for that is done when we write-protect a page,
> > > not an individual PTE. When we do so, our decision is based on both
> > > the dirty bit and the writable bit on each PTE mapping this page. So
> > > we only need to make sure we don't lose both on a PTE. And we don't
> > > here.
> >
> > Sorry, I don't follow what you're getting at here (page vs pte). Please can
> > you point me to the code you're referring to? The case I'm worried about is
> > code that holds sufficient locks (e.g. mmap_sem + ptl) finding an entry
> > where !pte_write() and assuming (despite pte_dirty()) that there can't be
> > any concurrent modifications to the mapped page. Granted, I haven't found
> > anything doing that, but I could not convince myself that it would be a bug
> > to write such code, either.
>
> Example 1: memory corruption is still possible with patch 4 & 6
>
> CPU0 CPU1 CPU2 CPU3
> ---- ---- ---- ----
> userspace page writeback
>
> [cache writable
> PTE in TLB]
>
> inc_tlb_flush_pending()
> clean_record_pte()
> pte_mkclean()
This path: ^^^^^ looks a bit weird to me and I _think_ only happens
via some vmware DRM driver (i.e. the only caller of
clean_record_shared_mapping_range()). Are you sure that's operating on
pages that can be reclaimed? I have a feeling it might all be pinned.
> tlb_gather_mmu()
> [set mm_tlb_flush_pending()]
> clear_refs_write()
> pte_wrprotect()
>
> page_mkclean_one()
> !pte_dirty() && !pte_write()
> [true, no flush]
>
> write page to disk
>
> Write to page
> [using stale PTE]
>
> drop clean page
> [data integrity compromised]
>
> flush_tlb_range()
>
> tlb_finish_mmu()
> [flush (with patch 4)]
Setting my earlier comment aside, I think a useful observation here
is that even with correct TLB invalidation, there is still a window
between modifying the page-table and flushing the TLB where another CPU
could see the updated page-table and incorrectly elide a flush. In these
cases we need to rely either on locking or use of tlb_flush_pending() to
ensure the correct behaviour.
> Example 2: why no flush when write-protecting is not a problem (after
> we fix the problem correctly by adding mm_tlb_flush_pending()).
So here you add an mm_tlb_flush_pending() check to the reclaim path
to resolve the race above.
> Case a:
>
> CPU0 CPU1 CPU2 CPU3
> ---- ---- ---- ----
> userspace page writeback
>
> [cache writable
> PTE in TLB]
>
> inc_tlb_flush_pending()
> clean_record_pte()
> pte_mkclean()
>
> clear_refs_write()
> pte_wrprotect()
>
> page_mkclean_one()
> !pte_dirty() && !pte_write() &&
> !mm_tlb_flush_pending()
> [false: flush]
>
> write page to disk
>
> Write to page
> [page fault]
>
> drop clean page
> [data integrity guaranteed]
>
> flush_tlb_range()
>
> Case b:
>
> CPU0 CPU1 CPU2
> ---- ---- ----
> userspace page writeback
>
> [cache writable
> PTE in TLB]
>
> clear_refs_write()
> pte_wrprotect()
> [pte_dirty() is false]
>
> page_mkclean_one()
> !pte_dirty() && !pte_write() &&
> !mm_tlb_flush_pending()
> [true: no flush]
>
> write page to disk
>
> Write to page
> [h/w tries to set
> the dirty bit
> but sees write-
> protected PTE,
> page fault]
I agree with you for this example, but I think if the page writeback ran
on CPU 1 after clear_refs_write() then we could have a problem: the updated
pte could sit in the store buffer of CPU1 and the walker on CPU0 would
be able to set the dirty bit. TLB invalidation in clear_refs_write()
would prevent that.
Will
next prev parent reply other threads:[~2020-11-24 14:31 UTC|newest]
Thread overview: 43+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-20 14:35 [PATCH 0/6] tlb: Fix access and (soft-)dirty bit management Will Deacon
2020-11-20 14:35 ` [PATCH 1/6] arm64: pgtable: Fix pte_accessible() Will Deacon
2020-11-20 16:03 ` Minchan Kim
2020-11-20 19:53 ` Yu Zhao
2020-11-23 13:27 ` Catalin Marinas
2020-11-24 10:02 ` Anshuman Khandual
2020-11-20 14:35 ` [PATCH 2/6] arm64: pgtable: Ensure dirty bit is preserved across pte_wrprotect() Will Deacon
2020-11-20 17:09 ` Minchan Kim
2020-11-23 14:31 ` Catalin Marinas
2020-11-23 14:22 ` Catalin Marinas
2020-11-20 14:35 ` [PATCH 3/6] tlb: mmu_gather: Remove unused start/end arguments from tlb_finish_mmu() Will Deacon
2020-11-20 17:20 ` Linus Torvalds
2020-11-23 16:48 ` Will Deacon
2020-11-20 14:35 ` [PATCH 4/6] mm: proc: Invalidate TLB after clearing soft-dirty page state Will Deacon
2020-11-20 15:00 ` Peter Zijlstra
2020-11-20 15:09 ` Peter Zijlstra
2020-11-20 15:15 ` Will Deacon
2020-11-20 15:27 ` Peter Zijlstra
2020-11-23 18:23 ` Will Deacon
2020-11-20 15:55 ` Minchan Kim
2020-11-23 18:41 ` Will Deacon
2020-11-25 22:51 ` Minchan Kim
2020-11-20 20:22 ` Yu Zhao
2020-11-21 2:49 ` Yu Zhao
2020-11-23 19:21 ` Yu Zhao
2020-11-23 22:04 ` Will Deacon
2020-11-20 14:35 ` [PATCH 5/6] tlb: mmu_gather: Introduce tlb_gather_mmu_fullmm() Will Deacon
2020-11-20 17:22 ` Linus Torvalds
2020-11-20 17:31 ` Linus Torvalds
2020-11-23 16:48 ` Will Deacon
2020-11-22 15:11 ` [tlb] e242a269fa: WARNING:at_mm/mmu_gather.c:#tlb_gather_mmu kernel test robot
2020-11-23 17:51 ` Will Deacon
2020-11-20 14:35 ` [PATCH 6/6] mm: proc: Avoid fullmm flush for young/dirty bit toggling Will Deacon
2020-11-20 17:41 ` Linus Torvalds
2020-11-20 17:45 ` Linus Torvalds
2020-11-20 20:40 ` Yu Zhao
2020-11-23 18:35 ` Will Deacon
2020-11-23 20:04 ` Yu Zhao
2020-11-23 21:17 ` Will Deacon
2020-11-24 1:13 ` Yu Zhao
2020-11-24 14:31 ` Will Deacon [this message]
2020-11-25 22:01 ` Minchan Kim
2020-11-24 14:46 ` Peter Zijlstra
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=20201124143116.GA13352@willie-the-truck \
--to=will@kernel.org \
--cc=anshuman.khandual@arm.com \
--cc=catalin.marinas@arm.com \
--cc=kernel-team@android.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=minchan@kernel.org \
--cc=peterz@infradead.org \
--cc=torvalds@linux-foundation.org \
--cc=yuzhao@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;
as well as URLs for NNTP newsgroup(s).