[PATCH] mm/memory: Fix missing pte marker for !page on pte zaps

[PATCH] mm/memory: Fix missing pte marker for !page on pte zaps

[peterx]

From: Peter Xu <peterx@redhat.com>

Commit 0cf18e839f64 of large folio zap work broke uffd-wp.  Now mm's uffd
unit test "wp-unpopulated" will trigger this WARN_ON_ONCE().

The WARN_ON_ONCE() asserts that an VMA cannot be registered with
userfaultfd-wp if it contains a !normal page, but it's actually possible.
One example is an anonymous vma, register with uffd-wp, read anything will
install a zero page.  Then when zap on it, this should trigger.

What's more, removing that WARN_ON_ONCE may not be enough either, because
we should also not rely on "whether it's a normal page" to decide whether
pte marker is needed.  For example, one can register wr-protect over some
DAX regions to track writes when UFFD_FEATURE_WP_ASYNC enabled, in which
case it can have page==NULL for a devmap but we may want to keep the marker
around.

Cc: Muhammad Usama Anjum <usama.anjum@collabora.com>
Cc: David Hildenbrand <david@redhat.com>
Fixes: 0cf18e839f64 ("mm/memory: handle !page case in zap_present_pte() separately")
Signed-off-by: Peter Xu <peterx@redhat.com>
---
 mm/memory.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/mm/memory.c b/mm/memory.c
index f2bc6dd15eb8..904f70b99498 100644
--- a/mm/memory.c
+++ b/mm/memory.c
@@ -1536,7 +1536,9 @@ static inline int zap_present_ptes(struct mmu_gather *tlb,
 		ptep_get_and_clear_full(mm, addr, pte, tlb->fullmm);
 		arch_check_zapped_pte(vma, ptent);
 		tlb_remove_tlb_entry(tlb, pte, addr);
-		VM_WARN_ON_ONCE(userfaultfd_wp(vma));
+		if (userfaultfd_pte_wp(vma, ptent))
+			zap_install_uffd_wp_if_needed(vma, addr, pte, 1,
+						      details, ptent);
 		ksm_might_unmap_zero_page(mm, ptent);
 		return 1;
 	}
-- 
2.44.0

Re: [PATCH] mm/memory: Fix missing pte marker for !page on pte zaps

[David Hildenbrand]

On 13.03.24 22:31, peterx@redhat.com wrote:
> From: Peter Xu <peterx@redhat.com>
> 
> Commit 0cf18e839f64 of large folio zap work broke uffd-wp.  Now mm's uffd
> unit test "wp-unpopulated" will trigger this WARN_ON_ONCE().

Good that I added the WARN_ON_ONCE() :)

> 
> The WARN_ON_ONCE() asserts that an VMA cannot be registered with
> userfaultfd-wp if it contains a !normal page, but it's actually possible.
> One example is an anonymous vma, register with uffd-wp, read anything will
> install a zero page.  Then when zap on it, this should trigger.

Are you sure? zap_install_uffd_wp_if_needed() contains right at the start:

	/* Zap on anonymous always means dropping everything */
	if (vma_is_anonymous(vma))
		return;

So if that's the case the unit test triggers, I'm confused.

> 
> What's more, removing that WARN_ON_ONCE may not be enough either, because
> we should also not rely on "whether it's a normal page" to decide whether
> pte marker is needed.  For example, one can register wr-protect over some
> DAX regions to track writes when UFFD_FEATURE_WP_ASYNC enabled, in which
> case it can have page==NULL for a devmap but we may want to keep the marker
> around.

I thought uffd-wp was limited to specific backends only. But looks like 
that changed with UFFD_FEATURE_WP_ASYNC, I guess?


Change itself looks, good. Not sure about the anon_vma example above.

Thanks!

Acked-by: David Hildenbrand <david@redhat.com>

> 
> Cc: Muhammad Usama Anjum <usama.anjum@collabora.com>
> Cc: David Hildenbrand <david@redhat.com>
> Fixes: 0cf18e839f64 ("mm/memory: handle !page case in zap_present_pte() separately")
> Signed-off-by: Peter Xu <peterx@redhat.com>
> ---
>   mm/memory.c | 4 +++-
>   1 file changed, 3 insertions(+), 1 deletion(-)
> 
> diff --git a/mm/memory.c b/mm/memory.c
> index f2bc6dd15eb8..904f70b99498 100644
> --- a/mm/memory.c
> +++ b/mm/memory.c
> @@ -1536,7 +1536,9 @@ static inline int zap_present_ptes(struct mmu_gather *tlb,
>   		ptep_get_and_clear_full(mm, addr, pte, tlb->fullmm);
>   		arch_check_zapped_pte(vma, ptent);
>   		tlb_remove_tlb_entry(tlb, pte, addr);
> -		VM_WARN_ON_ONCE(userfaultfd_wp(vma));
> +		if (userfaultfd_pte_wp(vma, ptent))
> +			zap_install_uffd_wp_if_needed(vma, addr, pte, 1,
> +						      details, ptent);
>   		ksm_might_unmap_zero_page(mm, ptent);
>   		return 1;
>   	}

-- 
Cheers,

David / dhildenb

Re: [PATCH] mm/memory: Fix missing pte marker for !page on pte zaps

[David Hildenbrand]

On 13.03.24 23:03, David Hildenbrand wrote:
> On 13.03.24 22:31, peterx@redhat.com wrote:
>> From: Peter Xu <peterx@redhat.com>
>>
>> Commit 0cf18e839f64 of large folio zap work broke uffd-wp.  Now mm's uffd
>> unit test "wp-unpopulated" will trigger this WARN_ON_ONCE().
> 
> Good that I added the WARN_ON_ONCE() :)
> 
>>
>> The WARN_ON_ONCE() asserts that an VMA cannot be registered with
>> userfaultfd-wp if it contains a !normal page, but it's actually possible.
>> One example is an anonymous vma, register with uffd-wp, read anything will
>> install a zero page.  Then when zap on it, this should trigger.
> 
> Are you sure? zap_install_uffd_wp_if_needed() contains right at the start:
> 
> 	/* Zap on anonymous always means dropping everything */
> 	if (vma_is_anonymous(vma))
> 		return;
> 
> So if that's the case the unit test triggers, I'm confused.


Ah, got it. It's not that we have to place a marker, just that it can 
happen. Of course it can. All makes sense.

Thanks!

-- 
Cheers,

David / dhildenb

Re: [PATCH] mm/memory: Fix missing pte marker for !page on pte zaps

[Peter Xu]

On Wed, Mar 13, 2024 at 11:03:04PM +0100, David Hildenbrand wrote:
> On 13.03.24 22:31, peterx@redhat.com wrote:
> > From: Peter Xu <peterx@redhat.com>
> > 
> > Commit 0cf18e839f64 of large folio zap work broke uffd-wp.  Now mm's uffd
> > unit test "wp-unpopulated" will trigger this WARN_ON_ONCE().
> 
> Good that I added the WARN_ON_ONCE() :)

To be explict, VM_WARN_ON_ONCE. :) And that's my guess that you didn't hit
it when you posted the series and did the tests, as I know latest distros
like Fedora dropped DEBUG_VM, so maybe you had your base config out of
there (but I normally have it irrelevant of that).

> 
> > 
> > The WARN_ON_ONCE() asserts that an VMA cannot be registered with
> > userfaultfd-wp if it contains a !normal page, but it's actually possible.
> > One example is an anonymous vma, register with uffd-wp, read anything will
> > install a zero page.  Then when zap on it, this should trigger.
> 
> Are you sure? zap_install_uffd_wp_if_needed() contains right at the start:
> 
> 	/* Zap on anonymous always means dropping everything */
> 	if (vma_is_anonymous(vma))
> 		return;

My example is not exactly how the test failed, but should be a simpler
version of it.  To trigger this warning I don't think it requires the zero
page to be wr-protected at all or have any pte marker involved.
UFFDIO_REGISTER should suffice, afaiu (feel free to read the example above
again; there's no mention of ioctl(UFFDIO_WRITEPROTECT)).

> 
> So if that's the case the unit test triggers, I'm confused.
> 
> > 
> > What's more, removing that WARN_ON_ONCE may not be enough either, because
> > we should also not rely on "whether it's a normal page" to decide whether
> > pte marker is needed.  For example, one can register wr-protect over some
> > DAX regions to track writes when UFFD_FEATURE_WP_ASYNC enabled, in which
> > case it can have page==NULL for a devmap but we may want to keep the marker
> > around.
> 
> I thought uffd-wp was limited to specific backends only. But looks like that
> changed with UFFD_FEATURE_WP_ASYNC, I guess?

Correct.  That was also what the new PAGEMAP ioctl relies on.

> 
> 
> Change itself looks, good. Not sure about the anon_vma example above.
> 
> Thanks!
> 
> Acked-by: David Hildenbrand <david@redhat.com>

Thanks,

-- 
Peter Xu

Re: [PATCH] mm/memory: Fix missing pte marker for !page on pte zaps

[David Hildenbrand]

On 13.03.24 23:26, Peter Xu wrote:
> On Wed, Mar 13, 2024 at 11:03:04PM +0100, David Hildenbrand wrote:
>> On 13.03.24 22:31, peterx@redhat.com wrote:
>>> From: Peter Xu <peterx@redhat.com>
>>>
>>> Commit 0cf18e839f64 of large folio zap work broke uffd-wp.  Now mm's uffd
>>> unit test "wp-unpopulated" will trigger this WARN_ON_ONCE().
>>
>> Good that I added the WARN_ON_ONCE() :)
> 
> To be explict, VM_WARN_ON_ONCE. :) And that's my guess that you didn't hit
> it when you posted the series and did the tests, as I know latest distros
> like Fedora dropped DEBUG_VM, so maybe you had your base config out of
> there (but I normally have it irrelevant of that).

It's rather surprising that this went unnoticed for quite a while. I 
usually do run with DEBUG_VM, but I have a bunch of different VMs where 
I tests stuff, likely it was off when I ran the selftests.

-- 
Cheers,

David / dhildenb