On 10/30/25 17:43, Lorenzo Stoakes wrote:

On Thu, Oct 30, 2025 at 04:31:56PM +0000, Pedro Falcato wrote:

quoted

On Thu, Oct 30, 2025 at 04:23:58PM +0000, Lorenzo Stoakes wrote:

quoted

On Thu, Oct 30, 2025 at 04:16:20PM +0000, Pedro Falcato wrote:

quoted

On Wed, Oct 29, 2025 at 04:50:31PM +0000, Lorenzo Stoakes wrote:

quoted

Currently, if a user needs to determine if guard regions are present in a
range, they have to scan all VMAs (or have knowledge of which ones might
have guard regions).

Since commit 8e2f2aeb8b48 ("fs/proc/task_mmu: add guard region bit to
pagemap") and the related commit a516403787e0 ("fs/proc: extend the
PAGEMAP_SCAN ioctl to report guard regions"), users can use either
/proc/$pid/pagemap or the PAGEMAP_SCAN functionality to perform this
operation at a virtual address level.

This is not ideal, and it gives no visibility at a /proc/$pid/smaps level
that guard regions exist in ranges.

This patch remedies the situation by establishing a new VMA flag,
VM_MAYBE_GUARD, to indicate that a VMA may contain guard regions (it is
uncertain because we cannot reasonably determine whether a
MADV_GUARD_REMOVE call has removed all of the guard regions in a VMA, and
additionally VMAs may change across merge/split).

We utilise 0x800 for this flag which makes it available to 32-bit
architectures also, a flag that was previously used by VM_DENYWRITE, which
was removed in commit 8d0920bde5eb ("mm: remove VM_DENYWRITE") and hasn't
bee reused yet.

The MADV_GUARD_INSTALL madvise() operation now must take an mmap write
lock (and also VMA write lock) whereas previously it did not, but this
seems a reasonable overhead.

Do you though? Could it be possible to simply atomically set the flag with
the read lock held? This would make it so we can't split the VMA (and tightly

VMA flags are not accessed atomically so no I don't think we can do that in any
workable way.

FWIW I think you could work it as an atomic flag and treat those races as benign
(this one, at least).

It's not benign as we need to ensure that page tables are correctly propagated
on fork.

Could we use MADVISE_VMA_READ_LOCK mode (would be actually an improvement
over the current MADVISE_MMAP_READ_LOCK), together with the atomic flag
setting? I think the places that could race with us to cause RMW use vma
write lock so that would be excluded. Fork AFAICS unfortunately doesn't (for
the oldmm) and it probably would't make sense to start doing it. Maybe we
could think of something to deal with this special case...

quoted

quoted

I also don't think it's at all necessary, see below.

quoted

define what "may have a guard page"), but it sounds much better than introducing
lock contention. I don't think it is reasonable to add a write lock to a feature
that may be used by such things as thread stack allocation, malloc, etc.

What lock contention? It's per-VMA so the contention is limited to the VMA in
question, and only over the span of time you are setting the gaurd region.

Don't we always need to take the mmap write lock when grabbing a VMA write
lock as well?

Yup. But at the same time you're doing the kind of operations that'd use this
you'd already be taking the lock anyway.

You don't hold it for long and you won't be doing this any more often than you'd
be doing other write operations, which you're also not going to be holding up
faults on other VMAs either (they can access other VMAs despite mmap write lock
being held), so I don't think there's ay issue here.

quoted

quoted

When allocating thread stacks you'll be mapping things into memory which... take
the write lock. malloc() if it goes to the kernel will also take the write lock.

But yes, good point, you're already serializing anyway. I don't think this is
a big deal.

Indeed

Besides thread stacks, I'm thinking of the userspace allocators usecase
(jemalloc etc) where guard regions were supposed to allow a cheap
use-after-free protection by replacing their existing
MADV_DONTNEED/MADV_FREE use with MADV_GUARD_INSTALL.
Now MADV_DONTNEED/MADV_FREE use MADVISE_VMA_READ_LOCK, and guard regions
moves from (worse but still reasonable) MADVISE_MMAP_READ_LOCK to the heavy
MADVISE_MMAP_WRITE_LOCK. I'm afraid this might be too heavy price for that
usecase :(

quoted

quoted

So I think you're overly worried about an operation that a. isn't going to be
something that happens all that often, b. when it does, it's at a time when
you'd be taking write locks anyway and c. won't contend important stuff like
page faults for any VMA other than the one having the the guard region
installed.

Yep, thanks.

No problemo, you can get yourself into sticky situations with lock contention
but I think this is not one! :)

quoted

--
Pedro

Cheers, Lorenzo