linux-kernel - Re: [PATCH 2/7] mm/munmap: Replace can_modify_mm with can_modify

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <CAKbZUD2pVQAyUNJjRxoS1VOnJ09winf79eXUDgn_1V4OH2UzDA@mail.gmail.com>
Date: Wed, 7 Aug 2024 14:13:19 +0100
From: Pedro Falcato <pedro.falcato@...il.com>
To: Lorenzo Stoakes <lorenzo.stoakes@...cle.com>
Cc: Andrew Morton <akpm@...ux-foundation.org>, "Liam R. Howlett" <Liam.Howlett@...cle.com>, 
	Vlastimil Babka <vbabka@...e.cz>, linux-mm@...ck.org, linux-kernel@...r.kernel.org, 
	oliver.sang@...el.com, torvalds@...ux-foundation.org, jeffxu@...gle.com, 
	Michael Ellerman <mpe@...erman.id.au>
Subject: Re: [PATCH 2/7] mm/munmap: Replace can_modify_mm with can_modify_vma

On Wed, Aug 7, 2024 at 2:02 PM Lorenzo Stoakes
<lorenzo.stoakes@...cle.com> wrote:
>
> On Tue, Aug 06, 2024 at 10:28:03PM GMT, Pedro Falcato wrote:
> > We were doing an extra mmap tree traversal just to check if the entire
> > range is modifiable. This can be done when we iterate through the VMAs
> > instead.
> >
> > Note that this removes the arch_unmap() callsites and therefore isn't
> > quite ready for Proper(tm) upstreaming.
>
> If this isn't ready for upstreaming which is it being submitted as a patch
> series and not an RFC or such?

Crap... I wasn't sure whether to mark this as RFC or not (I wasn't
sure if this could be applied as a hotfix, yes it's a little risky but
the changes themselves are simple, and fix an active regression).
I'll err on the side of caution next time :)

>
> Liam is likely to do some significant rework of this arch_unmap() stuff
> soon, and is certainly significantly reworking the munmap() logic, so to

FWIW there was a new series posted at
https://lore.kernel.org/linuxppc-dev/20240807124103.85644-1-mpe@ellerman.id.au/T/#m353eb23fc263033c9ca023ead6fa82d1a1ff3263
that do away with arch_unmap altogether (resulting from our exchange
on the regression thread).

> avoid conflicts it goes doubly that if this isn't meant for upstream then
> it should be RFC'd.
>
> >
> > Signed-off-by: Pedro Falcato <pedro.falcato@...il.com>
>
> This patch doesn't apply in the mm-unstable tree. If you want your series
> to come in through the mm tree you need to rebase on this.
>
> I made a major change to file structure which moves a bunch of mm/mmap.c
> stuff to mm/vma.c (similarly moving things around in headers), which is
> why.
>
> It also means I can't sensibly review it... :)

ACK. I'll rebase on mm-unstable for v2, sorry for the time lost.

-- 
Pedro