linux-kernel - Re: [PATCH v20 09/10] drm/shmem-helper: Switch drm_gem_shmem

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <710cdbd4-2c6e-48b7-b12b-972ab6d12abf@collabora.com>
Date: Wed, 2 Apr 2025 15:58:55 +0300
From: Dmitry Osipenko <dmitry.osipenko@...labora.com>
To: Thomas Zimmermann <tzimmermann@...e.de>, David Airlie
 <airlied@...il.com>, Simona Vetter <simona@...ll.ch>,
 Maarten Lankhorst <maarten.lankhorst@...ux.intel.com>,
 Maxime Ripard <mripard@...nel.org>,
 Christian König <christian.koenig@....com>,
 Gerd Hoffmann <kraxel@...hat.com>, Qiang Yu <yuq825@...il.com>,
 Steven Price <steven.price@....com>,
 Boris Brezillon <boris.brezillon@...labora.com>,
 Frank Binns <frank.binns@...tec.com>, Matt Coster <matt.coster@...tec.com>
Cc: dri-devel@...ts.freedesktop.org, linux-kernel@...r.kernel.org,
 kernel@...labora.com
Subject: Re: [PATCH v20 09/10] drm/shmem-helper: Switch
 drm_gem_shmem_vmap/vunmap to use pin/unpin

On 4/2/25 15:47, Thomas Zimmermann wrote:
> Hi
> 
> Am 22.03.25 um 22:26 schrieb Dmitry Osipenko:
>> The vmapped pages shall be pinned in memory and previously get/
>> put_pages()
>> were implicitly hard-pinning/unpinning the pages. This will no longer be
>> the case with addition of memory shrinker because pages_use_count > 0
>> won't
>> determine anymore whether pages are hard-pinned (they will be soft-
>> pinned),
>> while the new pages_pin_count will do the hard-pinning. Switch the
>> vmap/vunmap() to use pin/unpin() functions in a preparation of addition
>> of the memory shrinker support to drm-shmem.
> 
> I've meanwhile rediscovered this patch and I'm sure this is not correct.
> Vmap should not pin AFAIK. It is possible to vmap if the buffer has been
> pinned, but that's not automatic.  For other vmaps it is necessary to
> hold the reservation lock to prevent the buffer from moving.

Hi, with vmap() you're getting a kernel address. The GEM's memory should
be not movable while it's vmapped as we can't handle kernel page faults.

Not sure what you're meaning by the "other vmaps", please clarify.

-- 
Best regards,
Dmitry