All of lore.kernel.org
 help / color / mirror / Atom feed
From: Deborah Brouwer <deborah.brouwer@collabora.com>
To: Janne Grunau <j@jannau.net>
Cc: "Miguel Ojeda" <ojeda@kernel.org>,
	"Boqun Feng" <boqun@kernel.org>, "Gary Guo" <gary@garyguo.net>,
	"Björn Roy Baron" <bjorn3_gh@protonmail.com>,
	"Benno Lossin" <lossin@kernel.org>,
	"Andreas Hindborg" <a.hindborg@kernel.org>,
	"Alice Ryhl" <aliceryhl@google.com>,
	"Trevor Gross" <tmgross@umich.edu>,
	"Danilo Krummrich" <dakr@kernel.org>,
	"Daniel Almeida" <daniel.almeida@collabora.com>,
	"Tamir Duberstein" <tamird@kernel.org>,
	"Alexandre Courbot" <acourbot@nvidia.com>,
	"Onur Özkan" <work@onurozkan.dev>,
	"David Airlie" <airlied@gmail.com>,
	"Simona Vetter" <simona@ffwll.ch>, "Janne Grunau" <j@jananu.net>,
	"Sami Tolvanen" <samitolvanen@google.com>,
	"Lyude Paul" <lyude@redhat.com>,
	rust-for-linux@vger.kernel.org, linux-kernel@vger.kernel.org,
	dri-devel@lists.freedesktop.org, nova-gpu@lists.linux.dev,
	stable+noautosel@kernel.org
Subject: Re: [PATCH 0/3] Fix import of dma-buf buffers as DRM GEM shmem objects
Date: Mon, 16 Nov 2026 16:01:08 -0700	[thread overview]
Message-ID: <avuLNHy3bNAUpzNv@um790> (raw)
In-Reply-To: <20261026-rust-drm-gem-shmem-imported-sgt-v1-0-36f15ace04b1@jannau.net>

On Mon, Oct 26, 2026 at 12:14:07PM +0100, Janne Grunau wrote:
> Hej,
> 
> this is triggered by the discussion of gem/shmem.rs SGTableMap issues
> in [1] which has an additional issue with imported object.
> Imported objects are supported since GEM shmem's AllocOps use
> drm_gem_shmem_prime_import_sg_table() as gem_prime_import_sg_table which
> sets the same function pointer field in struct drm_driver.
> 
> The code is broken in two ways. SGTableMap::drop() calls
> __drm_gem_shmem_free_sgt_locked() uncondiftionally which it must not do
> for imported objects as the object does not own `sgt`.
> The second issue is that the imported object is allocated by
> __drm_gem_shmem_create() with sizeof(struct drm_gem_shmem_object) as
> size instead of the size of Rust's Object<T>.
> 
> The second issue is already fixed in tyr's downstream driver by adding a
> `gem_create_object` callback which is used instead of the minimal
> allocation in the C helper.
> The first issue is fixed in the Asahi downstream driver together with a
> drm_gem_is_imported() wrapper change taken from the tyr downstream
> driver.
> 
> All three patches are tested in the Asahi downstream driver. As this is
> unused code in-tree I marked the fixes as not suitable for stable
> backports. The known downstream users of the code are aware of the
> issues and carry fixes downstream. I don't expect new Rust DRM drivers
> to be based on the (old) stable trees.

Thank you Janne for this. The second issue was recently flagged
by sashiko [1] as a critical memory bug that needs to be addressed
before we upstream tyr's ioctl series [2]. I checked that this series
on my Rock5B with the proposed ioctl series and igt-gpu-tools
tests that exercise some of the panthor uapi and everything looks good.

For the whole series:
Tested-by: Deborah Brouwer <deborah.brouwer@collabora.com>

[1] https://sashiko.dev/#/patchset/20261109-tyr-ioctls-v6-0-b2896bf2b3c9%40kylinos.cn
[2] https://lore.kernel.org/rust-for-linux/20261109-tyr-ioctls-v6-0-b2896bf2b3c9@kylinos.cn/

> 
> Link: https://lore.kernel.org/rust-for-linux/20261019-rust-shmem-fixes-v1-1-3f7f6de5b5e7@google.com/ [1]
> Signed-off-by: Janne Grunau <j@jannau.net>
> ---
> Janne Grunau (1):
>       rust: drm: gem: shmem: Do not free SG table of imported objects
> 
> Sami Tolvanen (2):
>       rust: drm: gem: shmem: implement gem_create_object for prime imports
>       rust: drm: gem: add an is_imported() accessor
> 
>  drivers/gpu/drm/nova/gem.rs  |  4 +++
>  drivers/gpu/drm/tyr/gem.rs   |  4 +++
>  rust/helpers/drm.c           |  6 +++++
>  rust/kernel/drm/gem/mod.rs   | 16 ++++++++++++
>  rust/kernel/drm/gem/shmem.rs | 62 +++++++++++++++++++++++++++++++++++++++++++-
>  5 files changed, 91 insertions(+), 1 deletion(-)
> ---
> base-commit: 3497a403264b094ebd61b8b755165d56e7d9dc37
> change-id: 20261026-rust-drm-gem-shmem-imported-sgt-85c4a45be283
> 
> Best regards,
> -- 
> Janne Grunau <j@jannau.net>
> 

      parent reply	other threads:[~2026-11-16 23:01 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-10-26 11:14 [PATCH 0/3] Fix import of dma-buf buffers as DRM GEM shmem objects Janne Grunau
2026-10-26 11:14 ` [PATCH 1/3] rust: drm: gem: shmem: implement gem_create_object for prime imports Janne Grunau
2026-10-26 11:14 ` [PATCH 2/3] rust: drm: gem: add an is_imported() accessor Janne Grunau
2026-10-26 11:53   ` sashiko-bot
2026-10-26 11:14 ` [PATCH 3/3] rust: drm: gem: shmem: Do not free SG table of imported objects Janne Grunau
2026-11-16 23:01 ` Deborah Brouwer [this message]

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=avuLNHy3bNAUpzNv@um790 \
    --to=deborah.brouwer@collabora.com \
    --cc=a.hindborg@kernel.org \
    --cc=acourbot@nvidia.com \
    --cc=airlied@gmail.com \
    --cc=aliceryhl@google.com \
    --cc=bjorn3_gh@protonmail.com \
    --cc=boqun@kernel.org \
    --cc=dakr@kernel.org \
    --cc=daniel.almeida@collabora.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=gary@garyguo.net \
    --cc=j@jananu.net \
    --cc=j@jannau.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lossin@kernel.org \
    --cc=lyude@redhat.com \
    --cc=nova-gpu@lists.linux.dev \
    --cc=ojeda@kernel.org \
    --cc=rust-for-linux@vger.kernel.org \
    --cc=samitolvanen@google.com \
    --cc=simona@ffwll.ch \
    --cc=stable+noautosel@kernel.org \
    --cc=tamird@kernel.org \
    --cc=tmgross@umich.edu \
    --cc=work@onurozkan.dev \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.