Re: Patch "drm/sched: memset() 'job' in drm_sched_job_init()" has been added to the 6.12-stable tree

Re: Patch "drm/sched: memset() 'job' in drm_sched_job_init()" has been added to the 6.12-stable tree

[Philipp Stanner]

On Tue, 2024-12-10 at 15:41 -0500, Sasha Levin wrote:
> This is a note to let you know that I've just added the patch titled
> 
>     drm/sched: memset() 'job' in drm_sched_job_init()
> 
> to the 6.12-stable tree which can be found at:
>    
> http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
> 
> The filename of the patch is:
>      drm-sched-memset-job-in-drm_sched_job_init.patch
> and it can be found in the queue-6.12 subdirectory.
> 
> If you, or anyone else, feels it should not be added to the stable
> tree,
> please let <stable@vger.kernel.org> know about it.

Hi,

you can add it, it does improve things a bit.

But I'd like to use this opportunity to understand by what criteria you
found and selected this patch? Stable was not on CC, neither does the
patch contain a Fixes tag.


Regards,
P.


> 
> 
> 
> commit d0a6c893de0172427064e39be400a23b0ba5ffec
> Author: Philipp Stanner <pstanner@redhat.com>
> Date:   Mon Oct 21 12:50:28 2024 +0200
> 
>     drm/sched: memset() 'job' in drm_sched_job_init()
>     
>     [ Upstream commit 2320c9e6a768d135c7b0039995182bb1a4e4fd22 ]
>     
>     drm_sched_job_init() has no control over how users allocate
> struct
>     drm_sched_job. Unfortunately, the function can also not set some
> struct
>     members such as job->sched.
>     
>     This could theoretically lead to UB by users dereferencing the
> struct's
>     pointer members too early.
>     
>     It is easier to debug such issues if these pointers are
> initialized to
>     NULL, so dereferencing them causes a NULL pointer exception.
>     Accordingly, drm_sched_entity_init() does precisely that and
> initializes
>     its struct with memset().
>     
>     Initialize parameter "job" to 0 in drm_sched_job_init().
>     
>     Signed-off-by: Philipp Stanner <pstanner@redhat.com>
>     Link:
> https://patchwork.freedesktop.org/patch/msgid/20241021105028.19794-2-pstanner@redhat.com
>     Reviewed-by: Christian König <christian.koenig@amd.com>
>     Signed-off-by: Sasha Levin <sashal@kernel.org>
> 
> diff --git a/drivers/gpu/drm/scheduler/sched_main.c
> b/drivers/gpu/drm/scheduler/sched_main.c
> index e97c6c60bc96e..416590ea0dc3d 100644
> --- a/drivers/gpu/drm/scheduler/sched_main.c
> +++ b/drivers/gpu/drm/scheduler/sched_main.c
> @@ -803,6 +803,14 @@ int drm_sched_job_init(struct drm_sched_job
> *job,
>  		return -EINVAL;
>  	}
>  
> +	/*
> +	 * We don't know for sure how the user has allocated. Thus,
> zero the
> +	 * struct so that unallowed (i.e., too early) usage of
> pointers that
> +	 * this function does not set is guaranteed to lead to a
> NULL pointer
> +	 * exception instead of UB.
> +	 */
> +	memset(job, 0, sizeof(*job));
> +
>  	job->entity = entity;
>  	job->credits = credits;
>  	job->s_fence = drm_sched_fence_alloc(entity, owner);
> 

Re: Patch "drm/sched: memset() 'job' in drm_sched_job_init()" has been added to the 6.12-stable tree

[Greg KH]

On Wed, Dec 11, 2024 at 11:00:02AM +0100, Philipp Stanner wrote:
> On Tue, 2024-12-10 at 15:41 -0500, Sasha Levin wrote:
> > This is a note to let you know that I've just added the patch titled
> > 
> >     drm/sched: memset() 'job' in drm_sched_job_init()
> > 
> > to the 6.12-stable tree which can be found at:
> >    
> > http://www.kernel.org/git/?p=linux/kernel/git/stable/stable-queue.git;a=summary
> > 
> > The filename of the patch is:
> >      drm-sched-memset-job-in-drm_sched_job_init.patch
> > and it can be found in the queue-6.12 subdirectory.
> > 
> > If you, or anyone else, feels it should not be added to the stable
> > tree,
> > please let <stable@vger.kernel.org> know about it.
> 
> Hi,
> 
> you can add it, it does improve things a bit.
> 
> But I'd like to use this opportunity to understand by what criteria you
> found and selected this patch? Stable was not on CC, neither does the
> patch contain a Fixes tag.

You did see the AUTOSEL emails with this commit in it back on November
24, right?  that should explain this.

thanks,

greg k-h