All of lore.kernel.org
 help / color / mirror / Atom feed
From: Daniel Vetter <daniel.vetter@ffwll.ch>
To: "Christian König" <ckoenig.leichtzumerken@gmail.com>
Cc: matthew.brost@intel.com, thomas.hellstrom@linux.intel.com,
	dri-devel@lists.freedesktop.org
Subject: Re: [PATCH 1/8] drm/amdgpu: use GEM references instead of TTMs
Date: Thu, 25 Jul 2024 14:01:36 +0200	[thread overview]
Message-ID: <ZqI-oMrbSLSybC_p@phenom.ffwll.local> (raw)
In-Reply-To: <20240723121750.2086-1-christian.koenig@amd.com>

On Tue, Jul 23, 2024 at 02:17:43PM +0200, Christian König wrote:
> Instead of a TTM reference grab a GEM reference whenever necessary.
> 
> Signed-off-by: Christian König <christian.koenig@amd.com>
> Cc: Alex Deucher <alexander.deucher@amd.com>
> Cc: Felix Kuehling <felix.kuehling@amd.com>
> Cc: amd-gfx@lists.freedesktop.org

I was worried that changing the refcounting for the mmu_interval_notifier
that this patch does could result in a use-after-free, but the mmu
notifier code already does the required amount of mmgrab/mmdrop, so we're
good.

Reviewed-by: Daniel Vetter <daniel.vetter@ffwll.ch>

Cheers, Sima

> ---
>  drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c    | 8 ++++----
>  drivers/gpu/drm/amd/amdgpu/amdgpu_object.c | 7 ++-----
>  2 files changed, 6 insertions(+), 9 deletions(-)
> 
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
> index 67c234bcf89f..6be3d7cd1c51 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_gem.c
> @@ -87,11 +87,11 @@ static const struct vm_operations_struct amdgpu_gem_vm_ops = {
>  
>  static void amdgpu_gem_object_free(struct drm_gem_object *gobj)
>  {
> -	struct amdgpu_bo *robj = gem_to_amdgpu_bo(gobj);
> +	struct amdgpu_bo *aobj = gem_to_amdgpu_bo(gobj);
>  
> -	if (robj) {
> -		amdgpu_hmm_unregister(robj);
> -		amdgpu_bo_unref(&robj);
> +	if (aobj) {
> +		amdgpu_hmm_unregister(aobj);
> +		ttm_bo_put(&aobj->tbo);
>  	}
>  }
>  
> diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
> index 8d8c39be6129..6c187e310034 100644
> --- a/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
> +++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_object.c
> @@ -853,7 +853,7 @@ struct amdgpu_bo *amdgpu_bo_ref(struct amdgpu_bo *bo)
>  	if (bo == NULL)
>  		return NULL;
>  
> -	ttm_bo_get(&bo->tbo);
> +	drm_gem_object_get(&bo->tbo.base);
>  	return bo;
>  }
>  
> @@ -865,13 +865,10 @@ struct amdgpu_bo *amdgpu_bo_ref(struct amdgpu_bo *bo)
>   */
>  void amdgpu_bo_unref(struct amdgpu_bo **bo)
>  {
> -	struct ttm_buffer_object *tbo;
> -
>  	if ((*bo) == NULL)
>  		return;
>  
> -	tbo = &((*bo)->tbo);
> -	ttm_bo_put(tbo);
> +	drm_gem_object_get(&(*bo)->tbo.base);
>  	*bo = NULL;
>  }
>  
> -- 
> 2.34.1
> 

-- 
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch

      parent reply	other threads:[~2024-07-25 12:01 UTC|newest]

Thread overview: 16+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-07-23 12:17 [PATCH 1/8] drm/amdgpu: use GEM references instead of TTMs Christian König
2024-07-23 12:17 ` [PATCH 2/8] drm/radeon: " Christian König
2024-07-25 12:02   ` Daniel Vetter
2024-07-23 12:17 ` [PATCH 3/8] drm/loongson: " Christian König
2024-07-25 12:03   ` Daniel Vetter
2024-07-23 12:17 ` [PATCH 4/8] drm/vmwgfx: " Christian König
2024-07-25 12:16   ` Daniel Vetter
2024-07-23 12:17 ` [PATCH 5/8] drm/vmwgfx: partially revert "Adapt validation code for reference-free lookups" Christian König
2024-07-25 12:34   ` Daniel Vetter
2024-08-27 10:03   ` Thomas Hellström
2024-07-23 12:17 ` [PATCH 6/8] drm/ttm: revert "Export ttm_bo_get_unless_zero()" Christian König
2024-07-25 12:36   ` Daniel Vetter
2024-07-25 13:19   ` Danilo Krummrich
2024-07-23 12:17 ` [PATCH 7/8] drm/ttm: use GEM references for VM mappings Christian König
2024-07-23 12:17 ` [PATCH 8/8] drm/ttm: make ttm_bo_get internal Christian König
2024-07-25 12:01 ` Daniel Vetter [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=ZqI-oMrbSLSybC_p@phenom.ffwll.local \
    --to=daniel.vetter@ffwll.ch \
    --cc=ckoenig.leichtzumerken@gmail.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=matthew.brost@intel.com \
    --cc=thomas.hellstrom@linux.intel.com \
    /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.