[Intel-gfx] [PATCH 08/26] drm/i915/gem: Make eb_add_lut interruptible wait on object lock.

[Tvrtko Ursulin]

On 23/06/2020 15:28, Maarten Lankhorst wrote:
> The lock here should be interruptible, so we can backoff if needed.

I spied Chris posting "drm/i915/gem: Move obj->lut_list under its own 
lock" so maybe have a look at that.

My question here is..

> 
> Signed-off-by: Maarten Lankhorst <maarten.lankhorst at linux.intel.com>
> ---
>   drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c | 8 +++++++-
>   1 file changed, 7 insertions(+), 1 deletion(-)
> 
> diff --git a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
> index 2636a130fb57..aa441af81431 100644
> --- a/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
> +++ b/drivers/gpu/drm/i915/gem/i915_gem_execbuffer.c
> @@ -774,7 +774,12 @@ static int __eb_add_lut(struct i915_execbuffer *eb,
>   		if (err == 0) { /* And nor has this handle */
>   			struct drm_i915_gem_object *obj = vma->obj;
>   
> -			i915_gem_object_lock(obj, NULL);
> +			err = i915_gem_object_lock_interruptible(obj, NULL);

.. does this lock-unlock survive to the end of your series or gets 
completely subsumed by the ctx locking?

Regards,

Tvrtko

> +			if (err) {
> +				radix_tree_delete(&ctx->handles_vma, handle);
> +				goto unlock;
> +			}
> +
>   			if (idr_find(&eb->file->object_idr, handle) == obj) {
>   				list_add(&lut->obj_link, &obj->lut_list);
>   			} else {
> @@ -783,6 +788,7 @@ static int __eb_add_lut(struct i915_execbuffer *eb,
>   			}
>   			i915_gem_object_unlock(obj);
>   		}
> +unlock:
>   		mutex_unlock(&ctx->mutex);
>   	}
>   	if (unlikely(err))
>