stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
* [PATCH] drm/i915: Unconditionally flush any chipset buffers before execbuf
@ 2016-08-16 12:14 Chris Wilson
  2016-08-16 14:01 ` [Intel-gfx] " Mika Kuoppala
  0 siblings, 1 reply; 2+ messages in thread
From: Chris Wilson @ 2016-08-16 12:14 UTC (permalink / raw)
  To: intel-gfx; +Cc: Chris Wilson, Akash Goel, Daniel Vetter, Tvrtko Ursulin, stable

If userspace is asynchronously streaming into the batch or other
execobjects, we may not flush those writes along with a change in cache
domain (as there is no change). Therefore those writes may end up in
internal chipset buffers and not visible to the GPU upon execution. We
must issue a flush command or otherwise we encounter incoherency in the
batchbuffers and the GPU executing invalid commands (i.e. hanging) quite
regularly.

Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90841
Fixes: 1816f9236303 ("drm/i915: Support creation of unbound wc user...")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Akash Goel <akash.goel@intel.com>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
Tested-by: Matti Hämäläinen <ccr@tnsp.org>
Cc: stable@vger.kernel.org
---
 drivers/gpu/drm/i915/i915_gem_execbuffer.c | 10 +++-------
 1 file changed, 3 insertions(+), 7 deletions(-)

diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
index 699315304748..75957aef6219 100644
--- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
+++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
@@ -1016,7 +1016,6 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
 	const unsigned int other_rings = eb_other_engines(req);
 	struct i915_vma *vma;
 	uint32_t flush_domains = 0;
-	bool flush_chipset = false;
 	int ret;
 
 	list_for_each_entry(vma, vmas, exec_list) {
@@ -1029,16 +1028,13 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
 		}
 
 		if (obj->base.write_domain & I915_GEM_DOMAIN_CPU)
-			flush_chipset |= i915_gem_clflush_object(obj, false);
+			i915_gem_clflush_object(obj, false);
 
 		flush_domains |= obj->base.write_domain;
 	}
 
-	if (flush_chipset)
-		i915_gem_chipset_flush(req->engine->i915);
-
-	if (flush_domains & I915_GEM_DOMAIN_GTT)
-		wmb();
+	/* Unconditionally flush any chipset caches (for streaming writes). */
+	i915_gem_chipset_flush(req->engine->i915);
 
 	/* Unconditionally invalidate GPU caches and TLBs. */
 	return req->engine->emit_flush(req, EMIT_INVALIDATE);
-- 
2.8.1


^ permalink raw reply related	[flat|nested] 2+ messages in thread

* Re: [Intel-gfx] [PATCH] drm/i915: Unconditionally flush any chipset buffers before execbuf
  2016-08-16 12:14 [PATCH] drm/i915: Unconditionally flush any chipset buffers before execbuf Chris Wilson
@ 2016-08-16 14:01 ` Mika Kuoppala
  0 siblings, 0 replies; 2+ messages in thread
From: Mika Kuoppala @ 2016-08-16 14:01 UTC (permalink / raw)
  To: Chris Wilson, intel-gfx; +Cc: Daniel Vetter, Akash Goel, stable

Chris Wilson <chris@chris-wilson.co.uk> writes:

> If userspace is asynchronously streaming into the batch or other
> execobjects, we may not flush those writes along with a change in cache
> domain (as there is no change). Therefore those writes may end up in
> internal chipset buffers and not visible to the GPU upon execution. We
> must issue a flush command or otherwise we encounter incoherency in the
> batchbuffers and the GPU executing invalid commands (i.e. hanging) quite
> regularly.
>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=90841
> Fixes: 1816f9236303 ("drm/i915: Support creation of unbound wc user...")
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> Cc: Akash Goel <akash.goel@intel.com>
> Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
> Cc: Tvrtko Ursulin <tvrtko.ursulin@linux.intel.com>
> Tested-by: Matti Hämäläinen <ccr@tnsp.org>
> Cc: stable@vger.kernel.org
> ---
>  drivers/gpu/drm/i915/i915_gem_execbuffer.c | 10 +++-------
>  1 file changed, 3 insertions(+), 7 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_gem_execbuffer.c b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> index 699315304748..75957aef6219 100644
> --- a/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> +++ b/drivers/gpu/drm/i915/i915_gem_execbuffer.c
> @@ -1016,7 +1016,6 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
>  	const unsigned int other_rings = eb_other_engines(req);
>  	struct i915_vma *vma;
>  	uint32_t flush_domains = 0;

You don't need flush_domains, with this version.

> -	bool flush_chipset = false;
>  	int ret;
>  
>  	list_for_each_entry(vma, vmas, exec_list) {
> @@ -1029,16 +1028,13 @@ i915_gem_execbuffer_move_to_gpu(struct drm_i915_gem_request *req,
>  		}
>  
>  		if (obj->base.write_domain & I915_GEM_DOMAIN_CPU)
> -			flush_chipset |= i915_gem_clflush_object(obj, false);
> +			i915_gem_clflush_object(obj, false);
>  
>  		flush_domains |= obj->base.write_domain;
>  	}
>  
> -	if (flush_chipset)
> -		i915_gem_chipset_flush(req->engine->i915);
> -
> -	if (flush_domains & I915_GEM_DOMAIN_GTT)
> -		wmb();
> +	/* Unconditionally flush any chipset caches (for streaming writes). */
> +	i915_gem_chipset_flush(req->engine->i915);
>  
>  	/* Unconditionally invalidate GPU caches and TLBs. */
>  	return req->engine->emit_flush(req, EMIT_INVALIDATE);
> -- 
> 2.8.1
>
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> https://lists.freedesktop.org/mailman/listinfo/intel-gfx

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2016-08-16 14:03 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2016-08-16 12:14 [PATCH] drm/i915: Unconditionally flush any chipset buffers before execbuf Chris Wilson
2016-08-16 14:01 ` [Intel-gfx] " Mika Kuoppala

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).