[PATCH] drm/i915: Perform pageflip using mmio if the GPU is terminally wedged

[PATCH] drm/i915: Perform pageflip using mmio if the GPU is terminally wedged

[Chris Wilson]

After a hang and failed reset, we cannot use the GPU to execute the page
flip instructions. Instead we can force a synchronous mmio flip. (Later,
we can reduce the synchronicity of the mmio flip by moving some of the
delays off to a worker, like the current page flip code; see vblank
tasks.)

References: https://bugs.freedesktop.org/show_bug.cgi?id=72631
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/intel_display.c |   10 ++++++++++
 1 file changed, 10 insertions(+)

diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index 79ef515..d5eafd1 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -8717,6 +8717,9 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
 	     fb->pitches[0] != crtc->fb->pitches[0]))
 		return -EINVAL;
 
+	if (i915_terminally_wedged(&dev_priv->gpu_error))
+		goto out_hang;
+
 	work = kzalloc(sizeof(*work), GFP_KERNEL);
 	if (work == NULL)
 		return -ENOMEM;
@@ -8791,6 +8794,13 @@ cleanup:
 free_work:
 	kfree(work);
 
+	if (ret == -EIO) {
+out_hang:
+		intel_crtc_wait_for_pending_flips(crtc);
+		ret = intel_pipe_set_base(crtc, crtc->x, crtc->y, fb);
+		if (ret == 0 && event)
+			drm_send_vblank_event(dev, intel_crtc->pipe, event);
+	}
 	return ret;
 }
 
-- 
1.7.9.5

Re: [PATCH] drm/i915: Perform pageflip using mmio if the GPU is terminally wedged

[Ville Syrjälä]

On Thu, Feb 20, 2014 at 09:26:13AM +0000, Chris Wilson wrote:
> After a hang and failed reset, we cannot use the GPU to execute the page
> flip instructions. Instead we can force a synchronous mmio flip. (Later,
> we can reduce the synchronicity of the mmio flip by moving some of the
> delays off to a worker, like the current page flip code; see vblank
> tasks.)
> 
> References: https://bugs.freedesktop.org/show_bug.cgi?id=72631
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

> ---
>  drivers/gpu/drm/i915/intel_display.c |   10 ++++++++++
>  1 file changed, 10 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> index 79ef515..d5eafd1 100644
> --- a/drivers/gpu/drm/i915/intel_display.c
> +++ b/drivers/gpu/drm/i915/intel_display.c
> @@ -8717,6 +8717,9 @@ static int intel_crtc_page_flip(struct drm_crtc *crtc,
>  	     fb->pitches[0] != crtc->fb->pitches[0]))
>  		return -EINVAL;
>  
> +	if (i915_terminally_wedged(&dev_priv->gpu_error))
> +		goto out_hang;
> +
>  	work = kzalloc(sizeof(*work), GFP_KERNEL);
>  	if (work == NULL)
>  		return -ENOMEM;
> @@ -8791,6 +8794,13 @@ cleanup:
>  free_work:
>  	kfree(work);
>  
> +	if (ret == -EIO) {
> +out_hang:
> +		intel_crtc_wait_for_pending_flips(crtc);
> +		ret = intel_pipe_set_base(crtc, crtc->x, crtc->y, fb);
> +		if (ret == 0 && event)
> +			drm_send_vblank_event(dev, intel_crtc->pipe, event);
> +	}
>  	return ret;
>  }
>  
> -- 
> 1.7.9.5
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

-- 
Ville Syrjälä
Intel OTC

Re: [PATCH] drm/i915: Perform pageflip using mmio if the GPU is terminally wedged

[Daniel Vetter]

On Wed, Mar 05, 2014 at 03:08:27PM +0200, Ville Syrjälä wrote:
> On Thu, Feb 20, 2014 at 09:26:13AM +0000, Chris Wilson wrote:
> > After a hang and failed reset, we cannot use the GPU to execute the page
> > flip instructions. Instead we can force a synchronous mmio flip. (Later,
> > we can reduce the synchronicity of the mmio flip by moving some of the
> > delays off to a worker, like the current page flip code; see vblank
> > tasks.)
> > 
> > References: https://bugs.freedesktop.org/show_bug.cgi?id=72631
> > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> 
> Reviewed-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

Queued for -next, thanks for the patch.
-Daniel
-- 
Daniel Vetter
Software Engineer, Intel Corporation
+41 (0) 79 365 57 48 - http://blog.ffwll.ch