Re: [Intel-gfx] [PATCH 3/3] drm/i915: Re-enable vblank irqs for already active pipes

[Daniel Vetter <daniel@ffwll.ch>]

On Tue, May 20, 2014 at 05:20:05PM +0300, ville.syrjala@linux.intel.com wrote:
> From: Ville Syrjälä <ville.syrjala@linux.intel.com>
> 
> If a pipe is already active when we init/resume there might not be a
> full modeset afterwards so drm_vblank_on() may not get called. In such
> a case if someone is holding a vblank reference across a suspend/resume
> cycle drm_vblank_get() called after resuming won't re-enable the vblank
> interrupts.
> 
> So in order to make sure vblank interrupts get re-enabled post-resume,
> call drm_vblank_on() in intel_sanitize_crtc() if the crtc is already
> active.
> 
> v2: Also drm_vblank_off() if the pipe got disabled magically
> 
> Signed-off-by: Ville Syrjälä <ville.syrjala@linux.intel.com>

This seems to duct-tape over the funny failure I'm seeing on my snb where
kms_flip/vblank-vs-modeset|dpms-suspend work nicely, but
kms_flip/vblank-vs-suspend has some totally hilarious random vblank frame
counter after the modeset.

Testecase: igt/kms_flip/vblank-vs-suspend
Tested-by: Daniel Vetter <daniel.vetter@ffwll.ch>

Now the problem I have: We already call this in the crtc_enable hook. Why
does calling this here again add the necessary magic?

/me has no clue

Cheers, Daniel
> ---
>  drivers/gpu/drm/i915/intel_display.c | 6 ++++++
>  1 file changed, 6 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
> index 9420f4f..2e9f0b0 100644
> --- a/drivers/gpu/drm/i915/intel_display.c
> +++ b/drivers/gpu/drm/i915/intel_display.c
> @@ -11708,6 +11708,12 @@ static void intel_sanitize_crtc(struct intel_crtc *crtc)
>  	reg = PIPECONF(crtc->config.cpu_transcoder);
>  	I915_WRITE(reg, I915_READ(reg) & ~PIPECONF_FRAME_START_DELAY_MASK);
>  
> +	/* restore vblank interrupts to correct state */
> +	if (crtc->active)
> +		drm_vblank_on(dev, crtc->pipe);
> +	else
> +		drm_vblank_off(dev, crtc->pipe);
> +
>  	/* We need to sanitize the plane -> pipe mapping first because this will
>  	 * disable the crtc (and hence change the state) if it is wrong. Note
>  	 * that gen4+ has a fixed plane -> pipe mapping.  */
> -- 
> 1.8.5.5
> 
> _______________________________________________
> Intel-gfx mailing list
> Intel-gfx@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/intel-gfx

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