Re: [PATCH 3/4] drm/i915: Split pch irq handling to ack+handler

From: "Ville Syrjälä" <ville.syrjala@linux.intel.com>
To: Chris Wilson <chris@chris-wilson.co.uk>
Cc: intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH 3/4] drm/i915: Split pch irq handling to ack+handler
Date: Mon, 15 Apr 2019 19:56:34 +0300	[thread overview]
Message-ID: <20190415165634.GD24299@intel.com> (raw)
In-Reply-To: <155534688453.22690.17883214025402246034@skylake-alporthouse-com>

On Mon, Apr 15, 2019 at 05:48:04PM +0100, Chris Wilson wrote:
> Quoting Ville Syrjala (2019-04-15 16:49:03)
> > @@ -2563,15 +2613,20 @@ static void ilk_display_irq_handler(struct drm_i915_private *dev_priv,
> >  
> >         /* check event from PCH */
> >         if (de_iir & DE_PCH_EVENT) {
> > -               u32 pch_iir = I915_READ(SDEIIR);
> > +               struct pch_irq_regs pch = {};
> 
> If I am following along correctly, we don't need the memset here as we
> only ever check dependent members after a guard (such as the iir or
> hotplug trigger).

IIRC I did the zero inits to avoid false positives from the compiler
once these structs get hoisted up to the main irq handler. But I must
admit it's been a while since I wrote the basic form of this the memory
is getting hazy.

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

-- 
Ville Syrjälä
Intel
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx