[Intel-gfx] [PATCH] drm/i915: Initialize audio only when display is present

[Chris Wilson]

Quoting Mika Kahola (2017-11-10 14:06:38)
> On Fri, 2017-11-10 at 13:37 +0000, Chris Wilson wrote:
> > Quoting Mika Kahola (2017-11-10 13:31:54)
> > > 
> > > At least in Coffee Lake it happens that we start initiliazing audio
> > > when
> > > no display is connected. This was discovered by CI when running IGT
> > > test
> > > case
> > > 
> > > drv_module_reload --r basic-no-display
> > > 
> > > The issue here is that the 'intel_device_info_runtime_init()' sets
> > > num_pipes to 0 but before this happens the audio part has already
> > > started
> > > to initialize itself. Later on the num_pipes is updated to 0 in
> > > intel_device_info_runtime_init()
> > runtime_init happens twice? (I am confused by this pair of
> > sentences.)
> > 
> > and we hit that when audio part is digging
> This is tricky. i915_audio_component_get_eld() gets called from a sound
> driver. Even though, I disable intel_audio_init() call, these num_pipes
> warnings keep popping up in dmesg like this.

Ask for a kasan run. I don't see where i915 is unbound during cleanup.
component_del() only unbinds the master, and not itself. As I read the
code it seems like the old i915 device is not uncoupled.

First time I dipped my toe into the component code so take that with a
pinch of salt.
-Chris