[Intel-gfx] [PATCH] drm/i915/guc: Rename GuC irq trigger function

Michal Wajdeczko michal.wajdeczko at intel.com
Wed Aug 9 21:12:59 UTC 2017


On Wed, Aug 09, 2017 at 09:55:03PM +0100, Chris Wilson wrote:
> Quoting Michal Wajdeczko (2017-08-09 18:28:07)
> > We should emphasize that irq trigger function depends on Gen.
> > 
> > Signed-off-by: Michal Wajdeczko <michal.wajdeczko at intel.com>
> > Cc: Joonas Lahtinen <joonas.lahtinen at linux.intel.com>
> > ---
> >  drivers/gpu/drm/i915/intel_uc.c | 4 ++--
> >  1 file changed, 2 insertions(+), 2 deletions(-)
> > 
> > diff --git a/drivers/gpu/drm/i915/intel_uc.c b/drivers/gpu/drm/i915/intel_uc.c
> > index 27e072c..d78ecae 100644
> > --- a/drivers/gpu/drm/i915/intel_uc.c
> > +++ b/drivers/gpu/drm/i915/intel_uc.c
> > @@ -94,7 +94,7 @@ void intel_uc_sanitize_options(struct drm_i915_private *dev_priv)
> >                 i915.enable_guc_submission = HAS_GUC_SCHED(dev_priv);
> >  }
> >  
> > -static void guc_write_irq_trigger(struct intel_guc *guc)
> > +static void gen8_guc_trigger_irq(struct intel_guc *guc)
> >  {
> >         struct drm_i915_private *dev_priv = guc_to_i915(guc);
> >  
> > @@ -109,7 +109,7 @@ void intel_uc_init_early(struct drm_i915_private *dev_priv)
> >  
> >         mutex_init(&guc->send_mutex);
> >         guc->send = intel_guc_send_nop;
> > -       guc->notify = guc_write_irq_trigger;
> > +       guc->notify = gen8_guc_trigger_irq;
> 
> My vocab, so I may have this completely backwards,
> 
>   irq = pin/wire for asserting the interrupt (hw)
> 
>   interrupt = the act of interrupting the CPU
> 
> (so we have irq_handlers and interrupt context)
> 
> So gen8_guc_trigger_irq is
>   - from gen8 onwards
>   - interfacing with the guc
>   - trigger an irq
> It is the last part that causes a bit of confusion for me, I would have
> actually gone with guc->notify = gen8_guc_raise_irq or
> gen8_guc_send_interrupt.

>From above proposals I prefer gen8_guc_raise_irq, as
gen8_guc_send_"the act of interrupting the CPU" is also confusing

Thanks,
Michal
 
> 
> But definitely trigger_irq is already more meaningful (or at least less
> misleading) than write_irq_trigger.
> -Chris


More information about the Intel-gfx mailing list