diff mbox

drm/i915/selftests: Skip the execlists tests on !execlists machines

Message ID 20180504124202.24894-1-chris@chris-wilson.co.uk (mailing list archive)
State New, archived
Headers show

Commit Message

Chris Wilson May 4, 2018, 12:42 p.m. UTC
Ignore the tests looking at the innards of execlists and its submission
tasklets on machines that don't support execlists!

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
 drivers/gpu/drm/i915/selftests/intel_lrc.c | 4 ++++
 1 file changed, 4 insertions(+)

Comments

Tvrtko Ursulin May 4, 2018, 4:25 p.m. UTC | #1
On 04/05/2018 13:42, Chris Wilson wrote:
> Ignore the tests looking at the innards of execlists and its submission
> tasklets on machines that don't support execlists!
> 
> Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> ---
>   drivers/gpu/drm/i915/selftests/intel_lrc.c | 4 ++++
>   1 file changed, 4 insertions(+)
> 
> diff --git a/drivers/gpu/drm/i915/selftests/intel_lrc.c b/drivers/gpu/drm/i915/selftests/intel_lrc.c
> index ee7e22d18ff8..b7460b5dd4f7 100644
> --- a/drivers/gpu/drm/i915/selftests/intel_lrc.c
> +++ b/drivers/gpu/drm/i915/selftests/intel_lrc.c
> @@ -505,5 +505,9 @@ int intel_execlists_live_selftests(struct drm_i915_private *i915)
>   		SUBTEST(live_preempt),
>   		SUBTEST(live_late_preempt),
>   	};
> +
> +	if (!HAS_EXECLISTS(i915))
> +		return 0;
> +
>   	return i915_subtests(tests, i915);
>   }
> 

Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

Is it a mystery why it wasn't discovered when tests were added?

Regards,

Tvrtko
Chris Wilson May 4, 2018, 4:33 p.m. UTC | #2
Quoting Tvrtko Ursulin (2018-05-04 17:25:27)
> 
> On 04/05/2018 13:42, Chris Wilson wrote:
> > Ignore the tests looking at the innards of execlists and its submission
> > tasklets on machines that don't support execlists!
> > 
> > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> > ---
> >   drivers/gpu/drm/i915/selftests/intel_lrc.c | 4 ++++
> >   1 file changed, 4 insertions(+)
> > 
> > diff --git a/drivers/gpu/drm/i915/selftests/intel_lrc.c b/drivers/gpu/drm/i915/selftests/intel_lrc.c
> > index ee7e22d18ff8..b7460b5dd4f7 100644
> > --- a/drivers/gpu/drm/i915/selftests/intel_lrc.c
> > +++ b/drivers/gpu/drm/i915/selftests/intel_lrc.c
> > @@ -505,5 +505,9 @@ int intel_execlists_live_selftests(struct drm_i915_private *i915)
> >               SUBTEST(live_preempt),
> >               SUBTEST(live_late_preempt),
> >       };
> > +
> > +     if (!HAS_EXECLISTS(i915))
> > +             return 0;
> > +
> >       return i915_subtests(tests, i915);
> >   }
> > 
> 
> Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> 
> Is it a mystery why it wasn't discovered when tests were added?

No, they do feature tests more precise than HAS_EXECLISTS, namely
preemption. Just later on I'm planing some other tests that want
the general guard, hence nipping it in the bud before you ask for it to
be split out of the larger patch.
-Chris
Chris Wilson May 4, 2018, 9:03 p.m. UTC | #3
Quoting Chris Wilson (2018-05-04 17:33:57)
> Quoting Tvrtko Ursulin (2018-05-04 17:25:27)
> > 
> > On 04/05/2018 13:42, Chris Wilson wrote:
> > > Ignore the tests looking at the innards of execlists and its submission
> > > tasklets on machines that don't support execlists!
> > > 
> > > Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
> > > ---
> > >   drivers/gpu/drm/i915/selftests/intel_lrc.c | 4 ++++
> > >   1 file changed, 4 insertions(+)
> > > 
> > > diff --git a/drivers/gpu/drm/i915/selftests/intel_lrc.c b/drivers/gpu/drm/i915/selftests/intel_lrc.c
> > > index ee7e22d18ff8..b7460b5dd4f7 100644
> > > --- a/drivers/gpu/drm/i915/selftests/intel_lrc.c
> > > +++ b/drivers/gpu/drm/i915/selftests/intel_lrc.c
> > > @@ -505,5 +505,9 @@ int intel_execlists_live_selftests(struct drm_i915_private *i915)
> > >               SUBTEST(live_preempt),
> > >               SUBTEST(live_late_preempt),
> > >       };
> > > +
> > > +     if (!HAS_EXECLISTS(i915))
> > > +             return 0;
> > > +
> > >       return i915_subtests(tests, i915);
> > >   }
> > > 
> > 
> > Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> > 
> > Is it a mystery why it wasn't discovered when tests were added?
> 
> No, they do feature tests more precise than HAS_EXECLISTS, namely
> preemption. Just later on I'm planing some other tests that want
> the general guard, hence nipping it in the bud before you ask for it to
> be split out of the larger patch.

And pushed, thanks for the review.
-Chris
diff mbox

Patch

diff --git a/drivers/gpu/drm/i915/selftests/intel_lrc.c b/drivers/gpu/drm/i915/selftests/intel_lrc.c
index ee7e22d18ff8..b7460b5dd4f7 100644
--- a/drivers/gpu/drm/i915/selftests/intel_lrc.c
+++ b/drivers/gpu/drm/i915/selftests/intel_lrc.c
@@ -505,5 +505,9 @@  int intel_execlists_live_selftests(struct drm_i915_private *i915)
 		SUBTEST(live_preempt),
 		SUBTEST(live_late_preempt),
 	};
+
+	if (!HAS_EXECLISTS(i915))
+		return 0;
+
 	return i915_subtests(tests, i915);
 }