Re: [Intel-gfx] [PATCH 1/2] drm/i915/selftests: Add a cancel request selftest that triggers a reset

[Matthew Brost <matthew.brost@intel.com>]

On Thu, Jan 13, 2022 at 09:59:35AM -0800, John Harrison wrote:
> On 1/13/2022 09:34, Matthew Brost wrote:
> > On Thu, Jan 13, 2022 at 09:33:12AM -0800, John Harrison wrote:
> > > On 1/11/2022 15:11, Matthew Brost wrote:
> > > > Add a cancel request selftest that results in an engine reset to cancel
> > > > the request as it is non-preemptable. Also insert a NOP request after
> > > > the cancelled request and confirm that it completes successfully.
> > > > 
> > > > v2:
> > > >    (Tvrtko)
> > > >     - Skip test if preemption timeout compiled out
> > > >     - Skip test if engine reset isn't supported
> > > >     - Update debug prints to be more descriptive
> > > > v3:
> > > >     - Add comment explaining test
> > > > 
> > > > Signed-off-by: Matthew Brost <matthew.brost@intel.com>
> > > > ---
> > > >    drivers/gpu/drm/i915/selftests/i915_request.c | 117 ++++++++++++++++++
> > > >    1 file changed, 117 insertions(+)
> > > > 
> > > > diff --git a/drivers/gpu/drm/i915/selftests/i915_request.c b/drivers/gpu/drm/i915/selftests/i915_request.c
> > > > index 7f66f6d299b26..f78de99d5ae1e 100644
> > > > --- a/drivers/gpu/drm/i915/selftests/i915_request.c
> > > > +++ b/drivers/gpu/drm/i915/selftests/i915_request.c
> > > > @@ -782,6 +782,115 @@ static int __cancel_completed(struct intel_engine_cs *engine)
> > > >    	return err;
> > > >    }
> > > > +/*
> > > > + * Test to prove a non-preemptable request can be cancelled and a subsequent
> > > > + * request on the same context can successfully complete after cancallation.
> > > cancellation
> > > 
> > Yep.
> > 
> > > > + *
> > > > + * Testing methodology is to create non-preemptable request and submit it,
> > > a non-preemptible
> > > 
> > Yep.
> > 
> > > > + * wait for spinner to start, create a NOP request and submit it, cancel the
> > > > + * spinner, wait for spinner to complete and verify it failed with an error,
> > > > + * finally wait for NOP request to complete verify it succeeded without an
> > > > + * error. Preemption timeout also reduced / restored so test runs in a timely
> > > > + * maner.
> > > > + */
> > > > +static int __cancel_reset(struct drm_i915_private *i915,
> > > > +			  struct intel_engine_cs *engine)
> > > > +{
> > > > +	struct intel_context *ce;
> > > > +	struct igt_spinner spin;
> > > > +	struct i915_request *rq, *nop;
> > > > +	unsigned long preempt_timeout_ms;
> > > > +	int err = 0;
> > > > +
> > > > +	if (!CONFIG_DRM_I915_PREEMPT_TIMEOUT ||
> > > Does this matter? The test is overriding the default anyway.
> > > 
> > Yes. Execlists don't try to preempt anything if
> > CONFIG_DRM_I915_PREEMPT_TIMEOUT is turned off. If we wan't to avoid the
> > cancelation doing a full GT reset, CONFIG_DRM_I915_PREEMPT_TIMEOUT
> > should be turned on.
> Hmm, I would read that as a bug. The description for the config parameter
> is:
>           "This is adjustable via
>           /sys/class/drm/card?/engine/*/preempt_timeout_ms
> 
>           May be 0 to disable the timeout.
> 
>           The compiled in default may get overridden at driver probe time on
>           certain platforms and certain engines which will be reflected in
> the
>           sysfs control."
> 
> I would take that as meaning that even if the compiled in default is zero,
> the user or even the i915 driver itself could override that at runtime and
> enable pre-emption again. So having any code use this as a flag is broken.
> Indeed, any code other than 'engine->default_preempt_timeout =
> CONFIG_PREEMPT_TIMEOUT' is broken, IMHO.
> 

Can't really argue against you here.

> But maybe that's for a different patch. If the driver is already behaving
> badly and doing the correct thing here will actually cause test failures
> then you can't really do much other than follow the existing bad behaviour.
>

Yea, agree it is out of scope this patch / series. We can cleanup the
execlists code in a follow up patch if needed + loop in an execlists
expert for a reviewer. Maybe there is a unknown reason that code is
doing this?

Matt

> John.
> 
> 
> > > > +	    !intel_has_reset_engine(engine->gt))
> > > > +		return 0;
> > > > +
> > > > +	preempt_timeout_ms = engine->props.preempt_timeout_ms;
> > > > +	engine->props.preempt_timeout_ms = 100;
> > > > +
> > > > +	if (igt_spinner_init(&spin, engine->gt))
> > > > +		goto out_restore;
> > > > +
> > > > +	ce = intel_context_create(engine);
> > > > +	if (IS_ERR(ce)) {
> > > > +		err = PTR_ERR(ce);
> > > > +		goto out_spin;
> > > > +	}
> > > > +
> > > > +	rq = igt_spinner_create_request(&spin, ce, MI_NOOP);
> > > > +	if (IS_ERR(rq)) {
> > > > +		err = PTR_ERR(rq);
> > > > +		goto out_ce;
> > > > +	}
> > > > +
> > > > +	pr_debug("%s: Cancelling active non-preemptable request\n",
> > > > +		 engine->name);
> > > > +	i915_request_get(rq);
> > > > +	i915_request_add(rq);
> > > > +	if (!igt_wait_for_spinner(&spin, rq)) {
> > > > +		struct drm_printer p = drm_info_printer(engine->i915->drm.dev);
> > > > +
> > > > +		pr_err("Failed to start spinner on %s\n", engine->name);
> > > > +		intel_engine_dump(engine, &p, "%s\n", engine->name);
> > > > +		err = -ETIME;
> > > > +		goto out_rq;
> > > > +	}
> > > > +
> > > > +	nop = intel_context_create_request(ce);
> > > > +	if (IS_ERR(nop))
> > > > +		goto out_nop;
> > > Should be out_rq?
> > > 
> > Yes, it should.
> > 
> > Matt
> > 
> > > John.
> > > 
> > > 
> > > > +	i915_request_get(nop);
> > > > +	i915_request_add(nop);
> > > > +
> > > > +	i915_request_cancel(rq, -EINTR);
> > > > +
> > > > +	if (i915_request_wait(rq, 0, HZ) < 0) {
> > > > +		struct drm_printer p = drm_info_printer(engine->i915->drm.dev);
> > > > +
> > > > +		pr_err("%s: Failed to cancel hung request\n", engine->name);
> > > > +		intel_engine_dump(engine, &p, "%s\n", engine->name);
> > > > +		err = -ETIME;
> > > > +		goto out_nop;
> > > > +	}
> > > > +
> > > > +	if (rq->fence.error != -EINTR) {
> > > > +		pr_err("%s: fence not cancelled (%u)\n",
> > > > +		       engine->name, rq->fence.error);
> > > > +		err = -EINVAL;
> > > > +		goto out_nop;
> > > > +	}
> > > > +
> > > > +	if (i915_request_wait(nop, 0, HZ) < 0) {
> > > > +		struct drm_printer p = drm_info_printer(engine->i915->drm.dev);
> > > > +
> > > > +		pr_err("%s: Failed to complete nop request\n", engine->name);
> > > > +		intel_engine_dump(engine, &p, "%s\n", engine->name);
> > > > +		err = -ETIME;
> > > > +		goto out_nop;
> > > > +	}
> > > > +
> > > > +	if (nop->fence.error != 0) {
> > > > +		pr_err("%s: Nop request errored (%u)\n",
> > > > +		       engine->name, nop->fence.error);
> > > > +		err = -EINVAL;
> > > > +	}
> > > > +
> > > > +out_nop:
> > > > +	i915_request_put(nop);
> > > > +out_rq:
> > > > +	i915_request_put(rq);
> > > > +out_ce:
> > > > +	intel_context_put(ce);
> > > > +out_spin:
> > > > +	igt_spinner_fini(&spin);
> > > > +out_restore:
> > > > +	engine->props.preempt_timeout_ms = preempt_timeout_ms;
> > > > +	if (err)
> > > > +		pr_err("%s: %s error %d\n", __func__, engine->name, err);
> > > > +	return err;
> > > > +}
> > > > +
> > > >    static int live_cancel_request(void *arg)
> > > >    {
> > > >    	struct drm_i915_private *i915 = arg;
> > > > @@ -814,6 +923,14 @@ static int live_cancel_request(void *arg)
> > > >    			return err;
> > > >    		if (err2)
> > > >    			return err2;
> > > > +
> > > > +		/* Expects reset so call outside of igt_live_test_* */
> > > > +		err = __cancel_reset(i915, engine);
> > > > +		if (err)
> > > > +			return err;
> > > > +
> > > > +		if (igt_flush_test(i915))
> > > > +			return -EIO;
> > > >    	}
> > > >    	return 0;
>