Commit a7980a64 authored Apr 04, 2017 by Chris Wilson

drm/i915: Apply a cond_resched() to the saturated signaler

If the engine is continually completing nops, we can saturate the
signaler and keep it working indefinitely. This angers the NMI watchdog!

A good example is to disable semaphores on snb and run igt/gem_exec_nop -
the parallel, multi-engine workloads are more than sufficient to hog the
CPU, preventing the system from even processing ICMP echo replies.

v2: Tvrtko dug into cond_resched() on x86 and found that it only
depended upon preempt_count and not tif_need_resched() - which means
that we would always call schedule() at that point.

Fixes: c81d4613 ("drm/i915: Convert trace-irq to the breadcrumb waiter")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Mika Kuoppala <mika.kuoppala@intel.com>
Cc: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: http://patchwork.freedesktop.org/patch/msgid/20170404120531.10737-1-chris@chris-wilson.co.uk

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

parent b1becb88

drivers/gpu/drm/i915/intel_breadcrumbs.c

+14 −1

Original line number	Diff line number	Diff line
		@@ -580,6 +580,8 @@ static int intel_breadcrumbs_signaler(void *arg)
		signaler_set_rtpriority();

		do {
		bool do_schedule = true;

		set_current_state(TASK_INTERRUPTIBLE);

		/* We are either woken up by the interrupt bottom-half,
		@@ -626,7 +628,18 @@ static int intel_breadcrumbs_signaler(void *arg)
		spin_unlock_irq(&b->rb_lock);

		i915_gem_request_put(request);
		} else {

		/* If the engine is saturated we may be continually
		* processing completed requests. This angers the
		* NMI watchdog if we never let anything else
		* have access to the CPU. Let's pretend to be nice
		* and relinquish the CPU if we burn through the
		* entire RT timeslice!
		*/
		do_schedule = need_resched();
		}

		if (unlikely(do_schedule)) {
		DEFINE_WAIT(exec);

		if (kthread_should_park())