The continual resubmission model for execlists (and emulated over guc) requires that we keep feeding requests into the HW in order to generate more CS interrupts to drain the rest of the queue. Add a couple of asserts to ensure that we don't skip a cycle and come to a grinding halt. Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx> Cc: Michał Winiarski <michal.winiarski@xxxxxxxxx> --- drivers/gpu/drm/i915/intel_guc_submission.c | 6 ++++++ drivers/gpu/drm/i915/intel_lrc.c | 6 ++++++ 2 files changed, 12 insertions(+) diff --git a/drivers/gpu/drm/i915/intel_guc_submission.c b/drivers/gpu/drm/i915/intel_guc_submission.c index b43b58cc599b..946766b62459 100644 --- a/drivers/gpu/drm/i915/intel_guc_submission.c +++ b/drivers/gpu/drm/i915/intel_guc_submission.c @@ -747,6 +747,12 @@ static void guc_dequeue(struct intel_engine_cs *engine) execlists_set_active(execlists, EXECLISTS_ACTIVE_USER); guc_submit(engine); } + + /* We must always keep the beast fed if we have work piled up */ + GEM_BUG_ON(port_isset(execlists->port) && + !execlists_is_active(execlists, EXECLISTS_ACTIVE_USER)); + GEM_BUG_ON(execlists->first && !port_isset(execlists->port)); + unlock: spin_unlock_irq(&engine->timeline->lock); } diff --git a/drivers/gpu/drm/i915/intel_lrc.c b/drivers/gpu/drm/i915/intel_lrc.c index c2c8380a0121..6fbe1a8a37ad 100644 --- a/drivers/gpu/drm/i915/intel_lrc.c +++ b/drivers/gpu/drm/i915/intel_lrc.c @@ -642,6 +642,12 @@ static void execlists_dequeue(struct intel_engine_cs *engine) execlists->first = rb; if (submit) port_assign(port, last); + + /* We must always keep the beast fed if we have work piled up */ + GEM_BUG_ON(port_isset(execlists->port) && + !execlists_is_active(execlists, EXECLISTS_ACTIVE_USER)); + GEM_BUG_ON(execlists->first && !port_isset(execlists->port)); + unlock: spin_unlock_irq(&engine->timeline->lock); -- 2.16.1 _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
