drm/i915/execlists: HWACK checking superseded checking port[0].count
authorChris Wilson <chris@chris-wilson.co.uk>
Thu, 17 May 2018 11:56:47 +0000 (12:56 +0100)
committerChris Wilson <chris@chris-wilson.co.uk>
Thu, 17 May 2018 17:02:02 +0000 (18:02 +0100)
The HWACK bit more generically solves the problem of resubmitting ESLP
while the hardware is still processing the current ELSP write. We no
longer need to check port[0].count itself.

References: ba74cb10c775 ("drm/i915/execlists: Delay writing to ELSP until HW has processed the previous write")
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20180517115647.17205-1-chris@chris-wilson.co.uk
drivers/gpu/drm/i915/intel_lrc.c

index 493e6bd9e28adfc7a492922a9d213bc920c1f459..ae5adad7cc639aba0eb2a806bd994de454e9e973 100644 (file)
@@ -612,8 +612,6 @@ static bool __execlists_dequeue(struct intel_engine_cs *engine)
                GEM_BUG_ON(!execlists_is_active(execlists,
                                                EXECLISTS_ACTIVE_USER));
                GEM_BUG_ON(!port_count(&port[0]));
-               if (port_count(&port[0]) > 1)
-                       return false;
 
                /*
                 * If we write to ELSP a second time before the HW has had