drm/i915: Revert "drm/i915/tgl: Wa_1607138340"
diff mbox series

Message ID 20200108161954.29739-1-tvrtko.ursulin@linux.intel.com
State New
Headers show
Series
  • drm/i915: Revert "drm/i915/tgl: Wa_1607138340"
Related show

Commit Message

Tvrtko Ursulin Jan. 8, 2020, 4:19 p.m. UTC
From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>

This reverts commit 08fff7aeddc9dd72161b4c8fc27fbab12b4b9352.

For some yet unexplained reason not having this improves stability of some
media workloads.

Promise is that the media hang will be root caused properly and in the
meantime absence of this workaround is unlikely to cause problems.

Signed-off-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
Cc: Sudeep Dutt <sudeep.dutt@intel.com>
Cc: Francesco Balestrieri <francesco.balestrieri@intel.com>
Cc: Mika Kuoppala <mika.kuoppala@linux.intel.com>
Cc: Tony Ye <tony.ye@intel.com>
---
 drivers/gpu/drm/i915/gt/intel_lrc.c | 4 ----
 1 file changed, 4 deletions(-)

Comments

Chris Wilson Jan. 8, 2020, 4:27 p.m. UTC | #1
Quoting Tvrtko Ursulin (2020-01-08 16:19:54)
> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
> 
> This reverts commit 08fff7aeddc9dd72161b4c8fc27fbab12b4b9352.
> 
> For some yet unexplained reason not having this improves stability of some
> media workloads.
> 
> Promise is that the media hang will be root caused properly and in the
> meantime absence of this workaround is unlikely to cause problems.

I suspect this is shooting the messenger. The danger is that we rely on
FORCE_RESTORE for preemption, and so if this bit is triggering the issue,
their problem just becomes more sporadic.
-Chris
Mika Kuoppala Jan. 9, 2020, 8:46 a.m. UTC | #2
Chris Wilson <chris@chris-wilson.co.uk> writes:

> Quoting Tvrtko Ursulin (2020-01-08 16:19:54)
>> From: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
>> 
>> This reverts commit 08fff7aeddc9dd72161b4c8fc27fbab12b4b9352.
>> 
>> For some yet unexplained reason not having this improves stability of some
>> media workloads.
>> 
>> Promise is that the media hang will be root caused properly and in the
>> meantime absence of this workaround is unlikely to cause problems.
>
> I suspect this is shooting the messenger. The danger is that we rely on
> FORCE_RESTORE for preemption, and so if this bit is triggering the issue,
> their problem just becomes more sporadic.

Noted. But lets get a sample point with A0. We kind
of know where to focus our gaze from now on.

Acked-by: Mika Kuoppala <mika.kuoppala@linux.intel.com>

Patch
diff mbox series

diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c
index bd74b76c6403..2217c09ca227 100644
--- a/drivers/gpu/drm/i915/gt/intel_lrc.c
+++ b/drivers/gpu/drm/i915/gt/intel_lrc.c
@@ -1351,10 +1351,6 @@  static u64 execlists_update_context(struct i915_request *rq)
 	 */
 	wmb();
 
-	/* Wa_1607138340:tgl */
-	if (IS_TGL_REVID(rq->i915, TGL_REVID_A0, TGL_REVID_A0))
-		desc |= CTX_DESC_FORCE_RESTORE;
-
 	ce->lrc_desc &= ~CTX_DESC_FORCE_RESTORE;
 	return desc;
 }