On 09/10/2019 12:28, Chris Wilson wrote:
We set out-of-bound parameters inside the i915_requests.flags field,
such as disabling preemption or marking the end-of-context. We should
not coalesce consecutive requests if they have differing instructions
as we only inspect the last active request in a context. Thus if we
allow a later request to be merged into the same execution context, it
will mask any of the earlier flags.
References: 2a98f4e65bba ("drm/i915: add infrastructure to hold off preemption on a request")
Signed-off-by: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
Cc: Lionel Landwerlin <lionel.g.landwerlin@xxxxxxxxx>
Oops, I missed your patch :(I guess this works too, potentially providing better timeslicing than the update I made.
I'll give it a go later, but it sounds like it should work. Thanks, -Lionel
--- drivers/gpu/drm/i915/gt/intel_lrc.c | 3 +++ 1 file changed, 3 insertions(+) diff --git a/drivers/gpu/drm/i915/gt/intel_lrc.c b/drivers/gpu/drm/i915/gt/intel_lrc.c index 6db762c509b8..2e5192a9e796 100644 --- a/drivers/gpu/drm/i915/gt/intel_lrc.c +++ b/drivers/gpu/drm/i915/gt/intel_lrc.c @@ -1233,6 +1233,9 @@ static bool can_merge_rq(const struct i915_request *prev, if (i915_request_completed(next)) return true;+ if (unlikely(prev->flags ^ next->flags))+ return false; + if (!can_merge_ctx(prev->hw_context, next->hw_context)) return false;
_______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
