performance(ocl): program barrier pc in taskStream

Program barrier immediately to task stream.
This will reduce the number of batch buffer starts.

Related-To: NEO-8147

Signed-off-by: Dominik Dabek <dominik.dabek@intel.com>
This commit is contained in:
Dominik Dabek
2023-08-30 15:06:48 +00:00
committed by Compute-Runtime-Automation
parent a38ac3557b
commit 839c2d6737
16 changed files with 163 additions and 40 deletions

View File

@@ -811,7 +811,7 @@ HWTEST_TEMPLATED_F(BcsBufferTests, givenPipeControlRequestWhenDispatchingBlitEnq
}
}
HWTEST_TEMPLATED_F(BcsBufferTests, givenBarrierWithEmptyWaitlistWhenReleasingMultipleBlockedEnqueuesThenProgramBarrierOnce) {
HWTEST_TEMPLATED_F(BcsBufferTests, givenStallingCommandsOnNextFlushWhenReleasingMultipleBlockedEnqueuesThenProgramBarrierOnce) {
DebugManager.flags.OptimizeIoqBarriersHandling.set(0);
using PIPE_CONTROL = typename FamilyType::PIPE_CONTROL;
@@ -827,7 +827,7 @@ HWTEST_TEMPLATED_F(BcsBufferTests, givenBarrierWithEmptyWaitlistWhenReleasingMul
cl_event waitlist0[] = {&userEvent0};
cl_event waitlist1[] = {&userEvent1};
cmdQ->enqueueBarrierWithWaitList(0, nullptr, nullptr);
cmdQ->setStallingCommandsOnNextFlush(true);
cmdQ->enqueueWriteBuffer(buffer.get(), false, 0, 1, hostPtr, nullptr, 1, waitlist0, nullptr);
cmdQ->enqueueWriteBuffer(buffer.get(), false, 0, 1, hostPtr, nullptr, 1, waitlist1, nullptr);