From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga03.intel.com (mga03.intel.com [134.134.136.65]) by gabe.freedesktop.org (Postfix) with ESMTPS id 4757210E71E for ; Sat, 13 May 2023 02:22:42 +0000 (UTC) From: Umesh Nerlige Ramappa To: igt-dev@lists.freedesktop.org, Tvrtko Ursulin , Ashutosh Dixit Date: Fri, 12 May 2023 19:22:23 -0700 Message-Id: <20230513022234.2832233-5-umesh.nerlige.ramappa@intel.com> In-Reply-To: <20230513022234.2832233-1-umesh.nerlige.ramappa@intel.com> References: <20230513022234.2832233-1-umesh.nerlige.ramappa@intel.com> MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Subject: [igt-dev] [PATCH i-g-t 04/15] perf_pmu: Quiesce GPU if measuring idle busyness without spinner List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: igt-dev-bounces@lists.freedesktop.org Sender: "igt-dev" List-ID: The assumption in some tests is that the engines are not busy if no spinners are being run. This is not true in some cases where we see that the render is busy at the start of the test. Quiesce GPU to wait for such work to complete before checking for idle busyness. v2: Move gem_quiescent_gpu to beginning of test (Tvrtko) Signed-off-by: Umesh Nerlige Ramappa Reviewed-by: Ashutosh Dixit --- tests/i915/perf_pmu.c | 2 ++ 1 file changed, 2 insertions(+) diff --git a/tests/i915/perf_pmu.c b/tests/i915/perf_pmu.c index 155fc5f48..38bdee650 100644 --- a/tests/i915/perf_pmu.c +++ b/tests/i915/perf_pmu.c @@ -279,6 +279,7 @@ single(int gem_fd, const intel_ctx_t *ctx, int fd; uint64_t ahnd = get_reloc_ahnd(gem_fd, ctx->id); + gem_quiescent_gpu(gem_fd); fd = open_pmu(gem_fd, I915_PMU_ENGINE_BUSY(e->class, e->instance)); if (flags & TEST_BUSY) @@ -639,6 +640,7 @@ no_sema(int gem_fd, const intel_ctx_t *ctx, int fd[2]; uint64_t ahnd = get_reloc_ahnd(gem_fd, ctx->id); + gem_quiescent_gpu(gem_fd); fd[0] = open_group(gem_fd, I915_PMU_ENGINE_SEMA(e->class, e->instance), -1); fd[1] = open_group(gem_fd, I915_PMU_ENGINE_WAIT(e->class, e->instance), -- 2.36.1