Return-Path: X-Spam-Checker-Version: SpamAssassin 3.4.0 (2014-02-07) on aws-us-west-2-korg-lkml-1.web.codeaurora.org Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by smtp.lore.kernel.org (Postfix) with ESMTPS id 24737CCFA04 for ; Wed, 5 Nov 2025 03:51:39 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id A36CF10E108; Wed, 5 Nov 2025 03:51:38 +0000 (UTC) Authentication-Results: gabe.freedesktop.org; dkim=pass (2048-bit key; unprotected) header.d=intel.com header.i=@intel.com header.b="BA6w6xnH"; dkim-atps=neutral Received: from mgamail.intel.com (mgamail.intel.com [192.198.163.17]) by gabe.freedesktop.org (Postfix) with ESMTPS id 367B410E108 for ; Wed, 5 Nov 2025 03:51:37 +0000 (UTC) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/simple; d=intel.com; i=@intel.com; q=dns/txt; s=Intel; t=1762314697; x=1793850697; h=from:to:subject:date:message-id:mime-version: content-transfer-encoding; bh=nxtS6YdgZBOQNzs0Y0N93Q/8oKynlqZ3GDkZBCqjguc=; b=BA6w6xnHIHZiBlIkePPhJMm46AjzdbYJBqVpLUEjCmk0Nyd9YvfFUGek DzXo4qJk8+4pTqF5oYIKHK2/A9QCJ5VbfwBE3H1YY91daP9vK0s5NDe/v jNmFccDYcrQqqJP63+Q8/OeVF7Y4vPFUoKbnqVoKCbncHPw9jdang3su9 o3HUS6u3EShw2d6GDziQbRfk9h6M73LrWMSzfowH/edgHu6yx2Udj6tuZ 7WXQe0HmqpdYmaoq/MNDKpY15+/UwUlQD5B8UVWedzsZQVjrvzcNbSjoQ pAOxw6NZBrqOhyaMDPIGU8qRF/7q3gIfpM8Iv6fmTYpbqDHKuLgb3tjLD w==; X-CSE-ConnectionGUID: +UpjDaq2SrSA5ESVTWPe3g== X-CSE-MsgGUID: dGCjXJlwSAKSF0XQy7xplA== X-IronPort-AV: E=McAfee;i="6800,10657,11603"; a="64336567" X-IronPort-AV: E=Sophos;i="6.19,280,1754982000"; d="scan'208";a="64336567" Received: from fmviesa003.fm.intel.com ([10.60.135.143]) by fmvoesa111.fm.intel.com with ESMTP/TLS/ECDHE-RSA-AES256-GCM-SHA384; 04 Nov 2025 19:51:36 -0800 X-CSE-ConnectionGUID: euo3ciG0T8eon9yREc7YDg== X-CSE-MsgGUID: QEW3vw2MQM68IMjlcbR04Q== X-ExtLoop1: 1 Received: from dut2943pvc.iind.intel.com ([10.190.215.39]) by fmviesa003.fm.intel.com with ESMTP; 04 Nov 2025 19:51:35 -0800 From: nishit.sharma@intel.com To: igt-dev@lists.freedesktop.org, zbigniew.kempczynski@intel.com, nishit.sharma@intel.com Subject: [PATCH i-g-t v10 0/3] Multi-CCS support in IGT Date: Wed, 5 Nov 2025 03:51:31 +0000 Message-ID: <20251105035134.54594-1-nishit.sharma@intel.com> X-Mailer: git-send-email 2.43.0 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit X-BeenThere: igt-dev@lists.freedesktop.org X-Mailman-Version: 2.1.29 Precedence: list List-Id: Development mailing list for IGT GPU Tools List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: igt-dev-bounces@lists.freedesktop.org Sender: "igt-dev" From: Nishit Sharma Revision 1: Added loop_kernel_duration in available pipelines. loop_kernel_duration holds sleep duration and during sleep GPU will be workng. Patch-1/3: loop_kernel_duration addition in available pipelines in lib/intel_compute.c. igt_skip is called in pipeline which doesn't support loop_kernel shader. Patch-2/3: Added loop-duration-2s subtest in tests/intel/xe_compute to test if GPU is executing shader when sleep is called. This is verfied using gputop in another terminal. This also tests whether the test executed successfully in >= 2seconds of duration set Patch-3/3: Added eu-busy-10s test which creates multi threads depending on num_slices available and schedules workload on all ccs instances available. This also prints engine utlization in percentage while workload is executing. v2: - Addressed review comments from Zbyshek on description in Patch-1. - Added execution time check in Patch-2 regarding single iteration of scheduling and executing workload on single instance of CCS engine. v3: - Added missing code section in Patch-1. - Modified LOOP_DURATIOn macro to LOOP_DURATION_2s to avoid confusion. - Used LOOP_DURATION_2s in place of LOOP_DURATION and adjusted sleep durations. v4: - Test will skip if sriov is enabled and sriov_numfs > 0. - Test will continue runing if sriov_numvfs = 0 v5: - Incorporated review comments (Zbigniew) v6: - Simplified logic to capture EU busyness - Removed header which had pmu counter and engine instance logic - Moved pthread_data structure to xe_compute file which is used during thread creation v7: - collective counters are set per compute engine available in t->eci - fixed few nits v8: - Incorporated review comments (Zbigniew) - After workload execution set back ccs_mode to default value v9: - Moved get_gt_mask() in igt_fixture to avoid build failure v10: - Moved sriov enabled check in igt_fixture to skip test if sriov is enabled without initilializing or creation of threads. - Moved check for loop_kernel support in pipeline in igt_fixture v11: - Moved sriov enabled check in subtests which is setting ccs_mode value - loop_kernel support check moved in subtests which are using loop_kernel v12: - Added sriov enable check in subtest eu-busy-10s Nishit Sharma (3): lib/intel_compute: loop_kernel_duration in Pipeline tests/intel/xe_compute: Add loop_duration_2s subtest tests/intel/xe_compute: Add eu-busy-10s test lib/intel_compute.c | 26 ++- lib/intel_compute.h | 2 + tests/intel/xe_compute.c | 349 ++++++++++++++++++++++++++++++++++++++- tests/meson.build | 1 + 4 files changed, 374 insertions(+), 4 deletions(-) -- 2.43.0