From patchwork Fri May 13 14:36:33 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Dave Gordon X-Patchwork-Id: 9092011 Return-Path: X-Original-To: patchwork-intel-gfx@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork1.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork1.web.kernel.org (Postfix) with ESMTP id 243679F457 for ; Fri, 13 May 2016 14:37:24 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 29F9A2025A for ; Fri, 13 May 2016 14:37:23 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) by mail.kernel.org (Postfix) with ESMTP id 18FC920204 for ; Fri, 13 May 2016 14:37:22 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 1E3E16E5B1; Fri, 13 May 2016 14:37:21 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga01.intel.com (mga01.intel.com [192.55.52.88]) by gabe.freedesktop.org (Postfix) with ESMTP id 973AE6E5B1 for ; Fri, 13 May 2016 14:37:12 +0000 (UTC) Received: from orsmga002.jf.intel.com ([10.7.209.21]) by fmsmga101.fm.intel.com with ESMTP; 13 May 2016 07:37:05 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.24,614,1455004800"; d="scan'208";a="975791811" Received: from dsgordon-linux2.isw.intel.com ([10.102.226.88]) by orsmga002.jf.intel.com with ESMTP; 13 May 2016 07:37:02 -0700 From: Dave Gordon To: intel-gfx@lists.freedesktop.org Date: Fri, 13 May 2016 15:36:33 +0100 Message-Id: <1463150195-28805-6-git-send-email-david.s.gordon@intel.com> X-Mailer: git-send-email 1.9.1 In-Reply-To: <1463150195-28805-1-git-send-email-david.s.gordon@intel.com> References: <1463150195-28805-1-git-send-email-david.s.gordon@intel.com> Organization: Intel Corporation (UK) Ltd. - Co. Reg. #1134945 - Pipers Way, Swindon SN3 1RJ Subject: [Intel-gfx] [PATCH v4 5/7] drm/i915/guc: don't spinwait if the GuC's workqueue is full X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.18 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" X-Spam-Status: No, score=-5.6 required=5.0 tests=BAYES_00, RCVD_IN_DNSWL_MED, RP_MATCHES_RCVD, UNPARSEABLE_RELAY autolearn=unavailable version=3.3.1 X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on mail.kernel.org X-Virus-Scanned: ClamAV using ClamSMTP Rather than wait to see whether more space becomes available in the GuC submission workqueue, we can just return -EAGAIN and let the caller try again in a little while. This gets rid of an uninterruptable sleep in the polling code :) We'll also add a counter to the GuC client statistics, to see how often we find the WQ full. v2: Flag the likely() code path (Tvtrko Ursulin). v4: Add/update comments about failure counters (Tvtrko Ursulin). Signed-off-by: Dave Gordon Reviewed-by: Tvrtko Ursulin --- drivers/gpu/drm/i915/i915_debugfs.c | 1 + drivers/gpu/drm/i915/i915_guc_submission.c | 18 +++++++----------- drivers/gpu/drm/i915/intel_guc.h | 22 ++++++++++++++++------ 3 files changed, 24 insertions(+), 17 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_debugfs.c b/drivers/gpu/drm/i915/i915_debugfs.c index 24f4105..de05841 100644 --- a/drivers/gpu/drm/i915/i915_debugfs.c +++ b/drivers/gpu/drm/i915/i915_debugfs.c @@ -2506,6 +2506,7 @@ static void i915_guc_client_info(struct seq_file *m, seq_printf(m, "\tWQ size %d, offset: 0x%x, tail %d\n", client->wq_size, client->wq_offset, client->wq_tail); + seq_printf(m, "\tWork queue full: %u\n", client->no_wq_space); seq_printf(m, "\tFailed to queue: %u\n", client->q_fail); seq_printf(m, "\tFailed doorbell: %u\n", client->b_fail); seq_printf(m, "\tLast submission result: %d\n", client->retcode); diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c index 87cb739..85b2b89 100644 --- a/drivers/gpu/drm/i915/i915_guc_submission.c +++ b/drivers/gpu/drm/i915/i915_guc_submission.c @@ -468,26 +468,22 @@ static void guc_fini_ctx_desc(struct intel_guc *guc, */ int i915_guc_wq_check_space(struct drm_i915_gem_request *request) { - const size_t size = sizeof(struct guc_wq_item); + const size_t wqi_size = sizeof(struct guc_wq_item); struct i915_guc_client *gc = request->i915->guc.execbuf_client; struct guc_process_desc *desc; - int ret = -ETIMEDOUT, timeout_counter = 200; + u32 freespace; GEM_BUG_ON(gc == NULL); desc = gc->client_base + gc->proc_desc_offset; - while (timeout_counter-- > 0) { - if (CIRC_SPACE(gc->wq_tail, desc->head, gc->wq_size) >= size) { - ret = 0; - break; - } + freespace = CIRC_SPACE(gc->wq_tail, desc->head, gc->wq_size); + if (likely(freespace >= wqi_size)) + return 0; - if (timeout_counter) - usleep_range(1000, 2000); - }; + gc->no_wq_space += 1; - return ret; + return -EAGAIN; } static int guc_add_workqueue_item(struct i915_guc_client *gc, diff --git a/drivers/gpu/drm/i915/intel_guc.h b/drivers/gpu/drm/i915/intel_guc.h index 91f315c..380a743 100644 --- a/drivers/gpu/drm/i915/intel_guc.h +++ b/drivers/gpu/drm/i915/intel_guc.h @@ -48,9 +48,18 @@ struct drm_i915_gem_request; * queue (a circular array of work items), again described in the process * descriptor. Work queue pages are mapped momentarily as required. * - * Finally, we also keep a few statistics here, including the number of - * submissions to each engine, and a record of the last submission failure - * (if any). + * We also keep a few statistics on failures. Ideally, these should all + * be zero! + * no_wq_space: times that the submission pre-check found no space was + * available in the work queue (note, the queue is shared, + * not per-engine). It is OK for this to be nonzero, but + * it should not be huge! + * q_fail: failed to enqueue a work item. This should never happen, + * because we check for space beforehand. + * b_fail: failed to ring the doorbell. This should never happen, unless + * somehow the hardware misbehaves, or maybe if the GuC firmware + * crashes? We probably need to reset the GPU to recover. + * retcode: errno from last guc_submit() */ struct i915_guc_client { struct drm_i915_gem_object *client_obj; @@ -71,12 +80,13 @@ struct i915_guc_client { uint32_t wq_tail; uint32_t unused; /* Was 'wq_head' */ - /* GuC submission statistics & status */ - uint64_t submissions[GUC_MAX_ENGINES_NUM]; + uint32_t no_wq_space; uint32_t q_fail; uint32_t b_fail; int retcode; - int spare; /* pad to 32 DWords */ + + /* Per-engine counts of GuC submissions */ + uint64_t submissions[GUC_MAX_ENGINES_NUM]; }; enum intel_guc_fw_status {