From patchwork Tue Feb 16 05:27:18 2016 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: sourab.gupta@intel.com X-Patchwork-Id: 8321131 Return-Path: X-Original-To: patchwork-intel-gfx@patchwork.kernel.org Delivered-To: patchwork-parsemail@patchwork2.web.kernel.org Received: from mail.kernel.org (mail.kernel.org [198.145.29.136]) by patchwork2.web.kernel.org (Postfix) with ESMTP id 3E64EC02AA for ; Tue, 16 Feb 2016 05:25:54 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 2EB3C2025A for ; Tue, 16 Feb 2016 05:25:53 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) by mail.kernel.org (Postfix) with ESMTP id 057D220254 for ; Tue, 16 Feb 2016 05:25:52 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 9109D6E739; Tue, 16 Feb 2016 05:25:51 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by gabe.freedesktop.org (Postfix) with ESMTP id E35CA6E739 for ; Tue, 16 Feb 2016 05:25:49 +0000 (UTC) Received: from fmsmga001.fm.intel.com ([10.253.24.23]) by orsmga102.jf.intel.com with ESMTP; 15 Feb 2016 21:25:49 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.22,454,1449561600"; d="scan'208";a="903355392" Received: from sourab-desktop.iind.intel.com ([10.223.82.63]) by fmsmga001.fm.intel.com with ESMTP; 15 Feb 2016 21:25:47 -0800 From: sourab.gupta@intel.com To: intel-gfx@lists.freedesktop.org Date: Tue, 16 Feb 2016 10:57:18 +0530 Message-Id: <1455600439-18480-11-git-send-email-sourab.gupta@intel.com> X-Mailer: git-send-email 1.9.1 In-Reply-To: <1455600439-18480-1-git-send-email-sourab.gupta@intel.com> References: <1455600439-18480-1-git-send-email-sourab.gupta@intel.com> Cc: Jabin Wu , Sourab Gupta Subject: [Intel-gfx] [PATCH 10/11] drm/i915: Support opening multiple concurrent perf streams 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=-4.2 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 From: Sourab Gupta This patch adds support for opening multiple concurrent perf streams for different gpu engines, while having the restriction to open only a single stream open for a particular gpu engine. This enables userspace client to open multiple streams, one per engine, at any time to capture sample data for multiple gpu engines. Signed-off-by: Sourab Gupta --- drivers/gpu/drm/i915/i915_drv.h | 2 +- drivers/gpu/drm/i915/i915_perf.c | 65 +++++++++++++++++++++++----------------- 2 files changed, 38 insertions(+), 29 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h index b1c952c..bf65acb 100644 --- a/drivers/gpu/drm/i915/i915_drv.h +++ b/drivers/gpu/drm/i915/i915_drv.h @@ -2086,7 +2086,7 @@ struct drm_i915_private { spinlock_t hook_lock; struct hrtimer poll_check_timer; - struct i915_perf_stream *exclusive_stream; + struct i915_perf_stream *exclusive_stream[I915_NUM_RINGS]; wait_queue_head_t poll_wq[I915_NUM_RINGS]; struct { diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c index 1d2712d..3eb56d4 100644 --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -1061,7 +1061,7 @@ static void i915_perf_stream_destroy(struct i915_perf_stream *stream) { struct drm_i915_private *dev_priv = stream->dev_priv; - BUG_ON(stream != dev_priv->perf.exclusive_stream); + BUG_ON(stream != dev_priv->perf.exclusive_stream[stream->ring_id]); if (stream->using_oa) { dev_priv->perf.oa.ops.disable_metric_set(dev_priv); @@ -1075,7 +1075,7 @@ static void i915_perf_stream_destroy(struct i915_perf_stream *stream) if (stream->cs_mode) free_command_stream_buf(dev_priv, stream->ring_id); - dev_priv->perf.exclusive_stream = NULL; + dev_priv->perf.exclusive_stream[stream->ring_id] = NULL; } static void *vmap_oa_buffer(struct drm_i915_gem_object *obj) @@ -1434,17 +1434,17 @@ static void gen7_update_oacontrol_locked(struct drm_i915_private *dev_priv) { assert_spin_locked(&dev_priv->perf.hook_lock); - if (dev_priv->perf.exclusive_stream->enabled) { + if (dev_priv->perf.exclusive_stream[RCS]->enabled) { unsigned long ctx_id = 0; bool pinning_ok = false; - if (dev_priv->perf.exclusive_stream->ctx && + if (dev_priv->perf.exclusive_stream[RCS]->ctx && dev_priv->perf.oa.specific_ctx_id) { ctx_id = dev_priv->perf.oa.specific_ctx_id; pinning_ok = true; } - if (dev_priv->perf.exclusive_stream->ctx == NULL || + if (dev_priv->perf.exclusive_stream[RCS]->ctx == NULL || pinning_ok) { bool periodic = dev_priv->perf.oa.periodic; u32 period_exponent = dev_priv->perf.oa.period_exponent; @@ -1556,14 +1556,6 @@ static int i915_perf_stream_init(struct i915_perf_stream *stream, SAMPLE_TS); int ret; - /* To avoid the complexity of having to accurately filter - * counter reports and marshal to the appropriate client - * we currently only allow exclusive access */ - if (dev_priv->perf.exclusive_stream) { - DRM_ERROR("Stream already in use\n"); - return -EBUSY; - } - /* Ctx Id can be sampled in HSW only through command streamer mode */ if (IS_HASWELL(dev_priv->dev) && (props->sample_flags & SAMPLE_CTX_ID) && !props->cs_mode) { @@ -1576,6 +1568,13 @@ static int i915_perf_stream_init(struct i915_perf_stream *stream, if (require_oa_unit) { int format_size; + + /* Only allow exclusive access per stream */ + if (dev_priv->perf.exclusive_stream[RCS]) { + DRM_ERROR("Stream:0 already in use\n"); + return -EBUSY; + } + if (!dev_priv->perf.oa.ops.init_oa_buffer) { DRM_ERROR("OA unit not supported\n"); return -ENODEV; @@ -1673,6 +1672,13 @@ static int i915_perf_stream_init(struct i915_perf_stream *stream, } if (props->cs_mode) { + /* Only allow exclusive access per stream */ + if (dev_priv->perf.exclusive_stream[props->ring_id]) { + DRM_ERROR("Stream:%d already in use\n", props->ring_id); + ret = -EBUSY; + goto cs_error; + } + /* * The only time we should allow enabling CS mode if it's not * strictly required, is if SAMPLE_CTX_ID has been requested @@ -1709,7 +1715,7 @@ static int i915_perf_stream_init(struct i915_perf_stream *stream, goto cs_error; } - dev_priv->perf.exclusive_stream = stream; + dev_priv->perf.exclusive_stream[stream->ring_id] = stream; stream->destroy = i915_perf_stream_destroy; stream->enable = i915_perf_stream_enable; @@ -1751,8 +1757,8 @@ static void i915_oa_context_pin_notify_locked(struct drm_i915_private *dev_priv, dev_priv->perf.oa.ops.update_hw_ctx_id_locked == NULL) return; - if (dev_priv->perf.exclusive_stream && - dev_priv->perf.exclusive_stream->ctx == context) { + if (dev_priv->perf.exclusive_stream[RCS] && + dev_priv->perf.exclusive_stream[RCS]->ctx == context) { struct drm_i915_gem_object *obj = context->legacy_hw_ctx.rcs_state; u32 ctx_id = i915_gem_obj_ggtt_offset(obj); @@ -1820,8 +1826,8 @@ void i915_oa_legacy_ctx_switch_notify(struct drm_i915_gem_request *req) if (dev_priv->perf.oa.ops.legacy_ctx_switch_unlocked == NULL) return; - if (dev_priv->perf.exclusive_stream && - dev_priv->perf.exclusive_stream->enabled) { + if (dev_priv->perf.exclusive_stream[RCS] && + dev_priv->perf.exclusive_stream[RCS]->enabled) { /* XXX: We don't take a lock here and this may run * async with respect to stream methods. Notably we @@ -1944,23 +1950,26 @@ static ssize_t i915_perf_read(struct file *file, return ret; } -static enum hrtimer_restart poll_check_timer_cb(struct hrtimer *hrtimer) +static void wake_up_perf_streams(void *data, async_cookie_t cookie) { + struct drm_i915_private *dev_priv = data; struct i915_perf_stream *stream; - struct drm_i915_private *dev_priv = - container_of(hrtimer, typeof(*dev_priv), - perf.poll_check_timer); - - /* No need to protect the streams list here, since the hrtimer is - * disabled before the stream is removed from list, and currently a - * single exclusive_stream is supported. - * XXX: revisit this when multiple concurrent streams are supported. - */ + mutex_lock(&dev_priv->perf.streams_lock); list_for_each_entry(stream, &dev_priv->perf.streams, link) { if (stream_have_data__unlocked(stream)) wake_up(&dev_priv->perf.poll_wq[stream->ring_id]); } + mutex_unlock(&dev_priv->perf.streams_lock); +} + +static enum hrtimer_restart poll_check_timer_cb(struct hrtimer *hrtimer) +{ + struct drm_i915_private *dev_priv = + container_of(hrtimer, typeof(*dev_priv), + perf.poll_check_timer); + + async_schedule(wake_up_perf_streams, dev_priv); hrtimer_forward_now(hrtimer, ns_to_ktime(POLL_PERIOD));