From patchwork Tue Mar 13 15:23:03 2018 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Greg KH X-Patchwork-Id: 10279325 Return-Path: Received: from mail.wl.linuxfoundation.org (pdx-wl-mail.web.codeaurora.org [172.30.200.125]) by pdx-korg-patchwork.web.codeaurora.org (Postfix) with ESMTP id 86588602C2 for ; Tue, 13 Mar 2018 15:29:30 +0000 (UTC) Received: from mail.wl.linuxfoundation.org (localhost [127.0.0.1]) by mail.wl.linuxfoundation.org (Postfix) with ESMTP id 76E4F28F4F for ; Tue, 13 Mar 2018 15:29:30 +0000 (UTC) Received: by mail.wl.linuxfoundation.org (Postfix, from userid 486) id 6BA6528F6E; Tue, 13 Mar 2018 15:29:30 +0000 (UTC) X-Spam-Checker-Version: SpamAssassin 3.3.1 (2010-03-16) on pdx-wl-mail.web.codeaurora.org X-Spam-Level: X-Spam-Status: No, score=-4.2 required=2.0 tests=BAYES_00, RCVD_IN_DNSWL_MED autolearn=ham version=3.3.1 Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) (using TLSv1.2 with cipher DHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by mail.wl.linuxfoundation.org (Postfix) with ESMTPS id C895528F4F for ; Tue, 13 Mar 2018 15:29:29 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id C654D6E55C; Tue, 13 Mar 2018 15:29:28 +0000 (UTC) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mail.linuxfoundation.org (mail.linuxfoundation.org [140.211.169.12]) by gabe.freedesktop.org (Postfix) with ESMTPS id ADD4B6E55C for ; Tue, 13 Mar 2018 15:29:27 +0000 (UTC) Received: from localhost (LFbn-1-12258-90.w90-92.abo.wanadoo.fr [90.92.71.90]) by mail.linuxfoundation.org (Postfix) with ESMTPSA id D6EFFF8F; Tue, 13 Mar 2018 15:29:26 +0000 (UTC) From: Greg Kroah-Hartman To: linux-kernel@vger.kernel.org Date: Tue, 13 Mar 2018 16:23:03 +0100 Message-Id: <20180313152321.784391855@linuxfoundation.org> X-Mailer: git-send-email 2.16.2 In-Reply-To: <20180313152320.439085687@linuxfoundation.org> References: <20180313152320.439085687@linuxfoundation.org> User-Agent: quilt/0.65 X-stable: review MIME-Version: 1.0 Subject: [Intel-gfx] [PATCH 4.15 016/146] drm/i915/perf: fix perf stream opening lock X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Cc: Greg Kroah-Hartman , intel-gfx@lists.freedesktop.org, Rodrigo Vivi , stable@vger.kernel.org, Matthew Auld Errors-To: intel-gfx-bounces@lists.freedesktop.org Sender: "Intel-gfx" X-Virus-Scanned: ClamAV using ClamSMTP 4.15-stable review patch. If anyone has any objections, please let me know. ------------------ From: Lionel Landwerlin commit f616f2830c1ed79245cfeca900f7e8a3b3c08c06 upstream. We're seeing on CI that some contexts don't have the programmed OA period timer that directs the OA unit on how often to write reports. The issue is that we're not holding the drm lock from when we edit the context images down to when we set the exclusive_stream variable. This leaves a window for the deferred context allocation to call i915_oa_init_reg_state() that will not program the expected OA timer value, because we haven't set the exclusive_stream yet. v2: Drop need_lock from gen8_configure_all_contexts() (Matt) Signed-off-by: Lionel Landwerlin Reviewed-by: Matthew Auld Reviewed-by: Chris Wilson Fixes: 701f8231a2f ("drm/i915/perf: prune OA configs") Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=102254 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103715 Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=103755 Link: https://patchwork.freedesktop.org/patch/msgid/20180301110613.1737-1-lionel.g.landwerlin@intel.com Cc: Jani Nikula Cc: Joonas Lahtinen Cc: Rodrigo Vivi Cc: intel-gfx@lists.freedesktop.org Cc: # v4.14+ (cherry picked from commit 41d3fdcd15d5ecf29cc73e8b79c2327ebb54b960) Signed-off-by: Rodrigo Vivi Signed-off-by: Greg Kroah-Hartman --- drivers/gpu/drm/i915/i915_perf.c | 38 ++++++++++++-------------------------- 1 file changed, 12 insertions(+), 26 deletions(-) --- a/drivers/gpu/drm/i915/i915_perf.c +++ b/drivers/gpu/drm/i915/i915_perf.c @@ -1301,9 +1301,8 @@ static void i915_oa_stream_destroy(struc */ mutex_lock(&dev_priv->drm.struct_mutex); dev_priv->perf.oa.exclusive_stream = NULL; - mutex_unlock(&dev_priv->drm.struct_mutex); - dev_priv->perf.oa.ops.disable_metric_set(dev_priv); + mutex_unlock(&dev_priv->drm.struct_mutex); free_oa_buffer(dev_priv); @@ -1755,22 +1754,13 @@ static int gen8_switch_to_updated_kernel * Note: it's only the RCS/Render context that has any OA state. */ static int gen8_configure_all_contexts(struct drm_i915_private *dev_priv, - const struct i915_oa_config *oa_config, - bool interruptible) + const struct i915_oa_config *oa_config) { struct i915_gem_context *ctx; int ret; unsigned int wait_flags = I915_WAIT_LOCKED; - if (interruptible) { - ret = i915_mutex_lock_interruptible(&dev_priv->drm); - if (ret) - return ret; - - wait_flags |= I915_WAIT_INTERRUPTIBLE; - } else { - mutex_lock(&dev_priv->drm.struct_mutex); - } + lockdep_assert_held(&dev_priv->drm.struct_mutex); /* Switch away from any user context. */ ret = gen8_switch_to_updated_kernel_context(dev_priv, oa_config); @@ -1818,8 +1808,6 @@ static int gen8_configure_all_contexts(s } out: - mutex_unlock(&dev_priv->drm.struct_mutex); - return ret; } @@ -1862,7 +1850,7 @@ static int gen8_enable_metric_set(struct * to make sure all slices/subslices are ON before writing to NOA * registers. */ - ret = gen8_configure_all_contexts(dev_priv, oa_config, true); + ret = gen8_configure_all_contexts(dev_priv, oa_config); if (ret) return ret; @@ -1877,7 +1865,7 @@ static int gen8_enable_metric_set(struct static void gen8_disable_metric_set(struct drm_i915_private *dev_priv) { /* Reset all contexts' slices/subslices configurations. */ - gen8_configure_all_contexts(dev_priv, NULL, false); + gen8_configure_all_contexts(dev_priv, NULL); I915_WRITE(GDT_CHICKEN_BITS, (I915_READ(GDT_CHICKEN_BITS) & ~GT_NOA_ENABLE)); @@ -2127,6 +2115,10 @@ static int i915_oa_stream_init(struct i9 if (ret) goto err_oa_buf_alloc; + ret = i915_mutex_lock_interruptible(&dev_priv->drm); + if (ret) + goto err_lock; + ret = dev_priv->perf.oa.ops.enable_metric_set(dev_priv, stream->oa_config); if (ret) @@ -2134,23 +2126,17 @@ static int i915_oa_stream_init(struct i9 stream->ops = &i915_oa_stream_ops; - /* Lock device for exclusive_stream access late because - * enable_metric_set() might lock as well on gen8+. - */ - ret = i915_mutex_lock_interruptible(&dev_priv->drm); - if (ret) - goto err_lock; - dev_priv->perf.oa.exclusive_stream = stream; mutex_unlock(&dev_priv->drm.struct_mutex); return 0; -err_lock: +err_enable: dev_priv->perf.oa.ops.disable_metric_set(dev_priv); + mutex_unlock(&dev_priv->drm.struct_mutex); -err_enable: +err_lock: free_oa_buffer(dev_priv); err_oa_buf_alloc: