From patchwork Mon Nov 23 23:02:58 2015 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: yu.dai@intel.com X-Patchwork-Id: 7686961 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 1F575BF90C for ; Mon, 23 Nov 2015 23:05:38 +0000 (UTC) Received: from mail.kernel.org (localhost [127.0.0.1]) by mail.kernel.org (Postfix) with ESMTP id 45B092085E for ; Mon, 23 Nov 2015 23:05:37 +0000 (UTC) Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) by mail.kernel.org (Postfix) with ESMTP id 2922F2085B for ; Mon, 23 Nov 2015 23:05:36 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 1A54A6E1B5; Mon, 23 Nov 2015 15:05:35 -0800 (PST) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga02.intel.com (mga02.intel.com [134.134.136.20]) by gabe.freedesktop.org (Postfix) with ESMTP id 6E7E76E1B5 for ; Mon, 23 Nov 2015 15:05:33 -0800 (PST) Received: from fmsmga003.fm.intel.com ([10.253.24.29]) by orsmga101.jf.intel.com with ESMTP; 23 Nov 2015 15:05:14 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.20,338,1444719600"; d="scan'208";a="605982748" Received: from alex-hsw.fm.intel.com ([10.19.83.10]) by FMSMGA003.fm.intel.com with ESMTP; 23 Nov 2015 15:05:15 -0800 From: yu.dai@intel.com To: intel-gfx@lists.freedesktop.org Date: Mon, 23 Nov 2015 15:02:58 -0800 Message-Id: <1448319778-29282-1-git-send-email-yu.dai@intel.com> X-Mailer: git-send-email 2.5.0 In-Reply-To: <20151120084542.GU17050@phenom.ffwll.local> References: <20151120084542.GU17050@phenom.ffwll.local> Subject: [Intel-gfx] [PATCH] drm/i915/guc: Move wait for GuC out of spinlock/unlock 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=-4.8 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: Alex Dai When GuC Work Queue is full, driver will wait GuC for avaliable space by delaying 1ms. The wait needs to be out of spinlockirq / unlock. Otherwise, lockup happens because jiffies won't be updated dur to irq is disabled. Issue is found in igt/gem_close_race. Signed-off-by: Alex Dai --- drivers/gpu/drm/i915/i915_guc_submission.c | 27 +++++++++++++++++---------- 1 file changed, 17 insertions(+), 10 deletions(-) diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c index 0a6b007..1418397 100644 --- a/drivers/gpu/drm/i915/i915_guc_submission.c +++ b/drivers/gpu/drm/i915/i915_guc_submission.c @@ -201,10 +201,13 @@ static int guc_ring_doorbell(struct i915_guc_client *gc) union guc_doorbell_qw *db; void *base; int attempt = 2, ret = -EAGAIN; + unsigned long flags; base = kmap_atomic(i915_gem_object_get_page(gc->client_obj, 0)); desc = base + gc->proc_desc_offset; + spin_lock_irqsave(&gc->wq_lock, flags); + /* Update the tail so it is visible to GuC */ desc->tail = gc->wq_tail; @@ -248,7 +251,10 @@ static int guc_ring_doorbell(struct i915_guc_client *gc) db_exc.cookie = 1; } + spin_unlock_irqrestore(&gc->wq_lock, flags); + kunmap_atomic(base); + return ret; } @@ -487,16 +493,16 @@ static int guc_get_workqueue_space(struct i915_guc_client *gc, u32 *offset) struct guc_process_desc *desc; void *base; u32 size = sizeof(struct guc_wq_item); - int ret = 0, timeout_counter = 200; + int ret = -ETIMEDOUT, timeout_counter = 200; + unsigned long flags; base = kmap_atomic(i915_gem_object_get_page(gc->client_obj, 0)); desc = base + gc->proc_desc_offset; while (timeout_counter-- > 0) { - ret = wait_for_atomic(CIRC_SPACE(gc->wq_tail, desc->head, - gc->wq_size) >= size, 1); + spin_lock_irqsave(&gc->wq_lock, flags); - if (!ret) { + if (CIRC_SPACE(gc->wq_tail, desc->head, gc->wq_size) >= size) { *offset = gc->wq_tail; /* advance the tail for next workqueue item */ @@ -505,7 +511,13 @@ static int guc_get_workqueue_space(struct i915_guc_client *gc, u32 *offset) /* this will break the loop */ timeout_counter = 0; + ret = 0; } + + spin_unlock_irqrestore(&gc->wq_lock, flags); + + if (timeout_counter) + usleep_range(1000, 2000); }; kunmap_atomic(base); @@ -597,19 +609,17 @@ int i915_guc_submit(struct i915_guc_client *client, { struct intel_guc *guc = client->guc; enum intel_ring_id ring_id = rq->ring->id; - unsigned long flags; int q_ret, b_ret; /* Need this because of the deferred pin ctx and ring */ /* Shall we move this right after ring is pinned? */ lr_context_update(rq); - spin_lock_irqsave(&client->wq_lock, flags); - q_ret = guc_add_workqueue_item(client, rq); if (q_ret == 0) b_ret = guc_ring_doorbell(client); + spin_lock(&guc->host2guc_lock); client->submissions[ring_id] += 1; if (q_ret) { client->q_fail += 1; @@ -620,9 +630,6 @@ int i915_guc_submit(struct i915_guc_client *client, } else { client->retcode = 0; } - spin_unlock_irqrestore(&client->wq_lock, flags); - - spin_lock(&guc->host2guc_lock); guc->submissions[ring_id] += 1; guc->last_seqno[ring_id] = rq->seqno; spin_unlock(&guc->host2guc_lock);