From patchwork Wed Nov 28 15:18:45 2012 Content-Type: text/plain; charset="utf-8" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit X-Patchwork-Submitter: Mika Kuoppala X-Patchwork-Id: 1817101 Return-Path: X-Original-To: patchwork-intel-gfx@patchwork.kernel.org Delivered-To: patchwork-process-083081@patchwork1.kernel.org Received: from gabe.freedesktop.org (gabe.freedesktop.org [131.252.210.177]) by patchwork1.kernel.org (Postfix) with ESMTP id 460A93FD1A for ; Wed, 28 Nov 2012 15:19:03 +0000 (UTC) Received: from gabe.freedesktop.org (localhost [127.0.0.1]) by gabe.freedesktop.org (Postfix) with ESMTP id 38B7CE63E9 for ; Wed, 28 Nov 2012 07:19:03 -0800 (PST) X-Original-To: intel-gfx@lists.freedesktop.org Delivered-To: intel-gfx@lists.freedesktop.org Received: from mga14.intel.com (mga14.intel.com [143.182.124.37]) by gabe.freedesktop.org (Postfix) with ESMTP id CAFE2E5F39 for ; Wed, 28 Nov 2012 07:18:51 -0800 (PST) Received: from azsmga002.ch.intel.com ([10.2.17.35]) by azsmga102.ch.intel.com with ESMTP; 28 Nov 2012 07:18:51 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="4.84,179,1355126400"; d="scan'208";a="173568096" Received: from gaia.fi.intel.com (HELO gaia) ([10.237.72.66]) by AZSMGA002.ch.intel.com with ESMTP; 28 Nov 2012 07:18:50 -0800 Received: by gaia (Postfix, from userid 1000) id D9A1E413E7; Wed, 28 Nov 2012 17:18:47 +0200 (EET) From: Mika Kuoppala To: intel-gfx@lists.freedesktop.org Date: Wed, 28 Nov 2012 17:18:45 +0200 Message-Id: <1354115925-32630-1-git-send-email-mika.kuoppala@intel.com> X-Mailer: git-send-email 1.7.9.5 Subject: [Intel-gfx] [PATCH] drm/i915: Set sync_seqno properly after seqno wrap X-BeenThere: intel-gfx@lists.freedesktop.org X-Mailman-Version: 2.1.13 Precedence: list List-Id: Intel graphics driver community testing & development List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , MIME-Version: 1.0 Sender: intel-gfx-bounces+patchwork-intel-gfx=patchwork.kernel.org@lists.freedesktop.org Errors-To: intel-gfx-bounces+patchwork-intel-gfx=patchwork.kernel.org@lists.freedesktop.org i915_gem_handle_seqno_wrap() will zero all sync_seqnos but as the wrap can happen inside ring->sync_to(), pre wrap seqno was carried over and overwrote the zeroed sync_seqno. When wrap is handled, all outstanding requests will be retired and objects moved to inactive queue, causing their last_read_seqno to be zero. Use this to update the sync_seqno correctly. RING_SYNC registers after wrap will contain pre wrap values which are >= seqno. So injecting the semaphore wait into ring completes immediately. Original idea for using last_read_seqno from Chris Wilson. Signed-off-by: Mika Kuoppala Reviewed-by: Chris Wilson --- drivers/gpu/drm/i915/i915_gem.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-) diff --git a/drivers/gpu/drm/i915/i915_gem.c b/drivers/gpu/drm/i915/i915_gem.c index 85a0948..4ef8b57 100644 --- a/drivers/gpu/drm/i915/i915_gem.c +++ b/drivers/gpu/drm/i915/i915_gem.c @@ -2397,7 +2397,11 @@ i915_gem_object_sync(struct drm_i915_gem_object *obj, ret = to->sync_to(to, from, seqno); if (!ret) - from->sync_seqno[idx] = seqno; + /* We use last_read_seqno because sync_to() + * might have just caused seqno wrap under + * the radar. + */ + from->sync_seqno[idx] = obj->last_read_seqno; return ret; }