From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759772AbbIDNSk (ORCPT ); Fri, 4 Sep 2015 09:18:40 -0400 Received: from youngberry.canonical.com ([91.189.89.112]:50411 "EHLO youngberry.canonical.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933115AbbIDNKW (ORCPT ); Fri, 4 Sep 2015 09:10:22 -0400 From: Luis Henriques To: linux-kernel@vger.kernel.org, stable@vger.kernel.org, kernel-team@lists.ubuntu.com Cc: Thomas Hellstrom , Dave Airlie , Luis Henriques Subject: [PATCH 3.16.y-ckt 109/130] drm/vmwgfx: Fix execbuf locking issues Date: Fri, 4 Sep 2015 14:08:17 +0100 Message-Id: <1441372118-5933-110-git-send-email-luis.henriques@canonical.com> X-Mailer: git-send-email 2.1.4 In-Reply-To: <1441372118-5933-1-git-send-email-luis.henriques@canonical.com> References: <1441372118-5933-1-git-send-email-luis.henriques@canonical.com> X-Extended-Stable: 3.16 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org 3.16.7-ckt17 -stable review patch. If anyone has any objections, please let me know. ------------------ From: Thomas Hellstrom commit 3e04e2fe6d87807d27521ad6ebb9e7919d628f25 upstream. This addresses two issues that cause problems with viewperf maya-03 in situation with memory pressure. The first issue causes attempts to unreserve buffers if batched reservation fails due to, for example, a signal pending. While previously the ttm_eu api was resistant against this type of error, it is no longer and the lockdep code will complain about attempting to unreserve buffers that are not reserved. The issue is resolved by avoid calling ttm_eu_backoff_reservation in the buffer reserve error path. The second issue is that the binding_mutex may be held when user-space fence objects are created and hence during memory reclaims. This may cause recursive attempts to grab the binding mutex. The issue is resolved by not holding the binding mutex across fence creation and submission. Signed-off-by: Thomas Hellstrom Reviewed-by: Sinclair Yeh Signed-off-by: Dave Airlie Signed-off-by: Luis Henriques --- drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c index 6f6fa7d2db47..3a21633eba2f 100644 --- a/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c +++ b/drivers/gpu/drm/vmwgfx/vmwgfx_execbuf.c @@ -2467,7 +2467,7 @@ int vmw_execbuf_process(struct drm_file *file_priv, ret = ttm_eu_reserve_buffers(&ticket, &sw_context->validate_nodes); if (unlikely(ret != 0)) - goto out_err; + goto out_err_nores; ret = vmw_validate_buffers(dev_priv, sw_context); if (unlikely(ret != 0)) @@ -2511,6 +2511,7 @@ int vmw_execbuf_process(struct drm_file *file_priv, vmw_resource_relocations_free(&sw_context->res_relocations); vmw_fifo_commit(dev_priv, command_size); + mutex_unlock(&dev_priv->binding_mutex); vmw_query_bo_switch_commit(dev_priv, sw_context); ret = vmw_execbuf_fence_commands(file_priv, dev_priv, @@ -2526,7 +2527,6 @@ int vmw_execbuf_process(struct drm_file *file_priv, DRM_ERROR("Fence submission error. Syncing.\n"); vmw_resource_list_unreserve(&sw_context->resource_list, false); - mutex_unlock(&dev_priv->binding_mutex); ttm_eu_fence_buffer_objects(&ticket, &sw_context->validate_nodes, (void *) fence);