Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1752670AbdCNNlR (ORCPT ); Tue, 14 Mar 2017 09:41:17 -0400 Received: from mail.fireflyinternet.com ([109.228.58.192]:51115 "EHLO fireflyinternet.com" rhost-flags-OK-OK-OK-FAIL) by vger.kernel.org with ESMTP id S1751474AbdCNNlO (ORCPT ); Tue, 14 Mar 2017 09:41:14 -0400 X-Default-Received-SPF: pass (skip=forwardok (res=PASS)) x-ip-name=78.156.65.138; Date: Tue, 14 Mar 2017 13:41:03 +0000 From: Chris Wilson To: Sergey Senozhatsky Cc: Daniel Vetter , Sergey Senozhatsky , David Airlie , intel-gfx@lists.freedesktop.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org Subject: Re: [Intel-gfx] [linux-mmotm] i915_gem_userptr_get_pages: possible circular locking dependency detected Message-ID: <20170314134103.GI2118@nuc-i3427.alporthouse.com> Mail-Followup-To: Chris Wilson , Sergey Senozhatsky , Daniel Vetter , Sergey Senozhatsky , David Airlie , intel-gfx@lists.freedesktop.org, linux-kernel@vger.kernel.org, dri-devel@lists.freedesktop.org References: <20170314132109.GA388@tigerII.localdomain> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20170314132109.GA388@tigerII.localdomain> User-Agent: Mutt/1.5.21 (2010-09-15) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 1013 Lines: 22 On Tue, Mar 14, 2017 at 10:21:09PM +0900, Sergey Senozhatsky wrote: > Hello, > > [ 530.698622] ====================================================== > [ 530.698623] WARNING: possible circular locking dependency detected > [ 530.698626] 4.11.0-rc2-mm1-dbg-00167-gdb8a9941614c-dirty #222 Not tainted > [ 530.698627] ------------------------------------------------------ > [ 530.698628] Xorg/343 is trying to acquire lock: > [ 530.698630] (&mm->mmap_sem){++++++}, at: [] i915_gem_userptr_get_pages+0x60/0x29c [i915] > [ 530.698702] > but task is already holding lock: > [ 530.698703] (&obj->mm.lock){+.+.+.}, at: [] __i915_gem_object_get_pages+0x21/0x62 [i915] > [ 530.698763] > which lock already depends on the new lock. Yup, just seen it myself. This particular cycle is a non-issue, but there is another possiblity for a mmap_sem recursion in the ggtt fault handler. -Chris -- Chris Wilson, Intel Open Source Technology Centre