From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1754838AbZGNO21 (ORCPT ); Tue, 14 Jul 2009 10:28:27 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1754811AbZGNO20 (ORCPT ); Tue, 14 Jul 2009 10:28:26 -0400 Received: from cam-admin0.cambridge.arm.com ([193.131.176.58]:44072 "EHLO cam-admin0.cambridge.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754792AbZGNO2Z (ORCPT ); Tue, 14 Jul 2009 10:28:25 -0400 Subject: Re: possible memory leak on i915_gem_create_ioctl From: Catalin Marinas To: Alexey Fisher Cc: "kernel-testers-owner@vger.kernel.org" , "linux-kernel@vger.kernel.org" , "intel-gfx@lists.freedesktop.org" In-Reply-To: <4A5C93AD.5050403@fisher-privat.net> References: <4A5C93AD.5050403@fisher-privat.net> Content-Type: text/plain Organization: ARM Ltd Date: Tue, 14 Jul 2009 15:28:20 +0100 Message-Id: <1247581700.28240.106.camel@pc1117.cambridge.arm.com> Mime-Version: 1.0 X-Mailer: Evolution 2.22.3.1 Content-Transfer-Encoding: 7bit X-OriginalArrivalTime: 14 Jul 2009 14:28:21.0759 (UTC) FILETIME=[56DC40F0:01CA048F] Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 2009-07-14 at 16:18 +0200, Alexey Fisher wrote: > kmemleak report a lot of possible memory leaks if using any glxrelated > applications like glxgears, glxinfo, compiz, cairo-dock with g45 graphic. > > Here is a trace from /sys/kernel/debug/kmemleak: > > unreferenced object 0xffff8800b298e000 (size 544): > comm "openarena", pid 5002, jiffies 4296480664 > backtrace: > [] kmemleak_alloc+0x25/0x60 > [] kmem_cache_alloc+0xfb/0x180 > [] idr_pre_get+0x5f/0x90 > [] drm_gem_handle_create+0x2d/0xa0 > [] i915_gem_create_ioctl+0x60/0xc0 > [] drm_ioctl+0x184/0x3b0 > [] vfs_ioctl+0x7c/0xa0 > [] do_vfs_ioctl+0x8a/0x5a0 > [] sys_ioctl+0x81/0xa0 > [] system_call_fastpath+0x16/0x1b > [] 0xffffffffffffffff I already reported these - http://lkml.org/lkml/2009/7/9/110. They look like real leaks but someone who understands the code better would need to look at them. -- Catalin