From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S933102Ab1JDRAG (ORCPT ); Tue, 4 Oct 2011 13:00:06 -0400 Received: from mail-pz0-f42.google.com ([209.85.210.42]:49433 "EHLO mail-pz0-f42.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S932791Ab1JDRAD (ORCPT ); Tue, 4 Oct 2011 13:00:03 -0400 Date: Tue, 4 Oct 2011 09:59:58 -0700 From: Tejun Heo To: Catalin Marinas Cc: "linux-kernel@vger.kernel.org" , Huajun Li , Christoph Lameter Subject: Re: [PATCH 2/4] kmemleak: Handle percpu memory allocation Message-ID: <20111004165958.GA18371@dhcp-172-17-108-109.mtv.corp.google.com> References: <20110929105940.10660.76130.stgit@e102109-lin.cambridge.arm.com> <20110929110228.10660.87600.stgit@e102109-lin.cambridge.arm.com> <20110929192818.GB10425@mtj.dyndns.org> <20111003152126.GM18195@e102109-lin.cambridge.arm.com> <20111004075915.GE15637@dhcp-172-17-108-109.mtv.corp.google.com> <20111004090446.GA10071@e102109-lin.cambridge.arm.com> <20111004091314.GF15637@dhcp-172-17-108-109.mtv.corp.google.com> <20111004092624.GB10071@e102109-lin.cambridge.arm.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20111004092624.GB10071@e102109-lin.cambridge.arm.com> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Hello, Catalin. On Tue, Oct 04, 2011 at 10:26:24AM +0100, Catalin Marinas wrote: > Before kmemleak is initialised we still get memory allocations that > kmemleak stores in an early_log buffer (via the log_early() function > called from kmemleak_alloc_percpu). Later when kmemleak has all the data > structures in place, the kmemleak_init() function goes through the > early_log array and replays the previously recorded requests. The > early_alloc_percpu() function is used during early_log replaying and it > indeed registers every percpu memory block but the early_log is always > O(#PCPU_ALLOCS). > > The reason we don't call kmemleak_alloc_percpu() directly during > replaying is that early_alloc() also copies the previously recorded > stack trace into the newly created object (otherwise all early > allocations would be shown as done by kmemleak_init). Ah, okay, I was misreading the patch. For some reason, log_early() was nested inside for_each_possible_cpu(), but one other thing. kmemleak_free_percpu() is calling log_early() w/ KMEMLEAK_FREE. Shouldn't it be KMEMLEAK_FREE_PERCPU? Thanks. -- tejun