From: Ingo Molnar Subject: Re: [patch 003/152] jbd: fix commit of ordered data buffers Date: Fri, 29 Sep 2006 23:22:27 +0200 Message-ID: <20060929212227.GA14030@elte.hu> References: <200609260630.k8Q6UrvQ011999@shell0.pdx.osdl.net> <20060929122026.62ec29eb.akpm@osdl.org> <20060929191759.GA19304@elte.hu> <200609292154.30234.ak@suse.de> <20060929132057.3039bac8.akpm@osdl.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Andi Kleen , Badari Pulavarty , Jan Kara , torvalds@osdl.org, stable@kernel.org, ext4 Return-path: Received: from mx2.mail.elte.hu ([157.181.151.9]:38294 "EHLO mx2.mail.elte.hu") by vger.kernel.org with ESMTP id S1161419AbWI2Vax (ORCPT ); Fri, 29 Sep 2006 17:30:53 -0400 To: Andrew Morton Content-Disposition: inline In-Reply-To: <20060929132057.3039bac8.akpm@osdl.org> Sender: linux-ext4-owner@vger.kernel.org List-Id: linux-ext4.vger.kernel.org * Andrew Morton wrote: > + if (cpu_isset(cpu, backtrace_mask)) { > + cpu_clear(cpu, backtrace_mask); > + printk("NMI backtrace for cpu %d\n", cpu); > + dump_stack(); > + } yeah. There's just one issue: this will mix all the printks from all CPUs at once (the NMIs if IO-APIC triggered fire at once). The way i solved this was to add a private spinlock around the printk and dump_stack(). Ingo