From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759649AbXGTLSb (ORCPT ); Fri, 20 Jul 2007 07:18:31 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S1755281AbXGTLSZ (ORCPT ); Fri, 20 Jul 2007 07:18:25 -0400 Received: from andromeda.dapyr.net ([69.45.6.100]:33097 "EHLO andromeda.dapyr.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754872AbXGTLSY (ORCPT ); Fri, 20 Jul 2007 07:18:24 -0400 Date: Fri, 20 Jul 2007 07:18:23 -0400 From: Konrad Rzeszutek To: Andrew Morton Cc: darnok@68k.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] Inhibit NMI watchdog when Alt-SysRq-T operation is underway. Message-ID: <20070720111823.GA2334@andromeda.dapyr.net> References: <20070709140242.GA18098@andromeda.dapyr.net> <20070709155302.GB26570@andromeda.dapyr.net> <20070713164504.66b72e74.akpm@linux-foundation.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20070713164504.66b72e74.akpm@linux-foundation.org> User-Agent: Mutt/1.5.9i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hey Andrew, I tested your patch along with mine and found two things out: 1). Missing this patch (for i386 platform) diff --git a/arch/i386/kernel/traps.c b/arch/i386/kernel/traps.c index 90da057..9f3a7ff 100644 --- a/arch/i386/kernel/traps.c +++ b/arch/i386/kernel/traps.c @@ -207,6 +207,7 @@ static void print_trace_address(void *da { printk("%s [<%08lx>] ", (char *)data, addr); print_symbol("%s\n", addr); + touch_nmi_watchdog(); } static struct stacktrace_ops print_trace_ops = { 2). If I run Alt-SysRq-t about 5000 times in a loop, the slow down with this change is about 5%. Is this a big issue? (This was testing both i686 and x86_64). On Fri, Jul 13, 2007 at 04:45:04PM -0700, Andrew Morton wrote: > On Mon, 9 Jul 2007 11:53:02 -0400 > darnok@68k.org wrote: > > > static void print_trace_address(void *data, unsigned long addr) > > { > > + static int i = 0; > > + if (i && ((i % 8) == 0)) > > + touch_nmi_watchdog(); > > + i++; > > printk_address(addr); > > } > > I doubt if the "% 8" thing is really needed? printk_address() is pretty > slow and touch_nmi_watchdog is _reasonably_ fast. It could be made heaps > faster by: > > From: Andrew Morton > > Avoid dirtying remote cpu's memory if it already has the correct value. > > Cc: Andi Kleen > Cc: Konrad Rzeszutek > Signed-off-by: Andrew Morton > --- > > arch/i386/kernel/nmi.c | 8 +++++--- > x86_64/kernel/nmi.c | 0 > 2 files changed, 5 insertions(+), 3 deletions(-) > > diff -puN arch/i386/kernel/nmi.c~i386-speedup-touch_nmi_watchdog arch/i386/kernel/nmi.c > --- a/arch/i386/kernel/nmi.c~i386-speedup-touch_nmi_watchdog > +++ a/arch/i386/kernel/nmi.c > @@ -298,7 +298,7 @@ static unsigned int > last_irq_sums [NR_CPUS], > alert_counter [NR_CPUS]; > > -void touch_nmi_watchdog (void) > +void touch_nmi_watchdog(void) > { > if (nmi_watchdog > 0) { > unsigned cpu; > @@ -307,8 +307,10 @@ void touch_nmi_watchdog (void) > * Just reset the alert counters, (other CPUs might be > * spinning on locks we hold): > */ > - for_each_present_cpu (cpu) > - alert_counter[cpu] = 0; > + for_each_present_cpu(cpu) { > + if (alert_counter[cpu]) > + alert_counter[cpu] = 0; > + } > } > > /* > > So I'd be inclined to simplify your patch to a bare > > From: Konrad Rzeszutek > > On large memory configuration with not so fast CPUs the NMI watchdog is > triggered when memory addresses are being gathered and printed. The code > paths for Alt-SysRq-t are sprinkled with touch_nmi_watchdog in various > places but not in this routine (or in the loop that utilizes this > function). The patch has been tested for regression on large CPU+memory > configuration (128 logical CPUs + 224 GB) and 1,2,4,16-CPU sockets with > various memory sizes (1,2,4,6,20). > > Cc: Andi Kleen > Signed-off-by: Andrew Morton > --- > > arch/x86_64/kernel/traps.c | 1 + > 1 files changed, 1 insertion(+) > > diff -puN arch/x86_64/kernel/traps.c~inhibit-nmi-watchdog-when-alt-sysrq-t-operation-is-underway arch/x86_64/kernel/traps.c > --- a/arch/x86_64/kernel/traps.c~inhibit-nmi-watchdog-when-alt-sysrq-t-operation-is-underway > +++ a/arch/x86_64/kernel/traps.c > @@ -397,6 +397,7 @@ static int print_trace_stack(void *data, > > static void print_trace_address(void *data, unsigned long addr) > { > + touch_nmi_watchdog(); > printk_address(addr); > } > > _ > > > OK?