From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jens Axboe Date: Fri, 11 Jan 2008 09:56:54 +0000 Subject: Re: CONFIG_NO_HZ breaks blktrace timestamps Message-Id: <20080111095654.GR6258@kernel.dk> List-Id: References: <1199918912.8388.13.camel@lap75545.ornl.gov> <1199996752.9159.46.camel@lap75545.ornl.gov> <20080110234438.4826f658@inria.fr> <1200020661.5099.3.camel@obelisk.thedillows.org> <20080111090723.GI6258@kernel.dk> <20080111092334.GA8143@elte.hu> <20080111092513.GO6258@kernel.dk> <20080111094205.GE8143@elte.hu> In-Reply-To: <20080111094205.GE8143@elte.hu> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Ingo Molnar Cc: David Dillow , Guillaume Chazarain , linux-kernel@vger.kernel.org, linux-btrace@vger.kernel.org, mingo@redhat.com, tglx@linutronix.de On Fri, Jan 11 2008, Ingo Molnar wrote: > > * Jens Axboe wrote: > > > > they are from the scheduler git tree (except the first debug patch), > > > but queued up for v2.6.25 at the moment. > > > > So this means that blktrace will be broken with CONFIG_NO_HZ for > > 2.6.24? That's clearly a regression. > > 64-bit CONFIG_NO_HZ is a new feature in v2.6.24. If it happens on 32-bit > too and it didnt happen in v2.6.23 32-bit then it's a regression. If blktrace worked in 2.6.23 and it doesn't in 2.6.24 because of some option that isn't immediately apparent, then it's a regression. Period. > all this comes from blktrace's original decision of using sched_clock() > :-) It's not a global timesource and it's not trivial to turn it into a > halfways usable global timesource. Hey, it was a high res time source and the only one easily available :) I'm fine with using another timesource, I'll take suggestions or patches any day! -- Jens Axboe