linux-kernel - Re: cfq-iosched preempt issues

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:	Thu, 03 Mar 2011 08:46:16 +0800
From:	Shaohua Li <shaohua.li@...el.com>
To:	Jeff Moyer <jmoyer@...hat.com>
Cc:	"jaxboe@...ionio.com" <jaxboe@...ionio.com>,
	"vgoyal@...hat.com" <vgoyal@...hat.com>,
	"czoccolo@...il.com" <czoccolo@...il.com>,
	"guijianfeng@...fujitsu.com" <guijianfeng@...fujitsu.com>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: cfq-iosched preempt issues

On Thu, 2011-03-03 at 00:17 +0800, Jeff Moyer wrote:
> Shaohua Li <shaohua.li@...el.com> writes:
> 
> > queue preemption is good for some workloads and not for others. With commit
> > f8ae6e3eb825, the impact is amplified. I currently have two issues with it:
> > 1. In a multi-threaded workload, each thread runs a random read/write (for
> > example, mmap write) with iodepth 1. I found the queue depth gets smaller
> > with commit f8ae6e3eb825. The reason is write gets preempted, so more threads
> > are waitting for write, and on the other hand, there are less threads doing
> > read. This will make the queue depth small, so performance drops a little.
> > So in this case, speed up write can speed up read too, but we can't detect
> > it.
> 
> I don't fully understand your workload.  What is the aio-stress or fio
> command line/config file?
ha, sure, this is a fio workload. attached is the fio script. it's
basically a simple mmap rand write workload.

> > 2. cfq_may_dispatch doesn't limit queue depth if the queue is the sole queue.
> > What about if there are two queues, one sync and one async? If the sync queue's
> > think time is small, we can treat it as the sole queue, because the sync queue
> > will preempt async queue, so we don't need care about the async queue's latency.
> > The issue exists before, but f8ae6e3eb825 amplifies it. Below is a patch for it.
> 
> I'm not sure I entirely agree with that reasoning.  Do you have real
> workloads that are regressing due to this commit, or is it just these
> cooked up benchmarks?
this is an aiostress with option: '-O -l -L -s 1200m -i 1 -r 16k'
the random read throughput has impact with preemption.

Thanks,
Shaohua

View attachment "jobfile" of type "text/plain" (12205 bytes)