From mboxrd@z Thu Jan  1 00:00:00 1970
Return-Path: <axboe@kernel.dk>
Subject: Re: assert failing with latency log
References: <897324962.65623516.1464126010451.JavaMail.zimbra@redhat.com>
From: Jens Axboe <axboe@kernel.dk>
Message-ID: <5744F4C6.5060700@kernel.dk>
Date: Tue, 24 May 2016 18:41:42 -0600
MIME-Version: 1.0
In-Reply-To: <897324962.65623516.1464126010451.JavaMail.zimbra@redhat.com>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
To: Ben England <bengland@redhat.com>, fio@vger.kernel.org
Cc: Tim Wilkinson <twilkins@redhat.com>
List-ID: <fio@vger.kernel.org>

On 05/24/2016 03:40 PM, Ben England wrote:
> If I run this fio jobfile with absolute latest upstream code:
>
> [global]
> directory=/var/tmp
>
> [randreads]
> direct=1
> rate_iops=1000
> size=512m
> ioengine=sync
> bs=4k
> rw=randread
> filename=ben.dd
> runtime=30
> write_lat_log=foo
>
> It fails with
>
> fio: stat.c:1988: get_cur_log: Assertion `iolog->pending->nr_samples < iolog->pending->max_samples' failed.
>
> But if I take out the "write_lat_log=foo" line, it succeeds.  The assertion seems to be in the code path for latency logs, right?  Anyone else seen this?

Gah, that's my bad, I broke sync IO engines with logging. I pushed a fix:

http://git.kernel.dk/cgit/fio/commit/?id=cf2c8d5260c9676964854e79382cba3eb498fd2e

I'll tag a new version as well, this is pretty unfortunate.

-- 
Jens Axboe