On 03/26/2015 04:20 PM, Andy Moe wrote:
### Description:
FIO still seems to be returning zero and writing a file after bogus arguments have been provided. In my particular case, it's returning zero when I try to write data to a file with the 'sg' engine.
### Environment:
CentOS release 6.4 (Final)
Linux iosr80-0.us.cray.com 2.6.32-358.el6.x86_64 #1 SMP Fri Feb 22 00:31:26 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux
### FIO version:
fio-2.2.6-15-g3765
### How to reproduce:
Here is output demonstrating what I'm finding.
[moea@iosr80-0 dev-tests]$ ls -l | grep outfile0
[moea@iosr80-0 dev-tests]$ fio --filename outfile0 --rw=readwrite --ioengine=sg --blocksize=8m --size=128m --name=job0
job0: (g=0): rw=rw, bs=8M-8M/8M-8M/8M-8M, ioengine=sg, iodepth=1
fio-2.2.6-15-g3765
Starting 1 process
job0: Laying out IO file(s) (1 file(s) / 128MB)
ioengine sg only works on block devices
Run status group 0 (all jobs):
Disk stats (read/write):
dm-3: ios=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%, aggrios=0/0, aggrmerge=0/0, aggrticks=0/0, aggrin_queue=0, aggrutil=0.00%
sda: ios=0/0, merge=0/0, ticks=0/0, in_queue=0, util=0.00%
[moea@iosr80-0 dev-tests]$ echo $?
0
[moea@iosr80-0 dev-tests]$ ls -l outfile0
-rw-r--r--. 1 moea moea 134217728 Mar 26 17:06 outfile0
That's because fio will lay out the files first, regardless of IO
engine, before the engine opens them. So the file writing is really just
an artifact of that. It could be fixed, but I really don't see much
point to it.
It should not return 0, though, I checked in a fix for that just now.
--
Jens Axboe
--
To unsubscribe from this list: send the line "unsubscribe fio" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
