RE: fio 3.2

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



It's there any chance to fix it ?

dl560g10spmem01:/var/work # /usr/local/bin/fio --filename=/dev/dax0.0 --rw=randrw --refill_buffers --randrepeat=0 --ioengine=mmap --bssplit=4k/4:8k/7:16k/7:32k/15:64k/65:128k/1:256k/1 --rwmixread=5 --iodepth=1 --numjobs=16 --runtime=1800 --group_reporting --name=4-rand-rw-3xx --size=290g
4-rand-rw-3xx: (g=0): rw=randrw, bs=(R) 4096B-256KiB, (W) 4096B-256KiB, (T) 4096B-256KiB, ioengine=mmap, iodepth=1
...
fio-2.99
Starting 16 processes
4-rand-rw-3xx: failed to get file size of /dev/dax0.0

4-rand-rw-3xx: failed to get file size of /dev/dax0.0

4-rand-rw-3xx: failed to get file size of /dev/dax0.0

4-rand-rw-3xx: failed to get file size of /dev/dax0.0

4-rand-rw-3xx: failed to get file size of /dev/dax0.0

4-rand-rw-3xx: failed to get file size of /dev/dax0.0

4-rand-rw-3xx: failed to get file size of /dev/dax0.0

-----Original Message-----
From: Jens Axboe [mailto:axboe@xxxxxxxxx] 
Sent: Thursday, November 30, 2017 6:22 AM
To: Elliott, Robert (Persistent Memory) <elliott@xxxxxxx>; Gavriliuk, Anton (HPS Ukraine) <anton.gavriliuk@xxxxxxx>; Rebecca Cran <rebecca@xxxxxxxxxxxx>; Sitsofe Wheeler <sitsofe@xxxxxxxxx>
Cc: fio@xxxxxxxxxxxxxxx; Kani, Toshimitsu <toshi.kani@xxxxxxx>
Subject: Re: fio 3.2

On 11/28/2017 09:13 PM, Elliott, Robert (Persistent Memory) wrote:
> small_content_scramble has hardly been touched since 2011, so it 
> probably hasn't had much performance analysis.

That's fair, would be a good thing to look at, especially since it's on by default.

> One of the few changes made was to add an integer divide by 1000, 
> which is always slow (painfully slow on some CPUs):
> 
> 	offset = ((io_u->start_time.tv_nsec/1000) ^ boffset) & 511;
> 
> perf top doesn't show that taking time - I think the compiler realized 
> it could pull that calculation out of the loop and just do it once.  
> Different compilers and compiler options might not realize that.

Yeah, int divides suck. This didn't used to be like this, it got added with the nsec conversion. I have fixed this up now.

--
Jens Axboe

��.n��������+%������w��{.n�������^n�r������&��z�ޗ�zf���h���~����������_��+v���)ߣ�

[Index of Archives]     [Linux Kernel]     [Linux SCSI]     [Linux IDE]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux SCSI]

  Powered by Linux