(no subject)

[Minwoo Im]

[-- Attachment #1: Type: text/plain, Size: 14 bytes --]

subscribe fio

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

Re:

[Vincent Fu]

On 5/20/24 06:09, Minwoo Im wrote:
> subscribe fio
> 
> 

Minwoo, here are instructions for subscribing to this list:

http://vger.kernel.org/vger-lists.html#fio

Vincent

Re:

[Minwoo Im]

[-- Attachment #1: Type: text/plain, Size: 307 bytes --]

On 24-05-20 09:34:15, Vincent Fu wrote:
> On 5/20/24 06:09, Minwoo Im wrote:
> > subscribe fio
> > 
> > 
> 
> Minwoo, here are instructions for subscribing to this list:

Ah, my bad.  I should have sent to majordomo@vger.kernel.org.

Thanks!

> 
> http://vger.kernel.org/vger-lists.html#fio
> 
> Vincent
> 

[-- Attachment #2: Type: text/plain, Size: 0 bytes --]

(no subject)

[Nick Neumann]

I've filed the issue on github, but just thought I'd mention here too.
In real-world use it appears to be intermittent. I"m not yet sure how
intermittent, but I could see it being used in production and not
caught right away. I got lucky and stumbled on it when looking at
graphs of runs and noticed 15 seconds of no activity.

https://github.com/axboe/fio/issues/1457

With the null ioengine, I can make it reproduce very reliably, which
is encouraging as I move to debug.

I had just moved to using log compression as it is really powerful,
and the only way to store per I/O logs for a long run without pushing
up against the amount of physical memory in a system.

(Without compression, a GB of sequential writes at 128K block size is
on the order of 245KB of memory per log, so a TB is 245MB per log. Now
run a job to fill a 20TB drive and you're at 4.9GB for one log file.
If you record all 3 latency numbers too, you're talking close to
20GB.)

Re:

[Nick Neumann]

PR for this is up now.

On Sun, Aug 28, 2022 at 4:01 PM Nick Neumann <nick@pcpartpicker.com> wrote:
>
> I've filed the issue on github, but just thought I'd mention here too.
> In real-world use it appears to be intermittent. I"m not yet sure how
> intermittent, but I could see it being used in production and not
> caught right away. I got lucky and stumbled on it when looking at
> graphs of runs and noticed 15 seconds of no activity.
>
> https://github.com/axboe/fio/issues/1457
>
> With the null ioengine, I can make it reproduce very reliably, which
> is encouraging as I move to debug.
>
> I had just moved to using log compression as it is really powerful,
> and the only way to store per I/O logs for a long run without pushing
> up against the amount of physical memory in a system.
>
> (Without compression, a GB of sequential writes at 128K block size is
> on the order of 245KB of memory per log, so a TB is 245MB per log. Now
> run a job to fill a 20TB drive and you're at 4.9GB for one log file.
> If you record all 3 latency numbers too, you're talking close to
> 20GB.)

Re:

[Beierl, Mark]

On 2018-10-27, 12:55, "fio-owner@vger.kernel.org on behalf of Jens Axboe" <fio-owner@vger.kernel.org on behalf of axboe@kernel.dk> wrote:
    
    [EXTERNAL EMAIL] 
    Please report any suspicious attachments, links, or requests for sensitive information.
    
    
    On 10/26/18 6:54 AM, Mohanraj B wrote:
    > Hello,
    > 
    > I am trying to check how option --clocksource works.
    > 
    > 
    > bash# fio --name job1 --size 10m --clocksource 2
    >         valid values: gettimeofday Use gettimeofday(2) for timing
    >                     : clock_gettime Use clock_gettime(2) for timing
    >                     : cpu        Use CPU private clock
    > 
    > fio: failed parsing clocksource=2
    > 
    > bash# fio --name job1 --size 10m --clocksource gettimeofday(2)
    > bash: syntax error near unexpected token `('
    > 
    > Below command works fine.
    > bash# fio --name job1 --size 10m --clocksource gettimeofday
    > 
    > It runs without error but quiet not sure how to see the effect of this
    > option. also tried other options - clock_gettime, cpu gettimeofday and
    > dont see any difference.
    > 
    > Also is there any error in documentation passing gettimeofday(2)
    > throws parse error.
    
    The format is 'value' 'help', so you'd want to do:
    
    --clocksource=gettimeofday
    
    for instance.
    
    -- 
    Jens Axboe

Hello, Mohanraj

The help output that you see above states that using --clocksource=gettimeofday will use the gettimeofday function as defined in the man page in the section (2), which is where all the system calls manuals are stored.  The (2) is  not meant to be part of the command line, it is part of the description of the help text, which tells you where to find more information on what is being used to implement the clocksource.

Hope that helps clarify the help text.

Regards,
Mark

Re:

[Mohanraj B]

[-- Attachment #1: Type: text/plain, Size: 2053 bytes --]

Thanks Axboe and Mark.

On Mon 29 Oct, 2018, 7:50 PM Beierl, Mark, <Mark.Beierl@dell.com> wrote:

> On 2018-10-27, 12:55, "fio-owner@vger.kernel.org on behalf of Jens Axboe"
> <fio-owner@vger.kernel.org on behalf of axboe@kernel.dk> wrote:
>
>     [EXTERNAL EMAIL]
>     Please report any suspicious attachments, links, or requests for
> sensitive information.
>
>
>     On 10/26/18 6:54 AM, Mohanraj B wrote:
>     > Hello,
>     >
>     > I am trying to check how option --clocksource works.
>     >
>     >
>     > bash# fio --name job1 --size 10m --clocksource 2
>     >         valid values: gettimeofday Use gettimeofday(2) for timing
>     >                     : clock_gettime Use clock_gettime(2) for timing
>     >                     : cpu        Use CPU private clock
>     >
>     > fio: failed parsing clocksource=2
>     >
>     > bash# fio --name job1 --size 10m --clocksource gettimeofday(2)
>     > bash: syntax error near unexpected token `('
>     >
>     > Below command works fine.
>     > bash# fio --name job1 --size 10m --clocksource gettimeofday
>     >
>     > It runs without error but quiet not sure how to see the effect of
> this
>     > option. also tried other options - clock_gettime, cpu gettimeofday
> and
>     > dont see any difference.
>     >
>     > Also is there any error in documentation passing gettimeofday(2)
>     > throws parse error.
>
>     The format is 'value' 'help', so you'd want to do:
>
>     --clocksource=gettimeofday
>
>     for instance.
>
>     --
>     Jens Axboe
>
> Hello, Mohanraj
>
> The help output that you see above states that using
> --clocksource=gettimeofday will use the gettimeofday function as defined in
> the man page in the section (2), which is where all the system calls
> manuals are stored.  The (2) is  not meant to be part of the command line,
> it is part of the description of the help text, which tells you where to
> find more information on what is being used to implement the clocksource.
>
> Hope that helps clarify the help text.
>
> Regards,
> Mark
>
>
>
>

[-- Attachment #2: Type: text/html, Size: 2895 bytes --]

(no subject)

[Mohanraj B]

Hello,

I am trying to check how option --clocksource works.


bash# fio --name job1 --size 10m --clocksource 2
        valid values: gettimeofday Use gettimeofday(2) for timing
                    : clock_gettime Use clock_gettime(2) for timing
                    : cpu        Use CPU private clock

fio: failed parsing clocksource=2

bash# fio --name job1 --size 10m --clocksource gettimeofday(2)
bash: syntax error near unexpected token `('

Below command works fine.
bash# fio --name job1 --size 10m --clocksource gettimeofday

It runs without error but quiet not sure how to see the effect of this
option. also tried other options - clock_gettime, cpu gettimeofday and
dont see any difference.

Also is there any error in documentation passing gettimeofday(2)
throws parse error.

Thanks and Regards,
Mohan

Re:

[Jens Axboe]

On 10/26/18 6:54 AM, Mohanraj B wrote:
> Hello,
> 
> I am trying to check how option --clocksource works.
> 
> 
> bash# fio --name job1 --size 10m --clocksource 2
>         valid values: gettimeofday Use gettimeofday(2) for timing
>                     : clock_gettime Use clock_gettime(2) for timing
>                     : cpu        Use CPU private clock
> 
> fio: failed parsing clocksource=2
> 
> bash# fio --name job1 --size 10m --clocksource gettimeofday(2)
> bash: syntax error near unexpected token `('
> 
> Below command works fine.
> bash# fio --name job1 --size 10m --clocksource gettimeofday
> 
> It runs without error but quiet not sure how to see the effect of this
> option. also tried other options - clock_gettime, cpu gettimeofday and
> dont see any difference.
> 
> Also is there any error in documentation passing gettimeofday(2)
> throws parse error.

The format is 'value' 'help', so you'd want to do:

--clocksource=gettimeofday

for instance.

-- 
Jens Axboe