Re: [Bug report] Runtime, IOPS, bandwidth recorded incorrectly if small size with time_based

[Jens Axboe <axboe@kernel.dk>]

On 05/04/2015 05:17 PM, Brian Fulton wrote:
> Hi Jens,
>
> Akash and I both discovered this bug and work together.  He is
> currently on holiday for the next two weeks or so.  I am currently
> using the same tests that originally discovered the bug to verify it
> is fixed.

Ah perfect, I didn't realize that!

> Out of curiosity, why is the comma required after the zero in order to
> initialize all elements to zero?  I used this initializer instead of
> memset just as it was used here:
> https://github.com/axboe/fio/blob/master/cgroup.c#L26.  Are there
> certain compilers that require the comma?  I also checked the C99 spec
> http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf, Section
> 6.7.8, page 125, sections 1 and 21.  I just want to make sure I'm not
> loosing my mind.

Actually I might be mistaken, it's just an array. For structs you need 
the comma to ensure it zero fills the rest of the members, for arrays I 
believe your approach was actually fine.

-- 
Jens Axboe