On 9/19/18 1:29 PM, Dennis Zhou wrote:
> Hi Jens,
>
> On Wed, Sep 19, 2018 at 12:47:37PM -0600, Jens Axboe wrote:
>> On 9/19/18 12:25 PM, Dennis Zhou wrote:
>>> Running concurrent workloads via multiple jobs can lead to
>>> nondeterministic results as we are at the schedulers mercy. While the
>>> actual performance of the workload may vary regardless, this makes the
>>> order of events consistent.
>>>
>>> This patch introduces two flags: --merge_blktrace_file=<file> and
>>> --merge-blktrace-only. When the first is specified, files that are ':'
>>> separated in --read_iolog are passed to a merge function wich then
>>> produces a single sorted trace file. This file is then passed on in
>>> place of the sorted list or fio exists if --merge-blktrace-only is
>>> specified.
>>>
>>> During merging, events are filtered based on what fio cares about and
>>> the pdu is discarded as well.
>>
>> This looks fine to me, but it's missing documentation (fio.1 and HOWTO)
>> for added options. For the feature as a whole as well.
>>
>>> diff --git a/thread_options.h b/thread_options.h
>>> index 39315834..99552326 100644
>>> --- a/thread_options.h
>>> +++ b/thread_options.h
>>> @@ -258,6 +258,7 @@ struct thread_options {
>>> char *read_iolog_file;
>>> bool read_iolog_chunked;
>>> char *write_iolog_file;
>>> + char *merge_blktrace_file;
>>>
>>> unsigned int write_bw_log;
>>> unsigned int write_lat_log;
>>
>> This needs an accompanying change to struct thread_options_pack, and a
>> cconv.c addition to ensure it's converted properly over the wire for
>> client/server runs of fio.o
>>
>> Basically both comments here apply to the rest of the series as well :-)
>>
>
> Ah that makes sense. I wasn't quite sure what the _pack was for. Now I
> get it. I'll add the corresponding documentation and send out a v2!
> Thanks for looking at it so quickly.
No problem - also forgot to mention that you want to bump FIO_SERVER_VER
as well, if you modify thread_options_pack.
--
Jens Axboe
