linux-kernel - Re: Re: [PATCH 2/2] trace-cmd: Support trace

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:	Thu, 11 Apr 2013 19:06:12 +0900
From:	Yoshihiro YUNOMAE <yoshihiro.yunomae.ez@...achi.com>
To:	Steven Rostedt <rostedt@...dmis.org>
Cc:	Hidehiro Kawai <hidehiro.kawai.ez@...achi.com>,
	Masami Hiramatsu <masami.hiramatsu.pt@...achi.com>,
	linux-kernel@...r.kernel.org, yrl.pp-manager.tt@...achi.com
Subject: Re: Re: [PATCH 2/2] trace-cmd: Support trace_clock extraction

Hi Steven,

Thank you for reviewing my patch set.

(2013/04/10 1:00), Steven Rostedt wrote:
> On Fri, 2013-04-05 at 10:54 +0900, Yoshihiro YUNOMAE wrote:
>> In this patch, trace-cmd reads trace_clock on debugfs in the report/extract
>> modes and outputs the data to trace.dat file. Then, in the report mode,
>> trace-cmd reads trace_clock data from the file and switches outputting format
>> of timestamp for each trace_clock.
>>
>> Note that by applying this patch, the binary format of trace.data is changed
>> as follows:
>>
>> <Current binary format>
>> ...
>> [size of saved_cmdlines]
>> [saved_cmdlines contents]
>> [total cpu number]
>> ...
>>
>> <Changed binary format>
>> ...
>> [size of saved_cmdlines]
>> [saved_cmdlines contents]
>> [size of trace_clock]  <== add
>> [trace_clock contents] <== add
>> [total cpu number]
>> ...
>
> Hi Yoshihiro,
>
> I don't mind the feature addition, but I don't like the implementation.
> This will break backward compatibility with the trace.dat file, which
> I've been working hard not to have happen.

Yeah, I agree with you.
It is important that trace-cmd keeps backward compatibility.

> That doesn't mean that you can't implement this feature. This is what
> the options section is for. You can add an option that includes the
> trace_clock contents, or just simply say what type of clock is being
> used. Then a new trace-cmd can show the output like you want it to, and
> the old version will still work but show the old way, as trace-cmd is
> made to simply ignore options it does not know about.

Sounds good.
Maintenance of the implementation will be easier than that of my
implementation.

> I think it's time for me to push my latest updates to trace-cmd. As this
> will probably end up being the 3.0 version. I have examples there that
> use the options feature for more extensions that you can look at.

Thank you for pushing your latest patches!
I'll use your patches as a reference and submit the revised patches
using the options feature.

Thanks,
Yoshihiro YUNOMAE

-- 
Yoshihiro YUNOMAE
Software Platform Research Dept. Linux Technology Center
Hitachi, Ltd., Yokohama Research Laboratory
E-mail: yoshihiro.yunomae.ez@...achi.com


--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/