Re: [PATCHv v3] power: Include additional information in pm_print_times

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Saturday, June 22, 2013 06:05:50 PM Joe Perches wrote:
> On Sat, 2013-06-22 at 21:52 +0200, Rafael J. Wysocki wrote:
> > On Friday, June 21, 2013 07:27:22 PM Joe Perches wrote:
> > > On Sat, 2013-06-22 at 02:24 +0200, Rafael J. Wysocki wrote:
> > > > Namely, there are tools that use these messages to create suspend/resume time
> > > > charts and they will stop working after the proposed changes.
> > > 
> > > dmesg output isn't guaranteed to be stable.
> > 
> > So?
> 
> So even if new information was only appended to
> the existing line, the script could break.

In this particular case, if it is written sanely, it won't.

> If any script needs something stable it should
> depend on information available through other
> sources like trace or proc or sysfs.

That is clearly impossible in this particular case, though.

> Tools that use dmesg should adapt to whatever gets
> thrown at it and handle the output from whatever
> kernel versions the script supports.
> 
> For instance, what happens to the script when
> console_level is set to 1?

You know, these particular messages are not printed without initcall_debug in
the command line and the people who use them for diagnostics usually generate
them on purpose and *then* feed the log to the script (or tool).

> Requiring that no one can change a dmesg to
> add or improve the content for readability
> or intelligibility I think unreasonable.

In general, you'd be right, but this is not a general case.

Thanks,
Rafael


-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.