On Thu, Aug 16, 2012 at 06:06:48PM -0400, Steven Rostedt wrote:
> > /mnt/dbg/tracing/events/mce/mce_record/
> > |-- enable
> > |-- filter
> > |-- format
> > |-- id
> > `-- persistent
> 
> This looks rather hacky :-/

Ya think?! :-)

Well, this whole machinery is only for telling to userspace which is the
file descriptor of the persistent event. And yes, it could probably be
done with a lot less code.

Hell, I'm even fine with issuing it in dmesg and userspace parsing the
output - talk about efficient implementation.

> Can't this be added to the perf syscall, and have the user query the
> event instead?

Do you mean perf_ioctl? For that to work I need to have a file
descriptor of the "normal" event first and then query it to get the
persistent file descriptor... Or do you mean something easier?

> Note, this code will change drastically soon, and this change will
> break. I'll talk more about what will change at plumbers.

I'll try to get a video or something... :)

-- 
Regards/Gruss,
Boris.

Advanced Micro Devices GmbH
Einsteinring 24, 85609 Dornach
GM: Alberto Bozzo
Reg: Dornach, Landkreis Muenchen
HRB Nr. 43632 WEEE Registernr: 129 19551
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Reply via email to