[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH 06/14] xen: sched: tracing: enable TSC tracing for all events



On Mon, 2016-02-15 at 11:32 -0500, Konrad Rzeszutek Wilk wrote:
> On Fri, Feb 05, 2016 at 07:34:18PM +0100, Dario Faggioli wrote:
> > it is enabled for pretty much all of them already.
> > There were just a few that had it disabled.
> > 
> > When tracing a scheduler, timing information is
> > really important, so enable it everywhere scheduling
> > related.
> > 
> > Note that this was not really a problem if looking
> > at the traces with xenalyze, but it was if using
> > xentrace_format.
> 
> Was there any historical reason it was not used?
> 
I don't really know.

Funnily enough, for Credit1, it looks like it was me that did it
inÂcbcbe82f "xen: sched_credit: add some tracing". At the time, Credit2
had some tracing already, and the TRC_CSCHED2_TICKLE was not logging
the TSC.

The only reason I could think of is that, when doing TRC_CSCHED_TICKLE,
I "took inspiration" from what I found already in the code base (yeah,
I know, bad me! :-/).

As per why Credit2 does that in the first place, this is when it
started:Â1b561702 "credit2: Fix runq_tickle to use idle, tickled
masks", from George.

In that commit,ÂTRC_CSCHED2_TICKLE_CHECK is moved and, while doing so,
it looses the "TSC logging" property, while TRC_CSCHED2_TICKLE is
introduced from scratch, and that's directly with log_tsc=0.

I can't tell why that is the case, and I think we should change that.

As said in the commit message, xenalyze does not have much problems
with all this, so it's certainly not a super big deal (and it may even
be the reason why this went unnoticed until now, I guess).

However, if looking at the output of xentrace_format (which is a lot
less smart than xenalyze!) it indeed is quite useful to see the
timestamp (not to mention that, without it, the output itself is rather
disturbing to look at and post-process, for severe lack of alignment).

Maybe George knows better, but in any case, I think we should turn this
on. :-)

Thanks and Regards,
Dario
-- 
<<This happens because I choose it to happen!>> (Raistlin Majere)
-----------------------------------------------------------------
Dario Faggioli, Ph.D, http://about.me/dario.faggioli
Senior Software Engineer, Citrix Systems R&D Ltd., Cambridge (UK)

Attachment: signature.asc
Description: This is a digitally signed message part

_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxx
http://lists.xen.org/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.