Re: [PATCH] Add support for IBM Power event codes longer than sizeof int

[email protected] wrote on 11/21/2014 09:00:00 AM:

> From: [email protected]
> To: Henry May/Rochester/IBM@IBMUS, [email protected]
> Cc: oprofile-list <[email protected]>, William 
> Cohen <[email protected]>
> Date: 11/21/2014 09:00 AM
> Subject: Re: [PATCH] Add support for IBM Power event codes longer 
> than sizeof int
> 
> On 11/20/2014 06:37 PM, Henry May wrote:
> > 
> > 
> > Henry May
> > IBM InfoSphere Streams Performance
> > [email protected]
> > 720-342-8873
> > Tie: 963-8873
> > 
> > 
> > [email protected] wrote on 11/19/2014 04:51:24 PM:
> > 
> >> From: [email protected]
> >> To: oprofile-list <[email protected]>, Henry May/
> >> Rochester/IBM@IBMUS, William Cohen <[email protected]>
> >> Date: 11/19/2014 04:51 PM
> >> Subject: [PATCH] Add support for IBM Power event codes longer 
> than sizeof int
> >>
> >> Hi, Henry,
> >> The patch below addresses the issue you raised yesterday on the 
oprofile
> >> list (subject "PM_L1MISS_LAT_EXC_xxx events").  At first, I looked 
into
> >> what it would take to add general support for long event codes, but I 
soon
> >> discovered that it would entail a change to the sample file format, 
and
> >> I was loathe to do that just for eight events from IBM POWER8.  I 
verified
> >> that there are no other events for any architecture other than these 
eight
> >> that have event codes longer than an integer. So I implemented a bit 
of a
> >> hack, using the concept of a unit mask as an event qualifier. It's an 
OK
> >> fit, conceptually, it's just a bit hacky because no other ppc64 
events
> >> use unit masks.  But the fix was fairly small and simple, so I think 
it's
> >> fine.
> >>
> >> Please give it a test and let me know how it works for you.
> > I tested the patch and it only works with PM_L1MISS_LAT_EXC_256. 
> Does not work with 32, 1024, or 2048.  Currently trying another 
> iteration of the patch.
> Henry,
> My apologies. The testing I did for this patch used a full event 
> specification -- e.g, PM_L1MISS_LAT_EXC_256:0x0:1:1 -- and the patch
> worked for all events. But I get the same results you do if I use 
> ocount or operf with just the event name. I've found the bug in the 
> patch and as you alluded to above, you're already testing it. So 
> when you have your test results, please post here.
ocount accepts all counter specifications now, but the PM_L1MISS_EXC_LAT 
counts look bogus:

  latency | PM_L1MISS_EXC_LAT  | PM_RC_LIFETIME_EXC
       32 |        25,000,000  |          6,000,000
      256 |        30,000,000  |            660,000 
     1024 |        64,000,000  |             35,000
     2048 |        10,000,000  |              1,000

I would have expected the pattern exhibited by RC_LIFETIME_EXC, but it 
could be that I am misinterpreting the meaning of the counters.  Are the 
L1MISS_EXC_LAT counters counting the number of events that exceed the 
threshold, or the number of cycles accumulated by the events that exceed 
the threshold?
> 
> -Maynard
> >>
> >> Thanks!
> >> -Maynard
> >>
> >> 
------------------------------------------------------------------------
> >
> [snip]
>

------------------------------------------------------------------------------
Download BIRT iHub F-Type - The Free Enterprise-Grade BIRT Server
from Actuate! Instantly Supercharge Your Business Reports and Dashboards
with Interactivity, Sharing, Native Excel Exports, App Integration & more
Get technology previously reserved for billion-dollar corporations, FREE
http://pubads.g.doubleclick.net/gampad/clk?id=157005751&iu=/4140/ostg.clktrk

_______________________________________________
oprofile-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/oprofile-list