pcp
[Top] [All Lists]

Re: papi qa issue

To: Ken McDonell <kenj@xxxxxxxxxxxxxxxx>
Subject: Re: papi qa issue
From: "Frank Ch. Eigler" <fche@xxxxxxxxxx>
Date: Fri, 6 Mar 2015 15:26:31 -0500
Cc: "'PCP'" <pcp@xxxxxxxxxxx>
Delivered-to: pcp@xxxxxxxxxxx
In-reply-to: <00a901d0584a$1a1ac2d0$4e504870$@internode.on.net>
References: <007d01d057ed$6de34160$49a9c420$@internode.on.net> <y0msidi5k2y.fsf@xxxxxxxx> <00a901d0584a$1a1ac2d0$4e504870$@internode.on.net>
User-agent: Mutt/1.4.2.2i
Hi, Ken -

> > Could you run "papi_avail | head -22" and "papi_avail | grep TOT_INS"?
> 
> Output below.

Thanks.

> Understood.  813 is notrun on all my other VMs, but does run on the real
> hardware platforms (and passes).  So there is something different about vm14
> and the notrun guard that is allowing the test to be run on vm14.

> kenj@vm14:~$ papi_avail | head -22
> PAPI Version             : 5.1.1.0
> Vendor string and code   : AuthenticAMD (2)
> Model string and code    : AMD Opteron 23xx (Gen 3 Class Opteron) (6)

It's a bit tricky to read PAPI history, to see at what point support
for any particular cpu family & model were added.  I can't quite pick
out whether your modernish Opteron is or isn't supported by papi per
se, and whether your kernel lets it work (both may be necessary).

The clincher is:

> kenj@vm14:~$ papi_avail | grep TOT_INS
> PAPI_TOT_INS 0x80000032  No    No   Instructions completed

Methinks 813 should _notrun if that third column is "No".  (Arguably,
the papi pmda should have filtered out that from the listed metrics in
the dynamid pmns too.)

- FChE

<Prev in Thread] Current Thread [Next in Thread>