Re: data on devel code perf dip

From: markw(at)osdl(dot)org
To: maryedie(at)osdl(dot)org
Cc: "Mark Wong" <markw(at)osdl(dot)org>, "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us>, "Bruce Momjian" <pgman(at)candle(dot)pha(dot)pa(dot)us>, "Alvaro Herrera" <alvherre(at)alvh(dot)no-ip(dot)org>, "pgsql-hackers" <pgsql-hackers(at)postgresql(dot)org>, "Josh Berkus" <josh(at)agliodbs(dot)com>
Subject: Re: data on devel code perf dip
Date: 2005-08-16 23:57:58
Message-ID: 51627.63.224.228.211.1124236678.squirrel@www.osdl.org
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers pgsql-patches

> On Tue, 2005-08-16 at 16:07 -0700, Mark Wong wrote:
>> On Tue, 16 Aug 2005 18:53:55 -0400
>> Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
>>
>> > Mary Edie Meredith <maryedie(at)osdl(dot)org> writes:
>> > > I'm still very concerned about what I'm seeing in the oprofile:
>> > > namely: .CreateLWLocks is the second highest entry for postgres.
>> > > http://developer.osdl.org/maryedie/DBT2_PGSQL/59/oprofile.txt
>> >
>> > This says there's something wrong with your oprofile setup.
>> > CreateLWLocks is run *once* at postmaster startup, and it doesn't
>> > take all that long. Check to see if maybe your executables are
>> > out of sync with what oprofile is looking at?
>>
>> It is a POWER5 platform, if that has anything to do with it. It
>> certainly doesn't look sane, but for user apps oprofile is supposed to
>> be getting the symbols directly from the binaries its using. I'm
>> inclined to blame the platform support. ;)
>
> Could it have to do with the fact that I'm running in a Virtual Machine?
> Maybe they forgot to change the symbols to reflect the VM's "virtual
> physical address" if you get my drift. Who do you know on the IBM side,
> Mark, who could help us sort this out?

Not likely, that would imply the system knew it was running in a virtual
machine. You're not supposed to know you're in a virtual machine. I
don't know who at IBM to ask about the hypervisor. As for oprofile, the
oprofile list would be a start.

Mark

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Bruce Momjian 2005-08-17 00:09:17 Re: Missing CONCURRENT VACUUM (Was: Release notes for 8.1)
Previous Message Mary Edie Meredith 2005-08-16 23:41:11 Re: data on devel code perf dip

Browse pgsql-patches by date

  From Date Subject
Next Message Jim C. Nasby 2005-08-17 14:51:49 Re: pthread stack on FreeBSD WAS: HEAD doesn't cope with libraries in non-default
Previous Message Mary Edie Meredith 2005-08-16 23:41:11 Re: data on devel code perf dip