This is the mail archive of the
mailing list for the GCC project.
Re: Some (small) c++ compilation profiling data (oprofile)
- From: Will Cohen <wcohen at redhat dot com>
- To: Andi Kleen <ak at suse dot de>
- Cc: gcc at gcc dot gnu dot org
- Date: Fri, 17 May 2002 17:32:13 -0400
- Subject: Re: Some (small) c++ compilation profiling data (oprofile)
- Organization: Red Hat, Inc.
- References: <Pine.LNX.firstname.lastname@example.org> <3CE54609.email@example.com> <firstname.lastname@example.org>
The APIC may not be in the mobile Pentium III processor, but somewhere
in the supporting chipset used on the Dell Inspiron 4100 there is an
APIC. The /var/log/messages has entries indicating it is finding APIC
and the profiling I am doing is definitely using the Pentium III
performance monitoring hardware, not the RTC based sampling.
Andi Kleen wrote:
> Will Cohen <email@example.com> writes:
>>I have a Dell Inpiron 4100 with a 1 GHz mobile Pentium III and oprofile
>>is using the mobile Pentium III performance monitoring hardware.
> The mobile pentium III has no local APIC. This means oprofile cannot
> directly receive interrupts when counters overflow and has to do rtc
> based sampling instead.
> This is probably done by forcing your cpu type back.
Will Cohen, GCC Engineer ------ ,__o
Red Hat, 1801 Varsity Dr., Raleigh, NC 27606 ------ _-\_<,
Office (919)-754-3700 x44390 Fax (919)-754-3701------ (*)/'(*)