The Penryn-based CPUs have the same performance registers as older Core 2 chips, but they have a new model number (23 instead of 15). I suspect the same one-line patch from 5.2 is needed to oprofile code so oprofile will treat them the same way. I will test this out on Monday and post the patch.
This request mirrors the request that was accepted for RHEL 5.2.
Per Intel, this patch is necessaery to address a regression thart was introducced with the patch that resolves BX 313631.
Created attachment 302523 [details] backport of patch to 4.7
*** Bug 442634 has been marked as a duplicate of this bug. ***
Committed in 70.EL. RPMS are available at http://people.redhat.com/vgoyal/
The patch includes in 70.EL, but on bensley platform(Harpertown), oprofiled fail to start collect information. [root@Bensley-RHEL47 initial]# opcontrol --start No events given. Using default event: No events given.
An advisory has been issued which should help the problem described in this bug report. This report is therefore being closed with a resolution of ERRATA. For more information on therefore solution and/or where to find the updated files, please follow the link below. You may reopen this bug report if the solution does not work for you. http://rhn.redhat.com/errata/RHSA-2008-0665.html
Adding CPU_CLK_UNHALTED:1000000:0:1:1 parameter in /root/.oprofile/daemonrc, we also can collect the performance data.