[perfmon2] Question about cache-misses and r2E41

2014-09-13 Thread Hamid Reza Khaleghzadeh
Hello list, I ran an application on Intel Core 2 Duo twice as following: sudo perf stat -r 5 -e r2E41 -a my/application 3,131,111,625 r2e41 ( +- 0.65% ) sudo perf stat -r 5 -e cache-misses -a my/application 5,783,139 cache

[perfmon2] A problem with Umask values

2014-09-13 Thread Hamid Reza Khaleghzadeh
Hi all, I have a question about Umask value in Core 2 Duo processors. As you know, there is an umask value for each event number. I have considered two events from "Non-Architectural Performance Events in Core Duo Processors": (03H: LD_Blocks, Umask 00H) and (05H: Misalign_Mem_Ref, Umask: 00H). A

[perfmon2] Question about Umask and Core Specificity and MESI Qualification

2014-09-13 Thread Hamid Reza Khaleghzadeh
Hello, There are so many events that they have different Umask value. For example, evetn 2EH has two Umask values (4FH, 41H). Could you tell me how UMask can be used? In addition, a few events such as event 22H in Core Duo processors Requires core-specificity. May I know how these requirements can

Re: [perfmon2] A problem with Umask values

2014-09-13 Thread Stephane Eranian
On Sat, Sep 13, 2014 at 9:02 AM, Hamid Reza Khaleghzadeh wrote: > Hi all, > > I have a question about Umask value in Core 2 Duo processors. As you know, > there is an umask value for each event number. I have considered two events > from "Non-Architectural Performance Events in Core Duo Processors