[Ptools-perfapi] Count L3 cache misses on an AMD Opteron 6168

stephane eranian eranian at googlemail.com
Fri Sep 17 11:37:54 EDT 2010


On Fri, Sep 17, 2010 at 5:21 PM, Vince Weaver <vweaver1 at eecs.utk.edu> wrote:
> On Fri, 17 Sep 2010, Juri Lelli wrote:
>
>> I have no problems to read L1 and L2 cache misses, but I'd like to read L3 cache misses too. papi_avail
>> | grep L3 returns:
>
> L3 is tricky to measure, as the L3 is a shared resource used by multiple
> cores.  That's why it doesn't have a PAPI predefined event available.
>
> You can still measure things though, but you'll have to use a native
> event.
>
> from util/papi_native_avail on an Istanbul chip
>
> 0x40000079   L3_CACHE_MISSES  | L3 Cache Misses                                |
>  40000479   :READ_BLOCK_EXCLUSIVE  | Read Block Exclusive (Data cache read)   |
>  40000879   :READ_BLOCK_SHARED  | Read Block Shared (Instruction cache read)  |
>  40001079   :READ_BLOCK_MODIFY  | Read Block Modify                           |
>  40002079   :CORE_0_SELECT  | Core 0 Select                                   |
>  40004079   :CORE_1_SELECT  | Core 1 Select                                   |
>  40008079   :CORE_2_SELECT  | Core 2 Select                                   |
>  40010079   :CORE_3_SELECT  | Core 3 Select                                   |
>  40020079   :CORE_4_SELECT  | Core 4 Select                                   |
>  40040079   :CORE_5_SELECT  | Core 5 Select                                   |
>  40080079   :ANY_CORE  | Any core                                             |
>  40100079   :ALL  | All sub-events selected                                   |
>
Careful when using this event. There is an erratum published about many
Northbridge events with a CORE_SELECT type of umasks. They don't work.
The only valid choice you have is ANY_CORE.

I have updated libpfm-3.10 and GIT libpfm4 to take this into account.


> Vince
> _______________________________________________
> Ptools-perfapi mailing list
> Ptools-perfapi at eecs.utk.edu
> http://lists.eecs.utk.edu/mailman/listinfo/ptools-perfapi
>


More information about the Ptools-perfapi mailing list