Discussion:
help with sandybridge/ivybridge hwpmc NUMA DRAM counters
Adrian Chadd
2015-04-23 06:38:22 UTC
Permalink
hi all,

I'm having a spot of problem trying to get the local/remote dram
counters working on the NUMA sandybridge and ivybridge systems here.

Things work fine using intel-pcm, but those same counters don't work with hwpmc.

Sandybridge - there's apparently an MSR that needs to be fiddled if
the counters are active.

Ivybridge - the v1 and v2 chips have different local/remote dram
counters, and on my v2 setup there's actually /two/ LOCAL_DRAM:

***@testbox1:~/git/github/erikarn/freebsd/sys/dev/hwpmc %
pmccontrol -L | grep DRAM
MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM
MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM
MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM

Now, i may be able to get to figuring this out at some point in the
distant future, but I'd really appreciate any help I can get now. I'm
now at the point with the NUMA affinity API stuff where I'm now
chasing down when things are correctly working with local/remote RAM,
and I'd really like to use hwpmc in sampling mode to work on it.

Thanks for any help!



-adrian
Slawa Olhovchenkov
2015-04-23 06:50:06 UTC
Permalink
Post by Adrian Chadd
hi all,
I'm having a spot of problem trying to get the local/remote dram
counters working on the NUMA sandybridge and ivybridge systems here.
Things work fine using intel-pcm, but those same counters don't work with hwpmc.
Sandybridge - there's apparently an MSR that needs to be fiddled if
the counters are active.
Ivybridge - the v1 and v2 chips have different local/remote dram
# pmccontrol -L | grep DRAM
MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM
MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM
Post by Adrian Chadd
pmccontrol -L | grep DRAM
MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM
MEM_LOAD_UOPS_LLC_MISS_RETIRED.LOCAL_DRAM
MEM_LOAD_UOPS_LLC_MISS_RETIRED.REMOTE_DRAM
Now, i may be able to get to figuring this out at some point in the
distant future, but I'd really appreciate any help I can get now. I'm
now at the point with the NUMA affinity API stuff where I'm now
chasing down when things are correctly working with local/remote RAM,
and I'd really like to use hwpmc in sampling mode to work on it.
Thanks for any help!
-adrian
_______________________________________________
http://lists.freebsd.org/mailman/listinfo/freebsd-current
Adrian Chadd
2015-04-23 07:20:14 UTC
Permalink
Yeah, on stable/10. But on -HEAD it's different. There's two entries -
one for d3_01 and one for d3_03.



-adrian
Slawa Olhovchenkov
2015-04-23 07:22:43 UTC
Permalink
Post by Adrian Chadd
Yeah, on stable/10. But on -HEAD it's different. There's two entries -
one for d3_01 and one for d3_03.
What CPU model?
Adrian Chadd
2015-04-23 07:24:06 UTC
Permalink
Post by Slawa Olhovchenkov
Post by Adrian Chadd
Yeah, on stable/10. But on -HEAD it's different. There's two entries -
one for d3_01 and one for d3_03.
What CPU model?
CPU: Intel(R) Xeon(R) CPU E5-2650 v2 @ 2.60GHz (2600.06-MHz K8-class CPU)
Origin="GenuineIntel" Id=0x306e4 Family=0x6 Model=0x3e Stepping=4



-adrian
Slawa Olhovchenkov
2015-04-23 07:24:55 UTC
Permalink
Post by Adrian Chadd
Post by Slawa Olhovchenkov
Post by Adrian Chadd
Yeah, on stable/10. But on -HEAD it's different. There's two entries -
one for d3_01 and one for d3_03.
What CPU model?
Origin="GenuineIntel" Id=0x306e4 Family=0x6 Model=0x3e Stepping=4
Same with me?
Slawa Olhovchenkov
2015-04-23 08:07:05 UTC
Permalink
Post by Slawa Olhovchenkov
Post by Adrian Chadd
Post by Slawa Olhovchenkov
Post by Adrian Chadd
Yeah, on stable/10. But on -HEAD it's different. There's two entries -
one for d3_01 and one for d3_03.
What CPU model?
Origin="GenuineIntel" Id=0x306e4 Family=0x6 Model=0x3e Stepping=4
Same with me?
May be in you case E5-269x?

Loading...