Messages in this thread | ![/](/images/icornerl.gif) | | Date | Thu, 11 Nov 2010 19:42:03 +0100 | From | Andi Kleen <> | Subject | Re: [PATCH 1/3] perf-events: Add support for supplementary event registers |
| |
On 11/11/2010 7:06 PM, Stephane Eranian wrote:
Thanks for the review.
> One thing I noticed in your patch is that you don't special > case the configuration where HT is off. In that case, the > sharing problem goes away. I think you could override > either way during init.
The allocator should handle that transparently. With HT off the resource is always free. Or do you see any concrete problems with the current code path?
> Some more tidbits: > - OFFCORE_RESPONSE_0 is 0x01b7 > - OFFCORE_RESPONSE_1 is 0x01bb > > The umask is not zero but 1. Dont' know if you get > something meaningful is you pass a umask of zero.
Hmm I seem to get events that look meaningful with 0. But you're right 1 is better. I used that in manual tests, but it wasn't in the cache number mappings. I'll fix that for the next version.
-Andi
| ![\](/images/icornerr.gif) |