Messages in this thread | | | From | Andi Kleen <> | Subject | Re: [PATCH RFC] x86: avoid atomic operation in test_and_set_bit_lock if possible | Date | Thu, 24 Mar 2011 13:40:49 -0700 |
| |
Ingo Molnar <mingo@elte.hu> writes: > > One good way to see cache bounces is to run a misses/accesses ratio profile: > > perf top -e cache-misses -e cache-references --count-filter 10 > > Note the two events: this runs a 'weighted' profile, you'll see (LLC) > cache-misses of a function relative to cache-references it does, a > misses/references ratio in essence.
If anyone does this on a Nehalem please only use 2.6.39-rc*+ which includes the offcore patches. Anything before that will get you complete bogus numbers.
BTW with Lin-Ming's memory latency profiling code you can do this much more directly.
-Andi -- ak@linux.intel.com -- Speaking for myself only
| |