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 11:38:28 -0700 |
| |
Nikanth Karthikesan <knikanth@suse.de> writes:
> On x86_64 SMP with lots of CPU atomic instructions which assert the LOCK # > signal can stall other CPUs. And as the number of cores increase this penalty
This description is very wrong. No modern CPU still has a LOCK # signal or does global stalls for LOCK.
Do you actually have any data this is a problem and how much difference the patch makes?
Also there's the missing barrier now of course.
-Andi -- ak@linux.intel.com -- Speaking for myself only
| |