Messages in this thread | | | Date | Mon, 02 Jun 2014 10:38:18 -0700 | From | "H. Peter Anvin" <> | Subject | Re: [PATCH v2] introduce atomic_pointer to fix a race condition in cancelable mcs spinlocks |
| |
On 06/02/2014 10:25 AM, Waiman Long wrote: > > Doing an xchg is a very expensive operation compared with ACCESS_ONCE. I > will not suggest doing that to make it right for PA-RISC at the expense > of performance in other architectures. >
And of course, this gets into the toxic question: what are reasonable minimum requirements for Linux? How far do we need to stretch to support niche architectures which have very small (Linux) userbases?
-hpa
| |