Messages in this thread | | | Date | Fri, 1 Apr 2016 11:41:19 +0100 | From | Will Deacon <> | Subject | Re: [PATCH] locking/qrwlock: Allow multiple spinning readers |
| |
On Fri, Apr 01, 2016 at 12:31:43PM +0200, Peter Zijlstra wrote: > On Thu, Mar 31, 2016 at 06:12:38PM -0400, Waiman Long wrote: > > >>However, if we allow a limited number of readers to spin on the > > >>lock simultaneously, we can eliminates some of the reader-to-reader > > >>latencies at the expense of a bit more cacheline contention and > > >>probably more power consumption. > > >So the embedded people might not like that much. > > > > It could be. It is always a compromise. > > So ARM is the only one that currently waits without spinning and could > care; so Will might have an opinion. One 'solution' would be to make > this an optional feature.
Well, perhaps we could built this using the cmp-and-wait structure we spoke about a couple of months back. What happened with that? Is there something I need to go implement for ARM/arm64?
Will
| |