Messages in this thread | | | Date | Tue, 24 Sep 2013 10:21:28 +0200 | From | Peter Zijlstra <> | Subject | Re: [RFC GIT PULL] softirq: Consolidation and stack overrun fix |
| |
On Tue, Sep 24, 2013 at 06:16:53PM +1000, Benjamin Herrenschmidt wrote: > On Tue, 2013-09-24 at 10:04 +0200, Peter Zijlstra wrote: > > On Tue, Sep 24, 2013 at 11:52:07AM +1000, Benjamin Herrenschmidt wrote: > > > So if that holds, we have a solid way to do per-cpu. On one side, I tend > > > to think that r13 being task/thread/thread_info is probably a better > > > overall choice, I'm worried that going in a different direction than x86 > > > means generic code will get "tuned" to use per-cpu for performance > > > critical stuff rather than task/thread/thread_info in inflexible ways. > > > > The plus side of per-cpu over per-task is that one typically has a lot > > less cpus than tasks. Also, its far easier/cheaper to iterate cpus than > > it is to iterate tasks. > > I don't see how that relates to the above though...
It was a comment on the increase of per-cpu usage in generic code.
| |