On 01/05/14 14:51, Jon Medhurst (Tixy) wrote:
On Thu, 2014-05-01 at 13:59 +0100, Chris Redpath wrote: [...]
Changing the BUG to a simple return NULL and fixing up the callers for this case as you did in your testing patch is functionally correct and safe. The question for me is if we should bother to try to optimize idle_pull behavior during cpu_down - I'm open to opinions :)
My opinion is 'no', because unless I missed something, this case shouldn't really occur much or at all during normal use. If it did, surely you would have hit the BUG_ON before. ;-)
Yes, I tend to agree. You reverted all the patches in that release, right?
I can resend with a fixed-up final patch if you want, or you can fix it locally and I'll review it on this list if you prefer. Whatever is easiest for you.
Or does cpuidle hit this CPUs offlining case? (I'm a bit fuzzy on these things, but I would have thought not).
No it doesn't, you're right. The CPUs remain online during CPUIdle.