Messages in this thread | | | Date | Sat, 4 Apr 2015 11:34:00 +0200 | From | Ingo Molnar <> | Subject | Re: [PATCH] irq: revert non-working patch to affinity defaults |
| |
* Jesse Brandeburg <jesse.brandeburg@intel.com> wrote:
> > Now this is just a small annoyance that should not really matter - > > it would be nice to figure out the real reason for why the irqs > > move back to CPU#0. > > > > In theory the same could happen to 'irqbalanced' as well, if it > > calls shortly after an irq was registered - so this is not a bug > > we want to ignore. > > Let me know if I can do something to help, the IRQ code is a bit of > a steep learning curve, so the chances of me fixing it are small.
Well, as a starter, if you can reproduce it on a system (I cannot), then try to stick a few printks in there to print out the affinity mask as it gets changed plus dump_stack(), and see who changes it back?
Chances are it's irqbalanced? If not then the stack dump will tell. It shouldn't be too chatty.
(trace_printk() if you prefer traces.)
Thanks,
Ingo
| |