Messages in this thread | | | Subject | Re: RFC: futex_wait() can DoS the tick | From | Mike Galbraith <> | Date | Thu, 11 Jun 2015 09:35:59 +0200 |
| |
On Wed, 2015-06-10 at 17:12 +0200, Thomas Gleixner wrote:
> I just booted a SNB with lapic=notscdeadline and ran that test > program. All what happens is - as expected - that ksoftirqd runs more > than we would like it to. I cannot observe any anomality vs. local > timer interrupts at all.
Ditto on a E5-4620 box this morning, no sign of it. E5-2670 box I had verified "went away" when I put master on it this morning, but that doesn't matter much, having deadline timer feature alone definitely doesn't mean box can be convinced to reproduce the stall. Hohum.
-Mike
| |