tech-kern archive

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Old Index]

Re: high sys time, very very slow builds on new 24-core system



Andrew Doran <ad%netbsd.org@localhost> wrote:
> We have some algorhythms in the scheduler and mutual exclusion code that
> aren't designed for large numbers of cores, but I think they should be OK
> with 24 CPUs. <...>

I have committed mutex_oncpu patch, so it is O(1) in -current.

> <...>  (While I'm rambling about this I think the SPINLOCK_BACKOFF
> stuff should have some sort of randomness to it perhaps based off curlwp
> and cpu_counter() otherwise things could proceeed in lockstep, although
> again probably not the issue here).

Well, times of acquire attempts by other CPUs would normally differ, so
CPUs would start exponential back-off at different times, i.e. randomness
is already there.  Do you mean wake up path (e.g. cv_broadcast() hit)?

-- 
Mindaugas


Home | Main Index | Thread Index | Old Index