Re: Why is HZ on an Alpha 1024?
Perry Harrington (pedward@sun4.apsoft.com)
Fri, 1 May 1998 10:10:10 -0700 (PDT)
>
> And lo, Perry Harrington saith unto me:
> [re: HZ=1024 on alphas]
> > > 1024 has been chosen because 1000 is not a power of 2.
> Neither is 100, but HZ=100 works everywhere else...
>
> Personally, I think HZ=120 would be much nicer, since you can then
> do things every 1/20th, 1/24th, and 1/30th of a second without
> having to resort to the UTIME patch...
>
> Alphas with HZ > everyone else was okay when all Alphas were way
> faster than the competition. Now that all architectures have been
> around for a while, it's time that HZ either got fixed for everyone
> or got scaled to the actual power of the CPU...
Well, I'm trying an experiment. I've scaled HZ to 128. This still retains
the divisors to be exactly the same with 1024 (1024/56==18.2, 128/7==18.2).
I do recall that on the PC, certain things are expected to happen at 18.2
times a second, namely floppy stuff and such.
>
> Keith (HZ=floor(BogoMips / ln(year-1995)) ?)
>
--
Perry Harrington Linux rules all OSes. APSoft ()
email: perry@apsoft.com Think Blue. /\
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu