Messages in this thread | | | From | "Khimenko Victor" <> | Date | Wed, 2 Aug 2000 02:59:35 +0400 (MSD) | Subject | Re: sysconf (was Re: RLIM_INFINITY inconsistency between archs) |
| |
In <15481.965138601@cygnus.co.uk> David Woodhouse (dwmw2@infradead.org) wrote:
> hpa@zytor.com said: >> I don't think we want to do this! IMO, HZ should not get exported to >> user space *AT ALL*. Instead, for the few interfaces that need it, >> we'll export a "user space HZ" (USER_HZ) which is fixed. No need for >> a kernel hack. When we support nonstandard values for HZ, we need to >> fix the few interfaces that actually export jiffies values to convert >> from "user jiffies" to real jiffies.
> No. Why do the arithmetic in the kernel when we could just export the raw > data, with full accuracy, and let user-space sort it out?
Since then we'll need to fix 10'000 programs instead of one kernel when sheduling will be made more adaptive (without fixed HZ) ?
> User-space may want to know the value of HZ for other reasons anyway.
> if (HZ < 1000) { > perror("You must hack your kernel to improve scheduling latency"); > exit(1); > }
Gah. Yes. Such code is EXACTLY reason to NOT export HZ.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/
| |