Messages in this thread | | | Subject | Re: [patch] jiffies wraparound [Re: 2.1.125 Show stopper list: Draft] | From | Andi Kleen <> | Date | 23 Oct 1998 11:17:55 +0200 |
| |
In article <Pine.LNX.3.96.981022230121.23982E-100000@penguin.transmeta.com>, Linus Torvalds <torvalds@transmeta.com> writes:
> A kernel that exports internal knowledge is a BAD kernel. And I will > continue to refuse to make bad decisions like a sysctl() interface to HZ.
It already happened: most of the /proc/sys/net/ipv4/route* sysctls are in jiffies. Also if I understood correctly the idea behind POSIX sysconf(_CLK_TCK) is to give the application some minimum idea about the to be expected granuality from timers for realtime purposes. How well this works in practice I don't know, but I could imagine that it is useful to have at least a rough idea at runtime of how much of the tv_usec field of struct timerval is a lie.
if (sysconf(_CLK_TCK) < 500) { printf("Sorry, you kernel is too slow for me\n"); exit(1); }
For soft RT apps it makes sense to recompile the kernel with a higher HZ, and it would be nice if it wasn't required to recompile glibc too @)
-Andi > -
- 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/
| |