Messages in this thread | | | From | Andi Kleen <> | Subject | Re: sched_clock() uses are broken | Date | Sun, 7 May 2006 18:03:28 +0200 |
| |
On Sunday 07 May 2006 16:04, Nick Piggin wrote: > Russell King wrote: > > > Also, any comments on update_cpu_clock() and current_sched_time() both > > appearing to be buggy, or am I barking up the wrong tree with those? > > Can't remember off the top of my head who put those in, sorry. > > Aside from the fact that they appear to be fundamentally buggy anyway > because we don't require exact ns intervals from sched_clock() at the > best of times, I think your wrapping fix for them looks correct.
I must say I always hated them because adding code to the context switch for such an obscure POSIX bu^w"feature" is just a bad idea. Maybe it would be best to remove it again instead of adding the scaling needed to make it actually work
(since we got along with such a broken implementation for so long I would guess that nobody uses these timers anyways)
-Andi
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |