On Fri, 15 Apr 2005, Nick Piggin wrote:
> Jesper Juhl wrote:
> > On Fri, 15 Apr 2005, Nick Piggin wrote:
> >
> >
> > > Jesper Juhl wrote:
> > >
> > > > There are two expressions in kernel/sched.c that are always false since
> > > > they
> > > > test for <0 but the result of the expression is unsigned so they will
> > > > never
> > > > be less than zero. This patch implement the logic that I believe is
> > > > intended
> > > > without the signedness issue and without the nasty casts.
> > > > <disclaimer>patch is compile tested only</disclaimer>
> > > >
> > > This is not *quite* the intended behaviour. It is OK for prev->timestamp
> > > to be '0 - a bit' and now to be '0 + a bit' in the case of wrapping.
> > >
> > > Although considering they're 64-bit values, I'm not sure how much we care.
> > >
> >
> > How do you propose to fix this then? As the code is now the expressionsa
> > are always false - should we just remove the them? Or do you have a
> > sensible definition of "a bit" ? or ome other suggestion alltogether?
> >
>
> Make it a signed comparison?
>
As per this patch perhaps? :
Signed-off-by: Jesper Juhl <[EMAIL PROTECTED]>
--- linux-2.6.12-rc2-mm3-orig/kernel/sched.c 2005-04-11 21:20:56.000000000
+0200
+++ linux-2.6.12-rc2-mm3/kernel/sched.c 2005-04-15 02:21:34.000000000 +0200
@@ -2697,9 +2697,10 @@ need_resched_nonpreemptible:
schedstat_inc(rq, sched_cnt);
now = sched_clock();
if (likely((long long)now - prev->timestamp < NS_MAX_SLEEP_AVG)) {
- run_time = now - prev->timestamp;
- if (unlikely((long long)now - prev->timestamp < 0))
+ if (unlikely(((long long)now - (long long)prev->timestamp) < 0))
run_time = 0;
+ else
+ run_time = now - prev->timestamp;
} else
run_time = NS_MAX_SLEEP_AVG;
@@ -2776,7 +2777,7 @@ go_idle:
if (!rt_task(next) && next->activated > 0) {
unsigned long long delta = now - next->timestamp;
- if (unlikely((long long)now - next->timestamp < 0))
+ if (unlikely(((long long)now - (long long)next->timestamp) < 0))
delta = 0;
if (next->activated == 1)
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
