Re: [PATCH] 2.4.18 scheduler bugs

[Joe Korty]

>> - reschedule_idle() - smp_send_reschedule when setting idle's need_resched
>> 
>>     Idle tasks nowdays don't spin waiting for need->resched to change,
>>     they sleep on a halt insn instead.  Therefore any setting of
>>     need->resched on an idle task running on a remote CPU should be
>>     accompanied by a cross-processor interrupt.
> 
> this is broken as well. Check out the idle=poll feature i wrote some time
> ago.

The idle=poll stuff is a hack.  I'd like my idle cpus to sleep and still
have them wake up the moment work for them becomes available.  I see no
reason why an idle cpu should be forced to remain idle until the next
tick, nor why fixing that should be considered `broken'.

Joe