PLEASE TEST: IPI deadlock avoidance patch

[Willem Jan Withagen]

Julian Elischer wrote:

> Now to start work on the preemption hangs..
>

Well my dual AMD64 system seemed to be rather stable since I went with a 
Tyan board.
So I though lets be brave, and see how far we get with PREEMPTION....
I did not get anywhere.....

Right at boot time:
start_init: trying /sbin/init
panic: runq_choose: no proc on busy queue
cpuid = 0;
KDB: stack backtrace:
kdb_backtrace() at kdb_backtrace+0x37
panic() at panic+0x1e2
runq_choose() at runq_choose+0x6f
kseq_choose() at kseq_choose+0xc8
sched_choose() at sched_choose+0x5b
choosethread() at choosethread+0x47
sched_switch() at sched_switch+0x185
mi_switch() at mi_switch+0x28d
sleepq_switch() at sleepq_switch+0x123
sleepq_timedwait() at sleepq_timedwait+0xe
msleep() at msleep+0x425
g_io_schedule_down() at g_io_schedule_down+0x95
g_down_procbody() at g_down_procbody+0x40
fork_exit() at fork_exit+0xe9
fork_trampoline() at fork_trampoline+0xe

Dumping not possible since the system did not grow any kbd input.
Probably way to early to try this, but I was just  curious.

Question:
    Is this patch going into 6.0-CURRENT or in 5.3????

--WjW