diff options
| author | Martin Schwidefsky <schwidefsky@de.ibm.com> | 2010-05-17 10:00:03 +0200 |
|---|---|---|
| committer | Martin Schwidefsky <sky@mschwide.boeblingen.de.ibm.com> | 2010-05-17 10:00:15 +0200 |
| commit | 6377981faf1a4425b0531e577736ef03df97c8f6 (patch) | |
| tree | f690c5d357413cb6f26c8463519ad2b1f8800851 /drivers/s390 | |
| parent | 6a2df3a87276cdc08fd87070d09ea18d1fb9d622 (diff) | |
| download | op-kernel-dev-6377981faf1a4425b0531e577736ef03df97c8f6.zip op-kernel-dev-6377981faf1a4425b0531e577736ef03df97c8f6.tar.gz | |
[S390] idle time accounting vs. machine checks
A machine check can interrupt the i/o and external interrupt handler
anytime. If the machine check occurs while the interrupt handler is
waking up from idle vtime_start_cpu can get executed a second time
and the int_clock / async_enter_timer values in the lowcore get
clobbered. This can confuse the cpu time accounting.
To fix this problem two changes are needed. First the machine check
handler has to use its own copies of int_clock and async_enter_timer,
named mcck_clock and mcck_enter_timer. Second the nested execution
of vtime_start_cpu has to be prevented. This is done in s390_idle_check
by checking the wait bit in the program status word.
Signed-off-by: Martin Schwidefsky <schwidefsky@de.ibm.com>
Diffstat (limited to 'drivers/s390')
| -rw-r--r-- | drivers/s390/cio/cio.c | 3 |
1 files changed, 2 insertions, 1 deletions
diff --git a/drivers/s390/cio/cio.c b/drivers/s390/cio/cio.c index 5feea1a..f4e6cf3 100644 --- a/drivers/s390/cio/cio.c +++ b/drivers/s390/cio/cio.c @@ -616,7 +616,8 @@ void __irq_entry do_IRQ(struct pt_regs *regs) struct pt_regs *old_regs; old_regs = set_irq_regs(regs); - s390_idle_check(); + s390_idle_check(regs, S390_lowcore.int_clock, + S390_lowcore.async_enter_timer); irq_enter(); __get_cpu_var(s390_idle).nohz_delay = 1; if (S390_lowcore.int_clock >= S390_lowcore.clock_comparator) |
