On Aug 10, 2012, at 7:40 AM, Kumar Gala wrote:

=20
On Aug 2, 2012, at 5:04 AM, Zhao Chenhui wrote:
=20

quoted

On Sat, Jul 28, 2012 at 08:20:31AM +1000, Benjamin Herrenschmidt =

wrote:

quoted

quoted

On Fri, 2012-07-27 at 16:58 -0500, Kumar Gala wrote:

quoted

On Jul 20, 2012, at 7:47 AM, Zhao Chenhui wrote:
=20

quoted

During suspend, all interrupts including IPI will be disabled. In =

this case,

quoted

quoted

quoted

quoted

the suspend process will hang in SMP. To prevent this, pass the =

flag

quoted

quoted

quoted

quoted

IRQF_NO_SUSPEND when requesting IPI irq.
=20
Signed-off-by: Zhao Chenhui <redacted>
Signed-off-by: Li Yang <redacted>
---
arch/powerpc/kernel/smp.c |    2 +-
1 files changed, 1 insertions(+), 1 deletions(-)

=20
BenH,
=20
Can you ack?

=20
No I'll merge it but not until it's been in next for a bit unless =

you

quoted

quoted

have some strong emergency there, it's on my mental list of things =

to

quoted

quoted

shovel into next after rc1.
=20
Curiosity: didn't we use to disable all non-boot CPUs on suspend ?
=20
Cheers,
Ben.

=20
Yes, we disabled all non-boot CPUs on suspend by calling =

disable_nonboot_cpus().

quoted

The disable_nonboot_cpus() needs IPIs to work. But prior to
calling disable_nonboot_cpus(), the IPIs are disabled in =

dpm_suspend_noirq().

quoted

=20
-Chenhui

=20
Benh, Ack?

Ben?

- k=