Re: [PATCH 3.12 46/62] irq: Enable all irqs unconditionally in irq_resume

[Heiko Stübner]

Am Dienstag, 10. Dezember 2013, 10:08:47 schrieb Greg Kroah-Hartman:
> On Tue, Dec 10, 2013 at 10:01:13AM +0100, Pavel Machek wrote:
> > Hi!
> > 
> > > 3.12-stable review patch.  If anyone has any objections, please let
> > > 
> > >  me know.
> > 
> > Objections... well... I'm not sure if it meets stable criteria. In
> > particular, I don't think it was ever demonstrated as having bad
> > effect on the users.
> 
> Well, I'm guessing someone on the cc: here wanted it in stable@, so I'll
> let them speak up as to why...

As for bad effects on users ... any suspend_noirq callback returning an error 
and thus aborting the suspend will cause the error fixed by this, as the 
early-irqs won't get reenabled in the suspend-rollback [due to the syscore_ops 
doing their resume never being run].

You could for example lose keys connected thru the gpio-keys driver until the 
next reboot.

And it seems at least Laxman and myself were affected by it.


Heiko
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/