Re: [PATCH] cpufreq, powernow-k8: Remove usage of smp_processor_id() in preemptible code

["Rafael J. Wysocki" <rjw@sisk.pl>]

On Monday 15 of October 2012 10:40:11 Borislav Petkov wrote:
> On Mon, Oct 15, 2012 at 07:50:13AM +0200, Rafael J. Wysocki wrote:
> > Well, please don't tag patches for -stable, because -stable doesn't
> > take _patches_.
> 
> Really, I didn't know that?! :-)
> 
> > It takes commits from the Linus' tree and backports them and that's
> > maintainer's job to tag them for -stable, not yours.
> 
> You're not serious, right? This is not the case in at least 50% of the
> cases.
> 
> And this is OK because maintainers don't always know whether the patch
> should be tagged for stable. So yes, people should add the stable tag
> and yes, committers still have a veto over it.
> 
> And yes, Andreas and I *know* how stable patches get applied, thank you
> very much.

I didn't say you didn't know that.

> [ … ]
> 
> > Yes, they do, but that means it doesn't make sense to send them stuff
> > before it's been merged, right?
> 
> Ok, I get it, you don't want people to send patches to stable@vger
> *before* they've hit mainline.
> 
> Nothing in <Documentation/stable_kernel_rules.txt> states that
> stable@vger shouldn't get CCed on submissions unless the patch is
> upstream and besides, stable@vger gets CCed in a lot of discussions
> anyway so there's other traffic just the same.
> 
> Bottomline: If you think people shouldn't spam stable@vger, then tough
> luck - I don't think you can stop people from accidentally/due to the
> automated nature of the process, CC stable. Even if it said so in the
> above doc file.
> 
> As a result, stable maintainers simply rely on scripts which verify the
> patch is actually upstream before applying it to stable.

Well, perhaps I shouldn't care too, then. :-)

Thanks,
Rafael


-- 
I speak only for myself.
Rafael J. Wysocki, Intel Open Source Technology Center.