Re: [patch 0/4] XEN: Interrupt cleanups

[Thomas Gleixner]

on Tue, 8 Feb 2011, Konrad Rzeszutek Wilk wrote:

> On Tue, Feb 08, 2011 at 04:39:58PM +0100, Thomas Gleixner wrote:
> > On Tue, 8 Feb 2011, Ian Campbell wrote:
> > > On Tue, 2011-02-08 at 14:55 +0000, Thomas Gleixner wrote:
> > > > > So, with the fixes to 2/4 (irq_move_irq think from yesterday) and 4/4
> > > > > (below), the entire series is:
> > > > > Acked-by: Ian Campbell <ian.campbell@xxxxxxxxxx>
> > > > 
> > > > Cool. So what's the best way to proceed ? That code is not yet in
> > > > linus tree, right ?
> > > 
> > > Correct.
> > > 
> > > > So I guess the best way is that I add the core changes to a rc-4 based
> > > > branch and you can pull it in and apply the whole xen stuff to your
> > > > own tree.
> > > 
> > > My existing cleanup patches are in Konrad's tree (which is in linux-next
> > > etc) so that probably makes most sense as a home for this series. So
> > > unless Konrad has any objections I think it makes sense to pull your
> > > core changes into that branch and then apply your Xen bits on top.
> 
> Ok. Pulled in these patches and stuck Ack-ed by Ian on them.
> > > 
> > > Konrad's branch with my stuff is:
> > >   git://git.kernel.org/pub/scm/linux/kernel/git/konrad/xen.git stable/irq.rework
> > > 
> > > Konrad, this thread starts at <20110205200108.921707839@xxxxxxxxxxxxx>
> > > == http://thread.gmane.org/gmane.linux.kernel/1096437
> > > 
> > > > I base my pending patches on top of that so it wont be any problem
> > > > when merging the stuff together in next or linus later.
> > > 
> > > I don't think there will be much trouble with overlap between these and
> > > any Xen events.c changes for the next merge window but what you suggest
> > > should remove the risk.
> > 
> > Yes, and please talk to me next time before you hack around in the
> > guts of the interrupt code. I noticed just because I was skimming
> > -next, and that really conflicts with major cleanups I'm doing. If
> > there is a shortcoming in the generic code, then let me know.
> 
> <scratches his head> The rework was in Xen code not in generic, and
> the only generic changes that are in there ..  are your code?

The point is:

If you play with generic irq code in Xen in some weird way then you
basically block me to cleanup something in the core code w/o breaking
Xen. I wanted to move IRQ_SUSPENDED to a different field and
accidentally noticed that Xen was fiddling with in -next.

So that's what I'm grumpy about. You hack away in Xen and claim it's
confined to your code, while in reality it is _NOT_.

Again, if there is a problem with the generic code then talk to me.

That's going to be impossible anyway when I'm done with the
encapsulation.

Thanks,

	tglx
--
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/