GIC MULTI_IRQ_HANDLER branches

GIC MULTI_IRQ_HANDLER branches

[Marc Zyngier]

I've stashed my patches for MULTI_IRQ_HANDLER on the following branches:
git://github.com/mzyngier/arm-platforms.git gic-multi-irq
git://github.com/mzyngier/arm-platforms.git gic-multi-irq-asm

This is based on Russell's smp branch, plus my PPI patches that Russell
has already pulled for the upcoming merge window.

The difference between the two branches is that the second is keeping
the ASM version of the GIC handler instead of a C version.

The C version has been tested on VE, Realview, Tegra, and OMAP2/3/4.
The ASM version has only been tested on OMAP4.

Cheers,

	M.
-- 
I'm the slime oozin' out from your TV set...

GIC MULTI_IRQ_HANDLER branches

[Tony Lindgren]

* Marc Zyngier <maz@misterjones.org> [111024 16:59]:
> I've stashed my patches for MULTI_IRQ_HANDLER on the following branches:
> git://github.com/mzyngier/arm-platforms.git gic-multi-irq
> git://github.com/mzyngier/arm-platforms.git gic-multi-irq-asm
> 
> This is based on Russell's smp branch, plus my PPI patches that Russell
> has already pulled for the upcoming merge window.
> 
> The difference between the two branches is that the second is keeping
> the ASM version of the GIC handler instead of a C version.
> 
> The C version has been tested on VE, Realview, Tegra, and OMAP2/3/4.
> The ASM version has only been tested on OMAP4.

For the record, boots fine on omap1/2/3/4 like I replied earlier
with my Tested-by.

Can people please test this series on all subarchs so Marc can
then remove the remaining assembly code too.

Regards,

Tony

GIC MULTI_IRQ_HANDLER branches

[Russell King - ARM Linux]

On Tue, Oct 25, 2011 at 09:48:16AM +0200, Tony Lindgren wrote:
> * Marc Zyngier <maz@misterjones.org> [111024 16:59]:
> > I've stashed my patches for MULTI_IRQ_HANDLER on the following branches:
> > git://github.com/mzyngier/arm-platforms.git gic-multi-irq
> > git://github.com/mzyngier/arm-platforms.git gic-multi-irq-asm
> > 
> > This is based on Russell's smp branch, plus my PPI patches that Russell
> > has already pulled for the upcoming merge window.
> > 
> > The difference between the two branches is that the second is keeping
> > the ASM version of the GIC handler instead of a C version.
> > 
> > The C version has been tested on VE, Realview, Tegra, and OMAP2/3/4.
> > The ASM version has only been tested on OMAP4.
> 
> For the record, boots fine on omap1/2/3/4 like I replied earlier
> with my Tested-by.
> 
> Can people please test this series on all subarchs so Marc can
> then remove the remaining assembly code too.

My feeling is that this is too disruptive to take for 3.2 now that the
merge window is open: just adding Marc's PPI patches on Sunday caused
a number of _horrible_ conflicts crop up.  This probably is no
different.

So, let's postpone this kind of stuff until after this merge window has
closed.

(BTW, Linus won't be pulling major stuff which hasn't been in linux-next -
see the 3.1 announcement.)

GIC MULTI_IRQ_HANDLER branches

[Tony Lindgren]

* Russell King - ARM Linux <linux@arm.linux.org.uk> [111025 09:54]:
> On Tue, Oct 25, 2011 at 09:48:16AM +0200, Tony Lindgren wrote:
> > * Marc Zyngier <maz@misterjones.org> [111024 16:59]:
> > > I've stashed my patches for MULTI_IRQ_HANDLER on the following branches:
> > > git://github.com/mzyngier/arm-platforms.git gic-multi-irq
> > > git://github.com/mzyngier/arm-platforms.git gic-multi-irq-asm
> > > 
> > > This is based on Russell's smp branch, plus my PPI patches that Russell
> > > has already pulled for the upcoming merge window.
> > > 
> > > The difference between the two branches is that the second is keeping
> > > the ASM version of the GIC handler instead of a C version.
> > > 
> > > The C version has been tested on VE, Realview, Tegra, and OMAP2/3/4.
> > > The ASM version has only been tested on OMAP4.
> > 
> > For the record, boots fine on omap1/2/3/4 like I replied earlier
> > with my Tested-by.
> > 
> > Can people please test this series on all subarchs so Marc can
> > then remove the remaining assembly code too.
> 
> My feeling is that this is too disruptive to take for 3.2 now that the
> merge window is open: just adding Marc's PPI patches on Sunday caused
> a number of _horrible_ conflicts crop up.  This probably is no
> different.
> 
> So, let's postpone this kind of stuff until after this merge window has
> closed.

OK, I was hoping we could get it in without the assembly changes..
 
> (BTW, Linus won't be pulling major stuff which hasn't been in linux-next -
> see the 3.1 announcement.)

..but yeah I guess it's too late now.

Tony