2006-10-06 18:06:44

by Deepak Saxena

[permalink] [raw]
Subject: [PATCH 2.6.19-git] Fix ARM breakage due to no irq_regs.h


Signed-off-by: Deepak Saxena <[email protected]>


diff --git a/include/asm-arm/irq_regs.h b/include/asm-arm/irq_regs.h
new file mode 100644
index 0000000..3dd9c0b
--- /dev/null
+++ b/include/asm-arm/irq_regs.h
@@ -0,0 +1 @@
+#include <asm-generic/irq_regs.h>

--
Deepak Saxena - [email protected] - http://www.plexity.net

"An open heart has no possessions, only experiences" - Matt Bibbeau


2006-10-06 18:13:57

by Linus Torvalds

[permalink] [raw]
Subject: Re: [PATCH 2.6.19-git] Fix ARM breakage due to no irq_regs.h


That should not really be nearly enough.

I've pushed out a more complete (but still untested) version that should
hopefully be closer to fixing up any irq breakage on arm.

Linus

On Fri, 6 Oct 2006, Deepak Saxena wrote:
>
> diff --git a/include/asm-arm/irq_regs.h b/include/asm-arm/irq_regs.h
> new file mode 100644
> index 0000000..3dd9c0b
> --- /dev/null
> +++ b/include/asm-arm/irq_regs.h
> @@ -0,0 +1 @@
> +#include <asm-generic/irq_regs.h>

2006-10-06 18:22:55

by Ben Dooks

[permalink] [raw]
Subject: Re: [PATCH 2.6.19-git] Fix ARM breakage due to no irq_regs.h

On Fri, Oct 06, 2006 at 11:13:52AM -0700, Linus Torvalds wrote:
>
> That should not really be nearly enough.
>
> I've pushed out a more complete (but still untested) version that should
> hopefully be closer to fixing up any irq breakage on arm.

It should get built ky Kautobuild when the next -git is
released (http://armlinux.simtec.co.uk/kautobuild/) which
should at least tell us if it builds.

--
Ben ([email protected], http://www.fluff.org/)

'a smiley only costs 4 bytes'

2006-10-06 18:34:30

by Linus Torvalds

[permalink] [raw]
Subject: Re: [PATCH 2.6.19-git] Fix ARM breakage due to no irq_regs.h



On Fri, 6 Oct 2006, Ben Dooks wrote:
>
> It should get built ky Kautobuild when the next -git is
> released (http://armlinux.simtec.co.uk/kautobuild/) which
> should at least tell us if it builds.

Sure, but havign a 24-hour latency between test-builds is not exactly a
great way to do this ;)

For example, the one-liner that Deepak sent in will certainly fix the
particular build error that the autobuild site has right now. But doing it
one error at a time would be somewhat less than productive, considering
that the patch I committed was 2695 lines ;)

Of course, assuming I did actually get most of them (and that I didn't
introduce any stupid new syntax errors), the next autobuild report may be
more useful. Still, I know several people have arm build environments
already set up, so..

Linus