linux-next: Tree for Sept 24 (irqchip)

linux-next: Tree for Sept 24 (irqchip)

[Stephen Rothwell]

Hi Randy,

On Mon, 24 Sep 2012 08:39:05 -0700 Randy Dunlap <rdunlap@xenotime.net> wrote:
>
> On 09/24/2012 07:53 AM, Stephen Rothwell wrote:
> 
> > Today was a train wreck, with lots of new conflicts across several trees
> > and a few build failures as well.
> 
> I agree.  I mostly get this:
> 
> drivers/Kconfig:157: can't open file "drivers/irqchip/Kconfig"
> 
> i.e., no such file.
> 
> How does this happen?  :(
> 
> who handles irqchip?

Hmm, commit 89214f009c1d ("ARM: bcm2835: add interrupt controller
driver") from the arm-soc tree adds that file as an empty file (I have
the empty file in my tree).  Nothing adds anything to it.
-- 
Cheers,
Stephen Rothwell                    sfr at canb.auug.org.au
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 836 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20120925/a43debdc/attachment.sig>

linux-next: Tree for Sept 24 (irqchip)

[Stephen Warren]

On 09/24/2012 10:07 AM, Stephen Rothwell wrote:
> Hi Randy,
> 
> On Mon, 24 Sep 2012 08:39:05 -0700 Randy Dunlap
> <rdunlap@xenotime.net> wrote:
>> 
>> On 09/24/2012 07:53 AM, Stephen Rothwell wrote:
>> 
>>> Today was a train wreck, with lots of new conflicts across
>>> several trees and a few build failures as well.
>> 
>> I agree.  I mostly get this:
>> 
>> drivers/Kconfig:157: can't open file "drivers/irqchip/Kconfig"
>> 
>> i.e., no such file.
>> 
>> How does this happen?  :(
>> 
>> who handles irqchip?
> 
> Hmm, commit 89214f009c1d ("ARM: bcm2835: add interrupt controller 
> driver") from the arm-soc tree adds that file as an empty file (I
> have the empty file in my tree).  Nothing adds anything to it.

That's odd. I just checked out next-20120924, and that (empty) file
does appear on my disk as expected. Do some versions of git not handle
creating empty files?

Note that IIRC the Kconfig file is required, because otherwise the
build process won't recurse into drivers/irqchip/ and hence code won't
be linked into the kernel image.

The file is empty because the Makefile there keys of
CONFIG_ARCH_BCM2835 and hence doesn't need drivers/irqchip/Kconfig to
define any new Kconfig variables.

linux-next: Tree for Sept 24 (irqchip)

[Olof Johansson]

On Mon, Sep 24, 2012 at 9:07 AM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
> Hi Randy,
>
> On Mon, 24 Sep 2012 08:39:05 -0700 Randy Dunlap <rdunlap@xenotime.net> wrote:
>>
>> On 09/24/2012 07:53 AM, Stephen Rothwell wrote:
>>
>> > Today was a train wreck, with lots of new conflicts across several trees
>> > and a few build failures as well.
>>
>> I agree.  I mostly get this:
>>
>> drivers/Kconfig:157: can't open file "drivers/irqchip/Kconfig"
>>
>> i.e., no such file.
>>
>> How does this happen?  :(
>>
>> who handles irqchip?
>
> Hmm, commit 89214f009c1d ("ARM: bcm2835: add interrupt controller
> driver") from the arm-soc tree adds that file as an empty file (I have
> the empty file in my tree).  Nothing adds anything to it.

Yeah, the directory came in through arm-soc for-next (since the
bcm2835 driver was added), but there is an empty Kconfig in arm-soc. I
can see it in linux-next as well.

I wonder why it doesn't show up in your tree, Randy. You just did a
regular git checkout?


-Olof

linux-next: Tree for Sept 24 (irqchip)

[Randy Dunlap]

On 09/24/2012 09:46 AM, Olof Johansson wrote:

> On Mon, Sep 24, 2012 at 9:07 AM, Stephen Rothwell <sfr@canb.auug.org.au> wrote:
>> Hi Randy,
>>
>> On Mon, 24 Sep 2012 08:39:05 -0700 Randy Dunlap <rdunlap@xenotime.net> wrote:
>>>
>>> On 09/24/2012 07:53 AM, Stephen Rothwell wrote:
>>>
>>>> Today was a train wreck, with lots of new conflicts across several trees
>>>> and a few build failures as well.
>>>
>>> I agree.  I mostly get this:
>>>
>>> drivers/Kconfig:157: can't open file "drivers/irqchip/Kconfig"
>>>
>>> i.e., no such file.
>>>
>>> How does this happen?  :(
>>>
>>> who handles irqchip?
>>
>> Hmm, commit 89214f009c1d ("ARM: bcm2835: add interrupt controller
>> driver") from the arm-soc tree adds that file as an empty file (I have
>> the empty file in my tree).  Nothing adds anything to it.
> 
> Yeah, the directory came in through arm-soc for-next (since the
> bcm2835 driver was added), but there is an empty Kconfig in arm-soc. I
> can see it in linux-next as well.
> 
> I wonder why it doesn't show up in your tree, Randy. You just did a
> regular git checkout?


Sorry for the very delayed reply (and this is already fixed).

It doesn't show up in my kernel tree because I use linux-v3.x tarballs
and patch-3.x-rcY patches and linux-next patches, and
'patch' does not keep empty files around.

-- 
~Randy