master build: 2 failures 4 warnings (v4.8-11811-g35ff96d)

master build: 2 failures 4 warnings (v4.8-11811-g35ff96d)

[Mark Brown]

On Tue, Oct 11, 2016 at 07:30:35AM +0100, Build bot for Mark Brown wrote:

Linus' tree is currently failing to build arm and arm64 allmodconfigs
with:

> 	arm64-allmodconfig
> ERROR: "irq_set_parent" [drivers/mfd/tps65217.ko] undefined!

> 	arm-allmodconfig
> ERROR: "irq_set_parent" [drivers/mfd/tps65217.ko] undefined!

due to 6556bdacf646fc (mfd: tps65217: Add support for IRQs) since
irq_set_parent() isn't exported.  This has been present in -next
for getting on for a month, a patch was proposed adding the relevant
export but that isn't present in -next yet.

The function is being used in order to enable lazy IRQ disabling for
threaded interrupts:

	https://www.spinics.net/lists/arm-kernel/msg532864.html

What's the plan for getting this fixed in Linus' tree?
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 455 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20161011/37a83c95/attachment.sig>

master build: 2 failures 4 warnings (v4.8-11811-g35ff96d)

[Lee Jones]

On Tue, 11 Oct 2016, Mark Brown wrote:

> On Tue, Oct 11, 2016 at 07:30:35AM +0100, Build bot for Mark Brown wrote:
> 
> Linus' tree is currently failing to build arm and arm64 allmodconfigs
> with:
> 
> > 	arm64-allmodconfig
> > ERROR: "irq_set_parent" [drivers/mfd/tps65217.ko] undefined!
> 
> > 	arm-allmodconfig
> > ERROR: "irq_set_parent" [drivers/mfd/tps65217.ko] undefined!
> 
> due to 6556bdacf646fc (mfd: tps65217: Add support for IRQs) since
> irq_set_parent() isn't exported.  This has been present in -next
> for getting on for a month, a patch was proposed adding the relevant
> export but that isn't present in -next yet.
> 
> The function is being used in order to enable lazy IRQ disabling for
> threaded interrupts:
> 
> 	https://www.spinics.net/lists/arm-kernel/msg532864.html
> 
> What's the plan for getting this fixed in Linus' tree?

Here's the conversation:

  https://www.spinics.net/lists/arm-kernel/msg531850.html

I'm waiting on a firm answer from Arnd and Thomas before applying
anything.

-- 
Lee Jones
Linaro STMicroelectronics Landing Team Lead
Linaro.org ? Open source software for ARM SoCs
Follow Linaro: Facebook | Twitter | Blog

master build: 2 failures 4 warnings (v4.8-11811-g35ff96d)

[Arnd Bergmann]

On Friday, October 14, 2016 11:33:44 AM CEST Lee Jones wrote:
> On Tue, 11 Oct 2016, Mark Brown wrote:
> 
> > On Tue, Oct 11, 2016 at 07:30:35AM +0100, Build bot for Mark Brown wrote:
> > 
> > Linus' tree is currently failing to build arm and arm64 allmodconfigs
> > with:
> > 
> > >     arm64-allmodconfig
> > > ERROR: "irq_set_parent" [drivers/mfd/tps65217.ko] undefined!
> > 
> > >     arm-allmodconfig
> > > ERROR: "irq_set_parent" [drivers/mfd/tps65217.ko] undefined!
> > 
> > due to 6556bdacf646fc (mfd: tps65217: Add support for IRQs) since
> > irq_set_parent() isn't exported.  This has been present in -next
> > for getting on for a month, a patch was proposed adding the relevant
> > export but that isn't present in -next yet.
> > 
> > The function is being used in order to enable lazy IRQ disabling for
> > threaded interrupts:
> > 
> >       https://www.spinics.net/lists/arm-kernel/msg532864.html
> > 
> > What's the plan for getting this fixed in Linus' tree?
> 
> Here's the conversation:
> 
>   https://www.spinics.net/lists/arm-kernel/msg531850.html
> 
> I'm waiting on a firm answer from Arnd and Thomas before applying
> anything.

I don't understand what the hardware does, so I'm not really
able to answer this. Maybe just mark the driver as 'depends on
BROKEN' for now?

	Arnd

master build: 2 failures 4 warnings (v4.8-11811-g35ff96d)

[Mark Brown]

On Fri, Oct 14, 2016 at 10:12:04PM +0200, Arnd Bergmann wrote:
> On Friday, October 14, 2016 11:33:44 AM CEST Lee Jones wrote:

> > I'm waiting on a firm answer from Arnd and Thomas before applying
> > anything.

> I don't understand what the hardware does, so I'm not really
> able to answer this. Maybe just mark the driver as 'depends on
> BROKEN' for now?

There are a few boards using it, we should probably revert the broken
patches rather than just completely disable the driver.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 473 bytes
Desc: not available
URL: <http://lists.infradead.org/pipermail/linux-arm-kernel/attachments/20161017/3b8b7d09/attachment-0001.sig>

master build: 2 failures 4 warnings (v4.8-11811-g35ff96d)

[Thomas Gleixner]

On Fri, 14 Oct 2016, Lee Jones wrote:
> On Tue, 11 Oct 2016, Mark Brown wrote:
> 
> > On Tue, Oct 11, 2016 at 07:30:35AM +0100, Build bot for Mark Brown wrote:
> > 
> > Linus' tree is currently failing to build arm and arm64 allmodconfigs
> > with:
> > 
> > > 	arm64-allmodconfig
> > > ERROR: "irq_set_parent" [drivers/mfd/tps65217.ko] undefined!
> > 
> > > 	arm-allmodconfig
> > > ERROR: "irq_set_parent" [drivers/mfd/tps65217.ko] undefined!
> > 
> > due to 6556bdacf646fc (mfd: tps65217: Add support for IRQs) since
> > irq_set_parent() isn't exported.  This has been present in -next
> > for getting on for a month, a patch was proposed adding the relevant
> > export but that isn't present in -next yet.
> > 
> > The function is being used in order to enable lazy IRQ disabling for
> > threaded interrupts:
> > 
> > 	https://www.spinics.net/lists/arm-kernel/msg532864.html
> > 
> > What's the plan for getting this fixed in Linus' tree?
> 
> Here's the conversation:
> 
>   https://www.spinics.net/lists/arm-kernel/msg531850.html
> 
> I'm waiting on a firm answer from Arnd and Thomas before applying
> anything.

As I said, we can export it, if it's needed to make the driver modular. But
we definitely don't want to export it just to make the build happy.

Grigoryi has an excellent answer here:

   https://www.spinics.net/lists/arm-kernel/msg532864.html

Only the driver authors/maintainers can decide whether this is the case or
not.

Thanks,

	tglx