Russell King - ARM Linux | 29 Jun 2012 12:33
Picon

[BUILD BREAKAGE] versatile, LEDS updates

I've just added Versatile to the build system, and the result is not
encouraging.  The integration of the LEDS code into Versatile breaks
its build:

http://www.arm.linux.org.uk/developer/build/result.php?type=build&idx=1281

arch/arm/plat-versatile/leds.c: In function 'versatile_leds_init':
arch/arm/plat-versatile/leds.c:91: error: 'struct led_classdev' has no member named 'trigger_data'

This needs fixing before the LEDS code goes into mainline.

How well tested is the LEDS code?  Was it only tested with a set of
specifically updated configurations (it looks that way to me.)  Was
it ever tested with randconfig?
Russell King - ARM Linux | 29 Jun 2012 13:45
Picon

Re: [BUILD BREAKAGE] versatile, LEDS updates

On Fri, Jun 29, 2012 at 11:33:13AM +0100, Russell King - ARM Linux wrote:
> I've just added Versatile to the build system, and the result is not
> encouraging.  The integration of the LEDS code into Versatile breaks
> its build:
> 
> http://www.arm.linux.org.uk/developer/build/result.php?type=build&idx=1281
> 
> arch/arm/plat-versatile/leds.c: In function 'versatile_leds_init':
> arch/arm/plat-versatile/leds.c:91: error: 'struct led_classdev' has no member named 'trigger_data'
> 
> This needs fixing before the LEDS code goes into mainline.
> 
> How well tested is the LEDS code?  Was it only tested with a set of
> specifically updated configurations (it looks that way to me.)  Was
> it ever tested with randconfig?

BTW, this is the *EXACT* same error I reported on 3rd August 2011 with
the LEDS subsystem.

Arnd: please drop the LEDS updates from arm-soc until the point that I
raised 11 months ago is addressed and resolved.
Russell King - ARM Linux | 2 Jul 2012 10:39
Picon

Re: [BUILD BREAKAGE] versatile, LEDS updates

On Fri, Jun 29, 2012 at 12:45:32PM +0100, Russell King - ARM Linux wrote:
> On Fri, Jun 29, 2012 at 11:33:13AM +0100, Russell King - ARM Linux wrote:
> > I've just added Versatile to the build system, and the result is not
> > encouraging.  The integration of the LEDS code into Versatile breaks
> > its build:
> > 
> > http://www.arm.linux.org.uk/developer/build/result.php?type=build&idx=1281
> > 
> > arch/arm/plat-versatile/leds.c: In function 'versatile_leds_init':
> > arch/arm/plat-versatile/leds.c:91: error: 'struct led_classdev' has no member named 'trigger_data'
> > 
> > This needs fixing before the LEDS code goes into mainline.
> > 
> > How well tested is the LEDS code?  Was it only tested with a set of
> > specifically updated configurations (it looks that way to me.)  Was
> > it ever tested with randconfig?
> 
> BTW, this is the *EXACT* same error I reported on 3rd August 2011 with
> the LEDS subsystem.
> 
> Arnd: please drop the LEDS updates from arm-soc until the point that I
> raised 11 months ago is addressed and resolved.

Arnd,

You said on IRC that you would drop the LEDS patches until we had a working
version, but it still seems to be in arm-soc, and therefore the Versatile
build continues to fail.

When will you remove the LEDS patches?
(Continue reading)

Arnd Bergmann | 2 Jul 2012 17:40
Picon
Gravatar

Re: [BUILD BREAKAGE] versatile, LEDS updates

On Monday 02 July 2012, Russell King - ARM Linux wrote:
> 
> Arnd,
> 
> You said on IRC that you would drop the LEDS patches until we had a working
> version, but it still seems to be in arm-soc, and therefore the Versatile
> build continues to fail.
> 
> When will you remove the LEDS patches?
> 

Sorry for the delay, I've dropped them once more now.

Bryan,

Please fix this situation and address the problems that Russell has pointed
out multiple times if you want these patches to go into v3.6.

	Arnd

Gmane