Re: Build regressions/improvements in v3.0-rc5 (arch/cris)

[AmÃrico Wang]

On Thu, Jun 30, 2011 at 5:06 AM, Randy Dunlap <rdunlap@xxxxxxxxxxxx> wrote:
> On Wed, 29 Jun 2011 10:13:17 +0200 (CEST) Geert Uytterhoeven wrote:
>
>> *** ERRORS ***
>>
>> 698 regressions:
>
>> Â + net/mac80211/led.c: error: 'struct led_classdev' has no member named 'trig_list': Â=> 213, 277
>> Â + net/mac80211/led.c: error: dereferencing pointer to incomplete type: Â=> 73, 91, 100, 82
>> Â + net/mac80211/led.c: error: implicit declaration of function 'led_trigger_register': Â=> 74
>> Â + net/mac80211/led.c: error: implicit declaration of function 'led_trigger_unregister': Â=> 118
>> Â + net/mac80211/led.c: error: invalid application of 'sizeof' to incomplete type 'struct led_trigger' : Â=> 98, 71, 89, 80
>
>
> I think that this is what Andrew would call a cockup, or at least I would.
>
> These errors shouldn't be possible since net/mac80211/Kconfig says:
>
> config MAC80211_LEDS
> Â Â Â Âbool "Enable LED triggers"
> Â Â Â Âdepends on MAC80211
> Â Â Â Âdepends on LEDS_CLASS
> Â Â Â Âselect LEDS_TRIGGERS
>
>
>
> Preferably arch/cris/Kconfig would source drivers/Kconfig instead of sourcing
> individual drivers/foobar/Kconfig.
>
> Second option is for arch/cris/Kconfig to source drivers/leds/Kconfig
> unless there is some reason to avoid this (?).
>
> Or we should just prevent this particular driver from being built,
> but that's the least desirable choice here.

For me, net/mac80211/Kconfig should source drivers/leds/Kconfig
since it uses a Kconfig defined there...
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/