Re: mmotm 2008-12-24-01-20 uploaded (ath9k)

["Stefanik Gábor" <netrolller.3d@gmail.com>]

On Wed, Dec 24, 2008 at 10:43 PM, Randy Dunlap <randy.dunlap@oracle.com> wrote:
> Luis R. Rodriguez wrote:
>> On Wed, Dec 24, 2008 at 10:04:43AM -0800, Randy Dunlap wrote:
>>> akpm@linux-foundation.org wrote:
>>>> The mm-of-the-moment snapshot 2008-12-24-01-20 has been uploaded to
>>>>
>>>>    http://userweb.kernel.org/~akpm/mmotm/
>>>>
>>>> and will soon be available at
>>>>
>>>>    git://git.zen-sources.org/zen/mmotm.git
>>>
>>> ath9k build fails with:
>>>
>>> drivers/built-in.o: In function `ath_detach':
>>> main.c:(.text+0xee655): undefined reference to `rfkill_unregister'
>>> drivers/built-in.o: In function `ath_rfkill_poll':
>>> main.c:(.text+0xeedfa): undefined reference to `rfkill_force_state'
>>> drivers/built-in.o: In function `ath9k_start':
>>> main.c:(.text+0xefe79): undefined reference to `rfkill_register'
>>> main.c:(.text+0xeff20): undefined reference to `rfkill_free'
>>> drivers/built-in.o: In function `ath_pci_probe':
>>> main.c:(.text+0xf0f73): undefined reference to `rfkill_allocate'
>>>
>>> when ath9k is built-in and CONFIG_RFKILL=m.
>>>
>>> config attached.
>>
>> <-- snip -->
>>
>>> # CONFIG_MAC80211_DEBUGFS is not set
>>> # CONFIG_MAC80211_DEBUG_MENU is not set
>>> CONFIG_RFKILL=m
>>> CONFIG_RFKILL_INPUT=m
>>> CONFIG_RFKILL_LEDS=y
>>
>> Try this (already merged into wireless-testing):
>
> This seems to be already merged/used in mmotm and that's where the
> build failure happened.
>
> And this patch does not address how to call rfkill interfaces when
> CONFIG_RFKILL=m and ath9k is built into the kernel image...
>

IMO we should not allow compiling-in ath9k when rfkill is built as a
module, similar to how we do for mac80211 (CONFIG_MAC80211=m requires
all drivers to be modules).

-- 
Vista: [V]iruses, [I]ntruders, [S]pyware, [T]rojans and [A]dware. :-)