Re: [PATCH -next] irq/qcom-mpm: Fix build error without MAILBOX

[Marc Zyngier <maz@kernel.org>]

On 2022-03-17 13:59, Shawn Guo wrote:
> On Thu, Mar 17, 2022 at 09:19:56PM +0800, YueHaibing wrote:
>> If MAILBOX is n, building fails:
>> 
>> drivers/irqchip/irq-qcom-mpm.o: In function `mpm_pd_power_off':
>> irq-qcom-mpm.c:(.text+0x174): undefined reference to 
>> `mbox_send_message'
>> irq-qcom-mpm.c:(.text+0x174): relocation truncated to fit: 
>> R_AARCH64_CALL26 against undefined symbol `mbox_send_message'
>> 
>> Make QCOM_MPM depends on MAILBOX to fix this.
>> 
>> Fixes: a6199bb514d8 ("irqchip: Add Qualcomm MPM controller driver")
>> Signed-off-by: YueHaibing <yuehaibing@huawei.com>
> 
> Acked-by: Shawn Guo <shawn.guo@linaro.org>
> 
>> ---
>>  drivers/irqchip/Kconfig | 1 +
>>  1 file changed, 1 insertion(+)
>> 
>> diff --git a/drivers/irqchip/Kconfig b/drivers/irqchip/Kconfig
>> index 680d2fcf2686..15edb9a6fcae 100644
>> --- a/drivers/irqchip/Kconfig
>> +++ b/drivers/irqchip/Kconfig
>> @@ -433,6 +433,7 @@ config QCOM_PDC
>>  config QCOM_MPM
>>  	tristate "QCOM MPM"
>>  	depends on ARCH_QCOM
>> +	depends on MAILBOX
>>  	select IRQ_DOMAIN_HIERARCHY
>>  	help
>>  	  MSM Power Manager driver to manage and configure wakeup
>> --
>> 2.17.1
>> 

Hmmm... Even if that fix the build, this still isn't correct, is it?
You also need the mailbox *driver* to be selected. How is that
dependency enforced?

I'll queue this to fix -next, but this really is papering over
a bigger problem.

         M.
-- 
Jazz is not dead. It just smells funny...