diff mbox series

MIPS: ralink: of: fix build of rt2880_wdt watchdog module

Message ID 20210514131750.52867-1-krzysztof.kozlowski@canonical.com
State New
Headers show
Series MIPS: ralink: of: fix build of rt2880_wdt watchdog module | expand

Commit Message

Krzysztof Kozlowski May 14, 2021, 1:17 p.m. UTC
When rt2880_wdt watchdog driver is built as a module, the
rt_sysc_membase needs to be exported (it is being used via inlined
rt_sysc_r32):

  ERROR: modpost: "rt_sysc_membase" [drivers/watchdog/rt2880_wdt.ko] undefined!

Reported-by: kernel test robot <lkp@intel.com>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
---
 arch/mips/ralink/of.c | 2 ++
 1 file changed, 2 insertions(+)

Comments

Guenter Roeck May 14, 2021, 1:48 p.m. UTC | #1
On 5/14/21 6:17 AM, Krzysztof Kozlowski wrote:
> When rt2880_wdt watchdog driver is built as a module, the
> rt_sysc_membase needs to be exported (it is being used via inlined
> rt_sysc_r32):
> 
>    ERROR: modpost: "rt_sysc_membase" [drivers/watchdog/rt2880_wdt.ko] undefined!
> 
> Reported-by: kernel test robot <lkp@intel.com>
> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>

Acked-by: Guenter Roeck <linux@roeck-us.net>

I don't see a recent change in the code. Has that problem been there
all along ?

Thanks,
Guenter

> ---
>   arch/mips/ralink/of.c | 2 ++
>   1 file changed, 2 insertions(+)
> 
> diff --git a/arch/mips/ralink/of.c b/arch/mips/ralink/of.c
> index 0c5de07da097..b3ce706426c4 100644
> --- a/arch/mips/ralink/of.c
> +++ b/arch/mips/ralink/of.c
> @@ -24,6 +24,8 @@
>   #include "common.h"
>   
>   __iomem void *rt_sysc_membase;
> +EXPORT_SYMBOL_GPL(rt_sysc_membase);
> +
>   __iomem void *rt_memc_membase;
>   
>   __iomem void *plat_of_remap_node(const char *node)
>
John Crispin May 14, 2021, 6:31 p.m. UTC | #2
On 14.05.21 15:48, Guenter Roeck wrote:
> On 5/14/21 6:17 AM, Krzysztof Kozlowski wrote:
>> When rt2880_wdt watchdog driver is built as a module, the
>> rt_sysc_membase needs to be exported (it is being used via inlined
>> rt_sysc_r32):
>>
>>    ERROR: modpost: "rt_sysc_membase" [drivers/watchdog/rt2880_wdt.ko] 
>> undefined!
>>
>> Reported-by: kernel test robot <lkp@intel.com>
>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>
>
> Acked-by: Guenter Roeck <linux@roeck-us.net>
>
> I don't see a recent change in the code. Has that problem been there
> all along ?
>
>
yes it has ...

Acked-by: John Crispin <john@phrozen.org>
Krzysztof Kozlowski July 5, 2021, 3:15 p.m. UTC | #3
On 14/05/2021 16:04, Krzysztof Kozlowski wrote:
> On 14/05/2021 09:48, Guenter Roeck wrote:

>> On 5/14/21 6:17 AM, Krzysztof Kozlowski wrote:

>>> When rt2880_wdt watchdog driver is built as a module, the

>>> rt_sysc_membase needs to be exported (it is being used via inlined

>>> rt_sysc_r32):

>>>

>>>    ERROR: modpost: "rt_sysc_membase" [drivers/watchdog/rt2880_wdt.ko] undefined!

>>>

>>> Reported-by: kernel test robot <lkp@intel.com>

>>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>

>>

>> Acked-by: Guenter Roeck <linux@roeck-us.net>

>>

>> I don't see a recent change in the code. Has that problem been there

>> all along ?

> 

> I think the problem was there always but 0-day builder did not hit it

> until recently:

> https://lore.kernel.org/lkml/202105082122.yrF7ploN-lkp@intel.com/


Hi Guenter,

It seems that patch was not applied. Anything to fix here?


Best regards,
Krzysztof
Guenter Roeck July 5, 2021, 4:08 p.m. UTC | #4
On 7/5/21 8:15 AM, Krzysztof Kozlowski wrote:
> On 14/05/2021 16:04, Krzysztof Kozlowski wrote:

>> On 14/05/2021 09:48, Guenter Roeck wrote:

>>> On 5/14/21 6:17 AM, Krzysztof Kozlowski wrote:

>>>> When rt2880_wdt watchdog driver is built as a module, the

>>>> rt_sysc_membase needs to be exported (it is being used via inlined

>>>> rt_sysc_r32):

>>>>

>>>>     ERROR: modpost: "rt_sysc_membase" [drivers/watchdog/rt2880_wdt.ko] undefined!

>>>>

>>>> Reported-by: kernel test robot <lkp@intel.com>

>>>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>

>>>

>>> Acked-by: Guenter Roeck <linux@roeck-us.net>

>>>

>>> I don't see a recent change in the code. Has that problem been there

>>> all along ?

>>

>> I think the problem was there always but 0-day builder did not hit it

>> until recently:

>> https://lore.kernel.org/lkml/202105082122.yrF7ploN-lkp@intel.com/

> 

> Hi Guenter,

> 

> It seems that patch was not applied. Anything to fix here?

> 

I see commit fef532ea0cd8 ("MIPS: ralink: export rt_sysc_membase for
rt2880_wdt.c"). Does that not solve the problem ?

Guenter
Krzysztof Kozlowski July 5, 2021, 4:21 p.m. UTC | #5
On Mon, 5 Jul 2021 at 18:08, Guenter Roeck <linux@roeck-us.net> wrote:
>

> On 7/5/21 8:15 AM, Krzysztof Kozlowski wrote:

> > On 14/05/2021 16:04, Krzysztof Kozlowski wrote:

> >> On 14/05/2021 09:48, Guenter Roeck wrote:

> >>> On 5/14/21 6:17 AM, Krzysztof Kozlowski wrote:

> >>>> When rt2880_wdt watchdog driver is built as a module, the

> >>>> rt_sysc_membase needs to be exported (it is being used via inlined

> >>>> rt_sysc_r32):

> >>>>

> >>>>     ERROR: modpost: "rt_sysc_membase" [drivers/watchdog/rt2880_wdt.ko] undefined!

> >>>>

> >>>> Reported-by: kernel test robot <lkp@intel.com>

> >>>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>

> >>>

> >>> Acked-by: Guenter Roeck <linux@roeck-us.net>

> >>>

> >>> I don't see a recent change in the code. Has that problem been there

> >>> all along ?

> >>

> >> I think the problem was there always but 0-day builder did not hit it

> >> until recently:

> >> https://lore.kernel.org/lkml/202105082122.yrF7ploN-lkp@intel.com/

> >

> > Hi Guenter,

> >

> > It seems that patch was not applied. Anything to fix here?

> >

> I see commit fef532ea0cd8 ("MIPS: ralink: export rt_sysc_membase for

> rt2880_wdt.c"). Does that not solve the problem ?


Yes, that fixes it. It was sent later, though. :(

Anyway, thanks!

Best regards,
Krzysztof
Guenter Roeck July 5, 2021, 4:32 p.m. UTC | #6
On 7/5/21 9:21 AM, Krzysztof Kozlowski wrote:
> On Mon, 5 Jul 2021 at 18:08, Guenter Roeck <linux@roeck-us.net> wrote:

>>

>> On 7/5/21 8:15 AM, Krzysztof Kozlowski wrote:

>>> On 14/05/2021 16:04, Krzysztof Kozlowski wrote:

>>>> On 14/05/2021 09:48, Guenter Roeck wrote:

>>>>> On 5/14/21 6:17 AM, Krzysztof Kozlowski wrote:

>>>>>> When rt2880_wdt watchdog driver is built as a module, the

>>>>>> rt_sysc_membase needs to be exported (it is being used via inlined

>>>>>> rt_sysc_r32):

>>>>>>

>>>>>>      ERROR: modpost: "rt_sysc_membase" [drivers/watchdog/rt2880_wdt.ko] undefined!

>>>>>>

>>>>>> Reported-by: kernel test robot <lkp@intel.com>

>>>>>> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com>

>>>>>

>>>>> Acked-by: Guenter Roeck <linux@roeck-us.net>

>>>>>

>>>>> I don't see a recent change in the code. Has that problem been there

>>>>> all along ?

>>>>

>>>> I think the problem was there always but 0-day builder did not hit it

>>>> until recently:

>>>> https://lore.kernel.org/lkml/202105082122.yrF7ploN-lkp@intel.com/

>>>

>>> Hi Guenter,

>>>

>>> It seems that patch was not applied. Anything to fix here?

>>>

>> I see commit fef532ea0cd8 ("MIPS: ralink: export rt_sysc_membase for

>> rt2880_wdt.c"). Does that not solve the problem ?

> 

> Yes, that fixes it. It was sent later, though. :(

> 


Happens to me all the time. What is important is that things are getting
fixed. Personally I don't care that much for the credit.

Note though that your patch subject was a bit misleading, and it may well
be possible that the mips maintainers didn't realize that the fix needed
to be applied to the mips tree. The subject line in the applied patch
describes the change much better as it focuses on the change (what is done),
while the subject line in your patch focuses on the impact (what it fixes).

Guenter
diff mbox series

Patch

diff --git a/arch/mips/ralink/of.c b/arch/mips/ralink/of.c
index 0c5de07da097..b3ce706426c4 100644
--- a/arch/mips/ralink/of.c
+++ b/arch/mips/ralink/of.c
@@ -24,6 +24,8 @@ 
 #include "common.h"
 
 __iomem void *rt_sysc_membase;
+EXPORT_SYMBOL_GPL(rt_sysc_membase);
+
 __iomem void *rt_memc_membase;
 
 __iomem void *plat_of_remap_node(const char *node)