With ARCH=arm64, make allmodconfig && make W=1 C=1 reports:
WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/edac/layerscape_edac_mod.o

Add the missing invocation of the MODULE_DESCRIPTION() macro.

Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
---
 drivers/edac/layerscape_edac.c | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/edac/layerscape_edac.c b/drivers/edac/layerscape_edac.c
index d2f895033280..b70d5d258fcb 100644
--- a/drivers/edac/layerscape_edac.c
+++ b/drivers/edac/layerscape_edac.c
@@ -69,6 +69,7 @@ static void __exit fsl_ddr_mc_exit(void)
 
 module_exit(fsl_ddr_mc_exit);
 
+MODULE_DESCRIPTION("Freescale Layerscape EDAC module");
 MODULE_LICENSE("GPL");
 MODULE_AUTHOR("NXP Semiconductor");
 module_param(edac_op_state, int, 0444);

---
base-commit: 83a7eefedc9b56fe7bfeff13b6c7356688ffa670
change-id: 20240613-md-arm64-drivers-edac-2fa423340d75

On Thu, Jun 13, 2024 at 02:36:21PM -0700, Jeff Johnson wrote:
> With ARCH=arm64, make allmodconfig && make W=1 C=1 reports:
> WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/edac/layerscape_edac_mod.o
> 
> Add the missing invocation of the MODULE_DESCRIPTION() macro.
> 
> Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
> ---
>  drivers/edac/layerscape_edac.c | 1 +
>  1 file changed, 1 insertion(+)
> 
> diff --git a/drivers/edac/layerscape_edac.c b/drivers/edac/layerscape_edac.c
> index d2f895033280..b70d5d258fcb 100644
> --- a/drivers/edac/layerscape_edac.c
> +++ b/drivers/edac/layerscape_edac.c
> @@ -69,6 +69,7 @@ static void __exit fsl_ddr_mc_exit(void)
>  
>  module_exit(fsl_ddr_mc_exit);
>  
> +MODULE_DESCRIPTION("Freescale Layerscape EDAC module");
>  MODULE_LICENSE("GPL");
>  MODULE_AUTHOR("NXP Semiconductor");
>  module_param(edac_op_state, int, 0444);
> 
> ---

$ git grep -E "MODULE_(DESCRIPTION|LICENSE)" drivers/edac/

I'd expect to see regular pairs like this:

drivers/edac/al_mc_edac.c:348:MODULE_LICENSE("GPL v2");
drivers/edac/al_mc_edac.c:350:MODULE_DESCRIPTION("Amazon's Annapurna Lab's Memory Controller EDAC Driver");

drivers/edac/altera_edac.c:2216:MODULE_DESCRIPTION("EDAC Driver for Altera Memories");

drivers/edac/amd64_edac.c:4238:MODULE_LICENSE("GPL");
drivers/edac/amd64_edac.c:4240:MODULE_DESCRIPTION("MC support for AMD64 memory controllers");
...

but there are cases which need fixing.

How about you do them all with one patch?

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

On 6/16/2024 8:43 AM, Borislav Petkov wrote:
> On Thu, Jun 13, 2024 at 02:36:21PM -0700, Jeff Johnson wrote:
>> With ARCH=arm64, make allmodconfig && make W=1 C=1 reports:
>> WARNING: modpost: missing MODULE_DESCRIPTION() in drivers/edac/layerscape_edac_mod.o
>>
>> Add the missing invocation of the MODULE_DESCRIPTION() macro.
>>
>> Signed-off-by: Jeff Johnson <quic_jjohnson@quicinc.com>
>> ---
>>  drivers/edac/layerscape_edac.c | 1 +
>>  1 file changed, 1 insertion(+)
>>
>> diff --git a/drivers/edac/layerscape_edac.c b/drivers/edac/layerscape_edac.c
>> index d2f895033280..b70d5d258fcb 100644
>> --- a/drivers/edac/layerscape_edac.c
>> +++ b/drivers/edac/layerscape_edac.c
>> @@ -69,6 +69,7 @@ static void __exit fsl_ddr_mc_exit(void)
>>  
>>  module_exit(fsl_ddr_mc_exit);
>>  
>> +MODULE_DESCRIPTION("Freescale Layerscape EDAC module");
>>  MODULE_LICENSE("GPL");
>>  MODULE_AUTHOR("NXP Semiconductor");
>>  module_param(edac_op_state, int, 0444);
>>
>> ---
> 
> $ git grep -E "MODULE_(DESCRIPTION|LICENSE)" drivers/edac/
> 
> I'd expect to see regular pairs like this:
> 
> drivers/edac/al_mc_edac.c:348:MODULE_LICENSE("GPL v2");
> drivers/edac/al_mc_edac.c:350:MODULE_DESCRIPTION("Amazon's Annapurna Lab's Memory Controller EDAC Driver");
> 
> drivers/edac/altera_edac.c:2216:MODULE_DESCRIPTION("EDAC Driver for Altera Memories");
> 
> drivers/edac/amd64_edac.c:4238:MODULE_LICENSE("GPL");
> drivers/edac/amd64_edac.c:4240:MODULE_DESCRIPTION("MC support for AMD64 memory controllers");
> ...
> 
> but there are cases which need fixing.
> 
> How about you do them all with one patch?

My process has been, for the most part, to first fix the ones where I actually
observe the warning, unless there is just one or two others. For drivers/edac
there are more than a couple more that have a LICENSE but not a DESCRIPTION:
drivers/edac/mpc85xx_edac.c
drivers/edac/octeon_edac-l2c.c
drivers/edac/octeon_edac-lmc.c
drivers/edac/octeon_edac-pc.c
drivers/edac/octeon_edac-pci.c

So my preference is to first fix the one where I actually observed the
warning, and then later fix the ones which currently don't seem to produce a
warning. But a can make an exception and fix all of them in drivers/edac.

Also note I haven't even considered doing anything for the ones that have a
DESCRIPTION but not a LICENSE such as drivers/edac/altera_edac.c. Note that a
missing LICENSE would result in a build failure, not just a warning, so the
appropriate thing to do in that case is probably to remove the DESCRIPTION. It
has been enough of a job to fix the missing DESCRIPTIONs that actually
generate warnings (I've been making changes tree-wide for over a month,
touching almost 800 files). So I prefer to let others worry about removing
DESCRIPTION/LICENSE found in files that cannot be built as modules.

/jeff

On Sun, Jun 16, 2024 at 06:43:58PM -0700, Jeff Johnson wrote:
> My process has been, for the most part, to first fix the ones where I actually
> observe the warning, unless there is just one or two others. For drivers/edac
> there are more than a couple more that have a LICENSE but not a DESCRIPTION:

And my process is not to do excessive work of collecting silly oneliners because
some tool complains. Send me a whole patch which addresses *all* issues you've
found in drivers/edac/ or keep sending them one by one but I'll merge them all
into one. In any case, I won't do piecemeal silly oneliners.

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette