Re: [PATCH] fix gpu/drm/nouveau: fix return type in nouveau_dmem_migrate_to_ram()

["David Hildenbrand (Red Hat)" <david@kernel.org>]

On 13.11.25 21:51, Balbir Singh wrote:
> On 11/13/25 22:50, David Hildenbrand (Red Hat) wrote:
>> On 13.11.25 11:31, Balbir Singh wrote:
>>> On 11/13/25 20:09, David Hildenbrand (Red Hat) wrote:
>>>> On 13.11.25 06:13, Balbir Singh wrote:
>>>>> ret of type vm_fault_t is reused to capture the return value of
>>>>> nouveau_dmem_copy_folio(), which returns an int. Use a new copy_ret
>>>>> to fix the issue. The issue is not new, prior to this the function
>>>>> called was called nouveau_dmem_copy_one() and ret was used to capture
>>>>> it's value.
>>>>>
>>>>> Reported-by: kernel test robot <lkp@intel.com>
>>>>> Closes: https://lore.kernel.org/oe-kbuild-all/202511121922.oP20Lzb8-lkp@intel.com/
>>>>
>>>> It's a sparse warning, is there any runtime effect?
>>>>
>>>
>>> No impact really, it's just a sparse warning
>>
>> Okay, then please
>>
>> 1) Make that clear in the patch description
>>
>> 2) Make that clear in the patch subject.
>>
>> In particular, the current subject is weird. Should probably be
>>
>> "nouveau/dmem: fix sparse warning in nouveau_dmem_migrate_to_ram()"
>>
> 
> I did have "sparse warning" in the subject and checkpatch complained that
> tools should not be mentioned in the subject, but I'll give it a shot with that.

The important part is to be careful with the term "fix" when all we're 
fixing is some output of a static code analysis tool that does otherwise 
not result in any actual issues (build time, runtime).

-- 
Cheers

David