Since commit 8b41fc4454e ("kbuild: create modules.builtin without
Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
are used to identify modules. As a consequence, uses of the macro
in non-modules will cause modprobe to misidentify their containing
object file as a module when it is not (false positives), and modprobe
might succeed rather than failing with a suitable error message.
So remove it in the files in this commit, none of which can be built as
modules.
Signed-off-by: Nick Alcock <[email protected]>
Suggested-by: Luis Chamberlain <[email protected]>
Cc: Luis Chamberlain <[email protected]>
Cc: [email protected]
Cc: [email protected]
Cc: Hitomi Hasegawa <[email protected]>
Cc: Krzysztof Kozlowski <[email protected]>
Cc: Thierry Reding <[email protected]>
Cc: Jonathan Hunter <[email protected]>
Cc: [email protected]
---
drivers/memory/tegra/tegra186-emc.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/drivers/memory/tegra/tegra186-emc.c b/drivers/memory/tegra/tegra186-emc.c
index 26e763bde92a..e935ad4e95b6 100644
--- a/drivers/memory/tegra/tegra186-emc.c
+++ b/drivers/memory/tegra/tegra186-emc.c
@@ -280,4 +280,3 @@ module_platform_driver(tegra186_emc_driver);
MODULE_AUTHOR("Thierry Reding <[email protected]>");
MODULE_DESCRIPTION("NVIDIA Tegra186 External Memory Controller driver");
-MODULE_LICENSE("GPL v2");
--
2.39.1.268.g9de2f9a303
On 24/02/2023 16:07, Nick Alcock wrote:
> Since commit 8b41fc4454e ("kbuild: create modules.builtin without
> Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
> are used to identify modules. As a consequence, uses of the macro
> in non-modules will cause modprobe to misidentify their containing
> object file as a module when it is not (false positives), and modprobe
> might succeed rather than failing with a suitable error message.
>
> So remove it in the files in this commit, none of which can be built as
> modules.
>
Applied as well. Squashed with next one. Same subject messes with b4, so
prefix should be corrected if these were to stay as separate patches.
Best regards,
Krzysztof
On 06/03/2023 15:30, Krzysztof Kozlowski wrote:
> On 24/02/2023 16:07, Nick Alcock wrote:
>> Since commit 8b41fc4454e ("kbuild: create modules.builtin without
>> Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
>> are used to identify modules. As a consequence, uses of the macro
>> in non-modules will cause modprobe to misidentify their containing
>> object file as a module when it is not (false positives), and modprobe
>> might succeed rather than failing with a suitable error message.
>>
>> So remove it in the files in this commit, none of which can be built as
>> modules.
>>
>
> Applied as well. Squashed with next one. Same subject messes with b4, so
> prefix should be corrected if these were to stay as separate patches.
>
And all dropped. Run checkpatch before sending patches.
Best regards,
Krzysztof
On 06/03/2023 18:13, Nick Alcock wrote:
> On 6 Mar 2023, Krzysztof Kozlowski told this:
>
>> On 06/03/2023 15:30, Krzysztof Kozlowski wrote:
>>> On 24/02/2023 16:07, Nick Alcock wrote:
>>>> Since commit 8b41fc4454e ("kbuild: create modules.builtin without
>>>> Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
>>>> are used to identify modules. As a consequence, uses of the macro
>>>> in non-modules will cause modprobe to misidentify their containing
>>>> object file as a module when it is not (false positives), and modprobe
>>>> might succeed rather than failing with a suitable error message.
>>>>
>>>> So remove it in the files in this commit, none of which can be built as
>>>> modules.
>>>
>>> Applied as well. Squashed with next one. Same subject messes with b4, so
>>> prefix should be corrected if these were to stay as separate patches.
>>
>> And all dropped. Run checkpatch before sending patches.
>
> So... which of the 27 patches n this series is being dropped? It would
> also be nice to know what the checkpatch problems were, because all I
> can see from checkpatch is one error per patch, an apparent false
The ones I responded that I applied - so the memory controller ones.
> positive:
>
> ERROR: Please use git commit description style 'commit <12+ chars of sha1> ("<title line>")' - ie: 'commit 8b41fc4454e3 ("kbuild: create modules.builtin without Makefile.modbuiltin or tristate.conf")'
>
> The commit message says:
>
>> Since commit 8b41fc4454e ("kbuild: create modules.builtin without
>> Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
>
> ... which unless I am seeing things is precisely what checkpatch.pl is
> asking for. Perhaps it can't handle wrapped lines, but not wrapping a
> git commit log line *also* (correctly) elicits a checkpatch error...
It's not a false positive, but an error I am asking fix (assuming there
are no more ones) There is no problem with wrapping lines - checkpatch
handles it.
Best regards,
Krzysztof
On 6 Mar 2023, Krzysztof Kozlowski told this:
> On 06/03/2023 15:30, Krzysztof Kozlowski wrote:
>> On 24/02/2023 16:07, Nick Alcock wrote:
>>> Since commit 8b41fc4454e ("kbuild: create modules.builtin without
>>> Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
>>> are used to identify modules. As a consequence, uses of the macro
>>> in non-modules will cause modprobe to misidentify their containing
>>> object file as a module when it is not (false positives), and modprobe
>>> might succeed rather than failing with a suitable error message.
>>>
>>> So remove it in the files in this commit, none of which can be built as
>>> modules.
>>
>> Applied as well. Squashed with next one. Same subject messes with b4, so
>> prefix should be corrected if these were to stay as separate patches.
>
> And all dropped. Run checkpatch before sending patches.
So... which of the 27 patches n this series is being dropped? It would
also be nice to know what the checkpatch problems were, because all I
can see from checkpatch is one error per patch, an apparent false
positive:
ERROR: Please use git commit description style 'commit <12+ chars of sha1> ("<title line>")' - ie: 'commit 8b41fc4454e3 ("kbuild: create modules.builtin without Makefile.modbuiltin or tristate.conf")'
The commit message says:
> Since commit 8b41fc4454e ("kbuild: create modules.builtin without
> Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
... which unless I am seeing things is precisely what checkpatch.pl is
asking for. Perhaps it can't handle wrapped lines, but not wrapping a
git commit log line *also* (correctly) elicits a checkpatch error...
On 6 Mar 2023, Krzysztof Kozlowski stated:
> On 06/03/2023 18:13, Nick Alcock wrote:
>> On 6 Mar 2023, Krzysztof Kozlowski told this:
>>
>>> On 06/03/2023 15:30, Krzysztof Kozlowski wrote:
>>>> On 24/02/2023 16:07, Nick Alcock wrote:
>>>>> Since commit 8b41fc4454e ("kbuild: create modules.builtin without
>>>>> Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
>>>>> are used to identify modules. As a consequence, uses of the macro
>>>>> in non-modules will cause modprobe to misidentify their containing
>>>>> object file as a module when it is not (false positives), and modprobe
>>>>> might succeed rather than failing with a suitable error message.
>>>>>
>>>>> So remove it in the files in this commit, none of which can be built as
>>>>> modules.
>>>>
>>>> Applied as well. Squashed with next one. Same subject messes with b4, so
>>>> prefix should be corrected if these were to stay as separate patches.
>>>
>>> And all dropped. Run checkpatch before sending patches.
>>
>> So... which of the 27 patches n this series is being dropped? It would
>> also be nice to know what the checkpatch problems were, because all I
>> can see from checkpatch is one error per patch, an apparent false
>
> The ones I responded that I applied - so the memory controller ones.
OK, resent those three (now two), checkpatch-clean and fused identical
subjects together. (Not adjusted non-memory-controller patch commit
logs, to avoid causing disruption with those that are already flowing
into the tree.)
Hope this is better :)
--
NULL && (void)
On 08/03/2023 21:25, Nick Alcock wrote:
> On 6 Mar 2023, Krzysztof Kozlowski stated:
>
>> On 06/03/2023 18:13, Nick Alcock wrote:
>>> On 6 Mar 2023, Krzysztof Kozlowski told this:
>>>
>>>> On 06/03/2023 15:30, Krzysztof Kozlowski wrote:
>>>>> On 24/02/2023 16:07, Nick Alcock wrote:
>>>>>> Since commit 8b41fc4454e ("kbuild: create modules.builtin without
>>>>>> Makefile.modbuiltin or tristate.conf"), MODULE_LICENSE declarations
>>>>>> are used to identify modules. As a consequence, uses of the macro
>>>>>> in non-modules will cause modprobe to misidentify their containing
>>>>>> object file as a module when it is not (false positives), and modprobe
>>>>>> might succeed rather than failing with a suitable error message.
>>>>>>
>>>>>> So remove it in the files in this commit, none of which can be built as
>>>>>> modules.
>>>>>
>>>>> Applied as well. Squashed with next one. Same subject messes with b4, so
>>>>> prefix should be corrected if these were to stay as separate patches.
>>>>
>>>> And all dropped. Run checkpatch before sending patches.
>>>
>>> So... which of the 27 patches n this series is being dropped? It would
>>> also be nice to know what the checkpatch problems were, because all I
>>> can see from checkpatch is one error per patch, an apparent false
>>
>> The ones I responded that I applied - so the memory controller ones.
>
> OK, resent those three (now two), checkpatch-clean and fused identical
> subjects together. (Not adjusted non-memory-controller patch commit
> logs, to avoid causing disruption with those that are already flowing
> into the tree.)
>
> Hope this is better :)
If you send a new patch, mark it appropriately and include changelog.
Best regards,
Krzysztof