[PATCH 1/1] checkpatch: add MA_STATE to declaration_macros

[PATCH 1/1] checkpatch: add MA_STATE to declaration_macros

[Jim Cromie]

MA_STATE is used repeatedly as a declaration, add it to the list.

I left out the $Storage regex component, since I saw no cases wanting
it, via: $> grep -r -B1 '\bMA_STATE' .

This change fixes the following, reproduced locally.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202511071306.OJpTf7fK-lkp@intel.com/

Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
---
 scripts/checkpatch.pl | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
index e722dd6fa8ef..4b5e2e64dece 100755
--- a/scripts/checkpatch.pl
+++ b/scripts/checkpatch.pl
@@ -1096,7 +1096,8 @@ our $declaration_macros = qr{(?x:
 	(?:$Storage\s+)?(?:[A-Z_][A-Z0-9]*_){0,2}(?:DEFINE|DECLARE)(?:_[A-Z0-9]+){1,6}\s*\(|
 	(?:$Storage\s+)?[HLP]?LIST_HEAD\s*\(|
 	(?:SKCIPHER_REQUEST|SHASH_DESC|AHASH_REQUEST)_ON_STACK\s*\(|
-	(?:$Storage\s+)?(?:XA_STATE|XA_STATE_ORDER)\s*\(
+	(?:$Storage\s+)?(?:XA_STATE|XA_STATE_ORDER)\s*\(|
+	(?:MA_STATE)\s*\(
 )};
 
 our %allow_repeated_words = (
-- 
2.51.1

Re: [PATCH 1/1] checkpatch: add MA_STATE to declaration_macros

[Joe Perches]

On Sat, 2025-11-08 at 16:47 -0700, Jim Cromie wrote:
> MA_STATE is used repeatedly as a declaration, add it to the list.

OK but

> I left out the $Storage regex component, since I saw no cases wanting
> it, via: $> grep -r -B1 '\bMA_STATE' .
> 
> This change fixes the following, reproduced locally.

No following ?

And I looked at the MA_STATE macro definition that has
several NULL or zero unnecessary member initializations
but doesn't initialize the entire structure as all
non-named members are NULL or zero by default.

I wonder if the unnecessary NULL/0 initializations there
could actually increase the object size.

With gcc 15.2.1 allyesconfig & defconfig do not change
either way but I don't know if all supported versions
minimize the initialization properly.

> Reported-by: kernel test robot <lkp@intel.com>
> Closes: https://lore.kernel.org/oe-kbuild-all/202511071306.OJpTf7fK-lkp@intel.com/

This Closes: url is not found on lore

> 
> Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
> ---
>  scripts/checkpatch.pl | 3 ++-
>  1 file changed, 2 insertions(+), 1 deletion(-)
> 
> diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
> index e722dd6fa8ef..4b5e2e64dece 100755
> --- a/scripts/checkpatch.pl
> +++ b/scripts/checkpatch.pl
> @@ -1096,7 +1096,8 @@ our $declaration_macros = qr{(?x:
>  	(?:$Storage\s+)?(?:[A-Z_][A-Z0-9]*_){0,2}(?:DEFINE|DECLARE)(?:_[A-Z0-9]+){1,6}\s*\(|
>  	(?:$Storage\s+)?[HLP]?LIST_HEAD\s*\(|
>  	(?:SKCIPHER_REQUEST|SHASH_DESC|AHASH_REQUEST)_ON_STACK\s*\(|
> -	(?:$Storage\s+)?(?:XA_STATE|XA_STATE_ORDER)\s*\(
> +	(?:$Storage\s+)?(?:XA_STATE|XA_STATE_ORDER)\s*\(|
> +	(?:MA_STATE)\s*\(
>  )};
>  
>  our %allow_repeated_words = (

Re: [PATCH 1/1] checkpatch: add MA_STATE to declaration_macros

[jim.cromie]

On Sat, Nov 8, 2025 at 5:52 PM Joe Perches <joe@perches.com> wrote:
>
> On Sat, 2025-11-08 at 16:47 -0700, Jim Cromie wrote:
> > MA_STATE is used repeatedly as a declaration, add it to the list.
>
> OK but
>
> > I left out the $Storage regex component, since I saw no cases wanting
> > it, via: $> grep -r -B1 '\bMA_STATE' .
> >
> > This change fixes the following, reproduced locally.
>
> No following ?

the oe-kbuild report followed, but the link didnt work right..

>
> And I looked at the MA_STATE macro definition that has
> several NULL or zero unnecessary member initializations
> but doesn't initialize the entire structure as all
> non-named members are NULL or zero by default.
>
> I wonder if the unnecessary NULL/0 initializations there
> could actually increase the object size.
>
> With gcc 15.2.1 allyesconfig & defconfig do not change
> either way but I don't know if all supported versions
> minimize the initialization properly.
>
> > Reported-by: kernel test robot <lkp@intel.com>
> > Closes: https://lore.kernel.org/oe-kbuild-all/202511071306.OJpTf7fK-lkp@intel.com/
>
> This Closes: url is not found on lore

That is weird.
when I follow the Closes link, I get something unexpected - a "not
found - look here maybe",
AND a ../oe-kbuild/ link -
clicking that takes me to the report proper.

If you retry, you can edit the "-all" out of the URL and it works.

It sounds like a small bug somewhere in the lkp-robot CI chain.


>
> >
> > Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
> > ---
> >  scripts/checkpatch.pl | 3 ++-
> >  1 file changed, 2 insertions(+), 1 deletion(-)
> >
> > diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
> > index e722dd6fa8ef..4b5e2e64dece 100755
> > --- a/scripts/checkpatch.pl
> > +++ b/scripts/checkpatch.pl
> > @@ -1096,7 +1096,8 @@ our $declaration_macros = qr{(?x:
> >       (?:$Storage\s+)?(?:[A-Z_][A-Z0-9]*_){0,2}(?:DEFINE|DECLARE)(?:_[A-Z0-9]+){1,6}\s*\(|
> >       (?:$Storage\s+)?[HLP]?LIST_HEAD\s*\(|
> >       (?:SKCIPHER_REQUEST|SHASH_DESC|AHASH_REQUEST)_ON_STACK\s*\(|
> > -     (?:$Storage\s+)?(?:XA_STATE|XA_STATE_ORDER)\s*\(
> > +     (?:$Storage\s+)?(?:XA_STATE|XA_STATE_ORDER)\s*\(|
> > +     (?:MA_STATE)\s*\(
> >  )};
> >
> >  our %allow_repeated_words = (

Re: [PATCH 1/1] checkpatch: add MA_STATE to declaration_macros

[Philip Li]

On Sun, Nov 09, 2025 at 08:09:47AM -0700, jim.cromie@gmail.com wrote:
> On Sat, Nov 8, 2025 at 5:52 PM Joe Perches <joe@perches.com> wrote:
> >
> > On Sat, 2025-11-08 at 16:47 -0700, Jim Cromie wrote:
> > > MA_STATE is used repeatedly as a declaration, add it to the list.
> >
> > OK but
> >
> > > I left out the $Storage regex component, since I saw no cases wanting
> > > it, via: $> grep -r -B1 '\bMA_STATE' .
> > >
> > > This change fixes the following, reproduced locally.
> >
> > No following ?
> 
> the oe-kbuild report followed, but the link didnt work right..
> 
> >
> > And I looked at the MA_STATE macro definition that has
> > several NULL or zero unnecessary member initializations
> > but doesn't initialize the entire structure as all
> > non-named members are NULL or zero by default.
> >
> > I wonder if the unnecessary NULL/0 initializations there
> > could actually increase the object size.
> >
> > With gcc 15.2.1 allyesconfig & defconfig do not change
> > either way but I don't know if all supported versions
> > minimize the initialization properly.
> >
> > > Reported-by: kernel test robot <lkp@intel.com>
> > > Closes: https://lore.kernel.org/oe-kbuild-all/202511071306.OJpTf7fK-lkp@intel.com/
> >
> > This Closes: url is not found on lore
> 
> That is weird.
> when I follow the Closes link, I get something unexpected - a "not
> found - look here maybe",
> AND a ../oe-kbuild/ link -
> clicking that takes me to the report proper.
> 
> If you retry, you can edit the "-all" out of the URL and it works.
> 
> It sounds like a small bug somewhere in the lkp-robot CI chain.

Sorry, this is a bug in the bot, i will fix it to point to the right
link oe-kbuild for these manual check reports.

> 
> 
> >
> > >
> > > Signed-off-by: Jim Cromie <jim.cromie@gmail.com>
> > > ---
> > >  scripts/checkpatch.pl | 3 ++-
> > >  1 file changed, 2 insertions(+), 1 deletion(-)
> > >
> > > diff --git a/scripts/checkpatch.pl b/scripts/checkpatch.pl
> > > index e722dd6fa8ef..4b5e2e64dece 100755
> > > --- a/scripts/checkpatch.pl
> > > +++ b/scripts/checkpatch.pl
> > > @@ -1096,7 +1096,8 @@ our $declaration_macros = qr{(?x:
> > >       (?:$Storage\s+)?(?:[A-Z_][A-Z0-9]*_){0,2}(?:DEFINE|DECLARE)(?:_[A-Z0-9]+){1,6}\s*\(|
> > >       (?:$Storage\s+)?[HLP]?LIST_HEAD\s*\(|
> > >       (?:SKCIPHER_REQUEST|SHASH_DESC|AHASH_REQUEST)_ON_STACK\s*\(|
> > > -     (?:$Storage\s+)?(?:XA_STATE|XA_STATE_ORDER)\s*\(
> > > +     (?:$Storage\s+)?(?:XA_STATE|XA_STATE_ORDER)\s*\(|
> > > +     (?:MA_STATE)\s*\(
> > >  )};
> > >
> > >  our %allow_repeated_words = (
>