[PATCH] cgroup: drop unneeded semicolon

[PATCH] cgroup: drop unneeded semicolon

[Julia Lawall]

When a function-like macro expands to an expression, that expression
doesn't need a semicolon after it.  All uses have been verified to
have their own semicolons.

This was found using the following Coccinelle semantic patch:

@r@
identifier i : script:ocaml() { String.lowercase_ascii i = i };
expression e;
@@

*#define i(...) e;

Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>

---
 kernel/cgroup/cgroup.c |    2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
index 38f8d9df8..f87fc4550 100644
--- a/kernel/cgroup/cgroup.c
+++ b/kernel/cgroup/cgroup.c
@@ -104,7 +104,7 @@ DEFINE_PERCPU_RWSEM(cgroup_threadgroup_rwsem);
 #define cgroup_assert_mutex_or_rcu_locked()				\
 	RCU_LOCKDEP_WARN(!rcu_read_lock_held() &&			\
 			   !lockdep_is_held(&cgroup_mutex),		\
-			   "cgroup_mutex or RCU read lock required");
+			   "cgroup_mutex or RCU read lock required")
 
 /*
  * cgroup destruction makes heavy use of work items and there can be a lot

Re: [PATCH] cgroup: drop unneeded semicolon

[Waiman Long]

On 8/1/26 3:09 PM, Julia Lawall wrote:
> When a function-like macro expands to an expression, that expression
> doesn't need a semicolon after it.  All uses have been verified to
> have their own semicolons.
>
> This was found using the following Coccinelle semantic patch:
>
> @r@
> identifier i : script:ocaml() { String.lowercase_ascii i = i };
> expression e;
> @@
>
> *#define i(...) e;
>
> Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>
>
> ---
>   kernel/cgroup/cgroup.c |    2 +-
>   1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
> index 38f8d9df8..f87fc4550 100644
> --- a/kernel/cgroup/cgroup.c
> +++ b/kernel/cgroup/cgroup.c
> @@ -104,7 +104,7 @@ DEFINE_PERCPU_RWSEM(cgroup_threadgroup_rwsem);
>   #define cgroup_assert_mutex_or_rcu_locked()				\
>   	RCU_LOCKDEP_WARN(!rcu_read_lock_held() &&			\
>   			   !lockdep_is_held(&cgroup_mutex),		\
> -			   "cgroup_mutex or RCU read lock required");
> +			   "cgroup_mutex or RCU read lock required")
>   
>   /*
>    * cgroup destruction makes heavy use of work items and there can be a lot
>
The RCU_LOCKDEP_WARN() macro is basically a "do { ...} while (0...)" 
statement. It doesn't have a ';' at the end.  So your commit log isn't 
right.

An additional semicolon at the end doesn't harm. By removing the 
semicolon, even if not needed, may make people wonder if it is right 
when reading the code wasting their time to figure out if it is all 
right. So I don't see any advantage in removing it.

Cheers,
Longman

Re: [PATCH] cgroup: drop unneeded semicolon

[Tejun Heo]

Hello,

On Sat, Aug 01, 2026 at 07:37:13PM -0400, Waiman Long wrote:
> > diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
> > index 38f8d9df8..f87fc4550 100644
> > --- a/kernel/cgroup/cgroup.c
> > +++ b/kernel/cgroup/cgroup.c
> > @@ -104,7 +104,7 @@ DEFINE_PERCPU_RWSEM(cgroup_threadgroup_rwsem);
> >   #define cgroup_assert_mutex_or_rcu_locked()				\
> >   	RCU_LOCKDEP_WARN(!rcu_read_lock_held() &&			\
> >   			   !lockdep_is_held(&cgroup_mutex),		\
> > -			   "cgroup_mutex or RCU read lock required");
> > +			   "cgroup_mutex or RCU read lock required")
> >   /*
> >    * cgroup destruction makes heavy use of work items and there can be a lot
> > 
> The RCU_LOCKDEP_WARN() macro is basically a "do { ...} while (0...)"
> statement. It doesn't have a ';' at the end.  So your commit log isn't
> right.
> 
> An additional semicolon at the end doesn't harm. By removing the semicolon,
> even if not needed, may make people wonder if it is right when reading the
> code wasting their time to figure out if it is all right. So I don't see any
> advantage in removing it.

I think the extra semicolon can hurt. Imagine something like this:

        if (blah)
                cgroup_assert_mutex_or_rcu_locked();
        else
                something else;

This should work but with the extra semicolon, it doesn't. I don't see why
we wouldn't fix this. It's obviously wrong.

Thanks.

-- 
tejun

Re: [PATCH] cgroup: drop unneeded semicolon

[Waiman Long]

On 8/1/26 11:11 PM, Tejun Heo wrote:
> Hello,
>
> On Sat, Aug 01, 2026 at 07:37:13PM -0400, Waiman Long wrote:
>>> diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
>>> index 38f8d9df8..f87fc4550 100644
>>> --- a/kernel/cgroup/cgroup.c
>>> +++ b/kernel/cgroup/cgroup.c
>>> @@ -104,7 +104,7 @@ DEFINE_PERCPU_RWSEM(cgroup_threadgroup_rwsem);
>>>    #define cgroup_assert_mutex_or_rcu_locked()				\
>>>    	RCU_LOCKDEP_WARN(!rcu_read_lock_held() &&			\
>>>    			   !lockdep_is_held(&cgroup_mutex),		\
>>> -			   "cgroup_mutex or RCU read lock required");
>>> +			   "cgroup_mutex or RCU read lock required")
>>>    /*
>>>     * cgroup destruction makes heavy use of work items and there can be a lot
>>>
>> The RCU_LOCKDEP_WARN() macro is basically a "do { ...} while (0...)"
>> statement. It doesn't have a ';' at the end.  So your commit log isn't
>> right.
>>
>> An additional semicolon at the end doesn't harm. By removing the semicolon,
>> even if not needed, may make people wonder if it is right when reading the
>> code wasting their time to figure out if it is all right. So I don't see any
>> advantage in removing it.
> I think the extra semicolon can hurt. Imagine something like this:
>
>          if (blah)
>                  cgroup_assert_mutex_or_rcu_locked();
>          else
>                  something else;
>
> This should work but with the extra semicolon, it doesn't. I don't see why
> we wouldn't fix this. It's obviously wrong.

OK, I think I have misread the patch. Right, the extra ';' should be 
removed. Sorry for the noise.

Cheers,
Longman

Re: [PATCH] cgroup: drop unneeded semicolon

[Julia Lawall]

[-- Attachment #1: Type: text/plain, Size: 1875 bytes --]



On Sat, 1 Aug 2026, Waiman Long wrote:

> On 8/1/26 3:09 PM, Julia Lawall wrote:
> > When a function-like macro expands to an expression, that expression
> > doesn't need a semicolon after it.  All uses have been verified to
> > have their own semicolons.
> >
> > This was found using the following Coccinelle semantic patch:
> >
> > @r@
> > identifier i : script:ocaml() { String.lowercase_ascii i = i };
> > expression e;
> > @@
> >
> > *#define i(...) e;
> >
> > Signed-off-by: Julia Lawall <Julia.Lawall@inria.fr>
> >
> > ---
> >   kernel/cgroup/cgroup.c |    2 +-
> >   1 file changed, 1 insertion(+), 1 deletion(-)
> >
> > diff --git a/kernel/cgroup/cgroup.c b/kernel/cgroup/cgroup.c
> > index 38f8d9df8..f87fc4550 100644
> > --- a/kernel/cgroup/cgroup.c
> > +++ b/kernel/cgroup/cgroup.c
> > @@ -104,7 +104,7 @@ DEFINE_PERCPU_RWSEM(cgroup_threadgroup_rwsem);
> >   #define cgroup_assert_mutex_or_rcu_locked()				\
> >   	RCU_LOCKDEP_WARN(!rcu_read_lock_held() &&			\
> >   			   !lockdep_is_held(&cgroup_mutex),		\
> > -			   "cgroup_mutex or RCU read lock required");
> > +			   "cgroup_mutex or RCU read lock required")
> >     /*
> >    * cgroup destruction makes heavy use of work items and there can be a lot
> >
> The RCU_LOCKDEP_WARN() macro is basically a "do { ...} while (0...)"
> statement. It doesn't have a ';' at the end.  So your commit log isn't right.

OK.  Indeed I didn't take into account the definition of the macro.  The
usage context does have another semincolon, which is what is mentioned in
the commit log.

>
> An additional semicolon at the end doesn't harm. By removing the semicolon,
> even if not needed, may make people wonder if it is right when reading the
> code wasting their time to figure out if it is all right. So I don't see any
> advantage in removing it.

No problem to drop the patch if you don't think it's useful.

julia