keys: remove trailing semicolon in macro definition

Commit Message

Tom Rix Nov. 27, 2020, 7:15 p.m. UTC

From: Tom Rix <[e-mail]>

The macro use will already have a semicolon.

Signed-off-by: Tom Rix <[e-mail]>
---
 include/linux/key.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

Comments

Jarkko Sakkinen Nov. 29, 2020, 4:45 a.m. UTC | #1

On Fri, Nov 27, 2020 at 11:15:43AM -0800, [e-mail] wrote:
> From: Tom Rix <[e-mail]>
> 
> The macro use will already have a semicolon.
> 
> Signed-off-by: Tom Rix <[e-mail]>

I'm in-between whether this is worth of merging. The commit message
does not help with that decision too much.

/Jarkko

Joe Perches Nov. 29, 2020, 5:30 a.m. UTC | #2

On Sun, 2020-11-29 at 06:45 +0200, Jarkko Sakkinen wrote:
> On Fri, Nov 27, 2020 at 11:15:43AM -0800, [e-mail] wrote:
> > From: Tom Rix <[e-mail]>
> > 
> > The macro use will already have a semicolon.
> > 
> > Signed-off-by: Tom Rix <[e-mail]>
> 
> I'm in-between whether this is worth of merging. The commit message
> does not help with that decision too much.

It seems worthy of merging to me modulo whatver improvement is desired in
the commit message.

There are 3 existing uses of request_key_net.  All have a trailing semicolon.
There is 1 existing use of request_key_net_rcu.  It has a trailing semicolon.

No object change should occur.

David Howells Nov. 30, 2020, 8:58 a.m. UTC | #3

Jarkko Sakkinen <[e-mail]> wrote:

> I'm in-between whether this is worth of merging. The commit message
> does not help with that decision too much.

It is worth merging.  It might cause problems for someone using it in the body
of an if-statement, say, if they don't put braces in, but do put a semicolon
on the end.  I'll add it to my branch.

David

Jarkko Sakkinen Dec. 2, 2020, 4:45 p.m. UTC | #4

On Sat, Nov 28, 2020 at 09:30:44PM -0800, Joe Perches wrote:
> On Sun, 2020-11-29 at 06:45 +0200, Jarkko Sakkinen wrote:
> > On Fri, Nov 27, 2020 at 11:15:43AM -0800, [e-mail] wrote:
> > > From: Tom Rix <[e-mail]>
> > > 
> > > The macro use will already have a semicolon.
> > > 
> > > Signed-off-by: Tom Rix <[e-mail]>
> > 
> > I'm in-between whether this is worth of merging. The commit message
> > does not help with that decision too much.
> 
> It seems worthy of merging to me modulo whatver improvement is desired in
> the commit message.
> 
> There are 3 existing uses of request_key_net.  All have a trailing semicolon.
> There is 1 existing use of request_key_net_rcu.  It has a trailing semicolon.
> 
> No object change should occur.

OK, makes sense.

Acked-by: Jarkko Sakkinen <[e-mail]>

/Jarkko

Patch

diff --git a/include/linux/key.h b/include/linux/key.h
index 0f2e24f13c2b..1b0837c975b9 100644
--- a/include/linux/key.h
+++ b/include/linux/key.h
@@ -360,7 +360,7 @@  static inline struct key *request_key(struct key_type *type,
  * completion of keys undergoing construction with a non-interruptible wait.
  */
 #define request_key_net(type, description, net, callout_info) \
-	request_key_tag(type, description, net->key_domain, callout_info);
+	request_key_tag(type, description, net->key_domain, callout_info)
 
 /**
  * request_key_net_rcu - Request a key for a net namespace under RCU conditions
@@ -372,7 +372,7 @@  static inline struct key *request_key(struct key_type *type,
  * network namespace are used.
  */
 #define request_key_net_rcu(type, description, net) \
-	request_key_rcu(type, description, net->key_domain);
+	request_key_rcu(type, description, net->key_domain)
 #endif /* CONFIG_NET */
 
 extern int wait_for_key_construction(struct key *key, bool intr);