[2.6 patch] "extern inline" doesn't make sense

[2.6 patch] "extern inline" doesn't make sense

[Adrian Bunk]

"extern inline" doesn't make sense.


Signed-off-by: Adrian Bunk <bunk@stusta.de>

--- linux-2.6.13-rc6-mm1-full/fs/adfs/adfs.h.old	2005-08-19 23:21:33.000000000 +0200
+++ linux-2.6.13-rc6-mm1-full/fs/adfs/adfs.h	2005-08-19 23:22:07.000000000 +0200
@@ -97,7 +97,7 @@
 extern struct inode_operations adfs_file_inode_operations;
 extern struct file_operations adfs_file_operations;
 
-extern inline __u32 signed_asl(__u32 val, signed int shift)
+static inline __u32 signed_asl(__u32 val, signed int shift)
 {
 	if (shift >= 0)
 		val <<= shift;
@@ -112,7 +112,7 @@
  *
  * The root directory ID should always be looked up in the map [3.4]
  */
-extern inline int
+static inline int
 __adfs_block_map(struct super_block *sb, unsigned int object_id,
 		 unsigned int block)
 {

[2.6 patch] fs/adfs/adfs.h: "extern inline" doesn't make sense

[Adrian Bunk]

[ this time with a better subject ]

"extern inline" doesn't make sense.


Signed-off-by: Adrian Bunk <bunk@stusta.de>

--- linux-2.6.13-rc6-mm1-full/fs/adfs/adfs.h.old	2005-08-19 23:21:33.000000000 +0200
+++ linux-2.6.13-rc6-mm1-full/fs/adfs/adfs.h	2005-08-19 23:22:07.000000000 +0200
@@ -97,7 +97,7 @@
 extern struct inode_operations adfs_file_inode_operations;
 extern struct file_operations adfs_file_operations;
 
-extern inline __u32 signed_asl(__u32 val, signed int shift)
+static inline __u32 signed_asl(__u32 val, signed int shift)
 {
 	if (shift >= 0)
 		val <<= shift;
@@ -112,7 +112,7 @@
  *
  * The root directory ID should always be looked up in the map [3.4]
  */
-extern inline int
+static inline int
 __adfs_block_map(struct super_block *sb, unsigned int object_id,
 		 unsigned int block)
 {

Re: [2.6 patch] fs/adfs/adfs.h: "extern inline" doesn't make sense

[Russell King]

On Sat, Aug 20, 2005 at 01:44:43AM +0200, Adrian Bunk wrote:
> "extern inline" doesn't make sense.
> 
> Signed-off-by: Adrian Bunk <bunk@stusta.de>

Thanks Adrian - I've committed it to my tree.

-- 
Russell King
 Linux kernel    2.6 ARM Linux   - http://www.arm.linux.org.uk/
 maintainer of:  2.6 Serial core

Re: [2.6 patch] fs/adfs/adfs.h: "extern inline" doesn't make sense

[Pierre Ossman]

Adrian Bunk wrote:
> [ this time with a better subject ]
> 
> "extern inline" doesn't make sense.
> 
> 
> Signed-off-by: Adrian Bunk <bunk@stusta.de>
> 

Isn't 'extern inline' an old gcc trick to force inlining? (instead of
just hinting)

Re: [2.6 patch] fs/adfs/adfs.h: "extern inline" doesn't make sense

[Adrian Bunk]

On Sun, Aug 21, 2005 at 02:56:40PM +0200, Pierre Ossman wrote:
> Adrian Bunk wrote:
> > [ this time with a better subject ]
> > 
> > "extern inline" doesn't make sense.
> > 
> > 
> > Signed-off-by: Adrian Bunk <bunk@stusta.de>
> > 
> 
> Isn't 'extern inline' an old gcc trick to force inlining? (instead of
> just hinting)

For gcc >= 3.1, we are already telling the compiler that it should 
either inline the code or abort compilation.

"extern inline" doesn't make much sense. And it gives a warning with 
-Wmissing-prototypes.

I'm currently cleaning up warnings with -Wmissing-prototypes to get them 
to an acceptable level. -Wmissing-prototypes will help us to enforce a 
coding style that will avoid a class of nasty runtime errors that are 
currently sometimes hitting us.

cu
Adrian

-- 

       "Is there not promise of rain?" Ling Tan asked suddenly out
        of the darkness. There had been need of rain for many days.
       "Only a promise," Lao Er said.
                                       Pearl S. Buck - Dragon Seed