Fixes a warning when building with g++:
warning: deprecated conversion from string constant to 'char*'
And the file parameter use is constant, so mark it as such.
Signed-off-by: Simon Kagstrom <[email protected]>
---
include/linux/kernel.h | 5 +++--
kernel/sched.c | 2 +-
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 3fc9f5a..785d7d1 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -124,7 +124,7 @@ extern int _cond_resched(void);
#endif
#ifdef CONFIG_DEBUG_SPINLOCK_SLEEP
- void __might_sleep(char *file, int line, int preempt_offset);
+ void __might_sleep(const char *file, int line, int preempt_offset);
/**
* might_sleep - annotation for functions that can sleep
*
@@ -138,7 +138,8 @@ extern int _cond_resched(void);
# define might_sleep() \
do { __might_sleep(__FILE__, __LINE__, 0); might_resched(); } while (0)
#else
- static inline void __might_sleep(char *file, int line, int preempt_offset) { }
+ static inline void __might_sleep(const char *file, int line,
+ int preempt_offset) { }
# define might_sleep() do { might_resched(); } while (0)
#endif
diff --git a/kernel/sched.c b/kernel/sched.c
index 87f1f47..506e0ac 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -9695,7 +9695,7 @@ static inline int preempt_count_equals(int preempt_offset)
return (nested == PREEMPT_INATOMIC_BASE + preempt_offset);
}
-void __might_sleep(char *file, int line, int preempt_offset)
+void __might_sleep(const char *file, int line, int preempt_offset)
{
#ifdef in_atomic
static unsigned long prev_jiffy; /* ratelimiting */
--
1.6.0.4
Commit-ID: d894837f23f491aa7ed167aae767fc07cfe6e6e6
Gitweb: http://git.kernel.org/tip/d894837f23f491aa7ed167aae767fc07cfe6e6e6
Author: Simon Kagstrom <[email protected]>
AuthorDate: Wed, 23 Dec 2009 11:08:18 +0100
Committer: Ingo Molnar <[email protected]>
CommitDate: Mon, 28 Dec 2009 10:50:13 +0100
sched: might_sleep(): Make file parameter const char *
Fixes a warning when building with g++:
warning: deprecated conversion from string constant to 'char*'
And the file parameter use is constant, so mark it as such.
Signed-off-by: Simon Kagstrom <[email protected]>
Cc: [email protected]
LKML-Reference: <[email protected]>
Signed-off-by: Ingo Molnar <[email protected]>
---
include/linux/kernel.h | 5 +++--
kernel/sched.c | 2 +-
2 files changed, 4 insertions(+), 3 deletions(-)
diff --git a/include/linux/kernel.h b/include/linux/kernel.h
index 3fc9f5a..785d7d1 100644
--- a/include/linux/kernel.h
+++ b/include/linux/kernel.h
@@ -124,7 +124,7 @@ extern int _cond_resched(void);
#endif
#ifdef CONFIG_DEBUG_SPINLOCK_SLEEP
- void __might_sleep(char *file, int line, int preempt_offset);
+ void __might_sleep(const char *file, int line, int preempt_offset);
/**
* might_sleep - annotation for functions that can sleep
*
@@ -138,7 +138,8 @@ extern int _cond_resched(void);
# define might_sleep() \
do { __might_sleep(__FILE__, __LINE__, 0); might_resched(); } while (0)
#else
- static inline void __might_sleep(char *file, int line, int preempt_offset) { }
+ static inline void __might_sleep(const char *file, int line,
+ int preempt_offset) { }
# define might_sleep() do { might_resched(); } while (0)
#endif
diff --git a/kernel/sched.c b/kernel/sched.c
index c535cc4..64298a5 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -9694,7 +9694,7 @@ static inline int preempt_count_equals(int preempt_offset)
return (nested == PREEMPT_INATOMIC_BASE + preempt_offset);
}
-void __might_sleep(char *file, int line, int preempt_offset)
+void __might_sleep(const char *file, int line, int preempt_offset)
{
#ifdef in_atomic
static unsigned long prev_jiffy; /* ratelimiting */
On Mon, 2009-12-28 at 10:10 +0000, tip-bot for Simon Kagstrom wrote:
> Fixes a warning when building with g++:
However did you manage to build stuff with g++? The kernel very
frequently uses C++ keywords for other things.
* Peter Zijlstra <[email protected]> wrote:
> On Mon, 2009-12-28 at 10:10 +0000, tip-bot for Simon Kagstrom wrote:
>
> > Fixes a warning when building with g++:
>
> However did you manage to build stuff with g++? The kernel very
> frequently uses C++ keywords for other things.
yeah, good question. Alas the change itself (const propagation) looked good so
i applied it.
Ingo
(Back from vacation, hence the delayed response)
On Tue, 29 Dec 2009 13:11:48 +0100
Peter Zijlstra <[email protected]> wrote:
> On Mon, 2009-12-28 at 10:10 +0000, tip-bot for Simon Kagstrom wrote:
>
> > Fixes a warning when building with g++:
>
> However did you manage to build stuff with g++? The kernel very
> frequently uses C++ keywords for other things.
I expected this question :-)
Yes, there are some workarounds we have to make to get it compiling
(like #define new NEW prior to including the headers), but it works
out. I know this is something you shouldn't do, and I debated whether
to mention C++ at all.
Anyway, I think the change itself makes sense for plain C as well. The
comment just described the context where it showed up for us.
// Simon
On Mon, 2010-01-04 at 10:49 +0100, Simon Kagstrom wrote:
> (Back from vacation, hence the delayed response)
>
> On Tue, 29 Dec 2009 13:11:48 +0100
> Peter Zijlstra <[email protected]> wrote:
>
> > On Mon, 2009-12-28 at 10:10 +0000, tip-bot for Simon Kagstrom wrote:
> >
> > > Fixes a warning when building with g++:
> >
> > However did you manage to build stuff with g++? The kernel very
> > frequently uses C++ keywords for other things.
>
> I expected this question :-)
>
> Yes, there are some workarounds we have to make to get it compiling
> (like #define new NEW prior to including the headers), but it works
> out. I know this is something you shouldn't do, and I debated whether
> to mention C++ at all.
>
>
> Anyway, I think the change itself makes sense for plain C as well. The
> comment just described the context where it showed up for us.
Oh sure, the change is fine, just wondering what specific drug you were
using as it was not the regular kernel kind ;-)