Skip to contentSkip to conversation

sched: trivial sparse warning in sched.c

PATCH Applied Harvey Harrison 3 participants 5 messages Feb 15, 2008 – Feb 17, 2008

lore
Share & tools
lore
#0 lore raw

PATCH 2/3 sched: trivial sparse warning in sched.c

kernel/sched.c:3680:3: warning: returning void-valued expression

---
 kernel/sched.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)
kernel/sched.c +4 -2
index f28f19e..824d5a9 100644
@@ -3676,8 +3676,10 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
 	struct rq *rq = this_rq();
 	cputime64_t tmp;
 
-	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0))
-		return account_guest_time(p, cputime);
+	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) {
+		account_guest_time(p, cputime);
+		return;
+	}
 
 	p->stime = cputime_add(p->stime, cputime);
 
Signature · 1 line
1.5.4.1.1278.gc75be
#1 lore raw
Quoted 1 line · Harvey Harrison

kernel/sched.c:3680:3: warning: returning void-valued expression

I'm wondering,... why is this wrong?

Quoted 22 lines
---
 kernel/sched.c |    6 ++++--
 1 files changed, 4 insertions(+), 2 deletions(-)
kernel/sched.c +4 -2
index f28f19e..824d5a9 100644
@@ -3676,8 +3676,10 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
 	struct rq *rq = this_rq();
 	cputime64_t tmp;
 
-	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0))
-		return account_guest_time(p, cputime);
+	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) {
+		account_guest_time(p, cputime);
+		return;
+	}
 
 	p->stime = cputime_add(p->stime, cputime);
 
#2 lore raw

[Peter Zijlstra - Sat, Feb 16, 2008 at 10:11:29PM +0100]

| 
| On Fri, 2008-02-15 at 09:56 -0800, Harvey Harrison wrote:
| > kernel/sched.c:3680:3: warning: returning void-valued expression
| 
| I'm wondering,... why is this wrong?
| 
| > Signed-off-by: Harvey Harrison <[email protected]>
| > ---
| >  kernel/sched.c |    6 ++++--
| >  1 files changed, 4 insertions(+), 2 deletions(-)
| > 
| > diff --git a/kernel/sched.c b/kernel/sched.c
| > index f28f19e..824d5a9 100644
| > --- a/kernel/sched.c
| > +++ b/kernel/sched.c
| > @@ -3676,8 +3676,10 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
| >  	struct rq *rq = this_rq();
| >  	cputime64_t tmp;
| >  
| > -	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0))
| > -		return account_guest_time(p, cputime);
| > +	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) {
| > +		account_guest_time(p, cputime);
| > +		return;
| > +	}
| >  
| >  	p->stime = cputime_add(p->stime, cputime);
| >  

Hi Peter,

dont you find 'return (void)foo();' statement a bit strange (as it was in original code)? ;) Am I wrong? /it's night here, so half a brain already tuned off ;)/

- Cyrill -

#3 lore raw
Quoted 30 lines · Cyrill Gorcunov

[Peter Zijlstra - Sat, Feb 16, 2008 at 10:11:29PM +0100]

| 
| On Fri, 2008-02-15 at 09:56 -0800, Harvey Harrison wrote:
| > kernel/sched.c:3680:3: warning: returning void-valued expression
| 
| I'm wondering,... why is this wrong?
| 
| > Signed-off-by: Harvey Harrison <[email protected]>
| > ---
| >  kernel/sched.c |    6 ++++--
| >  1 files changed, 4 insertions(+), 2 deletions(-)
| > 
| > diff --git a/kernel/sched.c b/kernel/sched.c
| > index f28f19e..824d5a9 100644
| > --- a/kernel/sched.c
| > +++ b/kernel/sched.c
| > @@ -3676,8 +3676,10 @@ void account_system_time(struct task_struct *p, int hardirq_offset,
| >  	struct rq *rq = this_rq();
| >  	cputime64_t tmp;
| >  
| > -	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0))
| > -		return account_guest_time(p, cputime);
| > +	if ((p->flags & PF_VCPU) && (irq_count() - hardirq_offset == 0)) {
| > +		account_guest_time(p, cputime);
| > +		return;
| > +	}
| >  
| >  	p->stime = cputime_add(p->stime, cputime);
| >  
Quoted 5 lines

Hi Peter,

dont you find 'return (void)foo();' statement a bit strange (as it was in original code)? ;) Am I wrong? /it's night here, so half a brain already tuned off ;)/

void foo(void);

void bar(void)

{
	return foo();
}

Maybe I'm just weird an twisted, but no, I don't find it odd. In my mind its consistent with how all other return types function.

Not sure what the C std text says on the matter though.

#4 lore raw

[Peter Zijlstra - Sat, Feb 16, 2008 at 10:45:14PM +0100]

| 
| On Sun, 2008-02-17 at 00:27 +0300, Cyrill Gorcunov wrote:
| > [Peter Zijlstra - Sat, Feb 16, 2008 at 10:11:29PM +0100]
| > | 
[...]
| > 
| > Hi Peter,
| > 
| > dont you find 'return (void)foo();' statement a bit strange (as it was in
| > original code)? ;) Am I wrong? /it's night here, so half a brain already
| > tuned off ;)/
| 
| void foo(void);
| 
| void bar(void)
| {
| 	return foo();
| }
| 
| Maybe I'm just weird an twisted, but no, I don't find it odd. In my mind
| its consistent with how all other return types function.
| 
| Not sure what the C std text says on the matter though.
| 
| 

well, i've an obscure feeling about that. From one side I do understand that you're *absolutely right* but the second form of statement is more clear methink. That is what I prefer personally. Anyway the form of writting such a statement is taste of code author - so you choose ;)

- Cyrill -