Skip to contentSkip to conversation

kernel/kthread.c: make kthread_stop_sem() static

RFC PATCH In review Adrian Bunk 2 participants 4 messages Apr 23, 2006 – Apr 26, 2006

lore
Share & tools
lore
#0 lore raw

RFC PATCH 1/1 kernel/kthread.c: make kthread_stop_sem() static

This patch makes the needlessly global kthread_stop_sem() static.

---

 include/linux/kthread.h |   12 ------------
 kernel/kthread.c        |   26 +++++++++++++++++---------
 2 files changed, 17 insertions(+), 21 deletions(-)
include/linux/kthread.h renamed from include/linux/kthread.h.old +0 -12
@@ -70,18 +70,6 @@
 int kthread_stop(struct task_struct *k);
 
 /**
- * kthread_stop_sem: stop a thread created by kthread_create().
- * @k: thread created by kthread_create().
- * @s: semaphore that @k waits on while idle.
- *
- * Does essentially the same thing as kthread_stop() above, but wakes
- * @k by calling up(@s).
- *
- * Returns the result of threadfn(), or -EINTR if wake_up_process()
- * was never called. */
-int kthread_stop_sem(struct task_struct *k, struct semaphore *s);
-
-/**
  * kthread_should_stop: should this kthread return now?
  *
  * When someone calls kthread_stop on your kthread, it will be woken
kernel/kthread.c renamed from kernel/kthread.c.old +17 -9
@@ -164,15 +164,18 @@
 	set_task_cpu(k, cpu);
 	k->cpus_allowed = cpumask_of_cpu(cpu);
 }
-EXPORT_SYMBOL(kthread_bind);
 
-int kthread_stop(struct task_struct *k)
-{
-	return kthread_stop_sem(k, NULL);
-}
-EXPORT_SYMBOL(kthread_stop);
-
-int kthread_stop_sem(struct task_struct *k, struct semaphore *s)
+/**
+ * kthread_stop_sem: stop a thread created by kthread_create().
+ * @k: thread created by kthread_create().
+ * @s: semaphore that @k waits on while idle.
+ *
+ * Does essentially the same thing as kthread_stop() above, but wakes
+ * @k by calling up(@s).
+ *
+ * Returns the result of threadfn(), or -EINTR if wake_up_process()
+ * was never called. */
+static int kthread_stop_sem(struct task_struct *k, struct semaphore *s)
 {
 	int ret;
 
@@ -201,7 +204,12 @@
 
 	return ret;
 }
-EXPORT_SYMBOL(kthread_stop_sem);
+
+int kthread_stop(struct task_struct *k)
+{
+	return kthread_stop_sem(k, NULL);
+}
+EXPORT_SYMBOL(kthread_stop);
 
 static __init int helper_init(void)
 {
#1 lore raw

Hi Adrian,

Quoted 1 line · Adrian Bunk

This patch makes the needlessly global kthread_stop_sem() static.

Could you cleanup the code paths as well?

Now s is always NULL in kthread_stop_sem() and kthread_stop_sem() is degenerated to kthread_stop(). So it can be folded into the latter.

Regards

Ingo Oeser

#2 lore raw

RFC PATCH 1/1 kernel/kthread.c: possible cleanups

Quoted 10 lines · Ingo Oeser

Hi Adrian,

Quoted 1 line · Adrian Bunk

This patch makes the needlessly global kthread_stop_sem() static.

Could you cleanup the code paths as well?

Now s is always NULL in kthread_stop_sem() and kthread_stop_sem() is degenerated to kthread_stop(). So it can be folded into the latter.

Sounds reasonable, updated patch below.

Quoted 3 lines

Regards

Ingo Oeser

cu
Adrian

<-- snip -->

This patch contains the following possible cleanups:
- fold the otherwise unused kthread_stop_sem() into kthread_stop()
- remove the unused EXPORT_SYMBOL(kthread_bind)

---

 include/linux/kthread.h |   12 ------------
 kernel/kthread.c        |   14 ++------------
 2 files changed, 2 insertions(+), 24 deletions(-)
include/linux/kthread.h renamed from include/linux/kthread.h.old +0 -12
@@ -70,18 +70,6 @@
 int kthread_stop(struct task_struct *k);
 
 /**
- * kthread_stop_sem: stop a thread created by kthread_create().
- * @k: thread created by kthread_create().
- * @s: semaphore that @k waits on while idle.
- *
- * Does essentially the same thing as kthread_stop() above, but wakes
- * @k by calling up(@s).
- *
- * Returns the result of threadfn(), or -EINTR if wake_up_process()
- * was never called. */
-int kthread_stop_sem(struct task_struct *k, struct semaphore *s);
-
-/**
  * kthread_should_stop: should this kthread return now?
  *
  * When someone calls kthread_stop on your kthread, it will be woken
kernel/kthread.c renamed from kernel/kthread.c.old +2 -12
@@ -164,16 +164,9 @@
 	set_task_cpu(k, cpu);
 	k->cpus_allowed = cpumask_of_cpu(cpu);
 }
-EXPORT_SYMBOL(kthread_bind);
 
 int kthread_stop(struct task_struct *k)
 {
-	return kthread_stop_sem(k, NULL);
-}
-EXPORT_SYMBOL(kthread_stop);
-
-int kthread_stop_sem(struct task_struct *k, struct semaphore *s)
-{
 	int ret;
 
 	mutex_lock(&kthread_stop_lock);
@@ -187,10 +180,7 @@
 
 	/* Now set kthread_should_stop() to true, and wake it up. */
 	kthread_stop_info.k = k;
-	if (s)
-		up(s);
-	else
-		wake_up_process(k);
+	wake_up_process(k);
 	put_task_struct(k);
 
 	/* Once it dies, reset stop ptr, gather result and we're done. */
@@ -201,7 +191,7 @@
 
 	return ret;
 }
-EXPORT_SYMBOL(kthread_stop_sem);
+EXPORT_SYMBOL(kthread_stop);
 
 static __init int helper_init(void)
 {
Acked-by on v1 1/1
#3 lore raw

Re: [RFC: 2.6 patch] kernel/kthread.c: possible cleanups

Hi Adrian,

Quoted 8 lines · Adrian Bunk
Quoted 5 lines · Ingo Oeser

Could you cleanup the code paths as well?

Now s is always NULL in kthread_stop_sem() and kthread_stop_sem() is degenerated to kthread_stop(). So it can be folded into the latter.

Sounds reasonable, updated patch below.

Beautiful! Making Linux better line by line :-)

Regards

Ingo Oeser