Skip to contentSkip to conversation

x86_64 UP needs smp_call_function_single

PATCH In review Randy Dunlap 3 participants 10 messages Nov 30, 2006 Locking primitives

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 x86_64 UP needs smp_call_function_single

From: Randy Dunlap <[email protected]>

smp_call_function_single() needs to be visible in non-SMP builds, to fix:

arch/x86_64/kernel/vsyscall.c:283: warning: implicit declaration of function 'smp_call_function_single'

The (other/trivial) fix (instead of this one) is to add:
#include <asm/smp.h>
to linux-2.6.19-rc6-mm2/arch/x86_64/kernel/vsyscall.c

---
 include/asm-x86_64/smp.h |    7 -------
 include/linux/smp.h      |    7 +++++++
 2 files changed, 7 insertions(+), 7 deletions(-)
include/asm-x86_64/smp.h +0 -7
@@ -113,13 +113,6 @@ static __inline int logical_smp_processo
 #define cpu_physical_id(cpu)		x86_cpu_to_apicid[cpu]
 #else
 #define cpu_physical_id(cpu)		boot_cpu_id
-static inline int smp_call_function_single(int cpuid, void (*func) (void *info),
-				void *info, int retry, int wait)
-{
-	/* Disable interrupts here? */
-	func(info);
-	return 0;
-}
 #endif /* !CONFIG_SMP */
 #endif
 
include/linux/smp.h +7 -0
@@ -99,6 +99,13 @@ static inline int up_smp_call_function(v
 static inline void smp_send_reschedule(int cpu) { }
 #define num_booting_cpus()			1
 #define smp_prepare_boot_cpu()			do {} while (0)
+static inline int smp_call_function_single(int cpuid, void (*func) (void *info),
+				void *info, int retry, int wait)
+{
+	/* Disable interrupts here? */
+	func(info);
+	return 0;
+}
 
 #endif /* !SMP */
 

---

#1 lore raw
Quoted 38 lines · On Wed, 29 Nov 2006 17:01:11 -0800 Randy Dunlap <[email protected]> wrote:

From: Randy Dunlap <[email protected]>

smp_call_function_single() needs to be visible in non-SMP builds, to fix:

arch/x86_64/kernel/vsyscall.c:283: warning: implicit declaration of function 'smp_call_function_single'

The (other/trivial) fix (instead of this one) is to add:
#include <asm/smp.h>
to linux-2.6.19-rc6-mm2/arch/x86_64/kernel/vsyscall.c

---
 include/asm-x86_64/smp.h |    7 -------
 include/linux/smp.h      |    7 +++++++
 2 files changed, 7 insertions(+), 7 deletions(-)
include/asm-x86_64/smp.h +0 -7
@@ -113,13 +113,6 @@ static __inline int logical_smp_processo
 #define cpu_physical_id(cpu)		x86_cpu_to_apicid[cpu]
 #else
 #define cpu_physical_id(cpu)		boot_cpu_id
-static inline int smp_call_function_single(int cpuid, void (*func) (void *info),
-				void *info, int retry, int wait)
-{
-	/* Disable interrupts here? */
-	func(info);
-	return 0;
-}
 #endif /* !CONFIG_SMP */
 #endif
 
include/linux/smp.h +0 -0
@@ -99,6 +99,13 @@ static inline int up_smp_call_function(v
 static inline void smp_send_reschedule(int cpu) { }
 #define num_booting_cpus()			1
 #define smp_prepare_boot_cpu()			do {} while (0)

include/linux/smp.h static inline int up_smp_call_function(v · line 111

Quoted 10 lines
include/linux/smp.h +7 -0
+static inline int smp_call_function_single(int cpuid, void (*func) (void *info),
+				void *info, int retry, int wait)
+{
+	/* Disable interrupts here? */
+	func(info);
+	return 0;
+}
 
 #endif /* !SMP */
 

No, I think this patch is right - the declaration of the CONFIG_SMP smp_call_function_single() is in linux/smp.h so the !CONFIG_SMP declaration or definition should be there too.

It's still buggy though. It should disable local interrupts around the call to match the SMP version. I'll fix that separately.

maintainer
#2 lore raw
Quoted 8 lines · Andrew Morton

No, I think this patch is right - the declaration of the CONFIG_SMP smp_call_function_single() is in linux/smp.h so the !CONFIG_SMP declaration
or definition should be there too.

It's still buggy though. It should disable local interrupts around the
call to match the SMP version. I'll fix that separately.

hm, didnt i send an updated patch for that already? See the patch below, from many days ago. I sent it after the tsc-sync-rewrite patch.

Ingo

--------------->
Subject: x86_64: build fixes
From: Ingo Molnar <[email protected]>

x86_64 does not build cleanly on UP:

arch/x86_64/kernel/vsyscall.c: In function 'cpu_vsyscall_notifier':
arch/x86_64/kernel/vsyscall.c:282: warning: implicit declaration of function 'smp_call_function_single' arch/x86_64/kernel/vsyscall.c: At top level:
arch/x86_64/kernel/vsyscall.c:279: warning: 'cpu_vsyscall_notifier' defined but not used

this patch fixes it by making smp_call_function_single() globally available.

---
 include/asm-x86_64/smp.h |   11 ++---------
 include/linux/smp.h      |   10 +++++++---
 kernel/sched.c           |   19 +++++++++++++++++++
 3 files changed, 28 insertions(+), 12 deletions(-)
include/asm-x86_64/smp.h +1 -3
Index: linux/include/asm-x86_64/smp.h
===================================================================
@@ -115,16 +115,9 @@ static __inline int logical_smp_processo
 }
 
 #ifdef CONFIG_SMP
-#define cpu_physical_id(cpu)		x86_cpu_to_apicid[cpu]
+# define cpu_physical_id(cpu)		x86_cpu_to_apicid[cpu]
 #else
-#define cpu_physical_id(cpu)		boot_cpu_id
-static inline int smp_call_function_single(int cpuid, void (*func)

(void *info),
- void *info, int retry, int wait)
-{
- /* Disable interrupts here? */
- func(info);
- return 0;
-}
+# define cpu_physical_id(cpu) boot_cpu_id
#endif /* !CONFIG_SMP */
#endif

include/linux/smp.h +0 -0
Index: linux/include/linux/smp.h
===================================================================
@@ -53,9 +53,6 @@ extern void smp_cpus_done(unsigned int m
  */
 int smp_call_function(void(*func)(void *info), void *info, int retry,

int wait);

-int smp_call_function_single(int cpuid, void (*func) (void *info), void *info,
- int retry, int wait);
-
/*
* Call a function on all processors
*/
@@ -103,6 +100,13 @@ static inline void smp_send_reschedule(i
#endif /* !SMP */

/*
+ * Call a function on a specific CPU (on UP the function gets executed
+ * on the current CPU, immediately):
+ */
+int smp_call_function_single(int cpuid, void (*func) (void *info), void *info,
+ int retry, int wait);
+
+/*
* smp_processor_id(): get the current CPU ID.
*
* if DEBUG_PREEMPT is enabled the we check whether it is

kernel/sched.c +6 -0
Index: linux/kernel/sched.c
===================================================================
@@ -1110,6 +1110,25 @@ repeat:
 	task_rq_unlock(rq, &flags);
 }
 
+#ifndef CONFIG_SMP
+/*
+ * Call a function on a specific CPU (on UP the function gets executed
+ * on the current CPU, immediately):
+ */
+int smp_call_function_single(int cpuid, void (*func) (void *info), void

*info,
+ int retry, int wait)
+{
+ unsigned long flags;
+
+ local_irq_save(flags);
+ func(info);
+ local_irq_restore(flags);
+
+ return 0;
+}
+EXPORT_SYMBOL(smp_call_function_single);
+#endif
+
/***
* kick_process - kick a running thread to enter/exit the kernel
* @p: the to-be-kicked thread

#3 lore raw
Quoted 6 lines · On Thu, 30 Nov 2006 08:00:00 +0100 Ingo Molnar <[email protected]> wrote:
Quoted 5 lines · Andrew Morton

No, I think this patch is right - the declaration of the CONFIG_SMP smp_call_function_single() is in linux/smp.h so the !CONFIG_SMP declaration
or definition should be there too.

Quoted 7 lines
Quoted 3 lines

It's still buggy though. It should disable local interrupts around the
call to match the SMP version. I'll fix that separately.

hm, didnt i send an updated patch for that already? See the patch below, from many days ago. I sent it after the tsc-sync-rewrite patch.

Might have got lost.

Quoted 28 lines

--------------->
Subject: x86_64: build fixes
From: Ingo Molnar <[email protected]>

x86_64 does not build cleanly on UP:

arch/x86_64/kernel/vsyscall.c: In function 'cpu_vsyscall_notifier':
arch/x86_64/kernel/vsyscall.c:282: warning: implicit declaration of function 'smp_call_function_single' arch/x86_64/kernel/vsyscall.c: At top level:
arch/x86_64/kernel/vsyscall.c:279: warning: 'cpu_vsyscall_notifier' defined but not used

this patch fixes it by making smp_call_function_single() globally available.

---
 include/asm-x86_64/smp.h |   11 ++---------
 include/linux/smp.h      |   10 +++++++---
 kernel/sched.c           |   19 +++++++++++++++++++
 3 files changed, 28 insertions(+), 12 deletions(-)

Index: linux/include/asm-x86_64/smp.h
===================================================================

include/asm-x86_64/smp.h static __inline int logical_smp_processo · line 120

Quoted 9 lines
include/asm-x86_64/smp.h +1 -3
@@ -115,16 +115,9 @@ static __inline int logical_smp_processo
 }
 
 #ifdef CONFIG_SMP
-#define cpu_physical_id(cpu)		x86_cpu_to_apicid[cpu]
+# define cpu_physical_id(cpu)		x86_cpu_to_apicid[cpu]
 #else
-#define cpu_physical_id(cpu)		boot_cpu_id
-static inline int smp_call_function_single(int cpuid, void (*func)
 (void *info),

congratulations-your-first-wordwrapped-patch ;)

Quoted 13 lines
kernel/sched.c +6 -0
@@ -1110,6 +1110,25 @@ repeat:
 	task_rq_unlock(rq, &flags);
 }
 
+#ifndef CONFIG_SMP
+/*
+ * Call a function on a specific CPU (on UP the function gets executed
+ * on the current CPU, immediately):
+ */
+int smp_call_function_single(int cpuid, void (*func) (void *info), void
 *info,

kernel/sched.c repeat: · line 1129

Quoted 10 lines
kernel/sched.c +10 -0
+			     int retry, int wait)
+{
+	unsigned long flags;
+
+	local_irq_save(flags);
+	func(info);
+	local_irq_restore(flags);
+
+	return 0;
+}

yes, but a) calling the SMP version with local interrupts disabled is a bug, so we can use bare local_irq_disable() here and b) only two archictures call or use this function, so all the others don't want a copy of it.

So I did:

include/linux/smp.h renamed from include/linux/smp.h~up-smp_call_function_single-should-disable-interrupts +3 -1
@@ -15,6 +15,7 @@ extern void cpu_idle(void);
 #include <linux/kernel.h>
 #include <linux/compiler.h>
 #include <linux/thread_info.h>
+#include <linux/irqflags.h>
 #include <asm/smp.h>
 
 /*
@@ -102,8 +103,9 @@ static inline void smp_send_reschedule(i
 static inline int smp_call_function_single(int cpuid, void (*func) (void *info),
 				void *info, int retry, int wait)
 {
-	/* Disable interrupts here? */
+	local_irq_disable();	/* Match the SMP call environment */
 	func(info);
+	local_irq_enable();
 	return 0;
 }
 

_

which is somewhat unpleasant. I added a WARN_ON(irqs_disabled()) to the out-of-line SMP version.

btw, does anyone know why the SMP versions of this function use spin_lock_bh(&call_lock)?

maintainer
#4 lore raw
Quoted 3 lines · Andrew Morton

which is somewhat unpleasant. I added a WARN_ON(irqs_disabled()) to the
out-of-line SMP version.

ok.

Quoted 3 lines

btw, does anyone know why the SMP versions of this function use spin_lock_bh(&call_lock)?

that makes no sense (neither the get_cpu()/put_cpu() gymnastics) if this is called with irqs disabled all the time.

Ingo

Quoted 1 line
#5 lore raw
Quoted 6 lines · On Thu, 30 Nov 2006 10:22:20 +0100 Ingo Molnar <[email protected]> wrote:
Quoted 3 lines

btw, does anyone know why the SMP versions of this function use spin_lock_bh(&call_lock)?

that makes no sense (neither the get_cpu()/put_cpu() gymnastics) if this is called with irqs disabled all the time.

smp_call_function_single() must be called with local interrupts ENabled.

But why isn't it just spin_lock()?

<looks>

Eric simply copied that code from ia64, which added the spin_lock_bh() in 2.4.8. Ho-hum.

#6 lore raw
In reply to Ingo Molnar
Quoted 6 lines · On Thu, 30 Nov 2006 08:00:00 +0100 Ingo Molnar wrote:
Quoted 5 lines · Andrew Morton

No, I think this patch is right - the declaration of the CONFIG_SMP smp_call_function_single() is in linux/smp.h so the !CONFIG_SMP declaration
or definition should be there too.

Quoted 6 lines
Quoted 3 lines

It's still buggy though. It should disable local interrupts around the
call to match the SMP version. I'll fix that separately.

hm, didnt i send an updated patch for that already? See the patch below, from many days ago. I sent it after the tsc-sync-rewrite patch.

Hi Ingo,

Has there been a patch for this one? (UP again, not SMP)

drivers/input/ff-memless.c:384: warning: implicit declaration of function 'local_bh_disable' drivers/input/ff-memless.c:393: warning: implicit declaration of function 'local_bh_enable'

Thanks,
---
~Randy
config: http://oss.oracle.com/~rdunlap/configs/config-input-up-header

#7 lore raw
Quoted 19 lines · On Thu, 30 Nov 2006 14:11:40 -0800 Randy Dunlap <[email protected]> wrote:
Quoted 6 lines · On Thu, 30 Nov 2006 08:00:00 +0100 Ingo Molnar wrote:
Quoted 5 lines · Andrew Morton

No, I think this patch is right - the declaration of the CONFIG_SMP smp_call_function_single() is in linux/smp.h so the !CONFIG_SMP declaration
or definition should be there too.

Quoted 6 lines
Quoted 3 lines

It's still buggy though. It should disable local interrupts around the
call to match the SMP version. I'll fix that separately.

hm, didnt i send an updated patch for that already? See the patch below, from many days ago. I sent it after the tsc-sync-rewrite patch.

Hi Ingo,

Has there been a patch for this one? (UP again, not SMP)

Quoted 7 lines

drivers/input/ff-memless.c:384: warning: implicit declaration of function 'local_bh_disable' drivers/input/ff-memless.c:393: warning: implicit declaration of function 'local_bh_enable'

Thanks,
---
~Randy
config: http://oss.oracle.com/~rdunlap/configs/config-input-up-header

eww.. I guess linux/spinlock.h should really include linux/interrupt.h. But interrupt.h includes stuff like sched.h which will want spinlock.h.

This, maybe?

 include/linux/bottom_half.h |    5 +++++
 include/linux/interrupt.h   |    7 +------
 include/linux/spinlock.h    |    1 +
 3 files changed, 7 insertions(+), 6 deletions(-)
include/linux/bottom_half.h new file +5 -0
diff -puN /dev/null include/linux/bottom_half.h
@@ -0,0 +1,5 @@
+extern void local_bh_disable(void);
+extern void __local_bh_enable(void);
+extern void _local_bh_enable(void);
+extern void local_bh_enable(void);
+extern void local_bh_enable_ip(unsigned long ip);
include/linux/interrupt.h renamed from include/linux/interrupt.h~add-bottom_half.h +1 -6
diff -puN include/linux/interrupt.h~add-bottom_half.h include/linux/interrupt.h
@@ -11,6 +11,7 @@
 #include <linux/hardirq.h>
 #include <linux/sched.h>
 #include <linux/irqflags.h>
+#include <linux/bottom_half.h>
 #include <asm/atomic.h>
 #include <asm/ptrace.h>
 #include <asm/system.h>
@@ -217,12 +218,6 @@ static inline void __deprecated save_and
 #define save_and_cli(x)	save_and_cli(&x)
 #endif /* CONFIG_SMP */
 
-extern void local_bh_disable(void);
-extern void __local_bh_enable(void);
-extern void _local_bh_enable(void);
-extern void local_bh_enable(void);
-extern void local_bh_enable_ip(unsigned long ip);
-
 /* PLEASE, avoid to allocate new softirqs, if you need not _really_ high
    frequency threaded job scheduling. For almost all the purposes
    tasklets are more than enough. F.e. all serial device BHs et
include/linux/spinlock.h renamed from include/linux/spinlock.h~add-bottom_half.h +1 -0
diff -puN include/linux/spinlock.h~add-bottom_half.h include/linux/spinlock.h
@@ -52,6 +52,7 @@
 #include <linux/thread_info.h>
 #include <linux/kernel.h>
 #include <linux/stringify.h>
+#include <linux/bottom_half.h>
 
 #include <asm/system.h>
 

_

#8 lore raw
Quoted 23 lines · Andrew Morton
Quoted 20 lines · On Thu, 30 Nov 2006 14:11:40 -0800 Randy Dunlap <[email protected]> wrote:
Quoted 11 lines · On Thu, 30 Nov 2006 08:00:00 +0100 Ingo Molnar wrote:
Quoted 8 lines · Andrew Morton

No, I think this patch is right - the declaration of the CONFIG_SMP smp_call_function_single() is in linux/smp.h so the !CONFIG_SMP declaration
or definition should be there too.

It's still buggy though. It should disable local interrupts around the
call to match the SMP version. I'll fix that separately.

hm, didnt i send an updated patch for that already? See the patch below, from many days ago. I sent it after the tsc-sync-rewrite patch.

Hi Ingo,

Has there been a patch for this one? (UP again, not SMP)

drivers/input/ff-memless.c:384: warning: implicit declaration of function 'local_bh_disable' drivers/input/ff-memless.c:393: warning: implicit declaration of function 'local_bh_enable'

Quoted 8 lines
Quoted 4 lines

Thanks,
---
~Randy
config: http://oss.oracle.com/~rdunlap/configs/config-input-up-header

eww.. I guess linux/spinlock.h should really include linux/interrupt.h. But interrupt.h includes stuff like sched.h which will want spinlock.h.

This, maybe?

Looks good. I had already tried (and failed) adding interrupt.h to spinlock.h -- what a mess.

Signature · 1 line
~Randy
#9 lore raw
In reply to Andrew Morton
Quoted 30 lines · On Thu, 30 Nov 2006 14:27:19 -0800 Andrew Morton wrote:
Quoted 19 lines · On Thu, 30 Nov 2006 14:11:40 -0800 Randy Dunlap <[email protected]> wrote:
Quoted 6 lines · On Thu, 30 Nov 2006 08:00:00 +0100 Ingo Molnar wrote:
Quoted 5 lines · Andrew Morton

No, I think this patch is right - the declaration of the CONFIG_SMP smp_call_function_single() is in linux/smp.h so the !CONFIG_SMP declaration
or definition should be there too.

Quoted 6 lines
Quoted 3 lines

It's still buggy though. It should disable local interrupts around the
call to match the SMP version. I'll fix that separately.

hm, didnt i send an updated patch for that already? See the patch below, from many days ago. I sent it after the tsc-sync-rewrite patch.

Hi Ingo,

Has there been a patch for this one? (UP again, not SMP)

Quoted 7 lines

drivers/input/ff-memless.c:384: warning: implicit declaration of function 'local_bh_disable' drivers/input/ff-memless.c:393: warning: implicit declaration of function 'local_bh_enable'

Thanks,
---
~Randy
config: http://oss.oracle.com/~rdunlap/configs/config-input-up-header

Quoted 4 lines

eww.. I guess linux/spinlock.h should really include linux/interrupt.h. But interrupt.h includes stuff like sched.h which will want spinlock.h.

This, maybe?

Ack. Tested on UP and SMP x86_64.

Quoted 42 lines
 include/linux/bottom_half.h |    5 +++++
 include/linux/interrupt.h   |    7 +------
 include/linux/spinlock.h    |    1 +
 3 files changed, 7 insertions(+), 6 deletions(-)

diff -puN /dev/null include/linux/bottom_half.h

include/linux/bottom_half.h new file +5 -0
@@ -0,0 +1,5 @@
+extern void local_bh_disable(void);
+extern void __local_bh_enable(void);
+extern void _local_bh_enable(void);
+extern void local_bh_enable(void);
+extern void local_bh_enable_ip(unsigned long ip);

diff -puN include/linux/interrupt.h~add-bottom_half.h include/linux/interrupt.h

include/linux/interrupt.h renamed from include/linux/interrupt.h~add-bottom_half.h +1 -6
@@ -11,6 +11,7 @@
 #include <linux/hardirq.h>
 #include <linux/sched.h>
 #include <linux/irqflags.h>
+#include <linux/bottom_half.h>
 #include <asm/atomic.h>
 #include <asm/ptrace.h>
 #include <asm/system.h>
@@ -217,12 +218,6 @@ static inline void __deprecated save_and
 #define save_and_cli(x)	save_and_cli(&x)
 #endif /* CONFIG_SMP */
 
-extern void local_bh_disable(void);
-extern void __local_bh_enable(void);
-extern void _local_bh_enable(void);
-extern void local_bh_enable(void);
-extern void local_bh_enable_ip(unsigned long ip);
-
 /* PLEASE, avoid to allocate new softirqs, if you need not _really_ high
    frequency threaded job scheduling. For almost all the purposes
    tasklets are more than enough. F.e. all serial device BHs et

diff -puN include/linux/spinlock.h~add-bottom_half.h include/linux/spinlock.h

include/linux/spinlock.h line 57

Quoted 6 lines
include/linux/spinlock.h renamed from include/linux/spinlock.h~add-bottom_half.h +1 -0
@@ -52,6 +52,7 @@
 #include <linux/thread_info.h>
 #include <linux/kernel.h>
 #include <linux/stringify.h>
+#include <linux/bottom_half.h>
 
 #include <asm/system.h>

---
~Randy