[PATCH] Scheduler Profiling - Use Immediate Values

[PATCH] Scheduler Profiling - Use Immediate Values

[Mathieu Desnoyers]

Scheduler Profiling - Use Immediate Values

Use immediate values with lower d-cache hit in optimized version as a
condition for scheduler profiling call.

Depends on the Immediate values patches, applies to 2.6.23-rc4-mm1.

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
---
 drivers/kvm/svm.c       |    2 +-
 drivers/kvm/vmx.c       |    2 +-
 include/linux/profile.h |   10 ++++------
 kernel/profile.c        |   38 ++++++++++++++++++++++++++------------
 kernel/sched.c          |    3 ++-
 5 files changed, 34 insertions(+), 21 deletions(-)

Index: linux-2.6-lttng/kernel/profile.c
===================================================================
--- linux-2.6-lttng.orig/kernel/profile.c	2007-09-04 11:53:23.000000000 -0400
+++ linux-2.6-lttng/kernel/profile.c	2007-09-04 12:12:16.000000000 -0400
@@ -42,9 +42,6 @@ static int (*timer_hook)(struct pt_regs 
 static atomic_t *prof_buffer;
 static unsigned long prof_len, prof_shift;
 
-int prof_on __read_mostly;
-EXPORT_SYMBOL_GPL(prof_on);
-
 static cpumask_t prof_cpu_mask = CPU_MASK_ALL;
 #ifdef CONFIG_SMP
 static DEFINE_PER_CPU(struct profile_hit *[2], cpu_profile_hits);
@@ -52,6 +49,14 @@ static DEFINE_PER_CPU(int, cpu_profile_f
 static DEFINE_MUTEX(profile_flip_mutex);
 #endif /* CONFIG_SMP */
 
+/* Immediate values */
+immediate_char_t sleep_profiling __read_mostly,
+			sched_profiling __read_mostly,
+			kvm_profiling __read_mostly,
+			cpu_profiling __read_mostly;
+EXPORT_SYMBOL_GPL(kvm_profiling);
+EXPORT_SYMBOL_GPL(cpu_profiling);
+
 static int __init profile_setup(char * str)
 {
 	static char __initdata schedstr[] = "schedule";
@@ -60,7 +65,7 @@ static int __init profile_setup(char * s
 	int par;
 
 	if (!strncmp(str, sleepstr, strlen(sleepstr))) {
-		prof_on = SLEEP_PROFILING;
+		immediate_set_early(&sleep_profiling, 1);
 		if (str[strlen(sleepstr)] == ',')
 			str += strlen(sleepstr) + 1;
 		if (get_option(&str, &par))
@@ -69,7 +74,7 @@ static int __init profile_setup(char * s
 			"kernel sleep profiling enabled (shift: %ld)\n",
 			prof_shift);
 	} else if (!strncmp(str, schedstr, strlen(schedstr))) {
-		prof_on = SCHED_PROFILING;
+		immediate_set_early(&sched_profiling, 1);
 		if (str[strlen(schedstr)] == ',')
 			str += strlen(schedstr) + 1;
 		if (get_option(&str, &par))
@@ -78,7 +83,7 @@ static int __init profile_setup(char * s
 			"kernel schedule profiling enabled (shift: %ld)\n",
 			prof_shift);
 	} else if (!strncmp(str, kvmstr, strlen(kvmstr))) {
-		prof_on = KVM_PROFILING;
+		immediate_set_early(&kvm_profiling, 1);
 		if (str[strlen(kvmstr)] == ',')
 			str += strlen(kvmstr) + 1;
 		if (get_option(&str, &par))
@@ -88,7 +93,7 @@ static int __init profile_setup(char * s
 			prof_shift);
 	} else if (get_option(&str, &par)) {
 		prof_shift = par;
-		prof_on = CPU_PROFILING;
+		immediate_set_early(&cpu_profiling, 1);
 		printk(KERN_INFO "kernel profiling enabled (shift: %ld)\n",
 			prof_shift);
 	}
@@ -99,7 +104,10 @@ __setup("profile=", profile_setup);
 
 void __init profile_init(void)
 {
-	if (!prof_on) 
+	if (!_immediate_read(&sleep_profiling) &&
+		!_immediate_read(&sched_profiling) &&
+		!_immediate_read(&kvm_profiling) &&
+		!_immediate_read(&cpu_profiling))
 		return;
  
 	/* only text is profiled */
@@ -288,7 +296,7 @@ void profile_hits(int type, void *__pc, 
 	int i, j, cpu;
 	struct profile_hit *hits;
 
-	if (prof_on != type || !prof_buffer)
+	if (!prof_buffer)
 		return;
 	pc = min((pc - (unsigned long)_stext) >> prof_shift, prof_len - 1);
 	i = primary = (pc & (NR_PROFILE_GRP - 1)) << PROFILE_GRPSHIFT;
@@ -398,7 +406,7 @@ void profile_hits(int type, void *__pc, 
 {
 	unsigned long pc;
 
-	if (prof_on != type || !prof_buffer)
+	if (!prof_buffer)
 		return;
 	pc = ((unsigned long)__pc - (unsigned long)_stext) >> prof_shift;
 	atomic_add(nr_hits, &prof_buffer[min(pc, prof_len - 1)]);
@@ -555,7 +563,10 @@ static int __init create_hash_tables(voi
 	}
 	return 0;
 out_cleanup:
-	prof_on = 0;
+	immediate_set_early(&sleep_profiling, 0);
+	immediate_set_early(&sched_profiling, 0);
+	immediate_set_early(&kvm_profiling, 0);
+	immediate_set_early(&cpu_profiling, 0);
 	smp_mb();
 	on_each_cpu(profile_nop, NULL, 0, 1);
 	for_each_online_cpu(cpu) {
@@ -582,7 +593,10 @@ static int __init create_proc_profile(vo
 {
 	struct proc_dir_entry *entry;
 
-	if (!prof_on)
+	if (!_immediate_read(&sleep_profiling) &&
+		!_immediate_read(&sched_profiling) &&
+		!_immediate_read(&kvm_profiling) &&
+		!_immediate_read(&cpu_profiling))
 		return 0;
 	if (create_hash_tables())
 		return -1;
Index: linux-2.6-lttng/include/linux/profile.h
===================================================================
--- linux-2.6-lttng.orig/include/linux/profile.h	2007-09-04 11:53:23.000000000 -0400
+++ linux-2.6-lttng/include/linux/profile.h	2007-09-04 12:12:16.000000000 -0400
@@ -7,10 +7,12 @@
 #include <linux/init.h>
 #include <linux/cpumask.h>
 #include <linux/cache.h>
+#include <linux/immediate.h>
 
 #include <asm/errno.h>
 
-extern int prof_on __read_mostly;
+extern immediate_char_t sleep_profiling, sched_profiling, kvm_profiling,
+		cpu_profiling;
 
 #define CPU_PROFILING	1
 #define SCHED_PROFILING	2
@@ -35,11 +37,7 @@ void profile_hits(int, void *ip, unsigne
  */
 static inline void profile_hit(int type, void *ip)
 {
-	/*
-	 * Speedup for the common (no profiling enabled) case:
-	 */
-	if (unlikely(prof_on == type))
-		profile_hits(type, ip, 1);
+	profile_hits(type, ip, 1);
 }
 
 #ifdef CONFIG_PROC_FS
Index: linux-2.6-lttng/kernel/sched.c
===================================================================
--- linux-2.6-lttng.orig/kernel/sched.c	2007-09-04 11:54:05.000000000 -0400
+++ linux-2.6-lttng/kernel/sched.c	2007-09-04 12:12:17.000000000 -0400
@@ -3458,7 +3458,8 @@ static inline void schedule_debug(struct
 	if (unlikely(in_atomic_preempt_off()) && unlikely(!prev->exit_state))
 		__schedule_bug(prev);
 
-	profile_hit(SCHED_PROFILING, __builtin_return_address(0));
+	immediate_if (&sched_profiling)
+		profile_hit(SCHED_PROFILING, __builtin_return_address(0));
 
 	schedstat_inc(this_rq(), sched_cnt);
 }
Index: linux-2.6-lttng/drivers/kvm/svm.c
===================================================================
--- linux-2.6-lttng.orig/drivers/kvm/svm.c	2007-09-04 11:51:53.000000000 -0400
+++ linux-2.6-lttng/drivers/kvm/svm.c	2007-09-04 12:12:17.000000000 -0400
@@ -1570,7 +1570,7 @@ again:
 	/*
 	 * Profile KVM exit RIPs:
 	 */
-	if (unlikely(prof_on == KVM_PROFILING))
+	immediate_if (&kvm_profiling)
 		profile_hit(KVM_PROFILING,
 			(void *)(unsigned long)svm->vmcb->save.rip);
 
Index: linux-2.6-lttng/drivers/kvm/vmx.c
===================================================================
--- linux-2.6-lttng.orig/drivers/kvm/vmx.c	2007-09-04 11:51:53.000000000 -0400
+++ linux-2.6-lttng/drivers/kvm/vmx.c	2007-09-04 12:12:17.000000000 -0400
@@ -2234,7 +2234,7 @@ again:
 	/*
 	 * Profile KVM exit RIPs:
 	 */
-	if (unlikely(prof_on == KVM_PROFILING))
+	immediate_if (&kvm_profiling)
 		profile_hit(KVM_PROFILING, (void *)vmcs_readl(GUEST_RIP));
 
 	r = kvm_handle_exit(kvm_run, vcpu);
-- 
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68

Re: [PATCH] Scheduler Profiling - Use Immediate Values

[Randy Dunlap]

On Thu, 6 Sep 2007 16:05:37 -0400 Mathieu Desnoyers wrote:

> Scheduler Profiling - Use Immediate Values
> 
> Use immediate values with lower d-cache hit in optimized version as a
> condition for scheduler profiling call.
> 
> Depends on the Immediate values patches, applies to 2.6.23-rc4-mm1.
> 
> Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
> ---
>  drivers/kvm/svm.c       |    2 +-
>  drivers/kvm/vmx.c       |    2 +-
>  include/linux/profile.h |   10 ++++------
>  kernel/profile.c        |   38 ++++++++++++++++++++++++++------------
>  kernel/sched.c          |    3 ++-
>  5 files changed, 34 insertions(+), 21 deletions(-)
> 
> Index: linux-2.6-lttng/kernel/profile.c
> ===================================================================
> --- linux-2.6-lttng.orig/kernel/profile.c	2007-09-04 11:53:23.000000000 -0400
> +++ linux-2.6-lttng/kernel/profile.c	2007-09-04 12:12:16.000000000 -0400
> @@ -42,9 +42,6 @@ static int (*timer_hook)(struct pt_regs 
>  static atomic_t *prof_buffer;
>  static unsigned long prof_len, prof_shift;
>  
> -int prof_on __read_mostly;
> -EXPORT_SYMBOL_GPL(prof_on);
> -
>  static cpumask_t prof_cpu_mask = CPU_MASK_ALL;
>  #ifdef CONFIG_SMP
>  static DEFINE_PER_CPU(struct profile_hit *[2], cpu_profile_hits);
> @@ -52,6 +49,14 @@ static DEFINE_PER_CPU(int, cpu_profile_f
>  static DEFINE_MUTEX(profile_flip_mutex);
>  #endif /* CONFIG_SMP */
>  
> +/* Immediate values */
> +immediate_char_t sleep_profiling __read_mostly,
> +			sched_profiling __read_mostly,
> +			kvm_profiling __read_mostly,
> +			cpu_profiling __read_mostly;
> +EXPORT_SYMBOL_GPL(kvm_profiling);
> +EXPORT_SYMBOL_GPL(cpu_profiling);
> +
>  static int __init profile_setup(char * str)
>  {
>  	static char __initdata schedstr[] = "schedule";
> @@ -60,7 +65,7 @@ static int __init profile_setup(char * s
>  	int par;
>  
>  	if (!strncmp(str, sleepstr, strlen(sleepstr))) {
> -		prof_on = SLEEP_PROFILING;
> +		immediate_set_early(&sleep_profiling, 1);

These are booleans, eh?  so one could use            , true);

etc.

>  		if (str[strlen(sleepstr)] == ',')
>  			str += strlen(sleepstr) + 1;
>  		if (get_option(&str, &par))
> @@ -99,7 +104,10 @@ __setup("profile=", profile_setup);
>  
>  void __init profile_init(void)
>  {
> -	if (!prof_on) 
> +	if (!_immediate_read(&sleep_profiling) &&
> +		!_immediate_read(&sched_profiling) &&
> +		!_immediate_read(&kvm_profiling) &&
> +		!_immediate_read(&cpu_profiling))

Looks like this (in multiple places) needs to use prof_on, where
prof_on is set earlier as the logical 'or' of all (4 now, could be more)
profiling flags.
Or below, sometimes you just check prof_buffer.  Is that sufficient
here also?

>  		return;
>   
>  	/* only text is profiled */
> @@ -288,7 +296,7 @@ void profile_hits(int type, void *__pc, 
>  	int i, j, cpu;
>  	struct profile_hit *hits;
>  
> -	if (prof_on != type || !prof_buffer)
> +	if (!prof_buffer)
>  		return;
>  	pc = min((pc - (unsigned long)_stext) >> prof_shift, prof_len - 1);
>  	i = primary = (pc & (NR_PROFILE_GRP - 1)) << PROFILE_GRPSHIFT;
> @@ -398,7 +406,7 @@ void profile_hits(int type, void *__pc, 
>  {
>  	unsigned long pc;
>  
> -	if (prof_on != type || !prof_buffer)
> +	if (!prof_buffer)
>  		return;
>  	pc = ((unsigned long)__pc - (unsigned long)_stext) >> prof_shift;
>  	atomic_add(nr_hits, &prof_buffer[min(pc, prof_len - 1)]);
> @@ -555,7 +563,10 @@ static int __init create_hash_tables(voi
>  	}
>  	return 0;
>  out_cleanup:
> -	prof_on = 0;
> +	immediate_set_early(&sleep_profiling, 0);
> +	immediate_set_early(&sched_profiling, 0);
> +	immediate_set_early(&kvm_profiling, 0);
> +	immediate_set_early(&cpu_profiling, 0);
>  	smp_mb();
>  	on_each_cpu(profile_nop, NULL, 0, 1);
>  	for_each_online_cpu(cpu) {
> @@ -582,7 +593,10 @@ static int __init create_proc_profile(vo
>  {
>  	struct proc_dir_entry *entry;
>  
> -	if (!prof_on)
> +	if (!_immediate_read(&sleep_profiling) &&
> +		!_immediate_read(&sched_profiling) &&
> +		!_immediate_read(&kvm_profiling) &&
> +		!_immediate_read(&cpu_profiling))
>  		return 0;
>  	if (create_hash_tables())
>  		return -1;


---
~Randy
*** Remember to use Documentation/SubmitChecklist when testing your code ***

Re: [PATCH] Scheduler Profiling - Use Immediate Values

[Mathieu Desnoyers]

Hi Randy,

* Randy Dunlap (randy.dunlap@oracle.com) wrote:
> On Thu, 6 Sep 2007 16:05:37 -0400 Mathieu Desnoyers wrote:
> 
> > Scheduler Profiling - Use Immediate Values
> > 
> > Use immediate values with lower d-cache hit in optimized version as a
> > condition for scheduler profiling call.
> > 
> > Depends on the Immediate values patches, applies to 2.6.23-rc4-mm1.
> > 
> > Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
> > ---
> >  drivers/kvm/svm.c       |    2 +-
> >  drivers/kvm/vmx.c       |    2 +-
> >  include/linux/profile.h |   10 ++++------
> >  kernel/profile.c        |   38 ++++++++++++++++++++++++++------------
> >  kernel/sched.c          |    3 ++-
> >  5 files changed, 34 insertions(+), 21 deletions(-)
> > 
> > Index: linux-2.6-lttng/kernel/profile.c
> > ===================================================================
> > --- linux-2.6-lttng.orig/kernel/profile.c	2007-09-04 11:53:23.000000000 -0400
> > +++ linux-2.6-lttng/kernel/profile.c	2007-09-04 12:12:16.000000000 -0400
> > @@ -42,9 +42,6 @@ static int (*timer_hook)(struct pt_regs 
> >  static atomic_t *prof_buffer;
> >  static unsigned long prof_len, prof_shift;
> >  
> > -int prof_on __read_mostly;
> > -EXPORT_SYMBOL_GPL(prof_on);
> > -
> >  static cpumask_t prof_cpu_mask = CPU_MASK_ALL;
> >  #ifdef CONFIG_SMP
> >  static DEFINE_PER_CPU(struct profile_hit *[2], cpu_profile_hits);
> > @@ -52,6 +49,14 @@ static DEFINE_PER_CPU(int, cpu_profile_f
> >  static DEFINE_MUTEX(profile_flip_mutex);
> >  #endif /* CONFIG_SMP */
> >  
> > +/* Immediate values */
> > +immediate_char_t sleep_profiling __read_mostly,
> > +			sched_profiling __read_mostly,
> > +			kvm_profiling __read_mostly,
> > +			cpu_profiling __read_mostly;
> > +EXPORT_SYMBOL_GPL(kvm_profiling);
> > +EXPORT_SYMBOL_GPL(cpu_profiling);
> > +
> >  static int __init profile_setup(char * str)
> >  {
> >  	static char __initdata schedstr[] = "schedule";
> > @@ -60,7 +65,7 @@ static int __init profile_setup(char * s
> >  	int par;
> >  
> >  	if (!strncmp(str, sleepstr, strlen(sleepstr))) {
> > -		prof_on = SLEEP_PROFILING;
> > +		immediate_set_early(&sleep_profiling, 1);
> 
> These are booleans, eh?  so one could use            , true);
> 
> etc.
> 
> >  		if (str[strlen(sleepstr)] == ',')
> >  			str += strlen(sleepstr) + 1;
> >  		if (get_option(&str, &par))
> > @@ -99,7 +104,10 @@ __setup("profile=", profile_setup);
> >  
> >  void __init profile_init(void)
> >  {
> > -	if (!prof_on) 
> > +	if (!_immediate_read(&sleep_profiling) &&
> > +		!_immediate_read(&sched_profiling) &&
> > +		!_immediate_read(&kvm_profiling) &&
> > +		!_immediate_read(&cpu_profiling))
> 
> Looks like this (in multiple places) needs to use prof_on, where
> prof_on is set earlier as the logical 'or' of all (4 now, could be more)
> profiling flags.

I first did this patch when the immediate values only supported
booleans. I just redid it encoding the original prof_on values in a
signle immediate_char_t variable. It makes the change must less
intrusive.

Since there seem to be only one type of profiling possible at a given
time, there is no need for a bitfield : just doing the same as prof_on
currently does seems ok.

> Or below, sometimes you just check prof_buffer.  Is that sufficient
> here also?
> 

Yes, before we already do the check early in profile_hit (inline
function checking the same condition). In this patch, I moved the check
out of profile_hit, but since I now re-use prof_on with values, I can
put it back in profile_hit().

> >  		return;
> >   
> >  	/* only text is profiled */
> > @@ -288,7 +296,7 @@ void profile_hits(int type, void *__pc, 
> >  	int i, j, cpu;
> >  	struct profile_hit *hits;
> >  
> > -	if (prof_on != type || !prof_buffer)
> > +	if (!prof_buffer)
> >  		return;
> >  	pc = min((pc - (unsigned long)_stext) >> prof_shift, prof_len - 1);
> >  	i = primary = (pc & (NR_PROFILE_GRP - 1)) << PROFILE_GRPSHIFT;
> > @@ -398,7 +406,7 @@ void profile_hits(int type, void *__pc, 
> >  {
> >  	unsigned long pc;
> >  
> > -	if (prof_on != type || !prof_buffer)
> > +	if (!prof_buffer)
> >  		return;
> >  	pc = ((unsigned long)__pc - (unsigned long)_stext) >> prof_shift;
> >  	atomic_add(nr_hits, &prof_buffer[min(pc, prof_len - 1)]);
> > @@ -555,7 +563,10 @@ static int __init create_hash_tables(voi
> >  	}
> >  	return 0;
> >  out_cleanup:
> > -	prof_on = 0;
> > +	immediate_set_early(&sleep_profiling, 0);
> > +	immediate_set_early(&sched_profiling, 0);
> > +	immediate_set_early(&kvm_profiling, 0);
> > +	immediate_set_early(&cpu_profiling, 0);
> >  	smp_mb();
> >  	on_each_cpu(profile_nop, NULL, 0, 1);
> >  	for_each_online_cpu(cpu) {
> > @@ -582,7 +593,10 @@ static int __init create_proc_profile(vo
> >  {
> >  	struct proc_dir_entry *entry;
> >  
> > -	if (!prof_on)
> > +	if (!_immediate_read(&sleep_profiling) &&
> > +		!_immediate_read(&sched_profiling) &&
> > +		!_immediate_read(&kvm_profiling) &&
> > +		!_immediate_read(&cpu_profiling))
> >  		return 0;
> >  	if (create_hash_tables())
> >  		return -1;
> 
> 
> ---
> ~Randy
> *** Remember to use Documentation/SubmitChecklist when testing your code ***

-- 
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68

Re: [PATCH] Scheduler Profiling - Use Immediate Values

[Andi Kleen]

Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> writes:
>  
> -	profile_hit(SCHED_PROFILING, __builtin_return_address(0));
> +	immediate_if (&sched_profiling)

I must say I really dislike immediate_if(). You complained earlier
that something breaks coloring, but adding such macros will definitely
break a lot of editors (especially if you use it without {} like here)

It would be much nicer and readable to just use if 
(unlikely(immediate_read(&x)) or if you prefer to hide the unlikely
if (immediate_bool_test(&x)) with an implicit unlikely().

-Andi

Re: [PATCH] Scheduler Profiling - Use Immediate Values

[Mathieu Desnoyers]

* Andi Kleen (andi@firstfloor.org) wrote:
> Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> writes:
> >  
> > -	profile_hit(SCHED_PROFILING, __builtin_return_address(0));
> > +	immediate_if (&sched_profiling)
> 
> I must say I really dislike immediate_if(). You complained earlier
> that something breaks coloring, but adding such macros will definitely
> break a lot of editors (especially if you use it without {} like here)
> 
> It would be much nicer and readable to just use if 
> (unlikely(immediate_read(&x)) or if you prefer to hide the unlikely
> if (immediate_bool_test(&x)) with an implicit unlikely().
> 
> -Andi

Please refer to :

http://lkml.org/lkml/2007/7/3/301

The idea is not to hide the unlikely, but to leave the opportunity to
make this primitive evolve in something that won't depend on a load
immediate and only require patching of a jump, given the appropriate gcc
support (yet to come).

Mathieu


-- 
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68

Re: [PATCH] Scheduler Profiling - Use Immediate Values

[Andi Kleen]

> The idea is not to hide the unlikely, but to leave the opportunity to
> make this primitive evolve in something that won't depend on a load
> immediate and only require patching of a jump, given the appropriate gcc
> support (yet to come).

If that ever happens the code can be still changed. But i don't think
it's a good idea to uglify the code for something that if it 
ever exists will be a long time away.

Besides if gcc supports it I assume the gcc support could 
also be written in a way that makes it possible to hide it inside
a standard if ()

-Andi

[PATCH] Scheduler Profiling - Use Immediate Values

[Mathieu Desnoyers]

Use immediate values with lower d-cache hit in optimized version as a
condition for scheduler profiling call.

Changelog :
- Use imv_* instead of immediate_*.

It applies on 2.6.24-rc4-git3. It depends on the Immediate Values
(redux) patchset.

Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
---
 drivers/kvm/kvm_main.c  |    3 ++-
 include/linux/profile.h |    5 +++--
 kernel/profile.c        |   22 +++++++++++-----------
 kernel/sched_fair.c     |    6 +-----
 4 files changed, 17 insertions(+), 19 deletions(-)

Index: linux-2.6-lttng/kernel/profile.c
===================================================================
--- linux-2.6-lttng.orig/kernel/profile.c	2007-12-05 20:50:34.000000000 -0500
+++ linux-2.6-lttng/kernel/profile.c	2007-12-05 20:53:43.000000000 -0500
@@ -42,8 +42,8 @@ static int (*timer_hook)(struct pt_regs 
 static atomic_t *prof_buffer;
 static unsigned long prof_len, prof_shift;
 
-int prof_on __read_mostly;
-EXPORT_SYMBOL_GPL(prof_on);
+DEFINE_IMV(char, prof_on) __read_mostly;
+EXPORT_IMV_SYMBOL_GPL(prof_on);
 
 static cpumask_t prof_cpu_mask = CPU_MASK_ALL;
 #ifdef CONFIG_SMP
@@ -61,7 +61,7 @@ static int __init profile_setup(char * s
 
 	if (!strncmp(str, sleepstr, strlen(sleepstr))) {
 #ifdef CONFIG_SCHEDSTATS
-		prof_on = SLEEP_PROFILING;
+		imv_set(prof_on, SLEEP_PROFILING);
 		if (str[strlen(sleepstr)] == ',')
 			str += strlen(sleepstr) + 1;
 		if (get_option(&str, &par))
@@ -74,7 +74,7 @@ static int __init profile_setup(char * s
 			"kernel sleep profiling requires CONFIG_SCHEDSTATS\n");
 #endif /* CONFIG_SCHEDSTATS */
 	} else if (!strncmp(str, schedstr, strlen(schedstr))) {
-		prof_on = SCHED_PROFILING;
+		imv_set(prof_on, SCHED_PROFILING);
 		if (str[strlen(schedstr)] == ',')
 			str += strlen(schedstr) + 1;
 		if (get_option(&str, &par))
@@ -83,7 +83,7 @@ static int __init profile_setup(char * s
 			"kernel schedule profiling enabled (shift: %ld)\n",
 			prof_shift);
 	} else if (!strncmp(str, kvmstr, strlen(kvmstr))) {
-		prof_on = KVM_PROFILING;
+		imv_set(prof_on, KVM_PROFILING);
 		if (str[strlen(kvmstr)] == ',')
 			str += strlen(kvmstr) + 1;
 		if (get_option(&str, &par))
@@ -93,7 +93,7 @@ static int __init profile_setup(char * s
 			prof_shift);
 	} else if (get_option(&str, &par)) {
 		prof_shift = par;
-		prof_on = CPU_PROFILING;
+		imv_set(prof_on, CPU_PROFILING);
 		printk(KERN_INFO "kernel profiling enabled (shift: %ld)\n",
 			prof_shift);
 	}
@@ -104,7 +104,7 @@ __setup("profile=", profile_setup);
 
 void __init profile_init(void)
 {
-	if (!prof_on) 
+	if (!_imv_read(prof_on))
 		return;
  
 	/* only text is profiled */
@@ -293,7 +293,7 @@ void profile_hits(int type, void *__pc, 
 	int i, j, cpu;
 	struct profile_hit *hits;
 
-	if (prof_on != type || !prof_buffer)
+	if (!prof_buffer)
 		return;
 	pc = min((pc - (unsigned long)_stext) >> prof_shift, prof_len - 1);
 	i = primary = (pc & (NR_PROFILE_GRP - 1)) << PROFILE_GRPSHIFT;
@@ -403,7 +403,7 @@ void profile_hits(int type, void *__pc, 
 {
 	unsigned long pc;
 
-	if (prof_on != type || !prof_buffer)
+	if (!prof_buffer)
 		return;
 	pc = ((unsigned long)__pc - (unsigned long)_stext) >> prof_shift;
 	atomic_add(nr_hits, &prof_buffer[min(pc, prof_len - 1)]);
@@ -560,7 +560,7 @@ static int __init create_hash_tables(voi
 	}
 	return 0;
 out_cleanup:
-	prof_on = 0;
+	imv_set(prof_on, 0);
 	smp_mb();
 	on_each_cpu(profile_nop, NULL, 0, 1);
 	for_each_online_cpu(cpu) {
@@ -587,7 +587,7 @@ static int __init create_proc_profile(vo
 {
 	struct proc_dir_entry *entry;
 
-	if (!prof_on)
+	if (!_imv_read(prof_on))
 		return 0;
 	if (create_hash_tables())
 		return -1;
Index: linux-2.6-lttng/include/linux/profile.h
===================================================================
--- linux-2.6-lttng.orig/include/linux/profile.h	2007-12-05 20:50:34.000000000 -0500
+++ linux-2.6-lttng/include/linux/profile.h	2007-12-05 20:53:43.000000000 -0500
@@ -7,10 +7,11 @@
 #include <linux/init.h>
 #include <linux/cpumask.h>
 #include <linux/cache.h>
+#include <linux/immediate.h>
 
 #include <asm/errno.h>
 
-extern int prof_on __read_mostly;
+DECLARE_IMV(char, prof_on) __read_mostly;
 
 #define CPU_PROFILING	1
 #define SCHED_PROFILING	2
@@ -38,7 +39,7 @@ static inline void profile_hit(int type,
 	/*
 	 * Speedup for the common (no profiling enabled) case:
 	 */
-	if (unlikely(prof_on == type))
+	if (unlikely(imv_read(prof_on) == type))
 		profile_hits(type, ip, 1);
 }
 
Index: linux-2.6-lttng/drivers/kvm/kvm_main.c
===================================================================
--- linux-2.6-lttng.orig/drivers/kvm/kvm_main.c	2007-12-05 20:50:34.000000000 -0500
+++ linux-2.6-lttng/drivers/kvm/kvm_main.c	2007-12-05 20:53:43.000000000 -0500
@@ -2053,7 +2053,8 @@ again:
 	/*
 	 * Profile KVM exit RIPs:
 	 */
-	if (unlikely(prof_on == KVM_PROFILING)) {
+
+	if (unlikely(imv_read(prof_on) == KVM_PROFILING)) {
 		kvm_x86_ops->cache_regs(vcpu);
 		profile_hit(KVM_PROFILING, (void *)vcpu->rip);
 	}
Index: linux-2.6-lttng/kernel/sched_fair.c
===================================================================
--- linux-2.6-lttng.orig/kernel/sched_fair.c	2007-12-05 20:52:28.000000000 -0500
+++ linux-2.6-lttng/kernel/sched_fair.c	2007-12-05 20:53:43.000000000 -0500
@@ -461,12 +461,8 @@ static void enqueue_sleeper(struct cfs_r
 		 * get a milliseconds-range estimation of the amount of
 		 * time that the task spent sleeping:
 		 */
-		if (unlikely(prof_on == SLEEP_PROFILING)) {
-			struct task_struct *tsk = task_of(se);
-
-			profile_hits(SLEEP_PROFILING, (void *)get_wchan(tsk),
+		profile_hits(SLEEP_PROFILING, (void *)get_wchan(task_of(se)),
 				     delta >> 20);
-		}
 	}
 #endif
 }
-- 
Mathieu Desnoyers
Computer Engineering Ph.D. Student, Ecole Polytechnique de Montreal
OpenPGP key fingerprint: 8CD5 52C3 8E3C 4140 715F  BA06 3F25 A8FE 3BAE 9A68