Reduces overhead a bit and frees up a couple of registers.
Signed-off-by: Christoph Lameter <[email protected]>
Index: linux/arch/x86/include/asm/debugreg.h
===================================================================
--- linux.orig/arch/x86/include/asm/debugreg.h 2013-07-30 14:00:30.000000000 -0500
+++ linux/arch/x86/include/asm/debugreg.h 2013-07-30 14:00:57.503076270 -0500
@@ -97,11 +97,11 @@ extern void hw_breakpoint_restore(void);
DECLARE_PER_CPU(int, debug_stack_usage);
static inline void debug_stack_usage_inc(void)
{
- __get_cpu_var(debug_stack_usage)++;
+ __this_cpu_inc(debug_stack_usage);
}
static inline void debug_stack_usage_dec(void)
{
- __get_cpu_var(debug_stack_usage)--;
+ __this_cpu_dec(debug_stack_usage);
}
int is_debug_stack(unsigned long addr);
void debug_stack_set_zero(void);
On Tue, 2013-08-06 at 17:01 +0000, Christoph Lameter wrote:
> Reduces overhead a bit and frees up a couple of registers.
>
> Signed-off-by: Christoph Lameter <[email protected]>
>
> Index: linux/arch/x86/include/asm/debugreg.h
> ===================================================================
> --- linux.orig/arch/x86/include/asm/debugreg.h 2013-07-30 14:00:30.000000000 -0500
> +++ linux/arch/x86/include/asm/debugreg.h 2013-07-30 14:00:57.503076270 -0500
> @@ -97,11 +97,11 @@ extern void hw_breakpoint_restore(void);
> DECLARE_PER_CPU(int, debug_stack_usage);
> static inline void debug_stack_usage_inc(void)
> {
> - __get_cpu_var(debug_stack_usage)++;
> + __this_cpu_inc(debug_stack_usage);
I don't remember why I didn't use this in the first place. Perhaps I was
still in the "get_cpu" mind set.
Also, what's the difference between "__this_cpu_inc()" and
"this_cpu_inc()"?
> }
> static inline void debug_stack_usage_dec(void)
> {
> - __get_cpu_var(debug_stack_usage)--;
> + __this_cpu_dec(debug_stack_usage);
> }
> int is_debug_stack(unsigned long addr);
> void debug_stack_set_zero(void);
Might as well change is_debug_stack() to use __this_cpu_read() instead
of __get_cpu_var().
-- Steve
> Might as well change is_debug_stack() to use __this_cpu_read() instead
> of __get_cpu_var().
Subject: x86: use this_cpu for debug_stack_usage
Reduces overhead a bit and frees up a couple of registers.
Signed-off-by: Christoph Lameter <[email protected]>
Index: linux/arch/x86/include/asm/debugreg.h
===================================================================
--- linux.orig/arch/x86/include/asm/debugreg.h 2013-08-06 13:52:23.740092873 -0500
+++ linux/arch/x86/include/asm/debugreg.h 2013-08-06 13:52:23.740092873 -0500
@@ -97,11 +97,11 @@ extern void hw_breakpoint_restore(void);
DECLARE_PER_CPU(int, debug_stack_usage);
static inline void debug_stack_usage_inc(void)
{
- __get_cpu_var(debug_stack_usage)++;
+ __this_cpu_inc(debug_stack_usage);
}
static inline void debug_stack_usage_dec(void)
{
- __get_cpu_var(debug_stack_usage)--;
+ __this_cpu_dec(debug_stack_usage);
}
int is_debug_stack(unsigned long addr);
void debug_stack_set_zero(void);
Index: linux/arch/x86/kernel/cpu/common.c
===================================================================
--- linux.orig/arch/x86/kernel/cpu/common.c 2013-07-19 09:06:36.850047837 -0500
+++ linux/arch/x86/kernel/cpu/common.c 2013-08-06 13:54:36.426384889 -0500
@@ -1144,9 +1144,9 @@ DEFINE_PER_CPU(int, debug_stack_usage);
int is_debug_stack(unsigned long addr)
{
- return __get_cpu_var(debug_stack_usage) ||
- (addr <= __get_cpu_var(debug_stack_addr) &&
- addr > (__get_cpu_var(debug_stack_addr) - DEBUG_STKSZ));
+ return __this_cpu_read(debug_stack_usage) ||
+ (addr <= __this_cpu_read(debug_stack_addr) &&
+ addr > (__this_cpu_read(debug_stack_addr) - DEBUG_STKSZ));
}
DEFINE_PER_CPU(u32, debug_idt_ctr);
On Tue, 6 Aug 2013, Steven Rostedt wrote:
> I don't remember why I didn't use this in the first place. Perhaps I was
> still in the "get_cpu" mind set.
>
> Also, what's the difference between "__this_cpu_inc()" and
> "this_cpu_inc()"?
The fallback logic for arches not supporting segment prefixes is differnt.
this_cpu_inc() disables interrupts.
__this_cpu_inc() is used when you know that the scheduler cannot move the
process.
The code generated by both is the same on x86.
> > int is_debug_stack(unsigned long addr);
> > void debug_stack_set_zero(void);
>
> Might as well change is_debug_stack() to use __this_cpu_read() instead
> of __get_cpu_var().
True and that will also shave off some instructions.
On Tue, 2013-08-06 at 18:56 +0000, Christoph Lameter wrote:
> > Might as well change is_debug_stack() to use __this_cpu_read() instead
> > of __get_cpu_var().
I just noticed that you didn't include the x86 maintainers.
Reviewed-by: Steven Rostedt <[email protected]>
-- Steve
>
>
>
>
>
> Subject: x86: use this_cpu for debug_stack_usage
>
> Reduces overhead a bit and frees up a couple of registers.
>
> Signed-off-by: Christoph Lameter <[email protected]>
>
> Index: linux/arch/x86/include/asm/debugreg.h
> ===================================================================
> --- linux.orig/arch/x86/include/asm/debugreg.h 2013-08-06 13:52:23.740092873 -0500
> +++ linux/arch/x86/include/asm/debugreg.h 2013-08-06 13:52:23.740092873 -0500
> @@ -97,11 +97,11 @@ extern void hw_breakpoint_restore(void);
> DECLARE_PER_CPU(int, debug_stack_usage);
> static inline void debug_stack_usage_inc(void)
> {
> - __get_cpu_var(debug_stack_usage)++;
> + __this_cpu_inc(debug_stack_usage);
> }
> static inline void debug_stack_usage_dec(void)
> {
> - __get_cpu_var(debug_stack_usage)--;
> + __this_cpu_dec(debug_stack_usage);
> }
> int is_debug_stack(unsigned long addr);
> void debug_stack_set_zero(void);
> Index: linux/arch/x86/kernel/cpu/common.c
> ===================================================================
> --- linux.orig/arch/x86/kernel/cpu/common.c 2013-07-19 09:06:36.850047837 -0500
> +++ linux/arch/x86/kernel/cpu/common.c 2013-08-06 13:54:36.426384889 -0500
> @@ -1144,9 +1144,9 @@ DEFINE_PER_CPU(int, debug_stack_usage);
>
> int is_debug_stack(unsigned long addr)
> {
> - return __get_cpu_var(debug_stack_usage) ||
> - (addr <= __get_cpu_var(debug_stack_addr) &&
> - addr > (__get_cpu_var(debug_stack_addr) - DEBUG_STKSZ));
> + return __this_cpu_read(debug_stack_usage) ||
> + (addr <= __this_cpu_read(debug_stack_addr) &&
> + addr > (__this_cpu_read(debug_stack_addr) - DEBUG_STKSZ));
> }
>
> DEFINE_PER_CPU(u32, debug_idt_ctr);
On 08/06/2013 12:02 PM, Steven Rostedt wrote:
> On Tue, 2013-08-06 at 18:56 +0000, Christoph Lameter wrote:
>>> Might as well change is_debug_stack() to use __this_cpu_read() instead
>>> of __get_cpu_var().
>
> I just noticed that you didn't include the x86 maintainers.
>
> Reviewed-by: Steven Rostedt <[email protected]>
>
Looks good to me.
Do you want an ACK from me or do you want me to pick it up?
-hpa
On Tue, 2013-08-06 at 13:07 -0700, H. Peter Anvin wrote:
> On 08/06/2013 12:02 PM, Steven Rostedt wrote:
> > On Tue, 2013-08-06 at 18:56 +0000, Christoph Lameter wrote:
> >>> Might as well change is_debug_stack() to use __this_cpu_read() instead
> >>> of __get_cpu_var().
> >
> > I just noticed that you didn't include the x86 maintainers.
> >
> > Reviewed-by: Steven Rostedt <[email protected]>
> >
>
> Looks good to me.
>
> Do you want an ACK from me or do you want me to pick it up?
>
You can take it.
Thanks,
-- Steve