[PATCH] Use 64bit timer values while calibrating BogoMips

[PATCH] Use 64bit timer values while calibrating BogoMips

[Andi Kleen]

When calibrating BogoMips 32bit architectures using 
read_current_timer will truncate the timer value to 32bit.
This is a problem on x86 when the TSC timer is already beyond
2^32-1 or wraps during the calibration. The current code
has no wrap handling.  

This could potentially lead to wrong BogoMips, causing 
incorrect delays e.g. in hardware drivers and then difficult
to track down bugs.

Change read_current_timer() to always return a 64bit
value even on 32bit and do the computation in 64bit.

Only found by code inspection; i haven't seen a failure 
in the wild.

avr32/sparc64 read_current_timer() updated, but I haven't 
compiled/tested them.

Signed-off-by: Andi Kleen <ak@suse.de>

Index: linux-2.6.24-rc1-hack/arch/avr32/lib/delay.c
===================================================================
--- linux-2.6.24-rc1-hack.orig/arch/avr32/lib/delay.c
+++ linux-2.6.24-rc1-hack/arch/avr32/lib/delay.c
@@ -18,7 +18,7 @@
 #include <asm/processor.h>
 #include <asm/sysreg.h>
 
-int read_current_timer(unsigned long *timer_value)
+int read_current_timer(unsigned long long *timer_value)
 {
 	*timer_value = sysreg_read(COUNT);
 	return 0;
Index: linux-2.6.24-rc1-hack/arch/x86/lib/delay_32.c
===================================================================
--- linux-2.6.24-rc1-hack.orig/arch/x86/lib/delay_32.c
+++ linux-2.6.24-rc1-hack/arch/x86/lib/delay_32.c
@@ -60,10 +60,10 @@ void use_tsc_delay(void)
 	delay_fn = delay_tsc;
 }
 
-int read_current_timer(unsigned long *timer_val)
+int read_current_timer(unsigned long long *timer_val)
 {
 	if (delay_fn == delay_tsc) {
-		rdtscl(*timer_val);
+		rdtscll(*timer_val);
 		return 0;
 	}
 	return -1;
Index: linux-2.6.24-rc1-hack/arch/x86/lib/delay_64.c
===================================================================
--- linux-2.6.24-rc1-hack.orig/arch/x86/lib/delay_64.c
+++ linux-2.6.24-rc1-hack/arch/x86/lib/delay_64.c
@@ -18,7 +18,7 @@
 #include <asm/smp.h>
 #endif
 
-int read_current_timer(unsigned long *timer_value)
+int read_current_timer(unsigned long long *timer_value)
 {
 	rdtscll(*timer_value);
 	return 0;
Index: linux-2.6.24-rc1-hack/include/asm-avr32/timex.h
===================================================================
--- linux-2.6.24-rc1-hack.orig/include/asm-avr32/timex.h
+++ linux-2.6.24-rc1-hack/include/asm-avr32/timex.h
@@ -34,7 +34,7 @@ static inline cycles_t get_cycles (void)
 	return 0;
 }
 
-extern int read_current_timer(unsigned long *timer_value);
+extern int read_current_timer(unsigned long long *timer_value);
 #define ARCH_HAS_READ_CURRENT_TIMER	1
 
 #endif /* __ASM_AVR32_TIMEX_H */
Index: linux-2.6.24-rc1-hack/include/asm-x86/timex.h
===================================================================
--- linux-2.6.24-rc1-hack.orig/include/asm-x86/timex.h
+++ linux-2.6.24-rc1-hack/include/asm-x86/timex.h
@@ -12,7 +12,7 @@
 #endif
 #define CLOCK_TICK_RATE	PIT_TICK_RATE
 
-extern int read_current_timer(unsigned long *timer_value);
+extern int read_current_timer(unsigned long long *timer_value);
 #define ARCH_HAS_READ_CURRENT_TIMER	1
 
 #endif
Index: linux-2.6.24-rc1-hack/init/calibrate.c
===================================================================
--- linux-2.6.24-rc1-hack.orig/init/calibrate.c
+++ linux-2.6.24-rc1-hack/init/calibrate.c
@@ -31,8 +31,8 @@ __setup("lpj=", lpj_setup);
 
 static unsigned long __devinit calibrate_delay_direct(void)
 {
-	unsigned long pre_start, start, post_start;
-	unsigned long pre_end, end, post_end;
+	unsigned long long pre_start, start, post_start;
+	unsigned long long pre_end, end, post_end;
 	unsigned long start_jiffies;
 	unsigned long tsc_rate_min, tsc_rate_max;
 	unsigned long good_tsc_sum = 0;

Re: [PATCH] Use 64bit timer values while calibrating BogoMips

[Arjan van de Ven]

On Wed, 31 Oct 2007 16:52:16 +0100
Andi Kleen <ak@suse.de> wrote:

> 
> When calibrating BogoMips 32bit architectures using 
> read_current_timer will truncate the timer value to 32bit.
> This is a problem on x86 when the TSC timer is already beyond
> 2^32-1 or wraps during the calibration. The current code
> has no wrap handling.  
> 
> This could potentially lead to wrong BogoMips, causing 
> incorrect delays e.g. in hardware drivers and then difficult
> to track down bugs.

bogomips tends to not be used for that nowadays
> 
> Change read_current_timer() to always return a 64bit
> value even on 32bit and do the computation in 64bit.
> 
> Only found by code inspection; i haven't seen a failure 
> in the wild.
> 
> avr32/sparc64 read_current_timer() updated, but I haven't 
> compiled/tested them.

one general comment, if you really want 64 bit... you're using
"unsigned long long" all over, just use u64 instead.


-- 
If you want to reach me at my work email, use arjan@linux.intel.com
For development, discussion and tips for power savings, 
visit http://www.lesswatts.org

Re: [PATCH] Use 64bit timer values while calibrating BogoMips

[Andi Kleen]

> bogomips tends to not be used for that nowadays

udelay certainly uses loops_per_jiffy (= BogoMips), even
though it can be scaled.

-Andi