[PATCH] Platform SMIs and their interferance with tsc based delay calibration

Issue:
Current tsc based delay_calibration can result in significant errors in
loops_per_jiffy count when the platform events like SMIs
(System Management Interrupts that are non-maskable) are present. This could
lead to potential kernel panic(). This issue is becoming more visible with 2.6
kernel (as default HZ is 1000) and on platforms with higher SMI handling
latencies. During the boot time, SMIs are mostly used by BIOS (for things
like legacy keyboard emulation).

Description:
The psuedocode for current delay calibration with tsc based delay looks like
(0) Estimate a value for loops_per_jiffy
(1) While (loops_per_jiffy estimate is accurate enough)
(2)   wait for jiffy transition (jiffy1)
(3)   Note down current tsc (tsc1)
(4)   loop until tsc becomes tsc1 + loops_per_jiffy
(5)   check whether jiffy changed since jiffy1 or not and refine
loops_per_jiffy estimate

Consider the following cases
Case 1:
If SMIs happen between (2) and (3) above, we can end up with a
loops_per_jiffy value that is too low. This results in shorted delays and
kernel can panic () during boot (Mostly at IOAPIC timer initialization
timer_irq_works() as we don't have enough timer interrupts in a specified
interval).

Case 2:
If SMIs happen between (3) and (4) above, then we can end up with a
loops_per_jiffy value that is too high. And with current i386 code, too
high lpj value (greater than 17M) can result in a overflow in
delay.c:__const_udelay() again resulting in shorter delay and panic().

Solution:
The patch below makes the calibration routine aware of asynchronous events
like SMIs. We increase the delay calibration time and also identify any
significant errors (greater than 12.5%) in the calibration and notify it to
user.

Patch below changes both i386 and x86-64 architectures to use this
new and improved calibrate_delay_direct() routine.

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

6 files changed, 28 insertions, 0 deletions

diff --git a/arch/i386/kernel/timers/common.c b/arch/i386/kernel/timers/common.c
index 8e201219f525..b38cc0d0c71a 100644
--- a/arch/i386/kernel/timers/common.c
+++ b/arch/i386/kernel/timers/common.c
@@ -139,6 +139,15 @@ bad_calibration:
 }
 #endif
 
+
+unsigned long read_timer_tsc(void)
+{
+        unsigned long retval;
+        rdtscl(retval);
+        return retval;
+}
+
+
 /* calculate cpu_khz */
 void init_cpu_khz(void)
 {
diff --git a/arch/i386/kernel/timers/timer.c b/arch/i386/kernel/timers/timer.c
index a3d6a288088b..7e39ed8e33f8 100644
--- a/arch/i386/kernel/timers/timer.c
+++ b/arch/i386/kernel/timers/timer.c
@@ -64,3 +64,12 @@ struct timer_opts* __init select_timer(void)
         panic("select_timer: Cannot find a suitable timer\n");
         return NULL;
 }
+
+int read_current_timer(unsigned long *timer_val)
+{
+        if (cur_timer->read_timer) {
+                *timer_val = cur_timer->read_timer();
+                return 0;
+        }
+        return -1;
+}
diff --git a/arch/i386/kernel/timers/timer_hpet.c b/arch/i386/kernel/timers/timer_hpet.c
index f778f471a09a..15a7d727bd6f 100644
--- a/arch/i386/kernel/timers/timer_hpet.c
+++ b/arch/i386/kernel/timers/timer_hpet.c
@@ -186,6 +186,7 @@ static struct timer_opts timer_hpet = {
         .get_offset =           get_offset_hpet,
         .monotonic_clock =      monotonic_clock_hpet,
         .delay =                delay_hpet,
+        .read_timer =           read_timer_tsc,
 };
 
 struct init_timer_opts __initdata timer_hpet_init = {
diff --git a/arch/i386/kernel/timers/timer_pm.c b/arch/i386/kernel/timers/timer_pm.c
index d77f22030fe6..4ef20e663498 100644
--- a/arch/i386/kernel/timers/timer_pm.c
+++ b/arch/i386/kernel/timers/timer_pm.c
@@ -246,6 +246,7 @@ static struct timer_opts timer_pmtmr = {
         .get_offset             = get_offset_pmtmr,
         .monotonic_clock        = monotonic_clock_pmtmr,
         .delay                  = delay_pmtmr,
+        .read_timer             = read_timer_tsc,
 };
 
 struct init_timer_opts __initdata timer_pmtmr_init = {
diff --git a/arch/i386/kernel/timers/timer_tsc.c b/arch/i386/kernel/timers/timer_tsc.c
index 180444d87824..27f08ae9120b 100644
--- a/arch/i386/kernel/timers/timer_tsc.c
+++ b/arch/i386/kernel/timers/timer_tsc.c
@@ -572,6 +572,7 @@ static struct timer_opts timer_tsc = {
         .get_offset = get_offset_tsc,
         .monotonic_clock = monotonic_clock_tsc,
         .delay = delay_tsc,
+        .read_timer = read_timer_tsc,
 };
 
 struct init_timer_opts __initdata timer_tsc_init = {
diff --git a/arch/x86_64/lib/delay.c b/arch/x86_64/lib/delay.c
index aed61a668a1b..33a873a3c223 100644
--- a/arch/x86_64/lib/delay.c
+++ b/arch/x86_64/lib/delay.c
@@ -12,6 +12,7 @@
 #include <linux/sched.h>
 #include <linux/delay.h>
 #include <asm/delay.h>
+#include <asm/msr.h>
 
 #ifdef CONFIG_SMP
 #include <asm/smp.h>
@@ -19,6 +20,12 @@
 
 int x86_udelay_tsc = 0;         /* Delay via TSC */
 
+int read_current_timer(unsigned long *timer_value)
+{
+        rdtscll(*timer_value);
+        return 0;
+}
+
 void __delay(unsigned long loops)
 {
         unsigned bclock, now;