Merge branch 'master' of /pub/scm/linux/kernel/git/torvalds/linux-2.6

[linux-2.6-omap-h63xx.git] / arch / x86 / kernel / tsc_64.c

diff --git a/arch/x86/kernel/tsc_64.c b/arch/x86/kernel/tsc_64.c
index 947554ddabb6c7779f9e9adedfc07d7891dba283..fcc16e58609e17c42de7fae8a8309106df366423 100644 (file)
--- a/arch/x86/kernel/tsc_64.c
+++ b/arch/x86/kernel/tsc_64.c
@@ -11,6 +11,7 @@
 #include <asm/hpet.h>
 #include <asm/timex.h>
 #include <asm/timer.h>
+#include <asm/vgtod.h>
 
 static int notsc __initdata = 0;
 
@@ -44,8 +45,8 @@ DEFINE_PER_CPU(unsigned long, cyc2ns);
 
 static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu)
 {
-       unsigned long flags, prev_scale, *scale;
        unsigned long long tsc_now, ns_now;
+       unsigned long flags, *scale;
 
        local_irq_save(flags);
        sched_clock_idle_sleep_event();
@@ -55,7 +56,6 @@ static void set_cyc2ns_scale(unsigned long cpu_khz, int cpu)
        rdtscll(tsc_now);
        ns_now = __cycles_2_ns(tsc_now);
 
-       prev_scale = *scale;
        if (cpu_khz)
                *scale = (NSEC_PER_MSEC << CYC2NS_SCALE_FACTOR)/cpu_khz;
 
@@ -148,9 +148,7 @@ static int time_cpufreq_notifier(struct notifier_block *nb, unsigned long val,
                        mark_tsc_unstable("cpufreq changes");
        }
 
-       preempt_disable();
-       set_cyc2ns_scale(tsc_khz_ref, smp_processor_id());
-       preempt_enable();
+       set_cyc2ns_scale(tsc_khz_ref, freq->cpu);
 
        return 0;
 }
@@ -290,18 +288,34 @@ int __init notsc_setup(char *s)
 
 __setup("notsc", notsc_setup);
 
+static struct clocksource clocksource_tsc;
 
-/* clock source code: */
+/*
+ * We compare the TSC to the cycle_last value in the clocksource
+ * structure to avoid a nasty time-warp. This can be observed in a
+ * very small window right after one CPU updated cycle_last under
+ * xtime/vsyscall_gtod lock and the other CPU reads a TSC value which
+ * is smaller than the cycle_last reference value due to a TSC which
+ * is slighty behind. This delta is nowhere else observable, but in
+ * that case it results in a forward time jump in the range of hours
+ * due to the unsigned delta calculation of the time keeping core
+ * code, which is necessary to support wrapping clocksources like pm
+ * timer.
+ */
 static cycle_t read_tsc(void)
 {
        cycle_t ret = (cycle_t)get_cycles();
-       return ret;
+
+       return ret >= clocksource_tsc.cycle_last ?
+               ret : clocksource_tsc.cycle_last;
 }
 
 static cycle_t __vsyscall_fn vread_tsc(void)
 {
        cycle_t ret = (cycle_t)vget_cycles();
-       return ret;
+
+       return ret >= __vsyscall_gtod_data.clock.cycle_last ?
+               ret : __vsyscall_gtod_data.clock.cycle_last;
 }
 
 static struct clocksource clocksource_tsc = {