Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86

* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/x86/linux-2.6-x86: (77 commits)
  x86: UV startup of slave cpus
  x86: integrate pci-dma.c
  x86: don't do dma if mask is NULL.
  x86: return conditional to mmu
  x86: remove kludge from x86_64
  x86: unify gfp masks
  x86: retry allocation if failed
  x86: don't try to allocate from DMA zone at first
  x86: use a fallback dev for i386
  x86: use numa allocation function in i386
  x86: remove virt_to_bus in pci-dma_64.c
  x86: adjust dma_free_coherent for i386
  x86: move bad_dma_address
  x86: isolate coherent mapping functions
  x86: move dma_coherent functions to pci-dma.c
  x86: merge iommu initialization parameters
  x86: merge dma_supported
  x86: move pci fixup to pci-dma.c
  x86: move x86_64-specific to common code.
  x86: move initialization functions to pci-dma.c
  ...

1 files changed, 18 insertions, 5 deletions

diff --git a/arch/x86/kernel/tsc_32.c b/arch/x86/kernel/tsc_32.c
index 3d7e6e9fa6c..e4790728b22 100644
--- a/arch/x86/kernel/tsc_32.c
+++ b/arch/x86/kernel/tsc_32.c
@@ -221,9 +221,9 @@ EXPORT_SYMBOL(recalibrate_cpu_khz);
  * if the CPU frequency is scaled, TSC-based delays will need a different
  * loops_per_jiffy value to function properly.
  */
-static unsigned int ref_freq = 0;
-static unsigned long loops_per_jiffy_ref = 0;
-static unsigned long cpu_khz_ref = 0;
+static unsigned int ref_freq;
+static unsigned long loops_per_jiffy_ref;
+static unsigned long cpu_khz_ref;
 
 static int
 time_cpufreq_notifier(struct notifier_block *nb, unsigned long val, void *data)
@@ -283,15 +283,28 @@ core_initcall(cpufreq_tsc);
 
 /* clock source code */
 
-static unsigned long current_tsc_khz = 0;
+static unsigned long current_tsc_khz;
+static struct clocksource clocksource_tsc;
 
+/*
+ * We compare the TSC to the cycle_last value in the clocksource
+ * structure to avoid a nasty time-warp issue. This can be observed in
+ * a very small window right after one CPU updated cycle_last under
+ * xtime 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;
 
 	rdtscll(ret);
 
-	return ret;
+	return ret >= clocksource_tsc.cycle_last ?
+		ret : clocksource_tsc.cycle_last;
 }
 
 static struct clocksource clocksource_tsc = {