x86-64: Don't generate cmov in vread_tsc

[pandora-kernel.git] / arch / x86 / kernel / tsc.c

diff --git a/arch/x86/kernel/tsc.c b/arch/x86/kernel/tsc.c
index 1e62442..24249a5 100644 (file)
--- a/arch/x86/kernel/tsc.c
+++ b/arch/x86/kernel/tsc.c
@@ -767,6 +767,7 @@ static cycle_t read_tsc(struct clocksource *cs)
 static cycle_t __vsyscall_fn vread_tsc(void)
 {
        cycle_t ret;
+       u64 last;
 
        /*
         * Empirically, a fence (of type that depends on the CPU)
@@ -778,8 +779,21 @@ static cycle_t __vsyscall_fn vread_tsc(void)
        rdtsc_barrier();
        ret = (cycle_t)vget_cycles();
 
-       return ret >= VVAR(vsyscall_gtod_data).clock.cycle_last ?
-               ret : VVAR(vsyscall_gtod_data).clock.cycle_last;
+       last = VVAR(vsyscall_gtod_data).clock.cycle_last;
+
+       if (likely(ret >= last))
+               return ret;
+
+       /*
+        * GCC likes to generate cmov here, but this branch is extremely
+        * predictable (it's just a funciton of time and the likely is
+        * very likely) and there's a data dependence, so force GCC
+        * to generate a branch instead.  I don't barrier() because
+        * we don't actually need a barrier, and if this function
+        * ever gets inlined it will generate worse code.
+        */
+       asm volatile ("");
+       return last;
 }
 #endif