From: Greg KH on
2.6.32-stable review patch. If anyone has any objections, please let us know.


From: Andi Kleen <andi(a)>

commit 124482935fb7fb9303c8a8ab930149c6a93d9910 upstream.

This fixes the -Os breaks with gcc 4.5 bug. rdtsc_barrier needs to be
force inlined, otherwise user space will jump into kernel space and
kill init.

This also addresses
I believe.

Signed-off-by: Andi Kleen <ak(a)>
LKML-Reference: <20100618210859.GA10913(a)>
Signed-off-by: H. Peter Anvin <hpa(a)>
Signed-off-by: Greg Kroah-Hartman <gregkh(a)>

arch/x86/include/asm/system.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)

--- a/arch/x86/include/asm/system.h
+++ b/arch/x86/include/asm/system.h
@@ -449,7 +449,7 @@ void stop_this_cpu(void *dummy);
* (Could use an alternative three way for this if there was one.)
-static inline void rdtsc_barrier(void)
+static __always_inline void rdtsc_barrier(void)
alternative(ASM_NOP3, "mfence", X86_FEATURE_MFENCE_RDTSC);
alternative(ASM_NOP3, "lfence", X86_FEATURE_LFENCE_RDTSC);

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo(a)
More majordomo info at
Please read the FAQ at