x86: __force_order doesn't need to be an actual variable

It being static causes over a dozen instances to be scattered
across the kernel image, with non of them ever being referenced
in any way. Making the variable extern without ever defining it
works as well - all we need is to have the compiler think the
variable is being accessed.

Signed-off-by: Jan Beulich <jbeulich@suse.com>
Link: http://lkml.kernel.org/r/51A610B802000078000D99A0@nat28.tlf.novell.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>

diff --git a/arch/x86/include/asm/special_insns.h b/arch/x86/include/asm/special_insns.h
index 41fc93a2e225d1793465e1ccf5e5ebc0e354d8a5..2f4d924fe6c9fdf0181b309a160b2dae42b7c384 100644 (file)
--- a/arch/x86/include/asm/special_insns.h
+++ b/arch/x86/include/asm/special_insns.h
@@ -16,7 +16,7 @@ static inline void native_clts(void)
  * all loads stores around it, which can hurt performance. Solution is to
  * use a variable and mimic reads and writes to it to enforce serialization
  */
-static unsigned long __force_order;
+extern unsigned long __force_order;
 
 static inline unsigned long native_read_cr0(void)
 {