x86-64: remove locked instruction from switch_to()

Impact: micro-optimization

The patch below removes an unnecessary locked instruction from
switch_to().  TIF_FORK is only ever set in copy_thread() on initial
process creation, and gets cleared during the first scheduling of the
process.  As such, it is safe to use an unlocked test for the flag
within switch_to().
Signed-off-by: Benjamin LaHaise <bcrl@kvack.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>

arch/x86/include/asm/system.h

@@ -96,15 +96,15 @@ do {									\
 	     "thread_return:\n\t"					  \
 	     "movq %%gs:%P[pda_pcurrent],%%rsi\n\t"			  \
 	     "movq %P[thread_info](%%rsi),%%r8\n\t"			  \
-	     LOCK_PREFIX "btr  %[tif_fork],%P[ti_flags](%%r8)\n\t"	  \
 	     "movq %%rax,%%rdi\n\t" 					  \
-	     "jc   ret_from_fork\n\t"					  \
+	     "testl  %[_tif_fork],%P[ti_flags](%%r8)\n\t"	  \
+	     "jnz   ret_from_fork\n\t"					  \
 	     RESTORE_CONTEXT						  \
 	     : "=a" (last)					  	  \
 	     : [next] "S" (next), [prev] "D" (prev),			  \
 	       [threadrsp] "i" (offsetof(struct task_struct, thread.sp)), \
 	       [ti_flags] "i" (offsetof(struct thread_info, flags)),	  \
-	       [tif_fork] "i" (TIF_FORK),			  	  \
+	       [_tif_fork] "i" (_TIF_FORK),			  	  \
 	       [thread_info] "i" (offsetof(struct task_struct, stack)),   \
 	       [pda_pcurrent] "i" (offsetof(struct x8664_pda, pcurrent))  \
 	     : "memory", "cc" __EXTRA_CLOBBER)

arch/x86/kernel/entry_64.S

@@ -408,6 +408,8 @@ END(save_paranoid)
 ENTRY(ret_from_fork)
 	DEFAULT_FRAME
 
+	LOCK ; btr $TIF_FORK,TI_flags(%r8)
+
 	push kernel_eflags(%rip)
 	CFI_ADJUST_CFA_OFFSET 8
 	popf					# reset kernel eflags