mips/atomic: Fix loongson_llsc_mb() wreckage

The comment describing the loongson_llsc_mb() reorder case doesn't
make any sense what so ever. Instruction re-ordering is not an SMP
artifact, but rather a CPU local phenomenon. Clarify the comment by
explaining that these issue cause a coherence fail.

For the branch speculation case; if futex_atomic_cmpxchg_inatomic()
needs one at the bne branch target, then surely the normal
__cmpxch_asm() implementation does too. We cannot rely on the
barriers from cmpxchg() because cmpxchg_local() is implemented with
the same macro, and branch prediction and speculation are, too, CPU
local.

Fixes: e02e07e312 ("MIPS: Loongson: Introduce and use loongson_llsc_mb()")
Cc: Huacai Chen <chenhc@lemote.com>
Cc: Huang Pei <huangpei@loongson.cn>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Signed-off-by: Paul Burton <paul.burton@mips.com>

arch/mips/include/asm/atomic.h

@@ -193,6 +193,7 @@ static __inline__ int atomic_sub_if_positive(int i, atomic_t * v)
 	if (kernel_uses_llsc) {
 		int temp;
 
+		loongson_llsc_mb();
 		__asm__ __volatile__(
 		"	.set	push					\n"
 		"	.set	"MIPS_ISA_LEVEL"			\n"
@@ -200,12 +201,12 @@ static __inline__ int atomic_sub_if_positive(int i, atomic_t * v)
 		"	.set	pop					\n"
 		"	subu	%0, %1, %3				\n"
 		"	move	%1, %0					\n"
-		"	bltz	%0, 1f					\n"
+		"	bltz	%0, 2f					\n"
 		"	.set	push					\n"
 		"	.set	"MIPS_ISA_LEVEL"			\n"
 		"	sc	%1, %2					\n"
 		"\t" __scbeqz "	%1, 1b					\n"
-		"1:							\n"
+		"2:							\n"
 		"	.set	pop					\n"
 		: "=&r" (result), "=&r" (temp),
 		  "+" GCC_OFF_SMALL_ASM() (v->counter)