powerpc/64s/idle: Predict HMI wakeup as unlikely
authorNicholas Piggin <npiggin@gmail.com>
Tue, 13 Jun 2017 13:05:52 +0000 (23:05 +1000)
committerMichael Ellerman <mpe@ellerman.id.au>
Mon, 19 Jun 2017 09:46:29 +0000 (19:46 +1000)
In a busy system, idle wakeups can be expected from IPIs and device
interrupts.

Reviewed-by: Gautham R. Shenoy <ego@linux.vnet.ibm.com>
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
arch/powerpc/kernel/idle_book3s.S

index ebe80b5..1ea14b9 100644 (file)
@@ -306,7 +306,7 @@ FTR_SECTION_ELSE_NESTED(66);                                                \
        rlwinm  r0,r12,45-31,0xe;  /* P7 wake reason field is 3 bits */ \
 ALT_FTR_SECTION_END_NESTED_IFSET(CPU_FTR_ARCH_207S, 66);               \
        cmpwi   r0,0xa;                 /* Hypervisor maintenance ? */  \
-       bne     20f;                                                    \
+       bne+    20f;                                                    \
        /* Invoke opal call to handle hmi */                            \
        ld      r2,PACATOC(r13);                                        \
        ld      r1,PACAR1(r13);                                         \