powerpc/64s: idle POWER9 can execute stop in virtual mode
authorNicholas Piggin <npiggin@gmail.com>
Fri, 25 Aug 2017 04:30:35 +0000 (14:30 +1000)
committerMichael Ellerman <mpe@ellerman.id.au>
Tue, 29 Aug 2017 11:42:14 +0000 (21:42 +1000)
The hardware can execute stop in any context, and KVM does not
require real mode because siblings do not share MMU state. This
saves a switch to real-mode when going idle.

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

index 596b6ba8274ff86280d32a0ce3edccaea9d9450c..1125c9be9e067683c43c583d1672b1b6d3269840 100644 (file)
@@ -195,7 +195,16 @@ pnv_powersave_common:
        std     r5,_CCR(r1)
        std     r1,PACAR1(r13)
 
+BEGIN_FTR_SECTION
+       /*
+        * POWER9 does not require real mode to stop, and presently does not
+        * set hwthread_state for KVM (threads don't share MMU context), so
+        * we can remain in virtual mode for this.
+        */
+       bctr
+END_FTR_SECTION_IFSET(CPU_FTR_ARCH_300)
        /*
+        * POWER8
         * Go to real mode to do the nap, as required by the architecture.
         * Also, we need to be in real mode before setting hwthread_state,
         * because as soon as we do that, another thread can switch