From c6fe53a230e70bbb7a5582afe96cacf174eb0335 Mon Sep 17 00:00:00 2001
From: Dmitriy Vyukov <dvyukov@google.com>
Date: Sat, 16 Aug 2014 09:07:55 +0400
Subject: [PATCH] runtime: mark with non-atomic operations when GOMAXPROCS=1
 Perf builders show 3-5% GC pause increase with GOMAXPROCS=1 when marking with
 atomic ops:
 http://goperfd.appspot.com/perfdetail?commit=a8a6e765d6a87f7ccb71fd85a60eb5a821151f85&commit0=3b864e02b987171e05e2e9d0840b85b5b6476386&kind=builder&builder=linux-amd64&benchmark=http

LGTM=rlh
R=golang-codereviews, rlh
CC=dave, golang-codereviews, khr, rsc
https://golang.org/cl/128340043
---
 src/pkg/runtime/mgc0.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/pkg/runtime/mgc0.c b/src/pkg/runtime/mgc0.c
index 59dfd51e65c..f46d329017f 100644
--- a/src/pkg/runtime/mgc0.c
+++ b/src/pkg/runtime/mgc0.c
@@ -451,7 +451,8 @@ scanblock(byte *b, uintptr n, byte *ptrmask)
 			// quadruple is already marked. Otherwise we resort to CAS
 			// loop for marking.
 			bits8 = xbits>>(shift&~7);
-			if((bits8&(bitMask|(bitMask<<gcBits))) != (bitBoundary|(bitBoundary<<gcBits)))
+			if((bits8&(bitMask|(bitMask<<gcBits))) != (bitBoundary|(bitBoundary<<gcBits)) ||
+				work.nproc == 1)
 				((uint8*)bitp)[shift/8] = bits8 | (bitMarked<<(shift&7));
 			else {
 				for(;;) {