runtime: always give concurrent sweep some heap distance

Currently it's possible for the next_gc heap size trigger computed for
the next GC cycle to be less than the current allocated heap size.
This means the next cycle will start immediately, which means there's
no time to perform the concurrent sweep between GC cycles. This places
responsibility for finishing the sweep on GC itself, which delays GC
start-up and hence delays mutator assist.

Fix this by ensuring that next_gc is always at least a little higher
than the allocated heap size, so we won't trigger the next cycle
instantly.

Updates #11911.

Change-Id: I74f0b887bf187518d5fedffc7989817cbcf30592
Reviewed-on: https://go-review.googlesource.com/13043
Reviewed-by: Rick Hudson <rlh@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>

src/runtime/mgc.go

@@ -145,6 +145,11 @@ const (
 	// bookkeeping will use a large amount of each stack.
 	firstStackBarrierOffset = 1024
 	debugStackBarrier       = false
+
+	// sweepMinHeapDistance is a lower bound on the heap distance
+	// (in bytes) reserved for concurrent sweeping between GC
+	// cycles. This will be scaled by gcpercent/100.
+	sweepMinHeapDistance = 1024 * 1024
 )
 
 // heapminimum is the minimum heap size at which to trigger GC.
@@ -1489,6 +1494,21 @@ func gcMark(start_time int64) {
 	memstats.heap_marked = work.bytesMarked
 	memstats.heap_scan = uint64(gcController.scanWork)
 
+	minNextGC := memstats.heap_live + sweepMinHeapDistance*uint64(gcpercent)/100
+	if memstats.next_gc < minNextGC {
+		// The allocated heap is already past the trigger.
+		// This can happen if the triggerRatio is very low and
+		// the reachable heap estimate is less than the live
+		// heap size.
+		//
+		// Concurrent sweep happens in the heap growth from
+		// heap_live to next_gc, so bump next_gc up to ensure
+		// that concurrent sweep has some heap growth in which
+		// to perform sweeping before we start the next GC
+		// cycle.
+		memstats.next_gc = minNextGC
+	}
+
 	if trace.enabled {
 		traceHeapAlloc()
 		traceNextGC()