Increase default WORK_FIFO size to accommodate larger alltoall. (#1722)

src/init.cc

@@ -517,7 +517,7 @@ NCCL_PARAM(AggChannelSize, "AGG_CHANNEL_SIZE", -2);
 NCCL_PARAM(DisableGraphHelper, "GRAPH_HELPER_DISABLE", 0);
 // GDRCOPY support: FIFO_ENABLE when enabled locates a workFifo in CUDA memory
 NCCL_PARAM(GdrCopyFifoEnable, "GDRCOPY_FIFO_ENABLE", 1);
-#define NCCL_WORK_FIFO_BYTES_DEFAULT (1<<20)
+#define NCCL_WORK_FIFO_BYTES_DEFAULT (1<<22)
 NCCL_PARAM(WorkFifoBytes, "WORK_FIFO_BYTES", NCCL_WORK_FIFO_BYTES_DEFAULT);
 NCCL_PARAM(WorkArgsBytes, "WORK_ARGS_BYTES", INT64_MAX);
 enum ncclLaunchMode ncclParamLaunchMode;