[PATCH net-next v3 0/2] Add enhancements to RX path

This patch set fix gives some enhancements about RX path handling.
and thanks for Sergei Shtylyov helps reviewing during v2 to v3.

v1 -> v2: Fix message typos and add coverletter
  
v2 -> v3: 
Split from the previous series for submitting add enhancements 
as a series targeting 'net-next' and add indents before comments. 

Sean Wang (2):
  net: ethernet: mediatek: enhance RX path by reducing the frequency of
    the memory barrier used
  net: ethernet: mediatek: enhance RX path by aggregating more SKBs
    into NAPI

 drivers/net/ethernet/mediatek/mtk_eth_soc.c |   24 ++++++++++++------------
 1 file changed, 12 insertions(+), 12 deletions(-)

-- 
1.7.9.5

The patch adds support for aggregating more SKBs feed into NAPI in
order to get more benefits from generic receive offload (GRO) by
peeking at the RX ring status and moving more packets right before
returning from NAPI RX polling handler if NAPI budgets are still
available.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c |   15 +++++++--------
 1 file changed, 7 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 589ff50..d340673 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c

@@ -902,9 +902,6 @@ release_desc:
 	wmb();
 	mtk_w32(eth, ring->calc_idx, MTK_QRX_CRX_IDX0);
 
-	if (done < budget)
-		mtk_w32(eth, MTK_RX_DONE_INT, MTK_QMTK_INT_STATUS);
-
 	return done;
 }
 

@@ -1023,8 +1020,10 @@ static int mtk_napi_rx(struct napi_struct *napi, int budget)
 	struct mtk_eth *eth = container_of(napi, struct mtk_eth, rx_napi);
 	u32 status, mask;
 	int rx_done = 0;
+	int remain_budget = budget;
 
 	mtk_handle_status_irq(eth);
+poll_again:
 	mtk_w32(eth, MTK_RX_DONE_INT, MTK_QMTK_INT_STATUS);
 	rx_done = mtk_poll_rx(napi, budget, eth);
 

@@ -1035,14 +1034,14 @@ static int mtk_napi_rx(struct napi_struct *napi, int budget)
 			 "done rx %d, intr 0x%08x/0x%x\n",
 			 rx_done, status, mask);
 	}
-
-	if (rx_done == budget)
+	if (rx_done == remain_budget)
 		return budget;
 
 	status = mtk_r32(eth, MTK_QMTK_INT_STATUS);
-	if (status & MTK_RX_DONE_INT)
-		return budget;
-
+	if (status & MTK_RX_DONE_INT) {
+		remain_budget -= rx_done;
+		goto poll_again;
+	}
 	napi_complete(napi);
 	mtk_irq_enable(eth, MTK_RX_DONE_INT);
 

-- 
1.7.9.5

The patch makes moving wmb() to outside the loop that could help
RX path handling more faster although that RX descriptors aren't
freed for DMA to use as soon as possible, but based on my experiment
and the result shows it still can reach about 943Mbpis without
performance drop that is tested based on the setup with one port
using Giga PHY and 256 RX descriptors for DMA to move.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c |   11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index b782330..589ff50 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c

@@ -893,14 +893,15 @@ release_desc:
 		rxd->rxd2 = RX_DMA_PLEN0(ring->buf_size);
 
 		ring->calc_idx = idx;
-		/* make sure that all changes to the dma ring are flushed before
-		 * we continue
-		 */
-		wmb();
-		mtk_w32(eth, ring->calc_idx, MTK_QRX_CRX_IDX0);
 		done++;
 	}
 
+	/* make sure that all changes to the dma ring are flushed before
+	 * we continue
+	 */
+	wmb();
+	mtk_w32(eth, ring->calc_idx, MTK_QRX_CRX_IDX0);
+
 	if (done < budget)
 		mtk_w32(eth, MTK_RX_DONE_INT, MTK_QMTK_INT_STATUS);
 

-- 
1.7.9.5

You have to wait for me to merge the 'net' tree into 'net-next'
before you submit these changes.  These patches won't apply
cleanly otherwise.

That will happen the next time I merge my tree to Linus which
should be in the next day or two.