[PATCH 00/11] MediaTek SoC flow offload improvements + wireless support

[PATCH 00/11] MediaTek SoC flow offload improvements + wireless support

[Felix Fietkau]

This series contains the following improvements to mediatek ethernet flow
offload support:

- support dma-coherent on ethernet to improve performance
- add ipv6 offload support
- rework hardware flow table entry handling to improve dealing with hash
  collisions and competing flows
- support creating offload entries from user space
- support creating offload entries with just source/destination mac address,
  vlan and output device information
- add driver changes for supporting the Wireless Ethernet Dispatch core,
  which can be used to offload flows from ethernet to MT7915 PCIe WLAN
  devices

David Bentham (1):
  net: ethernet: mtk_eth_soc: add ipv6 flow offload support

Felix Fietkau (10):
  net: ethernet: mtk_eth_soc: add support for coherent DMA
  arm64: dts: mediatek: mt7622: add support for coherent DMA
  net: ethernet: mtk_eth_soc: add support for Wireless Ethernet Dispatch
    (WED)
  net: ethernet: mtk_eth_soc: implement flow offloading to WED devices
  arm64: dts: mediatek: mt7622: introduce nodes for Wireless Ethernet
    Dispatch
  net: ethernet: mtk_eth_soc: support TC_SETUP_BLOCK for PPE offload
  net: ethernet: mtk_eth_soc: allocate struct mtk_ppe separately
  net: ethernet: mtk_eth_soc: rework hardware flow table management
  net: ethernet: mtk_eth_soc: remove bridge flow offload type entry
    support
  net: ethernet: mtk_eth_soc: support creating mac address based offload
    entries

 arch/arm64/boot/dts/mediatek/mt7622.dtsi      |  32 +-
 drivers/net/ethernet/mediatek/Kconfig         |   4 +
 drivers/net/ethernet/mediatek/Makefile        |   5 +
 drivers/net/ethernet/mediatek/mtk_eth_soc.c   | 131 ++-
 drivers/net/ethernet/mediatek/mtk_eth_soc.h   |  14 +-
 drivers/net/ethernet/mediatek/mtk_ppe.c       | 368 +++++++-
 drivers/net/ethernet/mediatek/mtk_ppe.h       |  89 +-
 .../net/ethernet/mediatek/mtk_ppe_debugfs.c   |   1 -
 .../net/ethernet/mediatek/mtk_ppe_offload.c   | 189 +++-
 drivers/net/ethernet/mediatek/mtk_wed.c       | 875 ++++++++++++++++++
 drivers/net/ethernet/mediatek/mtk_wed.h       | 135 +++
 .../net/ethernet/mediatek/mtk_wed_debugfs.c   | 175 ++++
 drivers/net/ethernet/mediatek/mtk_wed_ops.c   |   8 +
 drivers/net/ethernet/mediatek/mtk_wed_regs.h  | 251 +++++
 include/linux/netdevice.h                     |   7 +
 include/linux/soc/mediatek/mtk_wed.h          | 131 +++
 net/core/dev.c                                |   4 +
 17 files changed, 2294 insertions(+), 125 deletions(-)
 create mode 100644 drivers/net/ethernet/mediatek/mtk_wed.c
 create mode 100644 drivers/net/ethernet/mediatek/mtk_wed.h
 create mode 100644 drivers/net/ethernet/mediatek/mtk_wed_debugfs.c
 create mode 100644 drivers/net/ethernet/mediatek/mtk_wed_ops.c
 create mode 100644 drivers/net/ethernet/mediatek/mtk_wed_regs.h
 create mode 100644 include/linux/soc/mediatek/mtk_wed.h

-- 
2.32.0 (Apple Git-132)

[PATCH 01/11] net: ethernet: mtk_eth_soc: add support for coherent DMA

[Felix Fietkau]

It improves performance by eliminating the need for a cache flush on rx and tx
In preparation for supporting WED (Wireless Ethernet Dispatch), also add a
function for disabling coherent DMA at runtime.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c | 95 +++++++++++++++------
 drivers/net/ethernet/mediatek/mtk_eth_soc.h |  9 ++
 2 files changed, 80 insertions(+), 24 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index f02d07ec5ccb..70db217ed831 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -9,6 +9,7 @@
 #include <linux/of_device.h>
 #include <linux/of_mdio.h>
 #include <linux/of_net.h>
+#include <linux/of_address.h>
 #include <linux/mfd/syscon.h>
 #include <linux/regmap.h>
 #include <linux/clk.h>
@@ -786,7 +787,7 @@ static int mtk_init_fq_dma(struct mtk_eth *eth)
 	dma_addr_t dma_addr;
 	int i;
 
-	eth->scratch_ring = dma_alloc_coherent(eth->dev,
+	eth->scratch_ring = dma_alloc_coherent(eth->dma_dev,
 					       cnt * sizeof(struct mtk_tx_dma),
 					       &eth->phy_scratch_ring,
 					       GFP_ATOMIC);
@@ -798,10 +799,10 @@ static int mtk_init_fq_dma(struct mtk_eth *eth)
 	if (unlikely(!eth->scratch_head))
 		return -ENOMEM;
 
-	dma_addr = dma_map_single(eth->dev,
+	dma_addr = dma_map_single(eth->dma_dev,
 				  eth->scratch_head, cnt * MTK_QDMA_PAGE_SIZE,
 				  DMA_FROM_DEVICE);
-	if (unlikely(dma_mapping_error(eth->dev, dma_addr)))
+	if (unlikely(dma_mapping_error(eth->dma_dev, dma_addr)))
 		return -ENOMEM;
 
 	phy_ring_tail = eth->phy_scratch_ring +
@@ -855,26 +856,26 @@ static void mtk_tx_unmap(struct mtk_eth *eth, struct mtk_tx_buf *tx_buf,
 {
 	if (MTK_HAS_CAPS(eth->soc->caps, MTK_QDMA)) {
 		if (tx_buf->flags & MTK_TX_FLAGS_SINGLE0) {
-			dma_unmap_single(eth->dev,
+			dma_unmap_single(eth->dma_dev,
 					 dma_unmap_addr(tx_buf, dma_addr0),
 					 dma_unmap_len(tx_buf, dma_len0),
 					 DMA_TO_DEVICE);
 		} else if (tx_buf->flags & MTK_TX_FLAGS_PAGE0) {
-			dma_unmap_page(eth->dev,
+			dma_unmap_page(eth->dma_dev,
 				       dma_unmap_addr(tx_buf, dma_addr0),
 				       dma_unmap_len(tx_buf, dma_len0),
 				       DMA_TO_DEVICE);
 		}
 	} else {
 		if (dma_unmap_len(tx_buf, dma_len0)) {
-			dma_unmap_page(eth->dev,
+			dma_unmap_page(eth->dma_dev,
 				       dma_unmap_addr(tx_buf, dma_addr0),
 				       dma_unmap_len(tx_buf, dma_len0),
 				       DMA_TO_DEVICE);
 		}
 
 		if (dma_unmap_len(tx_buf, dma_len1)) {
-			dma_unmap_page(eth->dev,
+			dma_unmap_page(eth->dma_dev,
 				       dma_unmap_addr(tx_buf, dma_addr1),
 				       dma_unmap_len(tx_buf, dma_len1),
 				       DMA_TO_DEVICE);
@@ -952,9 +953,9 @@ static int mtk_tx_map(struct sk_buff *skb, struct net_device *dev,
 	if (skb_vlan_tag_present(skb))
 		txd4 |= TX_DMA_INS_VLAN | skb_vlan_tag_get(skb);
 
-	mapped_addr = dma_map_single(eth->dev, skb->data,
+	mapped_addr = dma_map_single(eth->dma_dev, skb->data,
 				     skb_headlen(skb), DMA_TO_DEVICE);
-	if (unlikely(dma_mapping_error(eth->dev, mapped_addr)))
+	if (unlikely(dma_mapping_error(eth->dma_dev, mapped_addr)))
 		return -ENOMEM;
 
 	WRITE_ONCE(itxd->txd1, mapped_addr);
@@ -993,10 +994,10 @@ static int mtk_tx_map(struct sk_buff *skb, struct net_device *dev,
 
 
 			frag_map_size = min(frag_size, MTK_TX_DMA_BUF_LEN);
-			mapped_addr = skb_frag_dma_map(eth->dev, frag, offset,
+			mapped_addr = skb_frag_dma_map(eth->dma_dev, frag, offset,
 						       frag_map_size,
 						       DMA_TO_DEVICE);
-			if (unlikely(dma_mapping_error(eth->dev, mapped_addr)))
+			if (unlikely(dma_mapping_error(eth->dma_dev, mapped_addr)))
 				goto err_dma;
 
 			if (i == nr_frags - 1 &&
@@ -1274,18 +1275,18 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget,
 			netdev->stats.rx_dropped++;
 			goto release_desc;
 		}
-		dma_addr = dma_map_single(eth->dev,
+		dma_addr = dma_map_single(eth->dma_dev,
 					  new_data + NET_SKB_PAD +
 					  eth->ip_align,
 					  ring->buf_size,
 					  DMA_FROM_DEVICE);
-		if (unlikely(dma_mapping_error(eth->dev, dma_addr))) {
+		if (unlikely(dma_mapping_error(eth->dma_dev, dma_addr))) {
 			skb_free_frag(new_data);
 			netdev->stats.rx_dropped++;
 			goto release_desc;
 		}
 
-		dma_unmap_single(eth->dev, trxd.rxd1,
+		dma_unmap_single(eth->dma_dev, trxd.rxd1,
 				 ring->buf_size, DMA_FROM_DEVICE);
 
 		/* receive data */
@@ -1558,7 +1559,7 @@ static int mtk_tx_alloc(struct mtk_eth *eth)
 	if (!ring->buf)
 		goto no_tx_mem;
 
-	ring->dma = dma_alloc_coherent(eth->dev, MTK_DMA_SIZE * sz,
+	ring->dma = dma_alloc_coherent(eth->dma_dev, MTK_DMA_SIZE * sz,
 				       &ring->phys, GFP_ATOMIC);
 	if (!ring->dma)
 		goto no_tx_mem;
@@ -1576,7 +1577,7 @@ static int mtk_tx_alloc(struct mtk_eth *eth)
 	 * descriptors in ring->dma_pdma.
 	 */
 	if (!MTK_HAS_CAPS(eth->soc->caps, MTK_QDMA)) {
-		ring->dma_pdma = dma_alloc_coherent(eth->dev, MTK_DMA_SIZE * sz,
+		ring->dma_pdma = dma_alloc_coherent(eth->dma_dev, MTK_DMA_SIZE * sz,
 						    &ring->phys_pdma,
 						    GFP_ATOMIC);
 		if (!ring->dma_pdma)
@@ -1635,7 +1636,7 @@ static void mtk_tx_clean(struct mtk_eth *eth)
 	}
 
 	if (ring->dma) {
-		dma_free_coherent(eth->dev,
+		dma_free_coherent(eth->dma_dev,
 				  MTK_DMA_SIZE * sizeof(*ring->dma),
 				  ring->dma,
 				  ring->phys);
@@ -1643,7 +1644,7 @@ static void mtk_tx_clean(struct mtk_eth *eth)
 	}
 
 	if (ring->dma_pdma) {
-		dma_free_coherent(eth->dev,
+		dma_free_coherent(eth->dma_dev,
 				  MTK_DMA_SIZE * sizeof(*ring->dma_pdma),
 				  ring->dma_pdma,
 				  ring->phys_pdma);
@@ -1688,18 +1689,18 @@ static int mtk_rx_alloc(struct mtk_eth *eth, int ring_no, int rx_flag)
 			return -ENOMEM;
 	}
 
-	ring->dma = dma_alloc_coherent(eth->dev,
+	ring->dma = dma_alloc_coherent(eth->dma_dev,
 				       rx_dma_size * sizeof(*ring->dma),
 				       &ring->phys, GFP_ATOMIC);
 	if (!ring->dma)
 		return -ENOMEM;
 
 	for (i = 0; i < rx_dma_size; i++) {
-		dma_addr_t dma_addr = dma_map_single(eth->dev,
+		dma_addr_t dma_addr = dma_map_single(eth->dma_dev,
 				ring->data[i] + NET_SKB_PAD + eth->ip_align,
 				ring->buf_size,
 				DMA_FROM_DEVICE);
-		if (unlikely(dma_mapping_error(eth->dev, dma_addr)))
+		if (unlikely(dma_mapping_error(eth->dma_dev, dma_addr)))
 			return -ENOMEM;
 		ring->dma[i].rxd1 = (unsigned int)dma_addr;
 
@@ -1735,7 +1736,7 @@ static void mtk_rx_clean(struct mtk_eth *eth, struct mtk_rx_ring *ring)
 				continue;
 			if (!ring->dma[i].rxd1)
 				continue;
-			dma_unmap_single(eth->dev,
+			dma_unmap_single(eth->dma_dev,
 					 ring->dma[i].rxd1,
 					 ring->buf_size,
 					 DMA_FROM_DEVICE);
@@ -1746,7 +1747,7 @@ static void mtk_rx_clean(struct mtk_eth *eth, struct mtk_rx_ring *ring)
 	}
 
 	if (ring->dma) {
-		dma_free_coherent(eth->dev,
+		dma_free_coherent(eth->dma_dev,
 				  ring->dma_size * sizeof(*ring->dma),
 				  ring->dma,
 				  ring->phys);
@@ -2099,7 +2100,7 @@ static void mtk_dma_free(struct mtk_eth *eth)
 		if (eth->netdev[i])
 			netdev_reset_queue(eth->netdev[i]);
 	if (eth->scratch_ring) {
-		dma_free_coherent(eth->dev,
+		dma_free_coherent(eth->dma_dev,
 				  MTK_DMA_SIZE * sizeof(struct mtk_tx_dma),
 				  eth->scratch_ring,
 				  eth->phy_scratch_ring);
@@ -2448,6 +2449,8 @@ static void mtk_dim_tx(struct work_struct *work)
 
 static int mtk_hw_init(struct mtk_eth *eth)
 {
+	u32 dma_mask = ETHSYS_DMA_AG_MAP_PDMA | ETHSYS_DMA_AG_MAP_QDMA |
+		       ETHSYS_DMA_AG_MAP_PPE;
 	int i, val, ret;
 
 	if (test_and_set_bit(MTK_HW_INIT, &eth->state))
@@ -2460,6 +2463,10 @@ static int mtk_hw_init(struct mtk_eth *eth)
 	if (ret)
 		goto err_disable_pm;
 
+	if (eth->ethsys)
+		regmap_update_bits(eth->ethsys, ETHSYS_DMA_AG_MAP, dma_mask,
+				   of_dma_is_coherent(eth->dma_dev->of_node) * dma_mask);
+
 	if (MTK_HAS_CAPS(eth->soc->caps, MTK_SOC_MT7628)) {
 		ret = device_reset(eth->dev);
 		if (ret) {
@@ -3040,6 +3047,35 @@ static int mtk_add_mac(struct mtk_eth *eth, struct device_node *np)
 	return err;
 }
 
+void mtk_eth_set_dma_device(struct mtk_eth *eth, struct device *dma_dev)
+{
+	struct net_device *dev, *tmp;
+	LIST_HEAD(dev_list);
+	int i;
+
+	rtnl_lock();
+
+	for (i = 0; i < MTK_MAC_COUNT; i++) {
+		dev = eth->netdev[i];
+
+		if (!dev || !(dev->flags & IFF_UP))
+			continue;
+
+		list_add_tail(&dev->close_list, &dev_list);
+	}
+
+	dev_close_many(&dev_list, false);
+
+	eth->dma_dev = dma_dev;
+
+	list_for_each_entry_safe(dev, tmp, &dev_list, close_list) {
+		list_del_init(&dev->close_list);
+		dev_open(dev, NULL);
+	}
+
+	rtnl_unlock();
+}
+
 static int mtk_probe(struct platform_device *pdev)
 {
 	struct device_node *mac_np;
@@ -3053,6 +3089,7 @@ static int mtk_probe(struct platform_device *pdev)
 	eth->soc = of_device_get_match_data(&pdev->dev);
 
 	eth->dev = &pdev->dev;
+	eth->dma_dev = &pdev->dev;
 	eth->base = devm_platform_ioremap_resource(pdev, 0);
 	if (IS_ERR(eth->base))
 		return PTR_ERR(eth->base);
@@ -3101,6 +3138,16 @@ static int mtk_probe(struct platform_device *pdev)
 		}
 	}
 
+	if (of_dma_is_coherent(pdev->dev.of_node)) {
+		struct regmap *cci;
+
+		cci = syscon_regmap_lookup_by_phandle(pdev->dev.of_node,
+						      "mediatek,cci-control");
+		/* enable CPU/bus coherency */
+		if (!IS_ERR(cci))
+			regmap_write(cci, 0, 3);
+	}
+
 	if (MTK_HAS_CAPS(eth->soc->caps, MTK_SGMII)) {
 		eth->sgmii = devm_kzalloc(eth->dev, sizeof(*eth->sgmii),
 					  GFP_KERNEL);
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
index c9d42be314b5..e701544c4287 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -465,6 +465,12 @@
 #define RSTCTRL_FE		BIT(6)
 #define RSTCTRL_PPE		BIT(31)
 
+/* ethernet dma channel agent map */
+#define ETHSYS_DMA_AG_MAP	0x408
+#define ETHSYS_DMA_AG_MAP_PDMA	BIT(0)
+#define ETHSYS_DMA_AG_MAP_QDMA	BIT(1)
+#define ETHSYS_DMA_AG_MAP_PPE	BIT(2)
+
 /* SGMII subsystem config registers */
 /* Register to auto-negotiation restart */
 #define SGMSYS_PCS_CONTROL_1	0x0
@@ -882,6 +888,7 @@ struct mtk_sgmii {
 /* struct mtk_eth -	This is the main datasructure for holding the state
  *			of the driver
  * @dev:		The device pointer
+ * @dev:		The device pointer used for dma mapping/alloc
  * @base:		The mapped register i/o base
  * @page_lock:		Make sure that register operations are atomic
  * @tx_irq__lock:	Make sure that IRQ register operations are atomic
@@ -925,6 +932,7 @@ struct mtk_sgmii {
 
 struct mtk_eth {
 	struct device			*dev;
+	struct device			*dma_dev;
 	void __iomem			*base;
 	spinlock_t			page_lock;
 	spinlock_t			tx_irq_lock;
@@ -1023,6 +1031,7 @@ int mtk_gmac_rgmii_path_setup(struct mtk_eth *eth, int mac_id);
 int mtk_eth_offload_init(struct mtk_eth *eth);
 int mtk_eth_setup_tc(struct net_device *dev, enum tc_setup_type type,
 		     void *type_data);
+void mtk_eth_set_dma_device(struct mtk_eth *eth, struct device *dma_dev);
 
 
 #endif /* MTK_ETH_H */
-- 
2.32.0 (Apple Git-132)

[PATCH 02/11] arm64: dts: mediatek: mt7622: add support for coherent DMA

[Felix Fietkau]

It improves performance by eliminating the need for a cache flush on rx and tx

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 arch/arm64/boot/dts/mediatek/mt7622.dtsi | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/arch/arm64/boot/dts/mediatek/mt7622.dtsi b/arch/arm64/boot/dts/mediatek/mt7622.dtsi
index 6f8cb3ad1e84..a2257ec6d256 100644
--- a/arch/arm64/boot/dts/mediatek/mt7622.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt7622.dtsi
@@ -357,7 +357,7 @@ cci_control1: slave-if@4000 {
 		};
 
 		cci_control2: slave-if@5000 {
-			compatible = "arm,cci-400-ctrl-if";
+			compatible = "arm,cci-400-ctrl-if", "syscon";
 			interface-type = "ace";
 			reg = <0x5000 0x1000>;
 		};
@@ -945,6 +945,8 @@ eth: ethernet@1b100000 {
 		power-domains = <&scpsys MT7622_POWER_DOMAIN_ETHSYS>;
 		mediatek,ethsys = <&ethsys>;
 		mediatek,sgmiisys = <&sgmiisys>;
+		mediatek,cci-control = <&cci_control2>;
+		dma-coherent;
 		#address-cells = <1>;
 		#size-cells = <0>;
 		status = "disabled";
-- 
2.32.0 (Apple Git-132)

[PATCH 03/11] net: ethernet: mtk_eth_soc: add support for Wireless Ethernet Dispatch (WED)

[Felix Fietkau]

The Wireless Ethernet Dispatch subsystem on the MT7622 SoC can be
configured to intercept and handle access to the DMA queues and
PCIe interrupts for a MT7615/MT7915 wireless card.
It can manage the internal WDMA (Wireless DMA) controller, which allows
ethernet packets to be passed from the packet switch engine (PSE) to the
wireless card, bypassing the CPU entirely.
This can be used to implement hardware flow offloading from ethernet to
WLAN.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/ethernet/mediatek/Kconfig         |   4 +
 drivers/net/ethernet/mediatek/Makefile        |   5 +
 drivers/net/ethernet/mediatek/mtk_eth_soc.c   |  17 +
 drivers/net/ethernet/mediatek/mtk_eth_soc.h   |   3 +
 drivers/net/ethernet/mediatek/mtk_wed.c       | 875 ++++++++++++++++++
 drivers/net/ethernet/mediatek/mtk_wed.h       | 128 +++
 .../net/ethernet/mediatek/mtk_wed_debugfs.c   | 175 ++++
 drivers/net/ethernet/mediatek/mtk_wed_ops.c   |   8 +
 drivers/net/ethernet/mediatek/mtk_wed_regs.h  | 251 +++++
 include/linux/soc/mediatek/mtk_wed.h          | 131 +++
 10 files changed, 1597 insertions(+)
 create mode 100644 drivers/net/ethernet/mediatek/mtk_wed.c
 create mode 100644 drivers/net/ethernet/mediatek/mtk_wed.h
 create mode 100644 drivers/net/ethernet/mediatek/mtk_wed_debugfs.c
 create mode 100644 drivers/net/ethernet/mediatek/mtk_wed_ops.c
 create mode 100644 drivers/net/ethernet/mediatek/mtk_wed_regs.h
 create mode 100644 include/linux/soc/mediatek/mtk_wed.h

diff --git a/drivers/net/ethernet/mediatek/Kconfig b/drivers/net/ethernet/mediatek/Kconfig
index 86d356b4388d..da4ec235d146 100644
--- a/drivers/net/ethernet/mediatek/Kconfig
+++ b/drivers/net/ethernet/mediatek/Kconfig
@@ -7,6 +7,10 @@ config NET_VENDOR_MEDIATEK
 
 if NET_VENDOR_MEDIATEK
 
+config NET_MEDIATEK_SOC_WED
+	depends on ARCH_MEDIATEK || COMPILE_TEST
+	def_bool NET_MEDIATEK_SOC != n
+
 config NET_MEDIATEK_SOC
 	tristate "MediaTek SoC Gigabit Ethernet support"
 	depends on NET_DSA || !NET_DSA
diff --git a/drivers/net/ethernet/mediatek/Makefile b/drivers/net/ethernet/mediatek/Makefile
index 79d4cdbbcbf5..45ba0970504a 100644
--- a/drivers/net/ethernet/mediatek/Makefile
+++ b/drivers/net/ethernet/mediatek/Makefile
@@ -5,4 +5,9 @@
 
 obj-$(CONFIG_NET_MEDIATEK_SOC) += mtk_eth.o
 mtk_eth-y := mtk_eth_soc.o mtk_sgmii.o mtk_eth_path.o mtk_ppe.o mtk_ppe_debugfs.o mtk_ppe_offload.o
+mtk_eth-$(CONFIG_NET_MEDIATEK_SOC_WED) += mtk_wed.o
+ifdef CONFIG_DEBUG_FS
+mtk_eth-$(CONFIG_NET_MEDIATEK_SOC_WED) += mtk_wed_debugfs.o
+endif
+obj-$(CONFIG_NET_MEDIATEK_SOC_WED) += mtk_wed_ops.o
 obj-$(CONFIG_NET_MEDIATEK_STAR_EMAC) += mtk_star_emac.o
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 70db217ed831..4d7c542d89fb 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -24,6 +24,7 @@
 #include <net/dsa.h>
 
 #include "mtk_eth_soc.h"
+#include "mtk_wed.h"
 
 static int mtk_msg_level = -1;
 module_param_named(msg_level, mtk_msg_level, int, 0);
@@ -3170,6 +3171,22 @@ static int mtk_probe(struct platform_device *pdev)
 		}
 	}
 
+	for (i = 0;; i++) {
+		struct device_node *np = of_parse_phandle(pdev->dev.of_node,
+							  "mediatek,wed", i);
+		static const u32 wdma_regs[] = {
+			MTK_WDMA0_BASE,
+			MTK_WDMA1_BASE
+		};
+		void __iomem *wdma;
+
+		if (!np || i >= ARRAY_SIZE(wdma_regs))
+			break;
+
+		wdma = eth->base + wdma_regs[i];
+		mtk_wed_add_hw(np, eth, wdma, i);
+	}
+
 	for (i = 0; i < 3; i++) {
 		if (MTK_HAS_CAPS(eth->soc->caps, MTK_SHARED_INT) && i > 0)
 			eth->irq[i] = eth->irq[0];
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
index e701544c4287..74661682fd92 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -295,6 +295,9 @@
 #define MTK_GDM1_TX_GPCNT	0x2438
 #define MTK_STAT_OFFSET		0x40
 
+#define MTK_WDMA0_BASE		0x2800
+#define MTK_WDMA1_BASE		0x2c00
+
 /* QDMA descriptor txd4 */
 #define TX_DMA_CHKSUM		(0x7 << 29)
 #define TX_DMA_TSO		BIT(28)
diff --git a/drivers/net/ethernet/mediatek/mtk_wed.c b/drivers/net/ethernet/mediatek/mtk_wed.c
new file mode 100644
index 000000000000..f0eacf819cd9
--- /dev/null
+++ b/drivers/net/ethernet/mediatek/mtk_wed.c
@@ -0,0 +1,875 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2021 Felix Fietkau <nbd@nbd.name> */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/module.h>
+#include <linux/bitfield.h>
+#include <linux/dma-mapping.h>
+#include <linux/skbuff.h>
+#include <linux/of_platform.h>
+#include <linux/of_address.h>
+#include <linux/mfd/syscon.h>
+#include <linux/debugfs.h>
+#include <linux/soc/mediatek/mtk_wed.h>
+#include "mtk_eth_soc.h"
+#include "mtk_wed_regs.h"
+#include "mtk_wed.h"
+#include "mtk_ppe.h"
+
+#define MTK_PCIE_BASE(n)		(0x1a143000 + (n) * 0x2000)
+
+#define MTK_WED_PKT_SIZE		1900
+#define MTK_WED_BUF_SIZE		2048
+#define MTK_WED_BUF_PER_PAGE		(PAGE_SIZE / 2048)
+
+#define MTK_WED_TX_RING_SIZE		2048
+#define MTK_WED_WDMA_RING_SIZE		1024
+
+static struct mtk_wed_hw *hw_list[2];
+static DEFINE_MUTEX(hw_lock);
+
+static void
+wed_m32(struct mtk_wed_device *dev, u32 reg, u32 mask, u32 val)
+{
+	regmap_update_bits(dev->hw->regs, reg, mask | val, val);
+}
+
+static void
+wed_set(struct mtk_wed_device *dev, u32 reg, u32 mask)
+{
+	return wed_m32(dev, reg, 0, mask);
+}
+
+static void
+wed_clr(struct mtk_wed_device *dev, u32 reg, u32 mask)
+{
+	return wed_m32(dev, reg, mask, 0);
+}
+
+static void
+wdma_m32(struct mtk_wed_device *dev, u32 reg, u32 mask, u32 val)
+{
+	wdma_w32(dev, reg, (wdma_r32(dev, reg) & ~mask) | val);
+}
+
+static void
+wdma_set(struct mtk_wed_device *dev, u32 reg, u32 mask)
+{
+	wdma_m32(dev, reg, 0, mask);
+}
+
+static u32
+mtk_wed_read_reset(struct mtk_wed_device *dev)
+{
+	return wed_r32(dev, MTK_WED_RESET);
+}
+
+static void
+mtk_wed_reset(struct mtk_wed_device *dev, u32 mask)
+{
+	u32 status;
+
+	wed_w32(dev, MTK_WED_RESET, mask);
+	if (readx_poll_timeout(mtk_wed_read_reset, dev, status,
+			       !(status & mask), 0, 1000))
+		WARN_ON_ONCE(1);
+}
+
+static struct mtk_wed_hw *
+mtk_wed_assign(struct mtk_wed_device *dev)
+{
+	struct mtk_wed_hw *hw;
+
+	hw = hw_list[pci_domain_nr(dev->wlan.pci_dev->bus)];
+	if (!hw || hw->wed_dev)
+		return NULL;
+
+	hw->wed_dev = dev;
+	return hw;
+}
+
+static int
+mtk_wed_buffer_alloc(struct mtk_wed_device *dev)
+{
+	struct mtk_wdma_desc *desc;
+	dma_addr_t desc_phys;
+	void **page_list;
+	int token = dev->wlan.token_start;
+	int ring_size;
+	int n_pages;
+	int i, page_idx;
+
+	ring_size = dev->wlan.nbuf & ~(MTK_WED_BUF_PER_PAGE - 1);
+	n_pages = ring_size / MTK_WED_BUF_PER_PAGE;
+
+	page_list = kcalloc(n_pages, sizeof(*page_list), GFP_KERNEL);
+	if (!page_list)
+		return -ENOMEM;
+
+	dev->buf_ring.size = ring_size;
+	dev->buf_ring.pages = page_list;
+
+	desc = dma_alloc_coherent(dev->hw->dev, ring_size * sizeof(*desc),
+				  &desc_phys, GFP_KERNEL);
+	if (!desc)
+		return -ENOMEM;
+
+	dev->buf_ring.desc = desc;
+	dev->buf_ring.desc_phys = desc_phys;
+
+	for (i = 0, page_idx = 0; i < ring_size; i += MTK_WED_BUF_PER_PAGE) {
+		dma_addr_t page_phys, buf_phys;
+		struct page *page;
+		void *buf;
+		int s;
+
+		page = __dev_alloc_pages(GFP_KERNEL, 0);
+		if (!page)
+			return -ENOMEM;
+
+		page_phys = dma_map_page(dev->hw->dev, page, 0, PAGE_SIZE,
+					 DMA_BIDIRECTIONAL);
+		if (dma_mapping_error(dev->hw->dev, page_phys)) {
+			__free_page(page);
+			return -ENOMEM;
+		}
+
+		page_list[page_idx++] = page;
+		dma_sync_single_for_cpu(dev->hw->dev, page_phys, PAGE_SIZE,
+					DMA_BIDIRECTIONAL);
+
+		buf = page_to_virt(page);
+		buf_phys = page_phys;
+
+		for (s = 0; s < MTK_WED_BUF_PER_PAGE; s++) {
+			u32 txd_size;
+
+			txd_size = dev->wlan.init_buf(buf, buf_phys, token++);
+
+			desc->buf0 = buf_phys;
+			desc->buf1 = buf_phys + txd_size;
+			desc->ctrl = FIELD_PREP(MTK_WDMA_DESC_CTRL_LEN0,
+						txd_size) |
+				     FIELD_PREP(MTK_WDMA_DESC_CTRL_LEN1,
+						MTK_WED_BUF_SIZE - txd_size) |
+				     MTK_WDMA_DESC_CTRL_LAST_SEG1;
+			desc->info = 0;
+			desc++;
+
+			buf += MTK_WED_BUF_SIZE;
+			buf_phys += MTK_WED_BUF_SIZE;
+		}
+
+		dma_sync_single_for_device(dev->hw->dev, page_phys, PAGE_SIZE,
+					   DMA_BIDIRECTIONAL);
+	}
+
+	return 0;
+}
+
+static void
+mtk_wed_free_buffer(struct mtk_wed_device *dev)
+{
+	struct mtk_wdma_desc *desc = dev->buf_ring.desc;
+	void **page_list = dev->buf_ring.pages;
+	int page_idx;
+	int i;
+
+	if (!page_list)
+		return;
+
+	if (!desc)
+		goto free_pagelist;
+
+	for (i = 0, page_idx = 0; i < dev->buf_ring.size; i += MTK_WED_BUF_PER_PAGE) {
+		void *page = page_list[page_idx++];
+
+		if (!page)
+			break;
+
+		dma_unmap_page(dev->hw->dev, desc[i].buf0,
+			       PAGE_SIZE, DMA_BIDIRECTIONAL);
+		__free_page(page);
+	}
+
+	dma_free_coherent(dev->hw->dev, dev->buf_ring.size * sizeof(*desc),
+			  desc, dev->buf_ring.desc_phys);
+
+free_pagelist:
+	kfree(page_list);
+}
+
+static void
+mtk_wed_free_ring(struct mtk_wed_device *dev, struct mtk_wed_ring *ring)
+{
+	if (!ring->desc)
+		return;
+
+	dma_free_coherent(dev->hw->dev, ring->size * sizeof(*ring->desc),
+			  ring->desc, ring->desc_phys);
+}
+
+static void
+mtk_wed_free_tx_rings(struct mtk_wed_device *dev)
+{
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(dev->tx_ring); i++)
+		mtk_wed_free_ring(dev, &dev->tx_ring[i]);
+	for (i = 0; i < ARRAY_SIZE(dev->tx_wdma); i++)
+		mtk_wed_free_ring(dev, &dev->tx_wdma[i]);
+}
+
+static void
+mtk_wed_set_ext_int(struct mtk_wed_device *dev, bool en)
+{
+	u32 mask = MTK_WED_EXT_INT_STATUS_ERROR_MASK;
+
+	if (!dev->hw->num_flows)
+		mask &= ~MTK_WED_EXT_INT_STATUS_TKID_WO_PYLD;
+
+	wed_w32(dev, MTK_WED_EXT_INT_MASK, en ? mask : 0);
+	wed_r32(dev, MTK_WED_EXT_INT_MASK);
+}
+
+static void
+mtk_wed_stop(struct mtk_wed_device *dev)
+{
+	regmap_write(dev->hw->mirror, dev->hw->index * 4, 0);
+	mtk_wed_set_ext_int(dev, false);
+
+	wed_clr(dev, MTK_WED_CTRL,
+		MTK_WED_CTRL_WDMA_INT_AGENT_EN |
+		MTK_WED_CTRL_WPDMA_INT_AGENT_EN |
+		MTK_WED_CTRL_WED_TX_BM_EN |
+		MTK_WED_CTRL_WED_TX_FREE_AGENT_EN);
+	wed_w32(dev, MTK_WED_WPDMA_INT_TRIGGER, 0);
+	wed_w32(dev, MTK_WED_WDMA_INT_TRIGGER, 0);
+	wdma_w32(dev, MTK_WDMA_INT_MASK, 0);
+	wdma_w32(dev, MTK_WDMA_INT_GRP2, 0);
+	wed_w32(dev, MTK_WED_WPDMA_INT_MASK, 0);
+
+	wed_clr(dev, MTK_WED_GLO_CFG,
+		MTK_WED_GLO_CFG_TX_DMA_EN |
+		MTK_WED_GLO_CFG_RX_DMA_EN);
+	wed_clr(dev, MTK_WED_WPDMA_GLO_CFG,
+		MTK_WED_WPDMA_GLO_CFG_TX_DRV_EN |
+		MTK_WED_WPDMA_GLO_CFG_RX_DRV_EN);
+	wed_clr(dev, MTK_WED_WDMA_GLO_CFG,
+		MTK_WED_WDMA_GLO_CFG_RX_DRV_EN);
+}
+
+static void
+mtk_wed_detach(struct mtk_wed_device *dev)
+{
+	struct device_node *wlan_node = dev->wlan.pci_dev->dev.of_node;
+	struct mtk_wed_hw *hw = dev->hw;
+
+	mutex_lock(&hw_lock);
+
+	mtk_wed_stop(dev);
+
+	wdma_w32(dev, MTK_WDMA_RESET_IDX, MTK_WDMA_RESET_IDX_RX);
+	wdma_w32(dev, MTK_WDMA_RESET_IDX, 0);
+
+	mtk_wed_reset(dev, MTK_WED_RESET_WED);
+
+	mtk_wed_free_buffer(dev);
+	mtk_wed_free_tx_rings(dev);
+
+	if (of_dma_is_coherent(wlan_node))
+		regmap_update_bits(hw->hifsys, HIFSYS_DMA_AG_MAP,
+				   BIT(hw->index), BIT(hw->index));
+
+	if (!hw_list[!hw->index]->wed_dev &&
+	    hw->eth->dma_dev != hw->eth->dev)
+		mtk_eth_set_dma_device(hw->eth, hw->eth->dev);
+
+	memset(dev, 0, sizeof(*dev));
+	module_put(THIS_MODULE);
+
+	hw->wed_dev = NULL;
+	mutex_unlock(&hw_lock);
+}
+
+static void
+mtk_wed_hw_init_early(struct mtk_wed_device *dev)
+{
+	u32 mask, set;
+	u32 offset;
+
+	mtk_wed_stop(dev);
+	mtk_wed_reset(dev, MTK_WED_RESET_WED);
+
+	mask = MTK_WED_WDMA_GLO_CFG_BT_SIZE |
+	       MTK_WED_WDMA_GLO_CFG_DYNAMIC_DMAD_RECYCLE |
+	       MTK_WED_WDMA_GLO_CFG_RX_DIS_FSM_AUTO_IDLE;
+	set = FIELD_PREP(MTK_WED_WDMA_GLO_CFG_BT_SIZE, 2) |
+	      MTK_WED_WDMA_GLO_CFG_DYNAMIC_SKIP_DMAD_PREP |
+	      MTK_WED_WDMA_GLO_CFG_IDLE_DMAD_SUPPLY;
+	wed_m32(dev, MTK_WED_WDMA_GLO_CFG, mask, set);
+
+	wdma_set(dev, MTK_WDMA_GLO_CFG, MTK_WDMA_GLO_CFG_RX_INFO_PRERES);
+
+	offset = dev->hw->index ? 0x04000400 : 0;
+	wed_w32(dev, MTK_WED_WDMA_OFFSET0, 0x2a042a20 + offset);
+	wed_w32(dev, MTK_WED_WDMA_OFFSET1, 0x29002800 + offset);
+
+	wed_w32(dev, MTK_WED_PCIE_CFG_BASE, MTK_PCIE_BASE(dev->hw->index));
+	wed_w32(dev, MTK_WED_WPDMA_CFG_BASE, dev->wlan.wpdma_phys);
+}
+
+static void
+mtk_wed_hw_init(struct mtk_wed_device *dev)
+{
+	if (dev->init_done)
+		return;
+
+	dev->init_done = true;
+	mtk_wed_set_ext_int(dev, false);
+	wed_w32(dev, MTK_WED_TX_BM_CTRL,
+		MTK_WED_TX_BM_CTRL_PAUSE |
+		FIELD_PREP(MTK_WED_TX_BM_CTRL_VLD_GRP_NUM,
+			   dev->buf_ring.size / 128) |
+		FIELD_PREP(MTK_WED_TX_BM_CTRL_RSV_GRP_NUM,
+			   MTK_WED_TX_RING_SIZE / 256));
+
+	wed_w32(dev, MTK_WED_TX_BM_BASE, dev->buf_ring.desc_phys);
+
+	wed_w32(dev, MTK_WED_TX_BM_TKID,
+		FIELD_PREP(MTK_WED_TX_BM_TKID_START,
+			   dev->wlan.token_start) |
+		FIELD_PREP(MTK_WED_TX_BM_TKID_END,
+			   dev->wlan.token_start + dev->wlan.nbuf - 1));
+
+	wed_w32(dev, MTK_WED_TX_BM_BUF_LEN, MTK_WED_PKT_SIZE);
+
+	wed_w32(dev, MTK_WED_TX_BM_DYN_THR,
+		FIELD_PREP(MTK_WED_TX_BM_DYN_THR_LO, 1) |
+		MTK_WED_TX_BM_DYN_THR_HI);
+
+	mtk_wed_reset(dev, MTK_WED_RESET_TX_BM);
+
+	wed_set(dev, MTK_WED_CTRL,
+		MTK_WED_CTRL_WED_TX_BM_EN |
+		MTK_WED_CTRL_WED_TX_FREE_AGENT_EN);
+
+	wed_clr(dev, MTK_WED_TX_BM_CTRL, MTK_WED_TX_BM_CTRL_PAUSE);
+}
+
+static void
+mtk_wed_ring_reset(struct mtk_wdma_desc *desc, int size)
+{
+	int i;
+
+	for (i = 0; i < size; i++) {
+		desc[i].buf0 = 0;
+		desc[i].ctrl = cpu_to_le32(MTK_WDMA_DESC_CTRL_DMA_DONE);
+		desc[i].buf1 = 0;
+		desc[i].info = 0;
+	}
+}
+
+static u32
+mtk_wed_check_busy(struct mtk_wed_device *dev)
+{
+	if (wed_r32(dev, MTK_WED_GLO_CFG) & MTK_WED_GLO_CFG_TX_DMA_BUSY)
+		return true;
+
+	if (wed_r32(dev, MTK_WED_WPDMA_GLO_CFG) &
+	    MTK_WED_WPDMA_GLO_CFG_TX_DRV_BUSY)
+		return true;
+
+	if (wed_r32(dev, MTK_WED_CTRL) & MTK_WED_CTRL_WDMA_INT_AGENT_BUSY)
+		return true;
+
+	if (wed_r32(dev, MTK_WED_WDMA_GLO_CFG) &
+	    MTK_WED_WDMA_GLO_CFG_RX_DRV_BUSY)
+		return true;
+
+	if (wdma_r32(dev, MTK_WDMA_GLO_CFG) &
+	    MTK_WED_WDMA_GLO_CFG_RX_DRV_BUSY)
+		return true;
+
+	if (wed_r32(dev, MTK_WED_CTRL) &
+	    (MTK_WED_CTRL_WED_TX_BM_BUSY | MTK_WED_CTRL_WED_TX_FREE_AGENT_BUSY))
+		return true;
+
+	return false;
+}
+
+static int
+mtk_wed_poll_busy(struct mtk_wed_device *dev)
+{
+	int sleep = 15000;
+	int timeout = 100 * sleep;
+	u32 val;
+
+	return read_poll_timeout(mtk_wed_check_busy, val, !val, sleep,
+				 timeout, false, dev);
+}
+
+static void
+mtk_wed_reset_dma(struct mtk_wed_device *dev)
+{
+	bool busy = false;
+	u32 val;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(dev->tx_ring); i++) {
+		struct mtk_wdma_desc *desc = dev->tx_ring[i].desc;
+
+		if (!desc)
+			continue;
+
+		mtk_wed_ring_reset(desc, MTK_WED_TX_RING_SIZE);
+	}
+
+	if (mtk_wed_poll_busy(dev))
+		busy = mtk_wed_check_busy(dev);
+
+	if (busy) {
+		mtk_wed_reset(dev, MTK_WED_RESET_WED_TX_DMA);
+	} else {
+		wed_w32(dev, MTK_WED_RESET_IDX,
+			MTK_WED_RESET_IDX_TX |
+			MTK_WED_RESET_IDX_RX);
+		wed_w32(dev, MTK_WED_RESET_IDX, 0);
+	}
+
+	wdma_w32(dev, MTK_WDMA_RESET_IDX, MTK_WDMA_RESET_IDX_RX);
+	wdma_w32(dev, MTK_WDMA_RESET_IDX, 0);
+
+	if (busy) {
+		mtk_wed_reset(dev, MTK_WED_RESET_WDMA_INT_AGENT);
+		mtk_wed_reset(dev, MTK_WED_RESET_WDMA_RX_DRV);
+	} else {
+		wed_w32(dev, MTK_WED_WDMA_RESET_IDX,
+			MTK_WED_WDMA_RESET_IDX_RX | MTK_WED_WDMA_RESET_IDX_DRV);
+		wed_w32(dev, MTK_WED_WDMA_RESET_IDX, 0);
+
+		wed_set(dev, MTK_WED_WDMA_GLO_CFG,
+			MTK_WED_WDMA_GLO_CFG_RST_INIT_COMPLETE);
+
+		wed_clr(dev, MTK_WED_WDMA_GLO_CFG,
+			MTK_WED_WDMA_GLO_CFG_RST_INIT_COMPLETE);
+	}
+
+	for (i = 0; i < 100; i++) {
+		val = wed_r32(dev, MTK_WED_TX_BM_INTF);
+		if (FIELD_GET(MTK_WED_TX_BM_INTF_TKFIFO_FDEP, val) == 0x40)
+			break;
+	}
+
+	mtk_wed_reset(dev, MTK_WED_RESET_TX_FREE_AGENT);
+	mtk_wed_reset(dev, MTK_WED_RESET_TX_BM);
+
+	if (busy) {
+		mtk_wed_reset(dev, MTK_WED_RESET_WPDMA_INT_AGENT);
+		mtk_wed_reset(dev, MTK_WED_RESET_WPDMA_TX_DRV);
+		mtk_wed_reset(dev, MTK_WED_RESET_WPDMA_RX_DRV);
+	} else {
+		wed_w32(dev, MTK_WED_WPDMA_RESET_IDX,
+			MTK_WED_WPDMA_RESET_IDX_TX |
+			MTK_WED_WPDMA_RESET_IDX_RX);
+		wed_w32(dev, MTK_WED_WPDMA_RESET_IDX, 0);
+	}
+
+}
+
+static int
+mtk_wed_ring_alloc(struct mtk_wed_device *dev, struct mtk_wed_ring *ring,
+		   int size)
+{
+	ring->desc = dma_alloc_coherent(dev->hw->dev,
+					size * sizeof(*ring->desc),
+					&ring->desc_phys, GFP_KERNEL);
+	if (!ring->desc)
+		return -ENOMEM;
+
+	ring->size = size;
+	mtk_wed_ring_reset(ring->desc, size);
+
+	return 0;
+}
+
+static int
+mtk_wed_wdma_ring_setup(struct mtk_wed_device *dev, int idx, int size)
+{
+	struct mtk_wed_ring *wdma = &dev->tx_wdma[idx];
+
+	if (mtk_wed_ring_alloc(dev, wdma, MTK_WED_WDMA_RING_SIZE))
+		return -ENOMEM;
+
+	wdma_w32(dev, MTK_WDMA_RING_RX(idx) + MTK_WED_RING_OFS_BASE,
+		 wdma->desc_phys);
+	wdma_w32(dev, MTK_WDMA_RING_RX(idx) + MTK_WED_RING_OFS_COUNT,
+		 size);
+	wdma_w32(dev, MTK_WDMA_RING_RX(idx) + MTK_WED_RING_OFS_CPU_IDX, 0);
+
+	wed_w32(dev, MTK_WED_WDMA_RING_RX(idx) + MTK_WED_RING_OFS_BASE,
+		wdma->desc_phys);
+	wed_w32(dev, MTK_WED_WDMA_RING_RX(idx) + MTK_WED_RING_OFS_COUNT,
+		size);
+
+	return 0;
+}
+
+static void
+mtk_wed_start(struct mtk_wed_device *dev, u32 irq_mask)
+{
+	u32 wdma_mask;
+	u32 val;
+	int i;
+
+	for (i = 0; i < ARRAY_SIZE(dev->tx_wdma); i++)
+		if (!dev->tx_wdma[i].desc)
+			mtk_wed_wdma_ring_setup(dev, i, 16);
+
+	wdma_mask = FIELD_PREP(MTK_WDMA_INT_MASK_RX_DONE, GENMASK(1, 0));
+
+	mtk_wed_hw_init(dev);
+
+	wed_set(dev, MTK_WED_CTRL,
+		MTK_WED_CTRL_WDMA_INT_AGENT_EN |
+		MTK_WED_CTRL_WPDMA_INT_AGENT_EN |
+		MTK_WED_CTRL_WED_TX_BM_EN |
+		MTK_WED_CTRL_WED_TX_FREE_AGENT_EN);
+
+	wed_w32(dev, MTK_WED_PCIE_INT_TRIGGER, MTK_WED_PCIE_INT_TRIGGER_STATUS);
+
+	wed_w32(dev, MTK_WED_WPDMA_INT_TRIGGER,
+		MTK_WED_WPDMA_INT_TRIGGER_RX_DONE |
+		MTK_WED_WPDMA_INT_TRIGGER_TX_DONE);
+
+	wed_set(dev, MTK_WED_WPDMA_INT_CTRL,
+		MTK_WED_WPDMA_INT_CTRL_SUBRT_ADV);
+
+	wed_w32(dev, MTK_WED_WDMA_INT_TRIGGER, wdma_mask);
+	wed_clr(dev, MTK_WED_WDMA_INT_CTRL, wdma_mask);
+
+	wdma_w32(dev, MTK_WDMA_INT_MASK, wdma_mask);
+	wdma_w32(dev, MTK_WDMA_INT_GRP2, wdma_mask);
+
+	wed_w32(dev, MTK_WED_WPDMA_INT_MASK, irq_mask);
+	wed_w32(dev, MTK_WED_INT_MASK, irq_mask);
+
+	wed_set(dev, MTK_WED_GLO_CFG,
+		MTK_WED_GLO_CFG_TX_DMA_EN |
+		MTK_WED_GLO_CFG_RX_DMA_EN);
+	wed_set(dev, MTK_WED_WPDMA_GLO_CFG,
+		MTK_WED_WPDMA_GLO_CFG_TX_DRV_EN |
+		MTK_WED_WPDMA_GLO_CFG_RX_DRV_EN);
+	wed_set(dev, MTK_WED_WDMA_GLO_CFG,
+		MTK_WED_WDMA_GLO_CFG_RX_DRV_EN);
+
+	mtk_wed_set_ext_int(dev, true);
+	val = dev->wlan.wpdma_phys |
+	      MTK_PCIE_MIRROR_MAP_EN |
+	      FIELD_PREP(MTK_PCIE_MIRROR_MAP_WED_ID, dev->hw->index);
+
+	if (dev->hw->index)
+		val |= BIT(1);
+	val |= BIT(0);
+	regmap_write(dev->hw->mirror, dev->hw->index * 4, val);
+
+	dev->running = true;
+}
+
+static int
+mtk_wed_attach(struct mtk_wed_device *dev)
+	__releases(RCU)
+{
+	struct mtk_wed_hw *hw;
+	int ret = 0;
+
+	RCU_LOCKDEP_WARN(!rcu_read_lock_held(),
+			 "mtk_wed_attach without holding the RCU read lock");
+
+	if (pci_domain_nr(dev->wlan.pci_dev->bus) > 1 ||
+	    !try_module_get(THIS_MODULE))
+		ret = -ENODEV;
+
+	rcu_read_unlock();
+
+	if (ret)
+		return ret;
+
+	mutex_lock(&hw_lock);
+
+	hw = mtk_wed_assign(dev);
+	if (!hw) {
+		module_put(THIS_MODULE);
+		ret = -ENODEV;
+		goto out;
+	}
+
+	dev_info(&dev->wlan.pci_dev->dev, "attaching wed device %d\n", hw->index);
+
+	dev->hw = hw;
+	dev->dev = hw->dev;
+	dev->irq = hw->irq;
+	dev->wdma_idx = hw->index;
+
+	if (hw->eth->dma_dev == hw->eth->dev &&
+	    of_dma_is_coherent(hw->eth->dev->of_node))
+		mtk_eth_set_dma_device(hw->eth, hw->dev);
+
+	ret = mtk_wed_buffer_alloc(dev);
+	if (ret) {
+		mtk_wed_detach(dev);
+		goto out;
+	}
+
+	mtk_wed_hw_init_early(dev);
+	regmap_update_bits(hw->hifsys, HIFSYS_DMA_AG_MAP, BIT(hw->index), 0);
+
+out:
+	mutex_unlock(&hw_lock);
+
+	return ret;
+}
+
+static int
+mtk_wed_tx_ring_setup(struct mtk_wed_device *dev, int idx, void __iomem *regs)
+{
+	struct mtk_wed_ring *ring = &dev->tx_ring[idx];
+
+	/*
+	 * Tx ring redirection:
+	 * Instead of configuring the WLAN PDMA TX ring directly, the WLAN
+	 * driver allocated DMA ring gets configured into WED MTK_WED_RING_TX(n)
+	 * registers.
+	 *
+	 * WED driver posts its own DMA ring as WLAN PDMA TX and configures it
+	 * into MTK_WED_WPDMA_RING_TX(n) registers.
+	 * It gets filled with packets picked up from WED TX ring and from
+	 * WDMA RX.
+	 */
+
+	BUG_ON(idx > ARRAY_SIZE(dev->tx_ring));
+
+	if (mtk_wed_ring_alloc(dev, ring, MTK_WED_TX_RING_SIZE))
+		return -ENOMEM;
+
+	if (mtk_wed_wdma_ring_setup(dev, idx, MTK_WED_WDMA_RING_SIZE))
+		return -ENOMEM;
+
+	ring->reg_base = MTK_WED_RING_TX(idx);
+	ring->wpdma = regs;
+
+	/* WED -> WPDMA */
+	wpdma_tx_w32(dev, idx, MTK_WED_RING_OFS_BASE, ring->desc_phys);
+	wpdma_tx_w32(dev, idx, MTK_WED_RING_OFS_COUNT, MTK_WED_TX_RING_SIZE);
+	wpdma_tx_w32(dev, idx, MTK_WED_RING_OFS_CPU_IDX, 0);
+
+	wed_w32(dev, MTK_WED_WPDMA_RING_TX(idx) + MTK_WED_RING_OFS_BASE,
+		ring->desc_phys);
+	wed_w32(dev, MTK_WED_WPDMA_RING_TX(idx) + MTK_WED_RING_OFS_COUNT,
+		MTK_WED_TX_RING_SIZE);
+	wed_w32(dev, MTK_WED_WPDMA_RING_TX(idx) + MTK_WED_RING_OFS_CPU_IDX, 0);
+
+	return 0;
+}
+
+static int
+mtk_wed_txfree_ring_setup(struct mtk_wed_device *dev, void __iomem *regs)
+{
+	struct mtk_wed_ring *ring = &dev->txfree_ring;
+	int i;
+
+	/*
+	 * For txfree event handling, the same DMA ring is shared between WED
+	 * and WLAN. The WLAN driver accesses the ring index registers through
+	 * WED
+	 */
+	ring->reg_base = MTK_WED_RING_RX(1);
+	ring->wpdma = regs;
+
+	for (i = 0; i < 12; i += 4) {
+		u32 val = readl(regs + i);
+
+		wed_w32(dev, MTK_WED_RING_RX(1) + i, val);
+		wed_w32(dev, MTK_WED_WPDMA_RING_RX(1) + i, val);
+	}
+
+	return 0;
+}
+
+static u32
+mtk_wed_irq_get(struct mtk_wed_device *dev, u32 mask)
+{
+	u32 val;
+
+	val = wed_r32(dev, MTK_WED_EXT_INT_STATUS);
+	wed_w32(dev, MTK_WED_EXT_INT_STATUS, val);
+	val &= MTK_WED_EXT_INT_STATUS_ERROR_MASK;
+	if (!dev->hw->num_flows)
+		val &= ~MTK_WED_EXT_INT_STATUS_TKID_WO_PYLD;
+	if (val && net_ratelimit())
+		pr_err("mtk_wed%d: error status=%08x\n", dev->hw->index, val);
+
+	val = wed_r32(dev, MTK_WED_INT_STATUS);
+	val &= mask;
+	wed_w32(dev, MTK_WED_INT_STATUS, val); /* ACK */
+
+	return val;
+}
+
+static void
+mtk_wed_irq_set_mask(struct mtk_wed_device *dev, u32 mask)
+{
+	if (!dev->running)
+		return;
+
+	mtk_wed_set_ext_int(dev, !!mask);
+	wed_w32(dev, MTK_WED_INT_MASK, mask);
+}
+
+int mtk_wed_flow_add(int index)
+{
+	struct mtk_wed_hw *hw = hw_list[index];
+	int ret;
+
+	if (!hw || !hw->wed_dev)
+		return -ENODEV;
+
+	if (hw->num_flows) {
+		hw->num_flows++;
+		return 0;
+	}
+
+	mutex_lock(&hw_lock);
+	if (!hw->wed_dev) {
+		ret = -ENODEV;
+		goto out;
+	}
+
+	ret = hw->wed_dev->wlan.offload_enable(hw->wed_dev);
+	if (!ret)
+		hw->num_flows++;
+	mtk_wed_set_ext_int(hw->wed_dev, true);
+
+out:
+	mutex_unlock(&hw_lock);
+
+	return ret;
+}
+
+void mtk_wed_flow_remove(int index)
+{
+	struct mtk_wed_hw *hw = hw_list[index];
+
+	if (!hw)
+		return;
+
+	if (--hw->num_flows)
+		return;
+
+	mutex_lock(&hw_lock);
+	if (!hw->wed_dev)
+		goto out;
+
+	hw->wed_dev->wlan.offload_disable(hw->wed_dev);
+	mtk_wed_set_ext_int(hw->wed_dev, true);
+
+out:
+	mutex_unlock(&hw_lock);
+}
+
+void mtk_wed_add_hw(struct device_node *np, struct mtk_eth *eth,
+		    void __iomem *wdma, int index)
+{
+	static const struct mtk_wed_ops wed_ops = {
+		.attach = mtk_wed_attach,
+		.tx_ring_setup = mtk_wed_tx_ring_setup,
+		.txfree_ring_setup = mtk_wed_txfree_ring_setup,
+		.start = mtk_wed_start,
+		.stop = mtk_wed_stop,
+		.reset_dma = mtk_wed_reset_dma,
+		.reg_read = wed_r32,
+		.reg_write = wed_w32,
+		.irq_get = mtk_wed_irq_get,
+		.irq_set_mask = mtk_wed_irq_set_mask,
+		.detach = mtk_wed_detach,
+	};
+	struct device_node *eth_np = eth->dev->of_node;
+	struct platform_device *pdev;
+	struct mtk_wed_hw *hw;
+	struct regmap *regs;
+	int irq;
+
+	if (!np)
+		return;
+
+	pdev = of_find_device_by_node(np);
+	if (!pdev)
+		return;
+
+	get_device(&pdev->dev);
+	irq = platform_get_irq(pdev, 0);
+	if (irq < 0)
+		return;
+
+	regs = syscon_regmap_lookup_by_phandle(np, NULL);
+	if (!regs)
+		return;
+
+	rcu_assign_pointer(mtk_soc_wed_ops, &wed_ops);
+
+	mutex_lock(&hw_lock);
+
+	if (WARN_ON(hw_list[index]))
+		goto unlock;
+
+	hw = kzalloc(sizeof(*hw), GFP_KERNEL);
+	hw->node = np;
+	hw->regs = regs;
+	hw->eth = eth;
+	hw->dev = &pdev->dev;
+	hw->wdma = wdma;
+	hw->index = index;
+	hw->irq = irq;
+	hw->mirror = syscon_regmap_lookup_by_phandle(eth_np,
+						     "mediatek,pcie-mirror");
+	hw->hifsys = syscon_regmap_lookup_by_phandle(eth_np,
+						     "mediatek,hifsys");
+	if (IS_ERR(hw->mirror) || IS_ERR(hw->hifsys)) {
+		kfree(hw);
+		goto unlock;
+	}
+
+	if (!index) {
+		regmap_write(hw->mirror, 0, 0);
+		regmap_write(hw->mirror, 4, 0);
+	}
+	mtk_wed_hw_add_debugfs(hw);
+
+	hw_list[index] = hw;
+
+unlock:
+	mutex_unlock(&hw_lock);
+}
+
+void mtk_wed_exit(void)
+{
+	int i;
+
+	rcu_assign_pointer(mtk_soc_wed_ops, NULL);
+
+	synchronize_rcu();
+
+	for (i = 0; i < ARRAY_SIZE(hw_list); i++) {
+		struct mtk_wed_hw *hw;
+
+		hw = hw_list[i];
+		if (!hw)
+			continue;
+
+		hw_list[i] = NULL;
+		debugfs_remove(hw->debugfs_dir);
+		put_device(hw->dev);
+		kfree(hw);
+	}
+}
diff --git a/drivers/net/ethernet/mediatek/mtk_wed.h b/drivers/net/ethernet/mediatek/mtk_wed.h
new file mode 100644
index 000000000000..404c9a9b130d
--- /dev/null
+++ b/drivers/net/ethernet/mediatek/mtk_wed.h
@@ -0,0 +1,128 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2021 Felix Fietkau <nbd@nbd.name> */
+
+#ifndef __MTK_WED_PRIV_H
+#define __MTK_WED_PRIV_H
+
+#include <linux/soc/mediatek/mtk_wed.h>
+#include <linux/debugfs.h>
+#include <linux/regmap.h>
+
+struct mtk_eth;
+
+struct mtk_wed_hw {
+	struct device_node *node;
+	struct mtk_eth *eth;
+	struct regmap *regs;
+	struct regmap *hifsys;
+	struct device *dev;
+	void __iomem *wdma;
+	struct regmap *mirror;
+	struct dentry *debugfs_dir;
+	struct mtk_wed_device *wed_dev;
+	u32 debugfs_reg;
+	u32 num_flows;
+	char dirname[5];
+	int irq;
+	int index;
+};
+
+
+#ifdef CONFIG_NET_MEDIATEK_SOC_WED
+static inline void
+wed_w32(struct mtk_wed_device *dev, u32 reg, u32 val)
+{
+	regmap_write(dev->hw->regs, reg, val);
+}
+
+static inline u32
+wed_r32(struct mtk_wed_device *dev, u32 reg)
+{
+	unsigned int val;
+
+	regmap_read(dev->hw->regs, reg, &val);
+
+	return val;
+}
+
+static inline void
+wdma_w32(struct mtk_wed_device *dev, u32 reg, u32 val)
+{
+	writel(val, dev->hw->wdma + reg);
+}
+
+static inline u32
+wdma_r32(struct mtk_wed_device *dev, u32 reg)
+{
+	return readl(dev->hw->wdma + reg);
+}
+
+static inline u32
+wpdma_tx_r32(struct mtk_wed_device *dev, int ring, u32 reg)
+{
+	if (!dev->tx_ring[ring].wpdma)
+		return 0;
+
+	return readl(dev->tx_ring[ring].wpdma + reg);
+}
+
+static inline void
+wpdma_tx_w32(struct mtk_wed_device *dev, int ring, u32 reg, u32 val)
+{
+	if (!dev->tx_ring[ring].wpdma)
+		return;
+
+	writel(val, dev->tx_ring[ring].wpdma + reg);
+}
+
+static inline u32
+wpdma_txfree_r32(struct mtk_wed_device *dev, u32 reg)
+{
+	if (!dev->txfree_ring.wpdma)
+		return 0;
+
+	return readl(dev->txfree_ring.wpdma + reg);
+}
+
+static inline void
+wpdma_txfree_w32(struct mtk_wed_device *dev, u32 reg, u32 val)
+{
+	if (!dev->txfree_ring.wpdma)
+		return;
+
+	writel(val, dev->txfree_ring.wpdma + reg);
+}
+
+void mtk_wed_add_hw(struct device_node *np, struct mtk_eth *eth,
+		    void __iomem *wdma, int index);
+void mtk_wed_exit(void);
+int mtk_wed_flow_add(int index);
+void mtk_wed_flow_remove(int index);
+#else
+static inline void
+mtk_wed_add_hw(struct device_node *np, struct mtk_eth *eth,
+	       void __iomem *wdma, int index)
+{
+}
+static inline void
+mtk_wed_exit(void)
+{
+}
+static inline int mtk_wed_flow_add(int index)
+{
+	return -EINVAL;
+}
+static inline void mtk_wed_flow_remove(int index)
+{
+}
+#endif
+
+#ifdef CONFIG_DEBUG_FS
+void mtk_wed_hw_add_debugfs(struct mtk_wed_hw *hw);
+#else
+static inline void mtk_wed_hw_add_debugfs(struct mtk_wed_hw *hw)
+{
+}
+#endif
+
+#endif
diff --git a/drivers/net/ethernet/mediatek/mtk_wed_debugfs.c b/drivers/net/ethernet/mediatek/mtk_wed_debugfs.c
new file mode 100644
index 000000000000..a81d3fd1a439
--- /dev/null
+++ b/drivers/net/ethernet/mediatek/mtk_wed_debugfs.c
@@ -0,0 +1,175 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2021 Felix Fietkau <nbd@nbd.name> */
+
+#include <linux/seq_file.h>
+#include "mtk_wed.h"
+#include "mtk_wed_regs.h"
+
+struct reg_dump {
+	const char *name;
+	u16 offset;
+	u8 type;
+	u8 base;
+};
+
+enum {
+	DUMP_TYPE_STRING,
+	DUMP_TYPE_WED,
+	DUMP_TYPE_WDMA,
+	DUMP_TYPE_WPDMA_TX,
+	DUMP_TYPE_WPDMA_TXFREE,
+};
+
+#define DUMP_STR(_str) { _str, 0, DUMP_TYPE_STRING }
+#define DUMP_REG(_reg, ...) { #_reg, MTK_##_reg, __VA_ARGS__ }
+#define DUMP_RING(_prefix, _base, ...)				\
+	{ _prefix " BASE", _base, __VA_ARGS__ },		\
+	{ _prefix " CNT",  _base + 0x4, __VA_ARGS__ },	\
+	{ _prefix " CIDX", _base + 0x8, __VA_ARGS__ },	\
+	{ _prefix " DIDX", _base + 0xc, __VA_ARGS__ }
+
+#define DUMP_WED(_reg) DUMP_REG(_reg, DUMP_TYPE_WED)
+#define DUMP_WED_RING(_base) DUMP_RING(#_base, MTK_##_base, DUMP_TYPE_WED)
+
+#define DUMP_WDMA(_reg) DUMP_REG(_reg, DUMP_TYPE_WDMA)
+#define DUMP_WDMA_RING(_base) DUMP_RING(#_base, MTK_##_base, DUMP_TYPE_WDMA)
+
+#define DUMP_WPDMA_TX_RING(_n) DUMP_RING("WPDMA_TX" #_n, 0, DUMP_TYPE_WPDMA_TX, _n)
+#define DUMP_WPDMA_TXFREE_RING DUMP_RING("WPDMA_RX1", 0, DUMP_TYPE_WPDMA_TXFREE)
+
+static void
+print_reg_val(struct seq_file *s, const char *name, u32 val)
+{
+	seq_printf(s, "%-32s %08x\n", name, val);
+}
+
+static void
+dump_wed_regs(struct seq_file *s, struct mtk_wed_device *dev,
+	      const struct reg_dump *regs, int n_regs)
+{
+	const struct reg_dump *cur;
+	u32 val;
+
+	for (cur = regs; cur < &regs[n_regs]; cur++) {
+		switch (cur->type) {
+		case DUMP_TYPE_STRING:
+			seq_printf(s, "%s======== %s:\n",
+				   cur > regs ? "\n" : "",
+				   cur->name);
+			continue;
+		case DUMP_TYPE_WED:
+			val = wed_r32(dev, cur->offset);
+			break;
+		case DUMP_TYPE_WDMA:
+			val = wdma_r32(dev, cur->offset);
+			break;
+		case DUMP_TYPE_WPDMA_TX:
+			val = wpdma_tx_r32(dev, cur->base, cur->offset);
+			break;
+		case DUMP_TYPE_WPDMA_TXFREE:
+			val = wpdma_txfree_r32(dev, cur->offset);
+			break;
+		}
+		print_reg_val(s, cur->name, val);
+	}
+}
+
+
+static int
+wed_txinfo_show(struct seq_file *s, void *data)
+{
+	static const struct reg_dump regs[] = {
+		DUMP_STR("WED TX"),
+		DUMP_WED(WED_TX_MIB(0)),
+		DUMP_WED_RING(WED_RING_TX(0)),
+
+		DUMP_WED(WED_TX_MIB(1)),
+		DUMP_WED_RING(WED_RING_TX(1)),
+
+		DUMP_STR("WPDMA TX"),
+		DUMP_WED(WED_WPDMA_TX_MIB(0)),
+		DUMP_WED_RING(WED_WPDMA_RING_TX(0)),
+		DUMP_WED(WED_WPDMA_TX_COHERENT_MIB(0)),
+
+		DUMP_WED(WED_WPDMA_TX_MIB(1)),
+		DUMP_WED_RING(WED_WPDMA_RING_TX(1)),
+		DUMP_WED(WED_WPDMA_TX_COHERENT_MIB(1)),
+
+		DUMP_STR("WPDMA TX"),
+		DUMP_WPDMA_TX_RING(0),
+		DUMP_WPDMA_TX_RING(1),
+
+		DUMP_STR("WED WDMA RX"),
+		DUMP_WED(WED_WDMA_RX_MIB(0)),
+		DUMP_WED_RING(WED_WDMA_RING_RX(0)),
+		DUMP_WED(WED_WDMA_RX_THRES(0)),
+		DUMP_WED(WED_WDMA_RX_RECYCLE_MIB(0)),
+		DUMP_WED(WED_WDMA_RX_PROCESSED_MIB(0)),
+
+		DUMP_WED(WED_WDMA_RX_MIB(1)),
+		DUMP_WED_RING(WED_WDMA_RING_RX(1)),
+		DUMP_WED(WED_WDMA_RX_THRES(1)),
+		DUMP_WED(WED_WDMA_RX_RECYCLE_MIB(1)),
+		DUMP_WED(WED_WDMA_RX_PROCESSED_MIB(1)),
+
+		DUMP_STR("WDMA RX"),
+		DUMP_WDMA(WDMA_GLO_CFG),
+		DUMP_WDMA_RING(WDMA_RING_RX(0)),
+		DUMP_WDMA_RING(WDMA_RING_RX(1)),
+	};
+	struct mtk_wed_hw *hw = s->private;
+	struct mtk_wed_device *dev = hw->wed_dev;
+
+	if (!dev)
+		return 0;
+
+	dump_wed_regs(s, dev, regs, ARRAY_SIZE(regs));
+
+	return 0;
+}
+DEFINE_SHOW_ATTRIBUTE(wed_txinfo);
+
+
+static int
+mtk_wed_reg_set(void *data, u64 val)
+{
+	struct mtk_wed_hw *hw = data;
+
+	regmap_write(hw->regs, hw->debugfs_reg, val);
+
+	return 0;
+}
+
+static int
+mtk_wed_reg_get(void *data, u64 *val)
+{
+	struct mtk_wed_hw *hw = data;
+	unsigned int regval;
+	int ret;
+
+	ret = regmap_read(hw->regs, hw->debugfs_reg, &regval);
+	if (ret)
+		return ret;
+
+	*val = regval;
+
+	return 0;
+}
+
+DEFINE_DEBUGFS_ATTRIBUTE(fops_regval, mtk_wed_reg_get, mtk_wed_reg_set,
+             "0x%08llx\n");
+
+void mtk_wed_hw_add_debugfs(struct mtk_wed_hw *hw)
+{
+	struct dentry *dir;
+
+	snprintf(hw->dirname, sizeof(hw->dirname), "wed%d", hw->index);
+	dir = debugfs_create_dir(hw->dirname, NULL);
+	if (!dir)
+		return;
+
+	hw->debugfs_dir = dir;
+	debugfs_create_u32("regidx", 0600, dir, &hw->debugfs_reg);
+	debugfs_create_file_unsafe("regval", 0600, dir, hw, &fops_regval);
+	debugfs_create_file_unsafe("txinfo", 0400, dir, hw, &wed_txinfo_fops);
+}
diff --git a/drivers/net/ethernet/mediatek/mtk_wed_ops.c b/drivers/net/ethernet/mediatek/mtk_wed_ops.c
new file mode 100644
index 000000000000..a5d9d8a5bce2
--- /dev/null
+++ b/drivers/net/ethernet/mediatek/mtk_wed_ops.c
@@ -0,0 +1,8 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2020 Felix Fietkau <nbd@nbd.name> */
+
+#include <linux/kernel.h>
+#include <linux/soc/mediatek/mtk_wed.h>
+
+const struct mtk_wed_ops __rcu *mtk_soc_wed_ops;
+EXPORT_SYMBOL_GPL(mtk_soc_wed_ops);
diff --git a/drivers/net/ethernet/mediatek/mtk_wed_regs.h b/drivers/net/ethernet/mediatek/mtk_wed_regs.h
new file mode 100644
index 000000000000..0a0465ea58b4
--- /dev/null
+++ b/drivers/net/ethernet/mediatek/mtk_wed_regs.h
@@ -0,0 +1,251 @@
+// SPDX-License-Identifier: GPL-2.0-only
+/* Copyright (C) 2020 Felix Fietkau <nbd@nbd.name> */
+
+#ifndef __MTK_WED_REGS_H
+#define __MTK_WED_REGS_H
+
+#define MTK_WDMA_DESC_CTRL_LEN1			GENMASK(14, 0)
+#define MTK_WDMA_DESC_CTRL_LAST_SEG1		BIT(15)
+#define MTK_WDMA_DESC_CTRL_BURST		BIT(16)
+#define MTK_WDMA_DESC_CTRL_LEN0			GENMASK(29, 16)
+#define MTK_WDMA_DESC_CTRL_LAST_SEG0		BIT(30)
+#define MTK_WDMA_DESC_CTRL_DMA_DONE		BIT(31)
+
+struct mtk_wdma_desc {
+	__le32 buf0;
+	__le32 ctrl;
+	__le32 buf1;
+	__le32 info;
+} __packed __aligned(4);
+
+#define MTK_WED_RESET					0x008
+#define MTK_WED_RESET_TX_BM				BIT(0)
+#define MTK_WED_RESET_TX_FREE_AGENT			BIT(4)
+#define MTK_WED_RESET_WPDMA_TX_DRV			BIT(8)
+#define MTK_WED_RESET_WPDMA_RX_DRV			BIT(9)
+#define MTK_WED_RESET_WPDMA_INT_AGENT			BIT(11)
+#define MTK_WED_RESET_WED_TX_DMA			BIT(12)
+#define MTK_WED_RESET_WDMA_RX_DRV			BIT(17)
+#define MTK_WED_RESET_WDMA_INT_AGENT			BIT(19)
+#define MTK_WED_RESET_WED				BIT(31)
+
+#define MTK_WED_CTRL					0x00c
+#define MTK_WED_CTRL_WPDMA_INT_AGENT_EN			BIT(0)
+#define MTK_WED_CTRL_WPDMA_INT_AGENT_BUSY		BIT(1)
+#define MTK_WED_CTRL_WDMA_INT_AGENT_EN			BIT(2)
+#define MTK_WED_CTRL_WDMA_INT_AGENT_BUSY		BIT(3)
+#define MTK_WED_CTRL_WED_TX_BM_EN			BIT(8)
+#define MTK_WED_CTRL_WED_TX_BM_BUSY			BIT(9)
+#define MTK_WED_CTRL_WED_TX_FREE_AGENT_EN		BIT(10)
+#define MTK_WED_CTRL_WED_TX_FREE_AGENT_BUSY		BIT(11)
+#define MTK_WED_CTRL_RESERVE_EN				BIT(12)
+#define MTK_WED_CTRL_RESERVE_BUSY			BIT(13)
+#define MTK_WED_CTRL_FINAL_DIDX_READ			BIT(24)
+#define MTK_WED_CTRL_MIB_READ_CLEAR			BIT(28)
+
+#define MTK_WED_EXT_INT_STATUS				0x020
+#define MTK_WED_EXT_INT_STATUS_TF_LEN_ERR		BIT(0)
+#define MTK_WED_EXT_INT_STATUS_TKID_WO_PYLD		BIT(1)
+#define MTK_WED_EXT_INT_STATUS_TKID_TITO_INVALID	BIT(4)
+#define MTK_WED_EXT_INT_STATUS_TX_FBUF_LO_TH		BIT(8)
+#define MTK_WED_EXT_INT_STATUS_TX_FBUF_HI_TH		BIT(9)
+#define MTK_WED_EXT_INT_STATUS_RX_FBUF_LO_TH		BIT(12)
+#define MTK_WED_EXT_INT_STATUS_RX_FBUF_HI_TH		BIT(13)
+#define MTK_WED_EXT_INT_STATUS_RX_DRV_R_RESP_ERR	BIT(16)
+#define MTK_WED_EXT_INT_STATUS_RX_DRV_W_RESP_ERR	BIT(17)
+#define MTK_WED_EXT_INT_STATUS_RX_DRV_COHERENT		BIT(18)
+#define MTK_WED_EXT_INT_STATUS_RX_DRV_INIT_WDMA_EN	BIT(19)
+#define MTK_WED_EXT_INT_STATUS_RX_DRV_BM_DMAD_COHERENT	BIT(20)
+#define MTK_WED_EXT_INT_STATUS_TX_DRV_R_RESP_ERR	BIT(21)
+#define MTK_WED_EXT_INT_STATUS_TX_DRV_W_RESP_ERR	BIT(22)
+#define MTK_WED_EXT_INT_STATUS_RX_DRV_DMA_RECYCLE	BIT(24)
+#define MTK_WED_EXT_INT_STATUS_ERROR_MASK		(MTK_WED_EXT_INT_STATUS_TF_LEN_ERR | \
+							 MTK_WED_EXT_INT_STATUS_TKID_WO_PYLD | \
+							 MTK_WED_EXT_INT_STATUS_TKID_TITO_INVALID | \
+							 MTK_WED_EXT_INT_STATUS_RX_DRV_R_RESP_ERR | \
+							 MTK_WED_EXT_INT_STATUS_RX_DRV_W_RESP_ERR | \
+							 MTK_WED_EXT_INT_STATUS_RX_DRV_INIT_WDMA_EN | \
+							 MTK_WED_EXT_INT_STATUS_TX_DRV_R_RESP_ERR | \
+							 MTK_WED_EXT_INT_STATUS_TX_DRV_W_RESP_ERR)
+
+#define MTK_WED_EXT_INT_MASK				0x028
+
+#define MTK_WED_STATUS					0x060
+#define MTK_WED_STATUS_TX				GENMASK(15, 8)
+
+#define MTK_WED_TX_BM_CTRL				0x080
+#define MTK_WED_TX_BM_CTRL_VLD_GRP_NUM			GENMASK(6, 0)
+#define MTK_WED_TX_BM_CTRL_RSV_GRP_NUM			GENMASK(22, 16)
+#define MTK_WED_TX_BM_CTRL_PAUSE			BIT(28)
+
+#define MTK_WED_TX_BM_BASE				0x084
+
+#define MTK_WED_TX_BM_TKID				0x088
+#define MTK_WED_TX_BM_TKID_START			GENMASK(15, 0)
+#define MTK_WED_TX_BM_TKID_END				GENMASK(31, 16)
+
+#define MTK_WED_TX_BM_BUF_LEN				0x08c
+
+#define MTK_WED_TX_BM_INTF				0x09c
+#define MTK_WED_TX_BM_INTF_TKID				GENMASK(15, 0)
+#define MTK_WED_TX_BM_INTF_TKFIFO_FDEP			GENMASK(23, 16)
+#define MTK_WED_TX_BM_INTF_TKID_VALID			BIT(28)
+#define MTK_WED_TX_BM_INTF_TKID_READ			BIT(29)
+
+#define MTK_WED_TX_BM_DYN_THR				0x0a0
+#define MTK_WED_TX_BM_DYN_THR_LO			GENMASK(6, 0)
+#define MTK_WED_TX_BM_DYN_THR_HI			GENMASK(22, 16)
+
+#define MTK_WED_INT_STATUS				0x200
+#define MTK_WED_INT_MASK				0x204
+
+#define MTK_WED_GLO_CFG					0x208
+#define MTK_WED_GLO_CFG_TX_DMA_EN			BIT(0)
+#define MTK_WED_GLO_CFG_TX_DMA_BUSY			BIT(1)
+#define MTK_WED_GLO_CFG_RX_DMA_EN			BIT(2)
+#define MTK_WED_GLO_CFG_RX_DMA_BUSY			BIT(3)
+#define MTK_WED_GLO_CFG_RX_BT_SIZE			GENMASK(5, 4)
+#define MTK_WED_GLO_CFG_TX_WB_DDONE			BIT(6)
+#define MTK_WED_GLO_CFG_BIG_ENDIAN			BIT(7)
+#define MTK_WED_GLO_CFG_DIS_BT_SIZE_ALIGN		BIT(8)
+#define MTK_WED_GLO_CFG_TX_BT_SIZE_LO			BIT(9)
+#define MTK_WED_GLO_CFG_MULTI_DMA_EN			GENMASK(11, 10)
+#define MTK_WED_GLO_CFG_FIFO_LITTLE_ENDIAN		BIT(12)
+#define MTK_WED_GLO_CFG_MI_DEPTH_RD			GENMASK(21, 13)
+#define MTK_WED_GLO_CFG_TX_BT_SIZE_HI			GENMASK(23, 22)
+#define MTK_WED_GLO_CFG_SW_RESET			BIT(24)
+#define MTK_WED_GLO_CFG_FIRST_TOKEN_ONLY		BIT(26)
+#define MTK_WED_GLO_CFG_OMIT_RX_INFO			BIT(27)
+#define MTK_WED_GLO_CFG_OMIT_TX_INFO			BIT(28)
+#define MTK_WED_GLO_CFG_BYTE_SWAP			BIT(29)
+#define MTK_WED_GLO_CFG_RX_2B_OFFSET			BIT(31)
+
+#define MTK_WED_RESET_IDX				0x20c
+#define MTK_WED_RESET_IDX_TX				GENMASK(3, 0)
+#define MTK_WED_RESET_IDX_RX				GENMASK(17, 16)
+
+#define MTK_WED_TX_MIB(_n)				(0x2a0 + (_n) * 4)
+
+#define MTK_WED_RING_TX(_n)				(0x300 + (_n) * 0x10)
+
+#define MTK_WED_RING_RX(_n)				(0x400 + (_n) * 0x10)
+
+#define MTK_WED_WPDMA_INT_TRIGGER			0x504
+#define MTK_WED_WPDMA_INT_TRIGGER_RX_DONE		BIT(1)
+#define MTK_WED_WPDMA_INT_TRIGGER_TX_DONE		GENMASK(5, 4)
+
+#define MTK_WED_WPDMA_GLO_CFG				0x508
+#define MTK_WED_WPDMA_GLO_CFG_TX_DRV_EN			BIT(0)
+#define MTK_WED_WPDMA_GLO_CFG_TX_DRV_BUSY		BIT(1)
+#define MTK_WED_WPDMA_GLO_CFG_RX_DRV_EN			BIT(2)
+#define MTK_WED_WPDMA_GLO_CFG_RX_DRV_BUSY		BIT(3)
+#define MTK_WED_WPDMA_GLO_CFG_RX_BT_SIZE		GENMASK(5, 4)
+#define MTK_WED_WPDMA_GLO_CFG_TX_WB_DDONE		BIT(6)
+#define MTK_WED_WPDMA_GLO_CFG_BIG_ENDIAN		BIT(7)
+#define MTK_WED_WPDMA_GLO_CFG_DIS_BT_SIZE_ALIGN		BIT(8)
+#define MTK_WED_WPDMA_GLO_CFG_TX_BT_SIZE_LO		BIT(9)
+#define MTK_WED_WPDMA_GLO_CFG_MULTI_DMA_EN		GENMASK(11, 10)
+#define MTK_WED_WPDMA_GLO_CFG_FIFO_LITTLE_ENDIAN	BIT(12)
+#define MTK_WED_WPDMA_GLO_CFG_MI_DEPTH_RD		GENMASK(21, 13)
+#define MTK_WED_WPDMA_GLO_CFG_TX_BT_SIZE_HI		GENMASK(23, 22)
+#define MTK_WED_WPDMA_GLO_CFG_SW_RESET			BIT(24)
+#define MTK_WED_WPDMA_GLO_CFG_FIRST_TOKEN_ONLY		BIT(26)
+#define MTK_WED_WPDMA_GLO_CFG_OMIT_RX_INFO		BIT(27)
+#define MTK_WED_WPDMA_GLO_CFG_OMIT_TX_INFO		BIT(28)
+#define MTK_WED_WPDMA_GLO_CFG_BYTE_SWAP			BIT(29)
+#define MTK_WED_WPDMA_GLO_CFG_RX_2B_OFFSET		BIT(31)
+
+#define MTK_WED_WPDMA_RESET_IDX				0x50c
+#define MTK_WED_WPDMA_RESET_IDX_TX			GENMASK(3, 0)
+#define MTK_WED_WPDMA_RESET_IDX_RX			GENMASK(17, 16)
+
+#define MTK_WED_WPDMA_INT_CTRL				0x520
+#define MTK_WED_WPDMA_INT_CTRL_SUBRT_ADV		BIT(21)
+
+#define MTK_WED_WPDMA_INT_MASK				0x524
+
+#define MTK_WED_PCIE_CFG_BASE				0x560
+
+#define MTK_WED_PCIE_INT_TRIGGER			0x570
+#define MTK_WED_PCIE_INT_TRIGGER_STATUS			BIT(16)
+
+#define MTK_WED_WPDMA_CFG_BASE				0x580
+
+#define MTK_WED_WPDMA_TX_MIB(_n)			(0x5a0 + (_n) * 4)
+#define MTK_WED_WPDMA_TX_COHERENT_MIB(_n)		(0x5d0 + (_n) * 4)
+
+#define MTK_WED_WPDMA_RING_TX(_n)			(0x600 + (_n) * 0x10)
+#define MTK_WED_WPDMA_RING_RX(_n)			(0x700 + (_n) * 0x10)
+#define MTK_WED_WDMA_RING_RX(_n)			(0x900 + (_n) * 0x10)
+#define MTK_WED_WDMA_RX_THRES(_n)			(0x940 + (_n) * 0x4)
+
+#define MTK_WED_WDMA_GLO_CFG				0xa04
+#define MTK_WED_WDMA_GLO_CFG_TX_DRV_EN			BIT(0)
+#define MTK_WED_WDMA_GLO_CFG_RX_DRV_EN			BIT(2)
+#define MTK_WED_WDMA_GLO_CFG_RX_DRV_BUSY		BIT(3)
+#define MTK_WED_WDMA_GLO_CFG_BT_SIZE			GENMASK(5, 4)
+#define MTK_WED_WDMA_GLO_CFG_TX_WB_DDONE		BIT(6)
+#define MTK_WED_WDMA_GLO_CFG_RX_DIS_FSM_AUTO_IDLE	BIT(13)
+#define MTK_WED_WDMA_GLO_CFG_WCOMPLETE_SEL		BIT(16)
+#define MTK_WED_WDMA_GLO_CFG_INIT_PHASE_RXDMA_BYPASS	BIT(17)
+#define MTK_WED_WDMA_GLO_CFG_INIT_PHASE_BYPASS		BIT(18)
+#define MTK_WED_WDMA_GLO_CFG_FSM_RETURN_IDLE		BIT(19)
+#define MTK_WED_WDMA_GLO_CFG_WAIT_COHERENT		BIT(20)
+#define MTK_WED_WDMA_GLO_CFG_AXI_W_AFTER_AW		BIT(21)
+#define MTK_WED_WDMA_GLO_CFG_IDLE_DMAD_SUPPLY_SINGLE_W	BIT(22)
+#define MTK_WED_WDMA_GLO_CFG_IDLE_DMAD_SUPPLY		BIT(23)
+#define MTK_WED_WDMA_GLO_CFG_DYNAMIC_SKIP_DMAD_PREP	BIT(24)
+#define MTK_WED_WDMA_GLO_CFG_DYNAMIC_DMAD_RECYCLE	BIT(25)
+#define MTK_WED_WDMA_GLO_CFG_RST_INIT_COMPLETE		BIT(26)
+#define MTK_WED_WDMA_GLO_CFG_RXDRV_CLKGATE_BYPASS	BIT(30)
+
+#define MTK_WED_WDMA_RESET_IDX				0xa08
+#define MTK_WED_WDMA_RESET_IDX_RX			GENMASK(17, 16)
+#define MTK_WED_WDMA_RESET_IDX_DRV			GENMASK(25, 24)
+
+#define MTK_WED_WDMA_INT_TRIGGER			0xa28
+#define MTK_WED_WDMA_INT_TRIGGER_RX_DONE		GENMASK(17, 16)
+
+#define MTK_WED_WDMA_INT_CTRL				0xa2c
+#define MTK_WED_WDMA_INT_CTRL_POLL_SRC_SEL		GENMASK(17, 16)
+
+#define MTK_WED_WDMA_OFFSET0				0xaa4
+#define MTK_WED_WDMA_OFFSET1				0xaa8
+
+#define MTK_WED_WDMA_RX_MIB(_n)				(0xae0 + (_n) * 4)
+#define MTK_WED_WDMA_RX_RECYCLE_MIB(_n)			(0xae8 + (_n) * 4)
+#define MTK_WED_WDMA_RX_PROCESSED_MIB(_n)		(0xaf0 + (_n) * 4)
+
+#define MTK_WED_RING_OFS_BASE				0x00
+#define MTK_WED_RING_OFS_COUNT				0x04
+#define MTK_WED_RING_OFS_CPU_IDX			0x08
+#define MTK_WED_RING_OFS_DMA_IDX			0x0c
+
+#define MTK_WDMA_RING_RX(_n)				(0x100 + (_n) * 0x10)
+
+#define MTK_WDMA_GLO_CFG				0x204
+#define MTK_WDMA_GLO_CFG_RX_INFO_PRERES			GENMASK(28, 26)
+
+#define MTK_WDMA_RESET_IDX				0x208
+#define MTK_WDMA_RESET_IDX_TX				GENMASK(3, 0)
+#define MTK_WDMA_RESET_IDX_RX				GENMASK(17, 16)
+
+#define MTK_WDMA_INT_MASK				0x228
+#define MTK_WDMA_INT_MASK_TX_DONE			GENMASK(3, 0)
+#define MTK_WDMA_INT_MASK_RX_DONE			GENMASK(17, 16)
+#define MTK_WDMA_INT_MASK_TX_DELAY			BIT(28)
+#define MTK_WDMA_INT_MASK_TX_COHERENT			BIT(29)
+#define MTK_WDMA_INT_MASK_RX_DELAY			BIT(30)
+#define MTK_WDMA_INT_MASK_RX_COHERENT			BIT(31)
+
+#define MTK_WDMA_INT_GRP1				0x250
+#define MTK_WDMA_INT_GRP2				0x254
+
+#define MTK_PCIE_MIRROR_MAP(n)				((n) ? 0x4 : 0x0)
+#define MTK_PCIE_MIRROR_MAP_EN				BIT(0)
+#define MTK_PCIE_MIRROR_MAP_WED_ID			BIT(1)
+
+/* DMA channel mapping */
+#define HIFSYS_DMA_AG_MAP				0x008
+
+#endif
diff --git a/include/linux/soc/mediatek/mtk_wed.h b/include/linux/soc/mediatek/mtk_wed.h
new file mode 100644
index 000000000000..7e00cca06709
--- /dev/null
+++ b/include/linux/soc/mediatek/mtk_wed.h
@@ -0,0 +1,131 @@
+#ifndef __MTK_WED_H
+#define __MTK_WED_H
+
+#include <linux/kernel.h>
+#include <linux/rcupdate.h>
+#include <linux/regmap.h>
+#include <linux/pci.h>
+
+#define MTK_WED_TX_QUEUES		2
+
+struct mtk_wed_hw;
+struct mtk_wdma_desc;
+
+struct mtk_wed_ring {
+	struct mtk_wdma_desc *desc;
+	dma_addr_t desc_phys;
+	int size;
+
+	u32 reg_base;
+	void __iomem *wpdma;
+};
+
+struct mtk_wed_device {
+#ifdef CONFIG_NET_MEDIATEK_SOC_WED
+	const struct mtk_wed_ops *ops;
+	struct device *dev;
+	struct mtk_wed_hw *hw;
+	bool init_done, running;
+	int wdma_idx;
+	int irq;
+
+	struct mtk_wed_ring tx_ring[MTK_WED_TX_QUEUES];
+	struct mtk_wed_ring txfree_ring;
+	struct mtk_wed_ring tx_wdma[MTK_WED_TX_QUEUES];
+
+	struct {
+		int size;
+		void **pages;
+		struct mtk_wdma_desc *desc;
+		dma_addr_t desc_phys;
+	} buf_ring;
+
+	/* filled by driver: */
+	struct {
+		struct pci_dev *pci_dev;
+
+		u32 wpdma_phys;
+
+		u16 token_start;
+		unsigned int nbuf;
+
+		u32 (*init_buf)(void *ptr, dma_addr_t phys, int token_id);
+		int (*offload_enable)(struct mtk_wed_device *wed);
+		void (*offload_disable)(struct mtk_wed_device *wed);
+	} wlan;
+#endif
+};
+
+struct mtk_wed_ops {
+	int (*attach)(struct mtk_wed_device *dev);
+	int (*tx_ring_setup)(struct mtk_wed_device *dev, int ring,
+			     void __iomem *regs);
+	int (*txfree_ring_setup)(struct mtk_wed_device *dev,
+				 void __iomem *regs);
+	void (*detach)(struct mtk_wed_device *dev);
+
+	void (*stop)(struct mtk_wed_device *dev);
+	void (*start)(struct mtk_wed_device *dev, u32 irq_mask);
+	void (*reset_dma)(struct mtk_wed_device *dev);
+
+	u32 (*reg_read)(struct mtk_wed_device *dev, u32 reg);
+	void (*reg_write)(struct mtk_wed_device *dev, u32 reg, u32 val);
+
+	u32 (*irq_get)(struct mtk_wed_device *dev, u32 mask);
+	void (*irq_set_mask)(struct mtk_wed_device *dev, u32 mask);
+};
+
+extern const struct mtk_wed_ops __rcu *mtk_soc_wed_ops;
+
+static inline int
+mtk_wed_device_attach(struct mtk_wed_device *dev)
+{
+	int ret = -ENODEV;
+
+#ifdef CONFIG_NET_MEDIATEK_SOC_WED
+	rcu_read_lock();
+	dev->ops = rcu_dereference(mtk_soc_wed_ops);
+	if (dev->ops)
+		ret = dev->ops->attach(dev);
+	else
+		rcu_read_unlock();
+
+	if (ret)
+		dev->ops = NULL;
+#endif
+
+	return ret;
+}
+
+#ifdef CONFIG_NET_MEDIATEK_SOC_WED
+#define mtk_wed_device_active(_dev) !!(_dev)->ops
+#define mtk_wed_device_detach(_dev) (_dev)->ops->detach(_dev)
+#define mtk_wed_device_start(_dev, _mask) (_dev)->ops->start(_dev, _mask)
+#define mtk_wed_device_tx_ring_setup(_dev, _ring, _regs) \
+	(_dev)->ops->tx_ring_setup(_dev, _ring, _regs)
+#define mtk_wed_device_txfree_ring_setup(_dev, _regs) \
+	(_dev)->ops->txfree_ring_setup(_dev, _regs)
+#define mtk_wed_device_reg_read(_dev, _reg) \
+	(_dev)->ops->reg_read(_dev, _reg)
+#define mtk_wed_device_reg_write(_dev, _reg, _val) \
+	(_dev)->ops->reg_write(_dev, _reg, _val)
+#define mtk_wed_device_irq_get(_dev, _mask) \
+	(_dev)->ops->irq_get(_dev, _mask)
+#define mtk_wed_device_irq_set_mask(_dev, _mask) \
+	(_dev)->ops->irq_set_mask(_dev, _mask)
+#else
+static inline bool mtk_wed_device_active(struct mtk_wed_device *dev)
+{
+	return false;
+}
+#define mtk_wed_device_detach(_dev) do {} while (0)
+#define mtk_wed_device_start(_dev, _mask) do {} while (0)
+#define mtk_wed_device_tx_ring_setup(_dev, _ring, _regs) -ENODEV
+#define mtk_wed_device_txfree_ring_setup(_dev, _ring, _regs) -ENODEV
+#define mtk_wed_device_reg_read(_dev, _reg) 0
+#define mtk_wed_device_reg_write(_dev, _reg, _val) do {} while (0)
+#define mtk_wed_device_irq_get(_dev, _mask) 0
+#define mtk_wed_device_irq_set_mask(_dev, _mask) do {} while (0)
+#endif
+
+#endif
-- 
2.32.0 (Apple Git-132)

[PATCH 04/11] net: ethernet: mtk_eth_soc: implement flow offloading to WED devices

[Felix Fietkau]

This allows hardware flow offloading from Ethernet to WLAN on MT7622 SoC

Co-developed-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Lorenzo Bianconi <lorenzo@kernel.org>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/ethernet/mediatek/mtk_ppe.c       | 18 ++++++
 drivers/net/ethernet/mediatek/mtk_ppe.h       | 14 +++--
 .../net/ethernet/mediatek/mtk_ppe_offload.c   | 56 ++++++++++++++++++-
 drivers/net/ethernet/mediatek/mtk_wed.h       |  7 +++
 include/linux/netdevice.h                     |  7 +++
 net/core/dev.c                                |  4 ++
 6 files changed, 98 insertions(+), 8 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.c b/drivers/net/ethernet/mediatek/mtk_ppe.c
index 3ad10c793308..472bcd3269a7 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.c
@@ -329,6 +329,24 @@ int mtk_foe_entry_set_pppoe(struct mtk_foe_entry *entry, int sid)
 	return 0;
 }
 
+int mtk_foe_entry_set_wdma(struct mtk_foe_entry *entry, int wdma_idx, int txq,
+			   int bss, int wcid)
+{
+	struct mtk_foe_mac_info *l2 = mtk_foe_entry_l2(entry);
+	u32 *ib2 = mtk_foe_entry_ib2(entry);
+
+	*ib2 &= ~MTK_FOE_IB2_PORT_MG;
+	*ib2 |= MTK_FOE_IB2_WDMA_WINFO;
+	if (wdma_idx)
+		*ib2 |= MTK_FOE_IB2_WDMA_DEVIDX;
+
+	l2->vlan2 = FIELD_PREP(MTK_FOE_VLAN2_WINFO_BSS, bss) |
+		    FIELD_PREP(MTK_FOE_VLAN2_WINFO_WCID, wcid) |
+		    FIELD_PREP(MTK_FOE_VLAN2_WINFO_RING, txq);
+
+	return 0;
+}
+
 static inline bool mtk_foe_entry_usable(struct mtk_foe_entry *entry)
 {
 	return !(entry->ib1 & MTK_FOE_IB1_STATIC) &&
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.h b/drivers/net/ethernet/mediatek/mtk_ppe.h
index 242fb8f2ae65..df8ccaf48171 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.h
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.h
@@ -48,9 +48,9 @@ enum {
 #define MTK_FOE_IB2_DEST_PORT		GENMASK(7, 5)
 #define MTK_FOE_IB2_MULTICAST		BIT(8)
 
-#define MTK_FOE_IB2_WHNAT_QID2		GENMASK(13, 12)
-#define MTK_FOE_IB2_WHNAT_DEVIDX	BIT(16)
-#define MTK_FOE_IB2_WHNAT_NAT		BIT(17)
+#define MTK_FOE_IB2_WDMA_QID2		GENMASK(13, 12)
+#define MTK_FOE_IB2_WDMA_DEVIDX		BIT(16)
+#define MTK_FOE_IB2_WDMA_WINFO		BIT(17)
 
 #define MTK_FOE_IB2_PORT_MG		GENMASK(17, 12)
 
@@ -58,9 +58,9 @@ enum {
 
 #define MTK_FOE_IB2_DSCP		GENMASK(31, 24)
 
-#define MTK_FOE_VLAN2_WHNAT_BSS		GEMMASK(5, 0)
-#define MTK_FOE_VLAN2_WHNAT_WCID	GENMASK(13, 6)
-#define MTK_FOE_VLAN2_WHNAT_RING	GENMASK(15, 14)
+#define MTK_FOE_VLAN2_WINFO_BSS		GENMASK(5, 0)
+#define MTK_FOE_VLAN2_WINFO_WCID	GENMASK(13, 6)
+#define MTK_FOE_VLAN2_WINFO_RING	GENMASK(15, 14)
 
 enum {
 	MTK_FOE_STATE_INVALID,
@@ -281,6 +281,8 @@ int mtk_foe_entry_set_ipv6_tuple(struct mtk_foe_entry *entry,
 int mtk_foe_entry_set_dsa(struct mtk_foe_entry *entry, int port);
 int mtk_foe_entry_set_vlan(struct mtk_foe_entry *entry, int vid);
 int mtk_foe_entry_set_pppoe(struct mtk_foe_entry *entry, int sid);
+int mtk_foe_entry_set_wdma(struct mtk_foe_entry *entry, int wdma_idx, int txq,
+			   int bss, int wcid);
 int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_foe_entry *entry,
 			 u16 timestamp);
 int mtk_ppe_debugfs_init(struct mtk_ppe *ppe);
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
index 7bb1f20002b5..bcf342bb9051 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
@@ -10,6 +10,7 @@
 #include <net/pkt_cls.h>
 #include <net/dsa.h>
 #include "mtk_eth_soc.h"
+#include "mtk_wed.h"
 
 struct mtk_flow_data {
 	struct ethhdr eth;
@@ -39,6 +40,7 @@ struct mtk_flow_entry {
 	struct rhash_head node;
 	unsigned long cookie;
 	u16 hash;
+	s8 wed_index;
 };
 
 static const struct rhashtable_params mtk_flow_ht_params = {
@@ -80,6 +82,35 @@ mtk_flow_offload_mangle_eth(const struct flow_action_entry *act, void *eth)
 	memcpy(dest, src, act->mangle.mask ? 2 : 4);
 }
 
+static int
+mtk_flow_get_wdma_info(struct net_device *dev, const u8 *addr, struct mtk_wdma_info *info)
+{
+	struct net_device_path_ctx ctx = {
+		.dev = dev,
+		.daddr = addr,
+	};
+	struct net_device_path path = {};
+
+	if (!IS_ENABLED(CONFIG_NET_MEDIATEK_SOC_WED))
+		return -1;
+
+	if (!dev->netdev_ops->ndo_fill_forward_path)
+		return -1;
+
+	if (dev->netdev_ops->ndo_fill_forward_path(&ctx, &path))
+		return -1;
+
+	if (path.type != DEV_PATH_MTK_WDMA)
+		return -1;
+
+	info->wdma_idx = path.mtk_wdma.wdma_idx;
+	info->queue = path.mtk_wdma.queue;
+	info->bss = path.mtk_wdma.bss;
+	info->wcid = path.mtk_wdma.wcid;
+
+	return 0;
+}
+
 
 static int
 mtk_flow_mangle_ports(const struct flow_action_entry *act,
@@ -149,10 +180,20 @@ mtk_flow_get_dsa_port(struct net_device **dev)
 
 static int
 mtk_flow_set_output_device(struct mtk_eth *eth, struct mtk_foe_entry *foe,
-			   struct net_device *dev)
+			   struct net_device *dev, const u8 *dest_mac,
+			   int *wed_index)
 {
+	struct mtk_wdma_info info = {};
 	int pse_port, dsa_port;
 
+	if (mtk_flow_get_wdma_info(dev, dest_mac, &info) == 0) {
+		mtk_foe_entry_set_wdma(foe, info.wdma_idx, info.queue, info.bss,
+				       info.wcid);
+		pse_port = 3;
+		*wed_index = info.wdma_idx;
+		goto out;
+	}
+
 	dsa_port = mtk_flow_get_dsa_port(&dev);
 	if (dsa_port >= 0)
 		mtk_foe_entry_set_dsa(foe, dsa_port);
@@ -164,6 +205,7 @@ mtk_flow_set_output_device(struct mtk_eth *eth, struct mtk_foe_entry *foe,
 	else
 		return -EOPNOTSUPP;
 
+out:
 	mtk_foe_entry_set_pse_port(foe, pse_port);
 
 	return 0;
@@ -179,6 +221,7 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 	struct net_device *odev = NULL;
 	struct mtk_flow_entry *entry;
 	int offload_type = 0;
+	int wed_index = -1;
 	u16 addr_type = 0;
 	u32 timestamp;
 	u8 l4proto = 0;
@@ -326,10 +369,14 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 	if (data.pppoe.num == 1)
 		mtk_foe_entry_set_pppoe(&foe, data.pppoe.sid);
 
-	err = mtk_flow_set_output_device(eth, &foe, odev);
+	err = mtk_flow_set_output_device(eth, &foe, odev, data.eth.h_dest,
+					 &wed_index);
 	if (err)
 		return err;
 
+	if (wed_index >= 0 && (err = mtk_wed_flow_add(wed_index)) < 0)
+		return err;
+
 	entry = kzalloc(sizeof(*entry), GFP_KERNEL);
 	if (!entry)
 		return -ENOMEM;
@@ -343,6 +390,7 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 	}
 
 	entry->hash = hash;
+	entry->wed_index = wed_index;
 	err = rhashtable_insert_fast(&eth->flow_table, &entry->node,
 				     mtk_flow_ht_params);
 	if (err < 0)
@@ -353,6 +401,8 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 	mtk_foe_entry_clear(&eth->ppe, hash);
 free:
 	kfree(entry);
+	if (wed_index >= 0)
+	    mtk_wed_flow_remove(wed_index);
 	return err;
 }
 
@@ -369,6 +419,8 @@ mtk_flow_offload_destroy(struct mtk_eth *eth, struct flow_cls_offload *f)
 	mtk_foe_entry_clear(&eth->ppe, entry->hash);
 	rhashtable_remove_fast(&eth->flow_table, &entry->node,
 			       mtk_flow_ht_params);
+	if (entry->wed_index >= 0)
+		mtk_wed_flow_remove(entry->wed_index);
 	kfree(entry);
 
 	return 0;
diff --git a/drivers/net/ethernet/mediatek/mtk_wed.h b/drivers/net/ethernet/mediatek/mtk_wed.h
index 404c9a9b130d..981ec613f4b0 100644
--- a/drivers/net/ethernet/mediatek/mtk_wed.h
+++ b/drivers/net/ethernet/mediatek/mtk_wed.h
@@ -7,6 +7,7 @@
 #include <linux/soc/mediatek/mtk_wed.h>
 #include <linux/debugfs.h>
 #include <linux/regmap.h>
+#include <linux/netdevice.h>
 
 struct mtk_eth;
 
@@ -27,6 +28,12 @@ struct mtk_wed_hw {
 	int index;
 };
 
+struct mtk_wdma_info {
+	u8 wdma_idx;
+	u8 queue;
+	u16 wcid;
+	u8 bss;
+};
 
 #ifdef CONFIG_NET_MEDIATEK_SOC_WED
 static inline void
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index c79ee2296296..2a76b09b5cc9 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -853,6 +853,7 @@ enum net_device_path_type {
 	DEV_PATH_BRIDGE,
 	DEV_PATH_PPPOE,
 	DEV_PATH_DSA,
+	DEV_PATH_MTK_WDMA,
 };
 
 struct net_device_path {
@@ -878,6 +879,12 @@ struct net_device_path {
 			int port;
 			u16 proto;
 		} dsa;
+		struct {
+			u8 wdma_idx;
+			u8 queue;
+			u16 wcid;
+			u8 bss;
+		} mtk_wdma;
 	};
 };
 
diff --git a/net/core/dev.c b/net/core/dev.c
index 2d6771075720..e7b093e835a8 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -703,6 +703,10 @@ int dev_fill_forward_path(const struct net_device *dev, const u8 *daddr,
 		if (WARN_ON_ONCE(last_dev == ctx.dev))
 			return -1;
 	}
+
+	if (!ctx.dev)
+		return ret;
+
 	path = dev_fwd_path(stack);
 	if (!path)
 		return -1;
-- 
2.32.0 (Apple Git-132)

[PATCH 05/11] arm64: dts: mediatek: mt7622: introduce nodes for Wireless Ethernet Dispatch

[Felix Fietkau]

Introduce wed0 and wed1 nodes in order to enable offloading forwarding
between ethernet and wireless devices on the mt7622 chipset.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 arch/arm64/boot/dts/mediatek/mt7622.dtsi | 28 ++++++++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/arch/arm64/boot/dts/mediatek/mt7622.dtsi b/arch/arm64/boot/dts/mediatek/mt7622.dtsi
index a2257ec6d256..47d223e28f8d 100644
--- a/arch/arm64/boot/dts/mediatek/mt7622.dtsi
+++ b/arch/arm64/boot/dts/mediatek/mt7622.dtsi
@@ -901,6 +901,11 @@ sata_port: sata-phy@1a243000 {
 		};
 	};
 
+	hifsys: syscon@1af00000 {
+		compatible = "mediatek,mt7622-hifsys", "syscon";
+		reg = <0 0x1af00000 0 0x70>;
+	};
+
 	ethsys: syscon@1b000000 {
 		compatible = "mediatek,mt7622-ethsys",
 			     "syscon";
@@ -919,6 +924,26 @@ hsdma: dma-controller@1b007000 {
 		#dma-cells = <1>;
 	};
 
+	pcie_mirror: pcie-mirror@10000400 {
+		compatible = "mediatek,mt7622-pcie-mirror",
+			     "syscon";
+		reg = <0 0x10000400 0 0x10>;
+	};
+
+	wed0: wed@1020a000 {
+		compatible = "mediatek,mt7622-wed",
+			     "syscon";
+		reg = <0 0x1020a000 0 0x1000>;
+		interrupts = <GIC_SPI 214 IRQ_TYPE_LEVEL_LOW>;
+	};
+
+	wed1: wed@1020b000 {
+		compatible = "mediatek,mt7622-wed",
+			     "syscon";
+		reg = <0 0x1020b000 0 0x1000>;
+		interrupts = <GIC_SPI 215 IRQ_TYPE_LEVEL_LOW>;
+	};
+
 	eth: ethernet@1b100000 {
 		compatible = "mediatek,mt7622-eth",
 			     "mediatek,mt2701-eth",
@@ -946,6 +971,9 @@ eth: ethernet@1b100000 {
 		mediatek,ethsys = <&ethsys>;
 		mediatek,sgmiisys = <&sgmiisys>;
 		mediatek,cci-control = <&cci_control2>;
+		mediatek,wed = <&wed0>, <&wed1>;
+		mediatek,pcie-mirror = <&pcie_mirror>;
+		mediatek,hifsys = <&hifsys>;
 		dma-coherent;
 		#address-cells = <1>;
 		#size-cells = <0>;
-- 
2.32.0 (Apple Git-132)

[PATCH 06/11] net: ethernet: mtk_eth_soc: add ipv6 flow offload support

[Felix Fietkau]

From: David Bentham <db260179@gmail.com>

Add the missing IPv6 flow offloading support for routing only.
Hardware flow offloading is done by the packet processing engine (PPE)
of the Ethernet MAC and as it doesn't support mangling of IPv6 packets,
IPv6 NAT cannot be supported.

Signed-off-by: David Bentham <db260179@gmail.com>
Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 .../net/ethernet/mediatek/mtk_ppe_offload.c   | 28 +++++++++++++++++++
 1 file changed, 28 insertions(+)

diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
index bcf342bb9051..0113cddcebf4 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
@@ -6,6 +6,7 @@
 #include <linux/if_ether.h>
 #include <linux/rhashtable.h>
 #include <linux/ip.h>
+#include <linux/ipv6.h>
 #include <net/flow_offload.h>
 #include <net/pkt_cls.h>
 #include <net/dsa.h>
@@ -20,6 +21,11 @@ struct mtk_flow_data {
 			__be32 src_addr;
 			__be32 dst_addr;
 		} v4;
+
+		struct {
+			struct in6_addr src_addr;
+			struct in6_addr dst_addr;
+		} v6;
 	};
 
 	__be16 src_port;
@@ -65,6 +71,14 @@ mtk_flow_set_ipv4_addr(struct mtk_foe_entry *foe, struct mtk_flow_data *data,
 					    data->v4.dst_addr, data->dst_port);
 }
 
+static int
+mtk_flow_set_ipv6_addr(struct mtk_foe_entry *foe, struct mtk_flow_data *data)
+{
+	return mtk_foe_entry_set_ipv6_tuple(foe,
+					    data->v6.src_addr.s6_addr32, data->src_port,
+					    data->v6.dst_addr.s6_addr32, data->dst_port);
+}
+
 static void
 mtk_flow_offload_mangle_eth(const struct flow_action_entry *act, void *eth)
 {
@@ -296,6 +310,9 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 	case FLOW_DISSECTOR_KEY_IPV4_ADDRS:
 		offload_type = MTK_PPE_PKT_TYPE_IPV4_HNAPT;
 		break;
+	case FLOW_DISSECTOR_KEY_IPV6_ADDRS:
+		offload_type = MTK_PPE_PKT_TYPE_IPV6_ROUTE_5T;
+		break;
 	default:
 		return -EOPNOTSUPP;
 	}
@@ -331,6 +348,17 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 		mtk_flow_set_ipv4_addr(&foe, &data, false);
 	}
 
+	if (addr_type == FLOW_DISSECTOR_KEY_IPV6_ADDRS) {
+		struct flow_match_ipv6_addrs addrs;
+
+		flow_rule_match_ipv6_addrs(rule, &addrs);
+
+		data.v6.src_addr = addrs.key->src;
+		data.v6.dst_addr = addrs.key->dst;
+
+		mtk_flow_set_ipv6_addr(&foe, &data);
+	}
+
 	flow_action_for_each(i, act, &rule->action) {
 		if (act->id != FLOW_ACTION_MANGLE)
 			continue;
-- 
2.32.0 (Apple Git-132)

[PATCH 07/11] net: ethernet: mtk_eth_soc: support TC_SETUP_BLOCK for PPE offload

[Felix Fietkau]

This allows offload entries to be created from user space

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/ethernet/mediatek/mtk_ppe_offload.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
index 0113cddcebf4..da3bc93676f8 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
@@ -563,10 +563,13 @@ mtk_eth_setup_tc_block(struct net_device *dev, struct flow_block_offload *f)
 int mtk_eth_setup_tc(struct net_device *dev, enum tc_setup_type type,
 		     void *type_data)
 {
-	if (type == TC_SETUP_FT)
+	switch (type) {
+	case TC_SETUP_BLOCK:
+	case TC_SETUP_FT:
 		return mtk_eth_setup_tc_block(dev, type_data);
-
-	return -EOPNOTSUPP;
+	default:
+		return -EOPNOTSUPP;
+	}
 }
 
 int mtk_eth_offload_init(struct mtk_eth *eth)
-- 
2.32.0 (Apple Git-132)

[PATCH 08/11] net: ethernet: mtk_eth_soc: allocate struct mtk_ppe separately

[Felix Fietkau]

Preparation for adding more data to it, which will increase its size.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c     | 11 ++++++-----
 drivers/net/ethernet/mediatek/mtk_eth_soc.h     |  2 +-
 drivers/net/ethernet/mediatek/mtk_ppe.c         | 11 ++++++++---
 drivers/net/ethernet/mediatek/mtk_ppe.h         |  3 +--
 drivers/net/ethernet/mediatek/mtk_ppe_offload.c | 12 ++++++------
 5 files changed, 22 insertions(+), 17 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 4d7c542d89fb..5a2f3cb26e66 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -2269,7 +2269,7 @@ static int mtk_open(struct net_device *dev)
 		if (err)
 			return err;
 
-		if (eth->soc->offload_version && mtk_ppe_start(&eth->ppe) == 0)
+		if (eth->soc->offload_version && mtk_ppe_start(eth->ppe) == 0)
 			gdm_config = MTK_GDMA_TO_PPE;
 
 		mtk_gdm_config(eth, gdm_config);
@@ -2343,7 +2343,7 @@ static int mtk_stop(struct net_device *dev)
 	mtk_dma_free(eth);
 
 	if (eth->soc->offload_version)
-		mtk_ppe_stop(&eth->ppe);
+		mtk_ppe_stop(eth->ppe);
 
 	return 0;
 }
@@ -3262,10 +3262,11 @@ static int mtk_probe(struct platform_device *pdev)
 	}
 
 	if (eth->soc->offload_version) {
-		err = mtk_ppe_init(&eth->ppe, eth->dev,
-				   eth->base + MTK_ETH_PPE_BASE, 2);
-		if (err)
+		eth->ppe = mtk_ppe_init(eth->dev, eth->base + MTK_ETH_PPE_BASE, 2);
+		if (!eth->ppe) {
+			err = -ENOMEM;
 			goto err_free_dev;
+		}
 
 		err = mtk_eth_offload_init(eth);
 		if (err)
diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.h b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
index 74661682fd92..c98c7ee42c6f 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.h
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.h
@@ -985,7 +985,7 @@ struct mtk_eth {
 	u32				rx_dma_l4_valid;
 	int				ip_align;
 
-	struct mtk_ppe			ppe;
+	struct mtk_ppe			*ppe;
 	struct rhashtable		flow_table;
 };
 
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.c b/drivers/net/ethernet/mediatek/mtk_ppe.c
index 472bcd3269a7..c56518272729 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.c
@@ -384,10 +384,15 @@ int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_foe_entry *entry,
 	return hash;
 }
 
-int mtk_ppe_init(struct mtk_ppe *ppe, struct device *dev, void __iomem *base,
+struct mtk_ppe *mtk_ppe_init(struct device *dev, void __iomem *base,
 		 int version)
 {
 	struct mtk_foe_entry *foe;
+	struct mtk_ppe *ppe;
+
+	ppe = devm_kzalloc(dev, sizeof(*ppe), GFP_KERNEL);
+	if (!ppe)
+		return NULL;
 
 	/* need to allocate a separate device, since it PPE DMA access is
 	 * not coherent.
@@ -399,13 +404,13 @@ int mtk_ppe_init(struct mtk_ppe *ppe, struct device *dev, void __iomem *base,
 	foe = dmam_alloc_coherent(ppe->dev, MTK_PPE_ENTRIES * sizeof(*foe),
 				  &ppe->foe_phys, GFP_KERNEL);
 	if (!foe)
-		return -ENOMEM;
+		return NULL;
 
 	ppe->foe_table = foe;
 
 	mtk_ppe_debugfs_init(ppe);
 
-	return 0;
+	return ppe;
 }
 
 static void mtk_ppe_init_foe_table(struct mtk_ppe *ppe)
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.h b/drivers/net/ethernet/mediatek/mtk_ppe.h
index df8ccaf48171..190d2560ac86 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.h
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.h
@@ -246,8 +246,7 @@ struct mtk_ppe {
 	void *acct_table;
 };
 
-int mtk_ppe_init(struct mtk_ppe *ppe, struct device *dev, void __iomem *base,
-		 int version);
+struct mtk_ppe *mtk_ppe_init(struct device *dev, void __iomem *base, int version);
 int mtk_ppe_start(struct mtk_ppe *ppe);
 int mtk_ppe_stop(struct mtk_ppe *ppe);
 
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
index da3bc93676f8..76a94ec85232 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
@@ -411,7 +411,7 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 
 	entry->cookie = f->cookie;
 	timestamp = mtk_eth_timestamp(eth);
-	hash = mtk_foe_entry_commit(&eth->ppe, &foe, timestamp);
+	hash = mtk_foe_entry_commit(eth->ppe, &foe, timestamp);
 	if (hash < 0) {
 		err = hash;
 		goto free;
@@ -426,7 +426,7 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 
 	return 0;
 clear_flow:
-	mtk_foe_entry_clear(&eth->ppe, hash);
+	mtk_foe_entry_clear(eth->ppe, hash);
 free:
 	kfree(entry);
 	if (wed_index >= 0)
@@ -444,7 +444,7 @@ mtk_flow_offload_destroy(struct mtk_eth *eth, struct flow_cls_offload *f)
 	if (!entry)
 		return -ENOENT;
 
-	mtk_foe_entry_clear(&eth->ppe, entry->hash);
+	mtk_foe_entry_clear(eth->ppe, entry->hash);
 	rhashtable_remove_fast(&eth->flow_table, &entry->node,
 			       mtk_flow_ht_params);
 	if (entry->wed_index >= 0)
@@ -466,7 +466,7 @@ mtk_flow_offload_stats(struct mtk_eth *eth, struct flow_cls_offload *f)
 	if (!entry)
 		return -ENOENT;
 
-	timestamp = mtk_foe_entry_timestamp(&eth->ppe, entry->hash);
+	timestamp = mtk_foe_entry_timestamp(eth->ppe, entry->hash);
 	if (timestamp < 0)
 		return -ETIMEDOUT;
 
@@ -522,7 +522,7 @@ mtk_eth_setup_tc_block(struct net_device *dev, struct flow_block_offload *f)
 	struct flow_block_cb *block_cb;
 	flow_setup_cb_t *cb;
 
-	if (!eth->ppe.foe_table)
+	if (!eth->ppe || !eth->ppe->foe_table)
 		return -EOPNOTSUPP;
 
 	if (f->binder_type != FLOW_BLOCK_BINDER_TYPE_CLSACT_INGRESS)
@@ -574,7 +574,7 @@ int mtk_eth_setup_tc(struct net_device *dev, enum tc_setup_type type,
 
 int mtk_eth_offload_init(struct mtk_eth *eth)
 {
-	if (!eth->ppe.foe_table)
+	if (!eth->ppe || !eth->ppe->foe_table)
 		return 0;
 
 	return rhashtable_init(&eth->flow_table, &mtk_flow_ht_params);
-- 
2.32.0 (Apple Git-132)

[PATCH 09/11] net: ethernet: mtk_eth_soc: rework hardware flow table management

[Felix Fietkau]

The hardware was designed to handle flow detection and creation of flow entries
by itself, relying on the software primarily for filling in egress routing
information.
When there is a hash collision between multiple flows, this allows the hardware
to maintain the entry for the most active flow.
Additionally, the hardware only keeps offloading active for entries with at
least 30 packets per second.

With this rework, the code no longer creates a hardware entries directly.
Instead, the hardware entry is only created when the PPE reports a matching
unbound flow with the minimum target rate.
In order to reduce CPU overhead, looking for flows belonging to a hash entry
is rate limited to once every 100ms.

This rework is also used as preparation for emulating bridge offload by
managing L4 offload entries on demand.

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/ethernet/mediatek/mtk_eth_soc.c   |  10 +-
 drivers/net/ethernet/mediatek/mtk_ppe.c       | 134 ++++++++++++++++--
 drivers/net/ethernet/mediatek/mtk_ppe.h       |  38 ++++-
 .../net/ethernet/mediatek/mtk_ppe_offload.c   |  42 ++----
 4 files changed, 170 insertions(+), 54 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_eth_soc.c b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
index 5a2f3cb26e66..209d00f56f62 100644
--- a/drivers/net/ethernet/mediatek/mtk_eth_soc.c
+++ b/drivers/net/ethernet/mediatek/mtk_eth_soc.c
@@ -21,6 +21,7 @@
 #include <linux/pinctrl/devinfo.h>
 #include <linux/phylink.h>
 #include <linux/jhash.h>
+#include <linux/bitfield.h>
 #include <net/dsa.h>
 
 #include "mtk_eth_soc.h"
@@ -1239,7 +1240,7 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget,
 		struct net_device *netdev;
 		unsigned int pktlen;
 		dma_addr_t dma_addr;
-		u32 hash;
+		u32 hash, reason;
 		int mac;
 
 		ring = mtk_get_rx_ring(eth);
@@ -1315,6 +1316,11 @@ static int mtk_poll_rx(struct napi_struct *napi, int budget,
 			skb_set_hash(skb, hash, PKT_HASH_TYPE_L4);
 		}
 
+		reason = FIELD_GET(MTK_RXD4_PPE_CPU_REASON, trxd.rxd4);
+		if (reason == MTK_PPE_CPU_REASON_HIT_UNBIND_RATE_REACHED)
+			mtk_ppe_check_skb(eth->ppe, skb,
+					  trxd.rxd4 & MTK_RXD4_FOE_ENTRY);
+
 		if (netdev->features & NETIF_F_HW_VLAN_CTAG_RX &&
 		    (trxd.rxd2 & RX_DMA_VTAG))
 			__vlan_hwaccel_put_tag(skb, htons(ETH_P_8021Q),
@@ -3262,7 +3268,7 @@ static int mtk_probe(struct platform_device *pdev)
 	}
 
 	if (eth->soc->offload_version) {
-		eth->ppe = mtk_ppe_init(eth->dev, eth->base + MTK_ETH_PPE_BASE, 2);
+		eth->ppe = mtk_ppe_init(eth, eth->base + MTK_ETH_PPE_BASE, 2);
 		if (!eth->ppe) {
 			err = -ENOMEM;
 			goto err_free_dev;
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.c b/drivers/net/ethernet/mediatek/mtk_ppe.c
index c56518272729..a7fe33bbde63 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.c
@@ -6,9 +6,12 @@
 #include <linux/iopoll.h>
 #include <linux/etherdevice.h>
 #include <linux/platform_device.h>
+#include "mtk_eth_soc.h"
 #include "mtk_ppe.h"
 #include "mtk_ppe_regs.h"
 
+static DEFINE_SPINLOCK(ppe_lock);
+
 static void ppe_w32(struct mtk_ppe *ppe, u32 reg, u32 val)
 {
 	writel(val, ppe->base + reg);
@@ -41,6 +44,11 @@ static u32 ppe_clear(struct mtk_ppe *ppe, u32 reg, u32 val)
 	return ppe_m32(ppe, reg, val, 0);
 }
 
+static u32 mtk_eth_timestamp(struct mtk_eth *eth)
+{
+	return mtk_r32(eth, 0x0010) & MTK_FOE_IB1_BIND_TIMESTAMP;
+}
+
 static int mtk_ppe_wait_busy(struct mtk_ppe *ppe)
 {
 	int ret;
@@ -353,26 +361,59 @@ static inline bool mtk_foe_entry_usable(struct mtk_foe_entry *entry)
 	       FIELD_GET(MTK_FOE_IB1_STATE, entry->ib1) != MTK_FOE_STATE_BIND;
 }
 
-int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_foe_entry *entry,
-			 u16 timestamp)
+static bool
+mtk_flow_entry_match(struct mtk_flow_entry *entry, struct mtk_foe_entry *data)
+{
+	int type, len;
+
+	if ((data->ib1 ^ entry->data.ib1) & MTK_FOE_IB1_UDP)
+		return false;
+
+	type = FIELD_GET(MTK_FOE_IB1_PACKET_TYPE, entry->data.ib1);
+	if (type > MTK_PPE_PKT_TYPE_IPV4_DSLITE)
+		len = offsetof(struct mtk_foe_entry, ipv6._rsv);
+	else
+		len = offsetof(struct mtk_foe_entry, ipv4.ib2);
+
+	return !memcmp(&entry->data.data, &data->data, len - 4);
+}
+
+static void
+mtk_flow_entry_update(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
 {
 	struct mtk_foe_entry *hwe;
-	u32 hash;
+	struct mtk_foe_entry foe;
+
+	spin_lock_bh(&ppe_lock);
+	if (entry->hash == 0xffff)
+		goto out;
 
+	hwe = &ppe->foe_table[entry->hash];
+	memcpy(&foe, hwe, sizeof(foe));
+	if (!mtk_flow_entry_match(entry, &foe)) {
+		entry->hash = 0xffff;
+		goto out;
+	}
+
+	entry->data.ib1 = foe.ib1;
+
+out:
+	spin_unlock_bh(&ppe_lock);
+}
+
+static void
+__mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_foe_entry *entry,
+		       u16 hash)
+{
+	struct mtk_foe_entry *hwe;
+	u16 timestamp;
+
+	timestamp = mtk_eth_timestamp(ppe->eth);
 	timestamp &= MTK_FOE_IB1_BIND_TIMESTAMP;
 	entry->ib1 &= ~MTK_FOE_IB1_BIND_TIMESTAMP;
 	entry->ib1 |= FIELD_PREP(MTK_FOE_IB1_BIND_TIMESTAMP, timestamp);
 
-	hash = mtk_ppe_hash_entry(entry);
 	hwe = &ppe->foe_table[hash];
-	if (!mtk_foe_entry_usable(hwe)) {
-		hwe++;
-		hash++;
-
-		if (!mtk_foe_entry_usable(hwe))
-			return -ENOSPC;
-	}
-
 	memcpy(&hwe->data, &entry->data, sizeof(hwe->data));
 	wmb();
 	hwe->ib1 = entry->ib1;
@@ -380,13 +421,77 @@ int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_foe_entry *entry,
 	dma_wmb();
 
 	mtk_ppe_cache_clear(ppe);
+}
 
-	return hash;
+void mtk_foe_entry_clear(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
+{
+	spin_lock_bh(&ppe_lock);
+	hlist_del_init(&entry->list);
+	if (entry->hash != 0xffff) {
+		ppe->foe_table[entry->hash].ib1 &= ~MTK_FOE_IB1_STATE;
+		ppe->foe_table[entry->hash].ib1 |= FIELD_PREP(MTK_FOE_IB1_STATE,
+							      MTK_FOE_STATE_BIND);
+		dma_wmb();
+	}
+	entry->hash = 0xffff;
+	spin_unlock_bh(&ppe_lock);
+}
+
+int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
+{
+	u32 hash = mtk_ppe_hash_entry(&entry->data);
+
+	entry->hash = 0xffff;
+	spin_lock_bh(&ppe_lock);
+	hlist_add_head(&entry->list, &ppe->foe_flow[hash / 2]);
+	spin_unlock_bh(&ppe_lock);
+
+	return 0;
+}
+
+void __mtk_ppe_check_skb(struct mtk_ppe *ppe, struct sk_buff *skb, u16 hash)
+{
+	struct hlist_head *head = &ppe->foe_flow[hash / 2];
+	struct mtk_flow_entry *entry;
+	struct mtk_foe_entry *hwe = &ppe->foe_table[hash];
+	bool found = false;
+
+	if (hlist_empty(head))
+		return;
+
+	spin_lock_bh(&ppe_lock);
+	hlist_for_each_entry(entry, head, list) {
+		if (found || !mtk_flow_entry_match(entry, hwe)) {
+			if (entry->hash != 0xffff)
+				entry->hash = 0xffff;
+			continue;
+		}
+
+		entry->hash = hash;
+		__mtk_foe_entry_commit(ppe, &entry->data, hash);
+		found = true;
+	}
+	spin_unlock_bh(&ppe_lock);
+}
+
+int mtk_foe_entry_idle_time(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
+{
+	u16 now = mtk_eth_timestamp(ppe->eth) & MTK_FOE_IB1_BIND_TIMESTAMP;
+	u16 timestamp;
+
+	mtk_flow_entry_update(ppe, entry);
+	timestamp = entry->data.ib1 & MTK_FOE_IB1_BIND_TIMESTAMP;
+
+	if (timestamp > now)
+		return MTK_FOE_IB1_BIND_TIMESTAMP + 1 - timestamp + now;
+	else
+		return now - timestamp;
 }
 
-struct mtk_ppe *mtk_ppe_init(struct device *dev, void __iomem *base,
+struct mtk_ppe *mtk_ppe_init(struct mtk_eth *eth, void __iomem *base,
 		 int version)
 {
+	struct device *dev = eth->dev;
 	struct mtk_foe_entry *foe;
 	struct mtk_ppe *ppe;
 
@@ -398,6 +503,7 @@ struct mtk_ppe *mtk_ppe_init(struct device *dev, void __iomem *base,
 	 * not coherent.
 	 */
 	ppe->base = base;
+	ppe->eth = eth;
 	ppe->dev = dev;
 	ppe->version = version;
 
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.h b/drivers/net/ethernet/mediatek/mtk_ppe.h
index 190d2560ac86..217b44bed8b6 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.h
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.h
@@ -235,7 +235,17 @@ enum {
 	MTK_PPE_CPU_REASON_INVALID			= 0x1f,
 };
 
+struct mtk_flow_entry {
+	struct rhash_head node;
+	struct hlist_node list;
+	unsigned long cookie;
+	struct mtk_foe_entry data;
+	u16 hash;
+	s8 wed_index;
+};
+
 struct mtk_ppe {
+	struct mtk_eth *eth;
 	struct device *dev;
 	void __iomem *base;
 	int version;
@@ -243,18 +253,33 @@ struct mtk_ppe {
 	struct mtk_foe_entry *foe_table;
 	dma_addr_t foe_phys;
 
+	u16 foe_check_time[MTK_PPE_ENTRIES];
+	struct hlist_head foe_flow[MTK_PPE_ENTRIES / 2];
+
 	void *acct_table;
 };
 
-struct mtk_ppe *mtk_ppe_init(struct device *dev, void __iomem *base, int version);
+struct mtk_ppe *mtk_ppe_init(struct mtk_eth *eth, void __iomem *base, int version);
 int mtk_ppe_start(struct mtk_ppe *ppe);
 int mtk_ppe_stop(struct mtk_ppe *ppe);
 
+void __mtk_ppe_check_skb(struct mtk_ppe *ppe, struct sk_buff *skb, u16 hash);
+
 static inline void
-mtk_foe_entry_clear(struct mtk_ppe *ppe, u16 hash)
+mtk_ppe_check_skb(struct mtk_ppe *ppe, struct sk_buff *skb, u16 hash)
 {
-	ppe->foe_table[hash].ib1 = 0;
-	dma_wmb();
+	u16 now, diff;
+
+	if (!ppe)
+		return;
+
+	now = (u16)jiffies;
+	diff = now - ppe->foe_check_time[hash];
+	if (diff < HZ / 10)
+		return;
+
+	ppe->foe_check_time[hash] = now;
+	__mtk_ppe_check_skb(ppe, skb, hash);
 }
 
 static inline int
@@ -282,8 +307,9 @@ int mtk_foe_entry_set_vlan(struct mtk_foe_entry *entry, int vid);
 int mtk_foe_entry_set_pppoe(struct mtk_foe_entry *entry, int sid);
 int mtk_foe_entry_set_wdma(struct mtk_foe_entry *entry, int wdma_idx, int txq,
 			   int bss, int wcid);
-int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_foe_entry *entry,
-			 u16 timestamp);
+int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_flow_entry *entry);
+void mtk_foe_entry_clear(struct mtk_ppe *ppe, struct mtk_flow_entry *entry);
+int mtk_foe_entry_idle_time(struct mtk_ppe *ppe, struct mtk_flow_entry *entry);
 int mtk_ppe_debugfs_init(struct mtk_ppe *ppe);
 
 #endif
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
index 76a94ec85232..378685fe92b6 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
@@ -42,13 +42,6 @@ struct mtk_flow_data {
 	} pppoe;
 };
 
-struct mtk_flow_entry {
-	struct rhash_head node;
-	unsigned long cookie;
-	u16 hash;
-	s8 wed_index;
-};
-
 static const struct rhashtable_params mtk_flow_ht_params = {
 	.head_offset = offsetof(struct mtk_flow_entry, node),
 	.key_offset = offsetof(struct mtk_flow_entry, cookie),
@@ -56,12 +49,6 @@ static const struct rhashtable_params mtk_flow_ht_params = {
 	.automatic_shrinking = true,
 };
 
-static u32
-mtk_eth_timestamp(struct mtk_eth *eth)
-{
-	return mtk_r32(eth, 0x0010) & MTK_FOE_IB1_BIND_TIMESTAMP;
-}
-
 static int
 mtk_flow_set_ipv4_addr(struct mtk_foe_entry *foe, struct mtk_flow_data *data,
 		       bool egress)
@@ -237,10 +224,8 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 	int offload_type = 0;
 	int wed_index = -1;
 	u16 addr_type = 0;
-	u32 timestamp;
 	u8 l4proto = 0;
 	int err = 0;
-	int hash;
 	int i;
 
 	if (rhashtable_lookup(&eth->flow_table, &f->cookie, mtk_flow_ht_params))
@@ -410,23 +395,21 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 		return -ENOMEM;
 
 	entry->cookie = f->cookie;
-	timestamp = mtk_eth_timestamp(eth);
-	hash = mtk_foe_entry_commit(eth->ppe, &foe, timestamp);
-	if (hash < 0) {
-		err = hash;
+	memcpy(&entry->data, &foe, sizeof(entry->data));
+	entry->wed_index = wed_index;
+
+	if (mtk_foe_entry_commit(eth->ppe, entry) < 0)
 		goto free;
-	}
 
-	entry->hash = hash;
-	entry->wed_index = wed_index;
 	err = rhashtable_insert_fast(&eth->flow_table, &entry->node,
 				     mtk_flow_ht_params);
 	if (err < 0)
-		goto clear_flow;
+		goto clear;
 
 	return 0;
-clear_flow:
-	mtk_foe_entry_clear(eth->ppe, hash);
+
+clear:
+	mtk_foe_entry_clear(eth->ppe, entry);
 free:
 	kfree(entry);
 	if (wed_index >= 0)
@@ -444,7 +427,7 @@ mtk_flow_offload_destroy(struct mtk_eth *eth, struct flow_cls_offload *f)
 	if (!entry)
 		return -ENOENT;
 
-	mtk_foe_entry_clear(eth->ppe, entry->hash);
+	mtk_foe_entry_clear(eth->ppe, entry);
 	rhashtable_remove_fast(&eth->flow_table, &entry->node,
 			       mtk_flow_ht_params);
 	if (entry->wed_index >= 0)
@@ -458,7 +441,6 @@ static int
 mtk_flow_offload_stats(struct mtk_eth *eth, struct flow_cls_offload *f)
 {
 	struct mtk_flow_entry *entry;
-	int timestamp;
 	u32 idle;
 
 	entry = rhashtable_lookup(&eth->flow_table, &f->cookie,
@@ -466,11 +448,7 @@ mtk_flow_offload_stats(struct mtk_eth *eth, struct flow_cls_offload *f)
 	if (!entry)
 		return -ENOENT;
 
-	timestamp = mtk_foe_entry_timestamp(eth->ppe, entry->hash);
-	if (timestamp < 0)
-		return -ETIMEDOUT;
-
-	idle = mtk_eth_timestamp(eth) - timestamp;
+	idle = mtk_foe_entry_idle_time(eth->ppe, entry);
 	f->stats.lastused = jiffies - idle * HZ;
 
 	return 0;
-- 
2.32.0 (Apple Git-132)

[PATCH 10/11] net: ethernet: mtk_eth_soc: remove bridge flow offload type entry support

[Felix Fietkau]

According to MediaTek, this feature is not supported in current hardware

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/ethernet/mediatek/mtk_ppe.c         | 8 --------
 drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c | 1 -
 2 files changed, 9 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.c b/drivers/net/ethernet/mediatek/mtk_ppe.c
index a7fe33bbde63..aa0d190db65d 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.c
@@ -84,13 +84,6 @@ static u32 mtk_ppe_hash_entry(struct mtk_foe_entry *e)
 	u32 hash;
 
 	switch (FIELD_GET(MTK_FOE_IB1_PACKET_TYPE, e->ib1)) {
-		case MTK_PPE_PKT_TYPE_BRIDGE:
-			hv1 = e->bridge.src_mac_lo;
-			hv1 ^= ((e->bridge.src_mac_hi & 0xffff) << 16);
-			hv2 = e->bridge.src_mac_hi >> 16;
-			hv2 ^= e->bridge.dest_mac_lo;
-			hv3 = e->bridge.dest_mac_hi;
-			break;
 		case MTK_PPE_PKT_TYPE_IPV4_ROUTE:
 		case MTK_PPE_PKT_TYPE_IPV4_HNAPT:
 			hv1 = e->ipv4.orig.ports;
@@ -572,7 +565,6 @@ int mtk_ppe_start(struct mtk_ppe *ppe)
 	      MTK_PPE_FLOW_CFG_IP4_NAT |
 	      MTK_PPE_FLOW_CFG_IP4_NAPT |
 	      MTK_PPE_FLOW_CFG_IP4_DSLITE |
-	      MTK_PPE_FLOW_CFG_L2_BRIDGE |
 	      MTK_PPE_FLOW_CFG_IP4_NAT_FRAG;
 	ppe_w32(ppe, MTK_PPE_FLOW_CFG, val);
 
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c b/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c
index d4b482340cb9..eb0b598f14e4 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_debugfs.c
@@ -32,7 +32,6 @@ static const char *mtk_foe_pkt_type_str(int type)
 	static const char * const type_str[] = {
 		[MTK_PPE_PKT_TYPE_IPV4_HNAPT] = "IPv4 5T",
 		[MTK_PPE_PKT_TYPE_IPV4_ROUTE] = "IPv4 3T",
-		[MTK_PPE_PKT_TYPE_BRIDGE] = "L2",
 		[MTK_PPE_PKT_TYPE_IPV4_DSLITE] = "DS-LITE",
 		[MTK_PPE_PKT_TYPE_IPV6_ROUTE_3T] = "IPv6 3T",
 		[MTK_PPE_PKT_TYPE_IPV6_ROUTE_5T] = "IPv6 5T",
-- 
2.32.0 (Apple Git-132)

[PATCH 11/11] net: ethernet: mtk_eth_soc: support creating mac address based offload entries

[Felix Fietkau]

This will be used to implement a limited form of bridge offloading.
Since the hardware does not support flow table entries with just source
and destination MAC address, the driver has to emulate it.

The hardware automatically creates entries entries for incoming flows, even
when they are bridged instead of routed, and reports when packets for these
flows have reached the minimum PPS rate for offloading.

After this happens, we look up the L2 flow offload entry based on the MAC
header and fill in the output routing information in the flow table.
The dynamically created per-flow entries are automatically removed when
either the hardware flowtable entry expires, is replaced, or if the offload
rule they belong to is removed

Signed-off-by: Felix Fietkau <nbd@nbd.name>
---
 drivers/net/ethernet/mediatek/mtk_ppe.c       | 241 ++++++++++++++++--
 drivers/net/ethernet/mediatek/mtk_ppe.h       |  44 +++-
 .../net/ethernet/mediatek/mtk_ppe_offload.c   |  60 ++++-
 3 files changed, 299 insertions(+), 46 deletions(-)

diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.c b/drivers/net/ethernet/mediatek/mtk_ppe.c
index aa0d190db65d..282a1f34a88a 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.c
@@ -6,12 +6,22 @@
 #include <linux/iopoll.h>
 #include <linux/etherdevice.h>
 #include <linux/platform_device.h>
+#include <linux/if_ether.h>
+#include <linux/if_vlan.h>
+#include <net/dsa.h>
 #include "mtk_eth_soc.h"
 #include "mtk_ppe.h"
 #include "mtk_ppe_regs.h"
 
 static DEFINE_SPINLOCK(ppe_lock);
 
+static const struct rhashtable_params mtk_flow_l2_ht_params = {
+	.head_offset = offsetof(struct mtk_flow_entry, l2_node),
+	.key_offset = offsetof(struct mtk_flow_entry, data.bridge),
+	.key_len = offsetof(struct mtk_foe_bridge, key_end),
+	.automatic_shrinking = true,
+};
+
 static void ppe_w32(struct mtk_ppe *ppe, u32 reg, u32 val)
 {
 	writel(val, ppe->base + reg);
@@ -123,6 +133,9 @@ mtk_foe_entry_l2(struct mtk_foe_entry *entry)
 {
 	int type = FIELD_GET(MTK_FOE_IB1_PACKET_TYPE, entry->ib1);
 
+	if (type == MTK_PPE_PKT_TYPE_BRIDGE)
+		return &entry->bridge.l2;
+
 	if (type >= MTK_PPE_PKT_TYPE_IPV4_DSLITE)
 		return &entry->ipv6.l2;
 
@@ -134,6 +147,9 @@ mtk_foe_entry_ib2(struct mtk_foe_entry *entry)
 {
 	int type = FIELD_GET(MTK_FOE_IB1_PACKET_TYPE, entry->ib1);
 
+	if (type == MTK_PPE_PKT_TYPE_BRIDGE)
+		return &entry->bridge.ib2;
+
 	if (type >= MTK_PPE_PKT_TYPE_IPV4_DSLITE)
 		return &entry->ipv6.ib2;
 
@@ -168,7 +184,12 @@ int mtk_foe_entry_prepare(struct mtk_foe_entry *entry, int type, int l4proto,
 	if (type == MTK_PPE_PKT_TYPE_IPV6_ROUTE_3T)
 		entry->ipv6.ports = ports_pad;
 
-	if (type >= MTK_PPE_PKT_TYPE_IPV4_DSLITE) {
+	if (type == MTK_PPE_PKT_TYPE_BRIDGE) {
+		ether_addr_copy(entry->bridge.src_mac, src_mac);
+		ether_addr_copy(entry->bridge.dest_mac, dest_mac);
+		entry->bridge.ib2 = val;
+		l2 = &entry->bridge.l2;
+	} else if (type >= MTK_PPE_PKT_TYPE_IPV4_DSLITE) {
 		entry->ipv6.ib2 = val;
 		l2 = &entry->ipv6.l2;
 	} else {
@@ -371,6 +392,84 @@ mtk_flow_entry_match(struct mtk_flow_entry *entry, struct mtk_foe_entry *data)
 	return !memcmp(&entry->data.data, &data->data, len - 4);
 }
 
+static void
+__mtk_foe_entry_clear(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
+{
+	struct hlist_head *head;
+	struct hlist_node *tmp;
+
+	if (entry->type == MTK_FLOW_TYPE_L2) {
+		rhashtable_remove_fast(&ppe->l2_flows, &entry->l2_node,
+				       mtk_flow_l2_ht_params);
+
+		head = &entry->l2_flows;
+		hlist_for_each_entry_safe(entry, tmp, head, l2_data.list)
+			__mtk_foe_entry_clear(ppe, entry);
+		return;
+	}
+
+	hlist_del_init(&entry->list);
+	if (entry->hash != 0xffff) {
+		ppe->foe_table[entry->hash].ib1 &= ~MTK_FOE_IB1_STATE;
+		ppe->foe_table[entry->hash].ib1 |= FIELD_PREP(MTK_FOE_IB1_STATE,
+							      MTK_FOE_STATE_BIND);
+		dma_wmb();
+	}
+	entry->hash = 0xffff;
+
+	if (entry->type != MTK_FLOW_TYPE_L2_SUBFLOW)
+		return;
+
+	hlist_del_init(&entry->l2_data.list);
+	kfree(entry);
+}
+
+static int __mtk_foe_entry_idle_time(struct mtk_ppe *ppe, u32 ib1)
+{
+	u16 timestamp;
+	u16 now;
+
+	now = mtk_eth_timestamp(ppe->eth) & MTK_FOE_IB1_BIND_TIMESTAMP;
+	timestamp = ib1 & MTK_FOE_IB1_BIND_TIMESTAMP;
+
+	if (timestamp > now)
+		return MTK_FOE_IB1_BIND_TIMESTAMP + 1 - timestamp + now;
+	else
+		return now - timestamp;
+}
+
+static void
+mtk_flow_entry_update_l2(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
+{
+	struct mtk_flow_entry *cur;
+	struct mtk_foe_entry *hwe;
+	struct hlist_node *tmp;
+	int idle;
+
+	idle = __mtk_foe_entry_idle_time(ppe, entry->data.ib1);
+	hlist_for_each_entry_safe(cur, tmp, &entry->l2_flows, l2_data.list) {
+		int cur_idle;
+		u32 ib1;
+
+		hwe = &ppe->foe_table[cur->hash];
+		ib1 = READ_ONCE(hwe->ib1);
+
+		if (FIELD_GET(MTK_FOE_IB1_STATE, ib1) != MTK_FOE_STATE_BIND) {
+			cur->hash = 0xffff;
+			__mtk_foe_entry_clear(ppe, cur);
+			continue;
+		}
+
+		cur_idle = __mtk_foe_entry_idle_time(ppe, ib1);
+		if (cur_idle >= idle)
+			continue;
+
+		idle = cur_idle;
+		entry->data.ib1 &= ~MTK_FOE_IB1_BIND_TIMESTAMP;
+		entry->data.ib1 |= hwe->ib1 & MTK_FOE_IB1_BIND_TIMESTAMP;
+	}
+}
+
 static void
 mtk_flow_entry_update(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
 {
@@ -378,6 +477,12 @@ mtk_flow_entry_update(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
 	struct mtk_foe_entry foe;
 
 	spin_lock_bh(&ppe_lock);
+
+	if (entry->type == MTK_FLOW_TYPE_L2) {
+		mtk_flow_entry_update_l2(ppe, entry);
+		goto out;
+	}
+
 	if (entry->hash == 0xffff)
 		goto out;
 
@@ -419,21 +524,28 @@ __mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_foe_entry *entry,
 void mtk_foe_entry_clear(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
 {
 	spin_lock_bh(&ppe_lock);
-	hlist_del_init(&entry->list);
-	if (entry->hash != 0xffff) {
-		ppe->foe_table[entry->hash].ib1 &= ~MTK_FOE_IB1_STATE;
-		ppe->foe_table[entry->hash].ib1 |= FIELD_PREP(MTK_FOE_IB1_STATE,
-							      MTK_FOE_STATE_BIND);
-		dma_wmb();
-	}
-	entry->hash = 0xffff;
+	__mtk_foe_entry_clear(ppe, entry);
 	spin_unlock_bh(&ppe_lock);
 }
 
+static int
+mtk_foe_entry_commit_l2(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
+{
+	entry->type = MTK_FLOW_TYPE_L2;
+
+	return rhashtable_insert_fast(&ppe->l2_flows, &entry->l2_node,
+				      mtk_flow_l2_ht_params);
+}
+
 int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
 {
-	u32 hash = mtk_ppe_hash_entry(&entry->data);
+	int type = FIELD_GET(MTK_FOE_IB1_PACKET_TYPE, entry->data.ib1);
+	u32 hash;
+
+	if (type == MTK_PPE_PKT_TYPE_BRIDGE)
+		return mtk_foe_entry_commit_l2(ppe, entry);
 
+	hash = mtk_ppe_hash_entry(&entry->data);
 	entry->hash = 0xffff;
 	spin_lock_bh(&ppe_lock);
 	hlist_add_head(&entry->list, &ppe->foe_flow[hash / 2]);
@@ -442,18 +554,72 @@ int mtk_foe_entry_commit(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
 	return 0;
 }
 
+static void
+mtk_foe_entry_commit_subflow(struct mtk_ppe *ppe, struct mtk_flow_entry *entry,
+			     u16 hash)
+{
+	struct mtk_flow_entry *flow_info;
+	struct mtk_foe_entry foe, *hwe;
+	struct mtk_foe_mac_info *l2;
+	u32 ib1_mask = MTK_FOE_IB1_PACKET_TYPE | MTK_FOE_IB1_UDP;
+	int type;
+
+	flow_info = kzalloc(offsetof(struct mtk_flow_entry, l2_data.end),
+			    GFP_ATOMIC);
+	if (!flow_info)
+		return;
+
+	flow_info->l2_data.base_flow = entry;
+	flow_info->type = MTK_FLOW_TYPE_L2_SUBFLOW;
+	flow_info->hash = hash;
+	hlist_add_head(&flow_info->list, &ppe->foe_flow[hash / 2]);
+	hlist_add_head(&flow_info->l2_data.list, &entry->l2_flows);
+
+	hwe = &ppe->foe_table[hash];
+	memcpy(&foe, hwe, sizeof(foe));
+	foe.ib1 &= ib1_mask;
+	foe.ib1 |= entry->data.ib1 & ~ib1_mask;
+
+	l2 = mtk_foe_entry_l2(&foe);
+	memcpy(l2, &entry->data.bridge.l2, sizeof(*l2));
+
+	type = FIELD_GET(MTK_FOE_IB1_PACKET_TYPE, foe.ib1);
+	if (type == MTK_PPE_PKT_TYPE_IPV4_HNAPT)
+		memcpy(&foe.ipv4.new, &foe.ipv4.orig, sizeof(foe.ipv4.new));
+	else if (type >= MTK_PPE_PKT_TYPE_IPV6_ROUTE_3T && l2->etype == ETH_P_IP)
+		l2->etype = ETH_P_IPV6;
+
+	*mtk_foe_entry_ib2(&foe) = entry->data.bridge.ib2;
+
+	__mtk_foe_entry_commit(ppe, &foe, hash);
+}
+
 void __mtk_ppe_check_skb(struct mtk_ppe *ppe, struct sk_buff *skb, u16 hash)
 {
 	struct hlist_head *head = &ppe->foe_flow[hash / 2];
-	struct mtk_flow_entry *entry;
 	struct mtk_foe_entry *hwe = &ppe->foe_table[hash];
+	struct mtk_flow_entry *entry;
+	struct mtk_foe_bridge key = {};
+	struct ethhdr *eh;
 	bool found = false;
-
-	if (hlist_empty(head))
-		return;
+	u8 *tag;
 
 	spin_lock_bh(&ppe_lock);
+
+	if (FIELD_GET(MTK_FOE_IB1_STATE, hwe->ib1) == MTK_FOE_STATE_BIND)
+		goto out;
+
 	hlist_for_each_entry(entry, head, list) {
+		if (entry->type == MTK_FLOW_TYPE_L2_SUBFLOW) {
+			if (unlikely(FIELD_GET(MTK_FOE_IB1_STATE, hwe->ib1) ==
+				     MTK_FOE_STATE_BIND))
+				continue;
+
+			entry->hash = 0xffff;
+			__mtk_foe_entry_clear(ppe, entry);
+			continue;
+		}
+
 		if (found || !mtk_flow_entry_match(entry, hwe)) {
 			if (entry->hash != 0xffff)
 				entry->hash = 0xffff;
@@ -464,21 +630,50 @@ void __mtk_ppe_check_skb(struct mtk_ppe *ppe, struct sk_buff *skb, u16 hash)
 		__mtk_foe_entry_commit(ppe, &entry->data, hash);
 		found = true;
 	}
+
+	if (found)
+		goto out;
+
+	eh = eth_hdr(skb);
+	ether_addr_copy(key.dest_mac, eh->h_dest);
+	ether_addr_copy(key.src_mac, eh->h_source);
+	tag = skb->data - 2;
+	key.vlan = 0;
+	switch (skb->protocol) {
+#if IS_ENABLED(CONFIG_NET_DSA)
+	case htons(ETH_P_XDSA):
+		if (!netdev_uses_dsa(skb->dev) ||
+		    skb->dev->dsa_ptr->tag_ops->proto != DSA_TAG_PROTO_MTK)
+			goto out;
+
+		tag += 4;
+		if (get_unaligned_be16(tag) != ETH_P_8021Q)
+			break;
+
+		fallthrough;
+#endif
+	case htons(ETH_P_8021Q):
+		key.vlan = get_unaligned_be16(tag + 2) & VLAN_VID_MASK;
+		break;
+	default:
+		break;
+	}
+
+	entry = rhashtable_lookup_fast(&ppe->l2_flows, &key, mtk_flow_l2_ht_params);
+	if (!entry)
+		goto out;
+
+	mtk_foe_entry_commit_subflow(ppe, entry, hash);
+
+out:
 	spin_unlock_bh(&ppe_lock);
 }
 
 int mtk_foe_entry_idle_time(struct mtk_ppe *ppe, struct mtk_flow_entry *entry)
 {
-	u16 now = mtk_eth_timestamp(ppe->eth) & MTK_FOE_IB1_BIND_TIMESTAMP;
-	u16 timestamp;
-
 	mtk_flow_entry_update(ppe, entry);
-	timestamp = entry->data.ib1 & MTK_FOE_IB1_BIND_TIMESTAMP;
 
-	if (timestamp > now)
-		return MTK_FOE_IB1_BIND_TIMESTAMP + 1 - timestamp + now;
-	else
-		return now - timestamp;
+	return __mtk_foe_entry_idle_time(ppe, entry->data.ib1);
 }
 
 struct mtk_ppe *mtk_ppe_init(struct mtk_eth *eth, void __iomem *base,
@@ -492,6 +687,8 @@ struct mtk_ppe *mtk_ppe_init(struct mtk_eth *eth, void __iomem *base,
 	if (!ppe)
 		return NULL;
 
+	rhashtable_init(&ppe->l2_flows, &mtk_flow_l2_ht_params);
+
 	/* need to allocate a separate device, since it PPE DMA access is
 	 * not coherent.
 	 */
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe.h b/drivers/net/ethernet/mediatek/mtk_ppe.h
index 217b44bed8b6..1f5cf1c9a947 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe.h
+++ b/drivers/net/ethernet/mediatek/mtk_ppe.h
@@ -6,6 +6,7 @@
 
 #include <linux/kernel.h>
 #include <linux/bitfield.h>
+#include <linux/rhashtable.h>
 
 #define MTK_ETH_PPE_BASE		0xc00
 
@@ -84,19 +85,16 @@ struct mtk_foe_mac_info {
 	u16 src_mac_lo;
 };
 
+/* software-only entry type */
 struct mtk_foe_bridge {
-	u32 dest_mac_hi;
-
-	u16 src_mac_lo;
-	u16 dest_mac_lo;
+	u8 dest_mac[ETH_ALEN];
+	u8 src_mac[ETH_ALEN];
+	u16 vlan;
 
-	u32 src_mac_hi;
+	struct {} key_end;
 
 	u32 ib2;
 
-	u32 _rsv[5];
-
-	u32 udf_tsid;
 	struct mtk_foe_mac_info l2;
 };
 
@@ -235,13 +233,33 @@ enum {
 	MTK_PPE_CPU_REASON_INVALID			= 0x1f,
 };
 
+enum {
+	MTK_FLOW_TYPE_L4,
+	MTK_FLOW_TYPE_L2,
+	MTK_FLOW_TYPE_L2_SUBFLOW,
+};
+
 struct mtk_flow_entry {
+	union {
+		struct hlist_node list;
+		struct {
+			struct rhash_head l2_node;
+			struct hlist_head l2_flows;
+		};
+	};
+	u8 type;
+	s8 wed_index;
+	u16 hash;
+	union {
+		struct mtk_foe_entry data;
+		struct {
+			struct mtk_flow_entry *base_flow;
+			struct hlist_node list;
+			struct {} end;
+		} l2_data;
+	};
 	struct rhash_head node;
-	struct hlist_node list;
 	unsigned long cookie;
-	struct mtk_foe_entry data;
-	u16 hash;
-	s8 wed_index;
 };
 
 struct mtk_ppe {
@@ -256,6 +274,8 @@ struct mtk_ppe {
 	u16 foe_check_time[MTK_PPE_ENTRIES];
 	struct hlist_head foe_flow[MTK_PPE_ENTRIES / 2];
 
+	struct rhashtable l2_flows;
+
 	void *acct_table;
 };
 
diff --git a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
index 378685fe92b6..1fe31058b0f2 100644
--- a/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
+++ b/drivers/net/ethernet/mediatek/mtk_ppe_offload.c
@@ -31,6 +31,8 @@ struct mtk_flow_data {
 	__be16 src_port;
 	__be16 dst_port;
 
+	u16 vlan_in;
+
 	struct {
 		u16 id;
 		__be16 proto;
@@ -257,9 +259,45 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 		return -EOPNOTSUPP;
 	}
 
+	switch (addr_type) {
+	case 0:
+		offload_type = MTK_PPE_PKT_TYPE_BRIDGE;
+		if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_ETH_ADDRS)) {
+			struct flow_match_eth_addrs match;
+
+			flow_rule_match_eth_addrs(rule, &match);
+			memcpy(data.eth.h_dest, match.key->dst, ETH_ALEN);
+			memcpy(data.eth.h_source, match.key->src, ETH_ALEN);
+		} else {
+			return -EOPNOTSUPP;
+		}
+
+		if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_VLAN)) {
+			struct flow_match_vlan match;
+
+			flow_rule_match_vlan(rule, &match);
+
+			if (match.key->vlan_tpid != cpu_to_be16(ETH_P_8021Q))
+				return -EOPNOTSUPP;
+
+			data.vlan_in = match.key->vlan_id;
+		}
+		break;
+	case FLOW_DISSECTOR_KEY_IPV4_ADDRS:
+		offload_type = MTK_PPE_PKT_TYPE_IPV4_HNAPT;
+		break;
+	case FLOW_DISSECTOR_KEY_IPV6_ADDRS:
+		offload_type = MTK_PPE_PKT_TYPE_IPV6_ROUTE_5T;
+		break;
+	default:
+		return -EOPNOTSUPP;
+	}
+
 	flow_action_for_each(i, act, &rule->action) {
 		switch (act->id) {
 		case FLOW_ACTION_MANGLE:
+			if (offload_type == MTK_PPE_PKT_TYPE_BRIDGE)
+				return -EOPNOTSUPP;
 			if (act->mangle.htype == FLOW_ACT_MANGLE_HDR_TYPE_ETH)
 				mtk_flow_offload_mangle_eth(act, &data.eth);
 			break;
@@ -291,17 +329,6 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 		}
 	}
 
-	switch (addr_type) {
-	case FLOW_DISSECTOR_KEY_IPV4_ADDRS:
-		offload_type = MTK_PPE_PKT_TYPE_IPV4_HNAPT;
-		break;
-	case FLOW_DISSECTOR_KEY_IPV6_ADDRS:
-		offload_type = MTK_PPE_PKT_TYPE_IPV6_ROUTE_5T;
-		break;
-	default:
-		return -EOPNOTSUPP;
-	}
-
 	if (!is_valid_ether_addr(data.eth.h_source) ||
 	    !is_valid_ether_addr(data.eth.h_dest))
 		return -EINVAL;
@@ -315,10 +342,13 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 	if (flow_rule_match_key(rule, FLOW_DISSECTOR_KEY_PORTS)) {
 		struct flow_match_ports ports;
 
+		if (offload_type == MTK_PPE_PKT_TYPE_BRIDGE)
+			return -EOPNOTSUPP;
+
 		flow_rule_match_ports(rule, &ports);
 		data.src_port = ports.key->src;
 		data.dst_port = ports.key->dst;
-	} else {
+	} else if (offload_type != MTK_PPE_PKT_TYPE_BRIDGE) {
 		return -EOPNOTSUPP;
 	}
 
@@ -348,6 +378,9 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 		if (act->id != FLOW_ACTION_MANGLE)
 			continue;
 
+		if (offload_type == MTK_PPE_PKT_TYPE_BRIDGE)
+			return -EOPNOTSUPP;
+
 		switch (act->mangle.htype) {
 		case FLOW_ACT_MANGLE_HDR_TYPE_TCP:
 		case FLOW_ACT_MANGLE_HDR_TYPE_UDP:
@@ -373,6 +406,9 @@ mtk_flow_offload_replace(struct mtk_eth *eth, struct flow_cls_offload *f)
 			return err;
 	}
 
+	if (offload_type == MTK_PPE_PKT_TYPE_BRIDGE)
+		foe.bridge.vlan = data.vlan_in;
+
 	if (data.vlan.num == 1) {
 		if (data.vlan.proto != htons(ETH_P_8021Q))
 			return -EOPNOTSUPP;
-- 
2.32.0 (Apple Git-132)

Re: [PATCH 02/11] arm64: dts: mediatek: mt7622: add support for coherent DMA

[Felix Fietkau]

On 25.02.22 11:18, Felix Fietkau wrote:
> It improves performance by eliminating the need for a cache flush on rx and tx
> 
> Signed-off-by: Felix Fietkau <nbd@nbd.name>
Sorry, I accidentally left out the dt-bindings patches. They will be 
included in v2.

- Felix

Re: [PATCH 03/11] net: ethernet: mtk_eth_soc: add support for Wireless Ethernet Dispatch (WED)

[Jakub Kicinski]

On Fri, 25 Feb 2022 11:18:02 +0100 Felix Fietkau wrote:
> +	page_list = kcalloc(n_pages, sizeof(*page_list), GFP_KERNEL);
> +	if (!page_list)
> +		return -ENOMEM;
> +
> +	dev->buf_ring.size = ring_size;
> +	dev->buf_ring.pages = page_list;
> +
> +	desc = dma_alloc_coherent(dev->hw->dev, ring_size * sizeof(*desc),
> +				  &desc_phys, GFP_KERNEL);
> +	if (!desc)
> +		return -ENOMEM;
> +
> +	dev->buf_ring.desc = desc;
> +	dev->buf_ring.desc_phys = desc_phys;
> +
> +	for (i = 0, page_idx = 0; i < ring_size; i += MTK_WED_BUF_PER_PAGE) {
> +		dma_addr_t page_phys, buf_phys;
> +		struct page *page;
> +		void *buf;
> +		int s;
> +
> +		page = __dev_alloc_pages(GFP_KERNEL, 0);
> +		if (!page)
> +			return -ENOMEM;

I haven't looked at the code, yet, but this sure looks leaky.

Re: [PATCH 03/11] net: ethernet: mtk_eth_soc: add support for Wireless Ethernet Dispatch (WED)

[Felix Fietkau]

On 26.02.22 07:18, Jakub Kicinski wrote:
> On Fri, 25 Feb 2022 11:18:02 +0100 Felix Fietkau wrote:
>> +	page_list = kcalloc(n_pages, sizeof(*page_list), GFP_KERNEL);
>> +	if (!page_list)
>> +		return -ENOMEM;
>> +
>> +	dev->buf_ring.size = ring_size;
>> +	dev->buf_ring.pages = page_list;
>> +
>> +	desc = dma_alloc_coherent(dev->hw->dev, ring_size * sizeof(*desc),
>> +				  &desc_phys, GFP_KERNEL);
>> +	if (!desc)
>> +		return -ENOMEM;
>> +
>> +	dev->buf_ring.desc = desc;
>> +	dev->buf_ring.desc_phys = desc_phys;
>> +
>> +	for (i = 0, page_idx = 0; i < ring_size; i += MTK_WED_BUF_PER_PAGE) {
>> +		dma_addr_t page_phys, buf_phys;
>> +		struct page *page;
>> +		void *buf;
>> +		int s;
>> +
>> +		page = __dev_alloc_pages(GFP_KERNEL, 0);
>> +		if (!page)
>> +			return -ENOMEM;
> 
> I haven't looked at the code, yet, but this sure looks leaky.
It's not leaky. If the alloc fails, mtk_wed_detach is still called, 
which cleans up even incomplete ring allocations.

- Felix