From mboxrd@z Thu Jan 1 00:00:00 1970 Delivery-date: Wed, 13 Apr 2022 10:24:38 +0200 Received: from metis.ext.pengutronix.de ([2001:67c:670:201:290:27ff:fe1d:cc33]) by lore.white.stw.pengutronix.de with esmtps (TLS1.3) tls TLS_ECDHE_RSA_WITH_AES_256_GCM_SHA384 (Exim 4.94.2) (envelope-from ) id 1neYIp-000W6G-TD for lore@lore.pengutronix.de; Wed, 13 Apr 2022 10:24:38 +0200 Received: from bombadil.infradead.org ([2607:7c80:54:e::133]) by metis.ext.pengutronix.de with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1neYIm-0002VW-0I for lore@pengutronix.de; Wed, 13 Apr 2022 10:24:37 +0200 DKIM-Signature: v=1; a=rsa-sha256; q=dns/txt; c=relaxed/relaxed; d=lists.infradead.org; s=bombadil.20210309; h=Sender: Content-Transfer-Encoding:Content-Type:List-Subscribe:List-Help:List-Post: List-Archive:List-Unsubscribe:List-Id:MIME-Version:References:In-Reply-To: Message-Id:Date:Subject:Cc:To:From:Reply-To:Content-ID:Content-Description: Resent-Date:Resent-From:Resent-Sender:Resent-To:Resent-Cc:Resent-Message-ID: List-Owner; bh=wfx3VhlpV6tRVexLRC6jLHqRkrlmPLYxE1JyM0OJLXQ=; b=l9QtKqeFxtRVq0 D1minieGoN2ulw8nV7RRJp3rPLDSRA+N7WuLQKmFE+J0Aponk3BzHaNv6d7zufZko2p+0gFnXOD+h cJ7VAWL7cYxWxy1EHQr7JV/WsyLnOjO9TLxgYglg14tyZ4CDBA29mnjJ9M2rqc3wyRSTLDtc3/mOe fvUv69q2TQREqlUJLXXeUOgWrZrfzkU3mxfGMDhiC9aSl2FA8gbQLbU0s0+ycorIWjxIYfDUlxien I93uapyBlXvYWHOuldtRbkS6FqHl7bahxzMUAicdPsndF+yNrGYFeVgckYntrShLfGFHC+VuGeT46 KnspZjCxRFdCVO4BGodg==; Received: from localhost ([::1] helo=bombadil.infradead.org) by bombadil.infradead.org with esmtp (Exim 4.94.2 #2 (Red Hat Linux)) id 1neYHH-0000zc-MI; Wed, 13 Apr 2022 08:23:03 +0000 Received: from metis.ext.pengutronix.de ([2001:67c:670:201:290:27ff:fe1d:cc33]) by bombadil.infradead.org with esmtps (Exim 4.94.2 #2 (Red Hat Linux)) id 1neYGS-0000Tl-O7 for barebox@lists.infradead.org; Wed, 13 Apr 2022 08:22:18 +0000 Received: from drehscheibe.grey.stw.pengutronix.de ([2a0a:edc0:0:c01:1d::a2]) by metis.ext.pengutronix.de with esmtps (TLS1.3:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1neYGP-0001an-M6; Wed, 13 Apr 2022 10:22:09 +0200 Received: from [2a0a:edc0:0:1101:1d::ac] (helo=dude04.red.stw.pengutronix.de) by drehscheibe.grey.stw.pengutronix.de with esmtp (Exim 4.94.2) (envelope-from ) id 1neYGQ-002kDE-Bs; Wed, 13 Apr 2022 10:22:08 +0200 Received: from ore by dude04.red.stw.pengutronix.de with local (Exim 4.94.2) (envelope-from ) id 1neYGM-001nkp-Dz; Wed, 13 Apr 2022 10:22:06 +0200 From: Oleksij Rempel To: barebox@lists.infradead.org Cc: Oleksij Rempel Date: Wed, 13 Apr 2022 10:21:55 +0200 Message-Id: <20220413082205.429509-5-o.rempel@pengutronix.de> X-Mailer: git-send-email 2.30.2 In-Reply-To: <20220413082205.429509-1-o.rempel@pengutronix.de> References: <20220413082205.429509-1-o.rempel@pengutronix.de> MIME-Version: 1.0 X-CRM114-Version: 20100106-BlameMichelson ( TRE 0.8.0 (BSD) ) MR-646709E3 X-CRM114-CacheID: sfid-20220413_012213_124684_D8F3AF8C X-CRM114-Status: GOOD ( 26.44 ) X-BeenThere: barebox@lists.infradead.org X-Mailman-Version: 2.1.34 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Sender: "barebox" X-SA-Exim-Connect-IP: 2607:7c80:54:e::133 X-SA-Exim-Mail-From: barebox-bounces+lore=pengutronix.de@lists.infradead.org X-Spam-Checker-Version: SpamAssassin 3.4.2 (2018-09-13) on metis.ext.pengutronix.de X-Spam-Level: X-Spam-Status: No, score=-5.0 required=4.0 tests=AWL,BAYES_00,DKIMWL_WL_HIGH, DKIM_SIGNED,DKIM_VALID,HEADER_FROM_DIFFERENT_DOMAINS, MAILING_LIST_MULTI,RCVD_IN_DNSWL_MED,SPF_HELO_NONE,SPF_NONE, T_SCC_BODY_TEXT_LINE autolearn=unavailable autolearn_force=no version=3.4.2 Subject: [PATCH v5 04/14] net: add DSA framework to support basic switch functionality X-SA-Exim-Version: 4.2.1 (built Wed, 08 May 2019 21:11:16 +0000) X-SA-Exim-Scanned: Yes (on metis.ext.pengutronix.de) Add DSA based port multiplexing functionality for barebox. With this framework we will be able to use different ports of as switch separately. Signed-off-by: Oleksij Rempel --- drivers/net/Kconfig | 8 + drivers/net/Makefile | 1 + drivers/net/dsa.c | 444 +++++++++++++++++++++++++++++++++++++++++++ include/dsa.h | 90 +++++++++ 4 files changed, 543 insertions(+) create mode 100644 drivers/net/dsa.c create mode 100644 include/dsa.h diff --git a/drivers/net/Kconfig b/drivers/net/Kconfig index 65c93bbe84..4a72136b7f 100644 --- a/drivers/net/Kconfig +++ b/drivers/net/Kconfig @@ -300,5 +300,13 @@ config DRIVER_NET_LITEETH source "drivers/net/phy/Kconfig" source "drivers/net/usb/Kconfig" +menuconfig DSA + bool "Distributed Switch Architecture (DSA)" + select PHYLIB + +if DSA + +endif + endmenu diff --git a/drivers/net/Makefile b/drivers/net/Makefile index 40045bab04..406e13f1a0 100644 --- a/drivers/net/Makefile +++ b/drivers/net/Makefile @@ -1,4 +1,5 @@ # SPDX-License-Identifier: GPL-2.0-only +obj-$(CONFIG_DSA) += dsa.o obj-$(CONFIG_PHYLIB) += phy/ obj-$(CONFIG_NET_USB) += usb/ diff --git a/drivers/net/dsa.c b/drivers/net/dsa.c new file mode 100644 index 0000000000..f2420d306f --- /dev/null +++ b/drivers/net/dsa.c @@ -0,0 +1,444 @@ +// SPDX-License-Identifier: GPL-2.0-only + +#include +#include +#include +#include + +u32 dsa_user_ports(struct dsa_switch *ds) +{ + u32 mask = 0; + int i; + + for (i = 0; i < ds->num_ports; i++) { + if (ds->dp[i]) + mask |= BIT(ds->dp[i]->index); + } + + return mask; +} + +static int dsa_slave_phy_read(struct mii_bus *bus, int addr, int reg) +{ + struct dsa_switch *ds = bus->priv; + + if (ds->phys_mii_mask & BIT(addr)) + return ds->ops->phy_read(ds, addr, reg); + + return 0xffff; +} + +static int dsa_slave_phy_write(struct mii_bus *bus, int addr, int reg, u16 val) +{ + struct dsa_switch *ds = bus->priv; + + if (ds->phys_mii_mask & BIT(addr)) + return ds->ops->phy_write(ds, addr, reg, val); + + return 0; +} + +static int dsa_slave_mii_bus_init(struct dsa_switch *ds) +{ + ds->slave_mii_bus = xzalloc(sizeof(*ds->slave_mii_bus)); + ds->slave_mii_bus->priv = (void *)ds; + ds->slave_mii_bus->read = dsa_slave_phy_read; + ds->slave_mii_bus->write = dsa_slave_phy_write; + ds->slave_mii_bus->parent = ds->dev; + ds->slave_mii_bus->phy_mask = ~ds->phys_mii_mask; + + return mdiobus_register(ds->slave_mii_bus); +} + +static int dsa_port_probe(struct eth_device *edev) +{ + struct dsa_port *dp = edev->priv; + struct dsa_switch *ds = dp->ds; + const struct dsa_ops *ops = ds->ops; + phy_interface_t interface; + int ret; + + if (ops->port_probe) { + interface = of_get_phy_mode(dp->dev.device_node); + ret = ops->port_probe(dp, dp->index, interface); + if (ret) + return ret; + } + + return 0; +} + +static void dsa_port_set_ethaddr(struct eth_device *edev) +{ + struct dsa_port *dp = edev->priv; + struct dsa_switch *ds = dp->ds; + + if (is_valid_ether_addr(edev->ethaddr)) + return; + + if (!is_valid_ether_addr(ds->edev_master->ethaddr)) + return; + + eth_set_ethaddr(edev, ds->edev_master->ethaddr); +} + +static int dsa_port_start(struct eth_device *edev) +{ + struct dsa_port *dp = edev->priv; + struct dsa_switch *ds = dp->ds; + const struct dsa_ops *ops = ds->ops; + phy_interface_t interface; + int ret; + + if (dp->enabled) + return -EBUSY; + + interface = of_get_phy_mode(dp->dev.device_node); + + if (ops->port_pre_enable) { + /* In case of RMII interface we need to enable RMII clock + * before talking to the PHY. + */ + ret = ops->port_pre_enable(dp, dp->index, interface); + if (ret) + return ret; + } + + ret = phy_device_connect(edev, ds->slave_mii_bus, dp->index, NULL, 0, + interface); + if (ret) + return ret; + + dsa_port_set_ethaddr(edev); + + if (ops->port_enable) { + ret = ops->port_enable(dp, dp->index, dp->edev.phydev); + if (ret) + return ret; + } + + dp->enabled = true; + + if (!ds->cpu_port_users) { + struct dsa_port *dpc = ds->dp[ds->cpu_port]; + + if (ops->port_enable) { + ret = ops->port_enable(dpc, ds->cpu_port, + ds->cpu_port_fixed_phy); + if (ret) + return ret; + } + + eth_open(ds->edev_master); + } + + ds->cpu_port_users++; + + return 0; +} + +/* Stop the desired port, the CPU port and the master Eth interface */ +static void dsa_port_stop(struct eth_device *edev) +{ + struct dsa_port *dp = edev->priv; + struct dsa_switch *ds = dp->ds; + const struct dsa_ops *ops = ds->ops; + + if (!dp->enabled) + return; + + if (ops->port_disable) + ops->port_disable(dp, dp->index, dp->edev.phydev); + + dp->enabled = false; + ds->cpu_port_users--; + + if (!ds->cpu_port_users) { + struct dsa_port *dpc = ds->dp[ds->cpu_port]; + + if (ops->port_disable) + ops->port_disable(dpc, ds->cpu_port, + ds->cpu_port_fixed_phy); + + eth_close(ds->edev_master); + } +} + +static int dsa_port_send(struct eth_device *edev, void *packet, int length) +{ + struct dsa_port *dp = edev->priv; + struct dsa_switch *ds = dp->ds; + const struct dsa_ops *ops = ds->ops; + void *tx_buf = ds->tx_buf; + size_t full_length, stuff = 0; + int ret; + + if (length < 64) + stuff = 64 - length; + + full_length = length + ds->needed_headroom + ds->needed_tx_tailroom + + stuff; + + if (full_length > DSA_PKTSIZE) + return -ENOMEM; + + memset(tx_buf + full_length - stuff, 0, stuff); + memcpy(tx_buf + ds->needed_headroom, packet, length); + ret = ops->xmit(dp, dp->index, tx_buf, full_length); + if (ret) + return ret; + + return eth_send_raw(ds->edev_master, tx_buf, full_length); +} + +static int dsa_port_recv(struct eth_device *edev) +{ + struct dsa_port *dp = edev->priv; + int length; + + if (!dp->rx_buf_length) + return 0; + + net_receive(edev, dp->rx_buf, dp->rx_buf_length); + length = dp->rx_buf_length; + dp->rx_buf_length = 0; + + return length; +} + +static int dsa_ether_set_ethaddr(struct eth_device *edev, + const unsigned char *adr) +{ + struct dsa_port *dp = edev->priv; + struct dsa_switch *ds = dp->ds; + struct eth_device *edev_master; + + edev_master = ds->edev_master; + + return edev_master->set_ethaddr(edev_master, adr); +} + +static int dsa_ether_get_ethaddr(struct eth_device *edev, unsigned char *adr) +{ + struct dsa_port *dp = edev->priv; + struct dsa_switch *ds = dp->ds; + struct eth_device *edev_master; + + edev_master = ds->edev_master; + + return edev_master->get_ethaddr(edev_master, adr); +} + +static int dsa_switch_register_edev(struct dsa_switch *ds, + struct device_node *dn, int port) +{ + struct eth_device *edev; + struct device_d *dev; + struct dsa_port *dp; + int ret; + + ds->dp[port] = xzalloc(sizeof(*dp)); + + dp = ds->dp[port]; + dev = &dp->dev; + + dev_set_name(dev, "dsa_port"); + dev->id = DEVICE_ID_DYNAMIC; + dev->parent = ds->dev; + dev->device_node = dn; + + ret = register_device(dev); + if (ret) + return ret; + + dp->rx_buf = xmalloc(DSA_PKTSIZE); + dp->ds = ds; + dp->index = port; + + edev = &dp->edev; + edev->priv = dp; + edev->parent = dev; + edev->init = dsa_port_probe; + edev->open = dsa_port_start; + edev->send = dsa_port_send; + edev->recv = dsa_port_recv; + edev->halt = dsa_port_stop; + edev->get_ethaddr = dsa_ether_get_ethaddr; + edev->set_ethaddr = dsa_ether_set_ethaddr; + + return eth_register(edev); +} + +static int dsa_rx_preprocessor(struct eth_device *edev, unsigned char **packet, + int *length) +{ + struct dsa_switch *ds = edev->rx_preprocessor_priv; + const struct dsa_ops *ops = ds->ops; + struct dsa_port *dp; + int ret, port; + + ret = ops->rcv(ds, &port, *packet, *length); + if (ret) + return ret; + + *length -= ds->needed_headroom; + *packet += ds->needed_headroom; + *length -= ds->needed_rx_tailroom; + + if (port > DSA_MAX_PORTS) + return -ERANGE; + + dp = ds->dp[port]; + if (!dp) + return 0; + + if (*length > DSA_PKTSIZE) + return -ENOMEM; + + if (dp->rx_buf_length) + return -EIO; + + memcpy(dp->rx_buf, *packet, *length); + dp->rx_buf_length = *length; + + return -ENOMSG; +} + +static int dsa_switch_register_master(struct dsa_switch *ds, + struct device_node *np, + struct device_node *master, int port) +{ + struct device_node *phy_node; + struct phy_device *phydev; + struct dsa_port *dp; + int ret; + + if (ds->edev_master) { + dev_err(ds->dev, "master was already registered!\n"); + return -EINVAL; + } + + ds->edev_master = of_find_eth_device_by_node(master); + if (!ds->edev_master) { + dev_err(ds->dev, "can't find ethernet master device\n"); + return -ENODEV; + } + + ds->edev_master->rx_preprocessor = dsa_rx_preprocessor; + ds->edev_master->rx_preprocessor_priv = ds; + + ret = dev_set_param(&ds->edev_master->dev, "mode", "disabled"); + if (ret) + dev_warn(ds->dev, "Can't set disable master Ethernet device\n"); + + phy_node = of_get_child_by_name(np, "fixed-link"); + if (!phy_node) + return -ENODEV; + + phydev = of_phy_register_fixed_link(phy_node, ds->edev_master); + if (!phydev) + return -ENODEV; + + phydev->interface = of_get_phy_mode(np); + + ds->dp[port] = xzalloc(sizeof(*dp)); + dp = ds->dp[port]; + dp->ds = ds; + + ds->cpu_port = port; + ds->cpu_port_fixed_phy = phydev; + + return 0; +} + +static int dsa_switch_parse_ports_of(struct dsa_switch *ds, + struct device_node *dn) +{ + struct device_node *ports, *port; + int ret = 0; + u32 reg; + + ports = of_get_child_by_name(dn, "ports"); + if (!ports) { + /* The second possibility is "ethernet-ports" */ + ports = of_get_child_by_name(dn, "ethernet-ports"); + if (!ports) { + dev_err(ds->dev, "no ports child node found\n"); + return -EINVAL; + } + } + + /* At first step, find and register master/CPU interface */ + for_each_available_child_of_node(ports, port) { + struct device_node *master; + + ret = of_property_read_u32(port, "reg", ®); + if (ret) { + dev_err(ds->dev, "No or too many ports are configured\n"); + goto out_put_node; + } + + if (reg >= ds->num_ports) { + dev_err(ds->dev, "port %pOF index %u exceeds num_ports (%u)\n", + port, reg, ds->num_ports); + ret = -EINVAL; + goto out_put_node; + } + + master = of_parse_phandle(port, "ethernet", 0); + if (master) + dsa_switch_register_master(ds, port, master, reg); + } + + /* Now we can register regular switch ports */ + for_each_available_child_of_node(ports, port) { + of_property_read_u32(port, "reg", ®); + + if (of_parse_phandle(port, "ethernet", 0)) + continue; + + ret = dsa_switch_register_edev(ds, port, reg); + if (ret) { + dev_err(ds->dev, "Can't create edev for port %i\n", + reg); + return ret; + } + } + +out_put_node: + return ret; +} + +int dsa_register_switch(struct dsa_switch *ds) +{ + int ret; + + if (!ds->dev) { + pr_err("No dev is set\n"); + return -ENODEV; + } + + if (!ds->dev->device_node) + return -ENODEV; + + if (!ds->num_ports || ds->num_ports > DSA_MAX_PORTS) { + dev_err(ds->dev, "No or too many ports are configured\n"); + return -EINVAL; + } + + ret = dsa_switch_parse_ports_of(ds, ds->dev->device_node); + if (ret) + return ret; + + if (!ds->slave_mii_bus && ds->ops->phy_read) { + ret = dsa_slave_mii_bus_init(ds); + if (ret) + return ret; + } + + ds->tx_buf = dma_alloc(DSA_PKTSIZE); + + return 0; +} +EXPORT_SYMBOL_GPL(dsa_register_switch); + diff --git a/include/dsa.h b/include/dsa.h new file mode 100644 index 0000000000..75a939f2cb --- /dev/null +++ b/include/dsa.h @@ -0,0 +1,90 @@ +/* SPDX-License-Identifier: GPL-2.0-or-later */ +/* + * Copyright 2019-2021 NXP + */ + +#ifndef __DSA_H__ +#define __DSA_H__ + +#include +#include + +/** + * DSA stands for Distributed Switch Architecture and it is infrastructure + * intended to support drivers for Switches that rely on an intermediary + * Ethernet device for I/O. These switches may support cascading allowing + * them to be arranged as a tree. + * DSA is documented in detail in the Linux kernel documentation under + * Documentation/networking/dsa/dsa.txt + * The network layout of such a switch is shown below: + * + * |------| + * | eth0 | <--- master eth device (regular eth driver) + * |------| + * ^ | + * tag added by switch -->| | + * | | + * | |<-- tag added by DSA driver + * | v + * |--------------------------------------| + * | | CPU port | | <-- DSA (switch) device + * | ------------ | (DSA driver) + * | _________ _________ _________ | + * | | port0 | | port1 | ... | portn | | <-- ports as eth devices + * |-+-------+--+-------+-------+-------+-| ('dsa-port' eth driver) + * + */ + +#define DSA_MAX_PORTS 12 +#define DSA_PKTSIZE 1538 + +struct dsa_port; +struct dsa_switch; + +struct dsa_ops { + int (*port_probe)(struct dsa_port *dp, int port, + phy_interface_t phy_mode); + int (*port_pre_enable)(struct dsa_port *dp, int port, + phy_interface_t phy_mode); + int (*port_enable)(struct dsa_port *dp, int port, + struct phy_device *phy); + void (*port_disable)(struct dsa_port *dp, int port, + struct phy_device *phy); + int (*xmit)(struct dsa_port *dp, int port, void *packet, int length); + int (*rcv)(struct dsa_switch *ds, int *portp, void *packet, int length); + + int (*phy_read)(struct dsa_switch *ds, int port, int regnum); + int (*phy_write)(struct dsa_switch *ds, int port, int regnum, u16 val); +}; + +struct dsa_port { + struct device_d dev; + struct dsa_switch *ds; + unsigned int index; + struct eth_device edev; + unsigned char *rx_buf; + size_t rx_buf_length; + bool enabled; +}; + +struct dsa_switch { + struct device_d *dev; + const struct dsa_ops *ops; + size_t num_ports; + u32 cpu_port; + int cpu_port_users; + struct eth_device *edev_master; + struct phy_device *cpu_port_fixed_phy; + struct dsa_port *dp[DSA_MAX_PORTS]; + size_t needed_headroom; + size_t needed_rx_tailroom; + size_t needed_tx_tailroom; + void *tx_buf; + struct mii_bus *slave_mii_bus; + u32 phys_mii_mask; +}; + +int dsa_register_switch(struct dsa_switch *ds); +u32 dsa_user_ports(struct dsa_switch *ds); + +#endif /* __DSA_H__ */ -- 2.30.2 _______________________________________________ barebox mailing list barebox@lists.infradead.org http://lists.infradead.org/mailman/listinfo/barebox