linux/drivers/net/dsa/bcm_sf2.c

1620 lines
41 KiB
C
Raw Normal View History

// SPDX-License-Identifier: GPL-2.0-or-later
/*
* Broadcom Starfighter 2 DSA switch driver
*
* Copyright (C) 2014, Broadcom Corporation
*/
#include <linux/list.h>
#include <linux/module.h>
#include <linux/netdevice.h>
#include <linux/interrupt.h>
#include <linux/platform_device.h>
#include <linux/phy.h>
#include <linux/phy_fixed.h>
#include <linux/phylink.h>
#include <linux/mii.h>
#include <linux/clk.h>
#include <linux/of.h>
#include <linux/of_irq.h>
#include <linux/of_address.h>
#include <linux/of_net.h>
#include <linux/of_mdio.h>
#include <net/dsa.h>
#include <linux/ethtool.h>
#include <linux/if_bridge.h>
#include <linux/brcmphy.h>
#include <linux/etherdevice.h>
#include <linux/platform_data/b53.h>
#include "bcm_sf2.h"
#include "bcm_sf2_regs.h"
#include "b53/b53_priv.h"
#include "b53/b53_regs.h"
static u16 bcm_sf2_reg_rgmii_cntrl(struct bcm_sf2_priv *priv, int port)
{
switch (priv->type) {
case BCM4908_DEVICE_ID:
switch (port) {
case 7:
return REG_RGMII_11_CNTRL;
default:
break;
}
break;
default:
switch (port) {
case 0:
return REG_RGMII_0_CNTRL;
case 1:
return REG_RGMII_1_CNTRL;
case 2:
return REG_RGMII_2_CNTRL;
default:
break;
}
}
WARN_ONCE(1, "Unsupported port %d\n", port);
/* RO fallback reg */
return REG_SWITCH_STATUS;
}
static u16 bcm_sf2_reg_led_base(struct bcm_sf2_priv *priv, int port)
{
switch (port) {
case 0:
return REG_LED_0_CNTRL;
case 1:
return REG_LED_1_CNTRL;
case 2:
return REG_LED_2_CNTRL;
}
switch (priv->type) {
case BCM4908_DEVICE_ID:
switch (port) {
case 3:
return REG_LED_3_CNTRL;
case 7:
return REG_LED_4_CNTRL;
default:
break;
}
break;
default:
break;
}
WARN_ONCE(1, "Unsupported port %d\n", port);
/* RO fallback reg */
return REG_SWITCH_STATUS;
}
/* Return the number of active ports, not counting the IMP (CPU) port */
static unsigned int bcm_sf2_num_active_ports(struct dsa_switch *ds)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
unsigned int port, count = 0;
for (port = 0; port < ds->num_ports; port++) {
if (dsa_is_cpu_port(ds, port))
continue;
if (priv->port_sts[port].enabled)
count++;
}
return count;
}
static void bcm_sf2_recalc_clock(struct dsa_switch *ds)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
unsigned long new_rate;
unsigned int ports_active;
/* Frequenty in Mhz */
static const unsigned long rate_table[] = {
59220000,
60820000,
62500000,
62500000,
};
ports_active = bcm_sf2_num_active_ports(ds);
if (ports_active == 0 || !priv->clk_mdiv)
return;
/* If we overflow our table, just use the recommended operational
* frequency
*/
if (ports_active > ARRAY_SIZE(rate_table))
new_rate = 90000000;
else
new_rate = rate_table[ports_active - 1];
clk_set_rate(priv->clk_mdiv, new_rate);
}
static void bcm_sf2_imp_setup(struct dsa_switch *ds, int port)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
unsigned int i;
u32 reg, offset;
/* Enable the port memories */
reg = core_readl(priv, CORE_MEM_PSM_VDD_CTRL);
reg &= ~P_TXQ_PSM_VDD(port);
core_writel(priv, reg, CORE_MEM_PSM_VDD_CTRL);
/* Enable forwarding */
core_writel(priv, SW_FWDG_EN, CORE_SWMODE);
/* Enable IMP port in dumb mode */
reg = core_readl(priv, CORE_SWITCH_CTRL);
reg |= MII_DUMB_FWDG_EN;
core_writel(priv, reg, CORE_SWITCH_CTRL);
/* Configure Traffic Class to QoS mapping, allow each priority to map
* to a different queue number
*/
reg = core_readl(priv, CORE_PORT_TC2_QOS_MAP_PORT(port));
for (i = 0; i < SF2_NUM_EGRESS_QUEUES; i++)
reg |= i << (PRT_TO_QID_SHIFT * i);
core_writel(priv, reg, CORE_PORT_TC2_QOS_MAP_PORT(port));
b53_brcm_hdr_setup(ds, port);
if (port == 8) {
if (priv->type == BCM4908_DEVICE_ID ||
priv->type == BCM7445_DEVICE_ID)
offset = CORE_STS_OVERRIDE_IMP;
else
offset = CORE_STS_OVERRIDE_IMP2;
/* Force link status for IMP port */
reg = core_readl(priv, offset);
reg |= (MII_SW_OR | LINK_STS);
if (priv->type == BCM4908_DEVICE_ID)
reg |= GMII_SPEED_UP_2G;
else
reg &= ~GMII_SPEED_UP_2G;
core_writel(priv, reg, offset);
/* Enable Broadcast, Multicast, Unicast forwarding to IMP port */
reg = core_readl(priv, CORE_IMP_CTL);
reg |= (RX_BCST_EN | RX_MCST_EN | RX_UCST_EN);
reg &= ~(RX_DIS | TX_DIS);
core_writel(priv, reg, CORE_IMP_CTL);
} else {
reg = core_readl(priv, CORE_G_PCTL_PORT(port));
reg &= ~(RX_DIS | TX_DIS);
core_writel(priv, reg, CORE_G_PCTL_PORT(port));
}
priv->port_sts[port].enabled = true;
}
static void bcm_sf2_gphy_enable_set(struct dsa_switch *ds, bool enable)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
u32 reg;
reg = reg_readl(priv, REG_SPHY_CNTRL);
if (enable) {
reg |= PHY_RESET;
reg &= ~(EXT_PWR_DOWN | IDDQ_BIAS | IDDQ_GLOBAL_PWR | CK25_DIS);
reg_writel(priv, reg, REG_SPHY_CNTRL);
udelay(21);
reg = reg_readl(priv, REG_SPHY_CNTRL);
reg &= ~PHY_RESET;
} else {
reg |= EXT_PWR_DOWN | IDDQ_BIAS | PHY_RESET;
reg_writel(priv, reg, REG_SPHY_CNTRL);
mdelay(1);
reg |= CK25_DIS;
}
reg_writel(priv, reg, REG_SPHY_CNTRL);
/* Use PHY-driven LED signaling */
if (!enable) {
u16 led_ctrl = bcm_sf2_reg_led_base(priv, 0);
if (priv->type == BCM7278_DEVICE_ID ||
priv->type == BCM7445_DEVICE_ID) {
reg = reg_led_readl(priv, led_ctrl, 0);
reg |= LED_CNTRL_SPDLNK_SRC_SEL;
reg_led_writel(priv, reg, led_ctrl, 0);
}
}
}
static inline void bcm_sf2_port_intr_enable(struct bcm_sf2_priv *priv,
int port)
{
unsigned int off;
switch (port) {
case 7:
off = P7_IRQ_OFF;
break;
case 0:
/* Port 0 interrupts are located on the first bank */
intrl2_0_mask_clear(priv, P_IRQ_MASK(P0_IRQ_OFF));
return;
default:
off = P_IRQ_OFF(port);
break;
}
intrl2_1_mask_clear(priv, P_IRQ_MASK(off));
}
static inline void bcm_sf2_port_intr_disable(struct bcm_sf2_priv *priv,
int port)
{
unsigned int off;
switch (port) {
case 7:
off = P7_IRQ_OFF;
break;
case 0:
/* Port 0 interrupts are located on the first bank */
intrl2_0_mask_set(priv, P_IRQ_MASK(P0_IRQ_OFF));
intrl2_0_writel(priv, P_IRQ_MASK(P0_IRQ_OFF), INTRL2_CPU_CLEAR);
return;
default:
off = P_IRQ_OFF(port);
break;
}
intrl2_1_mask_set(priv, P_IRQ_MASK(off));
intrl2_1_writel(priv, P_IRQ_MASK(off), INTRL2_CPU_CLEAR);
}
static int bcm_sf2_port_setup(struct dsa_switch *ds, int port,
struct phy_device *phy)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
unsigned int i;
u32 reg;
if (!dsa_is_user_port(ds, port))
return 0;
priv->port_sts[port].enabled = true;
bcm_sf2_recalc_clock(ds);
/* Clear the memory power down */
reg = core_readl(priv, CORE_MEM_PSM_VDD_CTRL);
reg &= ~P_TXQ_PSM_VDD(port);
core_writel(priv, reg, CORE_MEM_PSM_VDD_CTRL);
/* Enable Broadcom tags for that port if requested */
if (priv->brcm_tag_mask & BIT(port))
b53_brcm_hdr_setup(ds, port);
/* Configure Traffic Class to QoS mapping, allow each priority to map
* to a different queue number
*/
reg = core_readl(priv, CORE_PORT_TC2_QOS_MAP_PORT(port));
for (i = 0; i < SF2_NUM_EGRESS_QUEUES; i++)
reg |= i << (PRT_TO_QID_SHIFT * i);
core_writel(priv, reg, CORE_PORT_TC2_QOS_MAP_PORT(port));
/* Re-enable the GPHY and re-apply workarounds */
if (priv->int_phy_mask & 1 << port && priv->hw_params.num_gphy == 1) {
bcm_sf2_gphy_enable_set(ds, true);
if (phy) {
/* if phy_stop() has been called before, phy
* will be in halted state, and phy_start()
* will call resume.
*
* the resume path does not configure back
* autoneg settings, and since we hard reset
* the phy manually here, we need to reset the
* state machine also.
*/
phy->state = PHY_READY;
phy_init_hw(phy);
}
}
/* Enable MoCA port interrupts to get notified */
if (port == priv->moca_port)
bcm_sf2_port_intr_enable(priv, port);
/* Set per-queue pause threshold to 32 */
core_writel(priv, 32, CORE_TXQ_THD_PAUSE_QN_PORT(port));
/* Set ACB threshold to 24 */
for (i = 0; i < SF2_NUM_EGRESS_QUEUES; i++) {
reg = acb_readl(priv, ACB_QUEUE_CFG(port *
SF2_NUM_EGRESS_QUEUES + i));
reg &= ~XOFF_THRESHOLD_MASK;
reg |= 24;
acb_writel(priv, reg, ACB_QUEUE_CFG(port *
SF2_NUM_EGRESS_QUEUES + i));
}
return b53_enable_port(ds, port, phy);
}
static void bcm_sf2_port_disable(struct dsa_switch *ds, int port)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
u32 reg;
/* Disable learning while in WoL mode */
if (priv->wol_ports_mask & (1 << port)) {
reg = core_readl(priv, CORE_DIS_LEARN);
reg |= BIT(port);
core_writel(priv, reg, CORE_DIS_LEARN);
return;
}
if (port == priv->moca_port)
bcm_sf2_port_intr_disable(priv, port);
if (priv->int_phy_mask & 1 << port && priv->hw_params.num_gphy == 1)
bcm_sf2_gphy_enable_set(ds, false);
b53_disable_port(ds, port);
/* Power down the port memory */
reg = core_readl(priv, CORE_MEM_PSM_VDD_CTRL);
reg |= P_TXQ_PSM_VDD(port);
core_writel(priv, reg, CORE_MEM_PSM_VDD_CTRL);
priv->port_sts[port].enabled = false;
bcm_sf2_recalc_clock(ds);
}
static int bcm_sf2_sw_indir_rw(struct bcm_sf2_priv *priv, int op, int addr,
int regnum, u16 val)
{
int ret = 0;
u32 reg;
reg = reg_readl(priv, REG_SWITCH_CNTRL);
reg |= MDIO_MASTER_SEL;
reg_writel(priv, reg, REG_SWITCH_CNTRL);
/* Page << 8 | offset */
reg = 0x70;
reg <<= 2;
core_writel(priv, addr, reg);
/* Page << 8 | offset */
reg = 0x80 << 8 | regnum << 1;
reg <<= 2;
if (op)
ret = core_readl(priv, reg);
else
core_writel(priv, val, reg);
reg = reg_readl(priv, REG_SWITCH_CNTRL);
reg &= ~MDIO_MASTER_SEL;
reg_writel(priv, reg, REG_SWITCH_CNTRL);
return ret & 0xffff;
}
static int bcm_sf2_sw_mdio_read(struct mii_bus *bus, int addr, int regnum)
{
struct bcm_sf2_priv *priv = bus->priv;
/* Intercept reads from Broadcom pseudo-PHY address, else, send
* them to our master MDIO bus controller
*/
if (addr == BRCM_PSEUDO_PHY_ADDR && priv->indir_phy_mask & BIT(addr))
return bcm_sf2_sw_indir_rw(priv, 1, addr, regnum, 0);
else
return mdiobus_read_nested(priv->master_mii_bus, addr, regnum);
}
static int bcm_sf2_sw_mdio_write(struct mii_bus *bus, int addr, int regnum,
u16 val)
{
struct bcm_sf2_priv *priv = bus->priv;
/* Intercept writes to the Broadcom pseudo-PHY address, else,
* send them to our master MDIO bus controller
*/
if (addr == BRCM_PSEUDO_PHY_ADDR && priv->indir_phy_mask & BIT(addr))
return bcm_sf2_sw_indir_rw(priv, 0, addr, regnum, val);
else
return mdiobus_write_nested(priv->master_mii_bus, addr,
regnum, val);
}
static irqreturn_t bcm_sf2_switch_0_isr(int irq, void *dev_id)
{
struct dsa_switch *ds = dev_id;
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
priv->irq0_stat = intrl2_0_readl(priv, INTRL2_CPU_STATUS) &
~priv->irq0_mask;
intrl2_0_writel(priv, priv->irq0_stat, INTRL2_CPU_CLEAR);
return IRQ_HANDLED;
}
static irqreturn_t bcm_sf2_switch_1_isr(int irq, void *dev_id)
{
struct dsa_switch *ds = dev_id;
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
priv->irq1_stat = intrl2_1_readl(priv, INTRL2_CPU_STATUS) &
~priv->irq1_mask;
intrl2_1_writel(priv, priv->irq1_stat, INTRL2_CPU_CLEAR);
if (priv->irq1_stat & P_LINK_UP_IRQ(P7_IRQ_OFF)) {
priv->port_sts[7].link = true;
dsa_port_phylink_mac_change(ds, 7, true);
}
if (priv->irq1_stat & P_LINK_DOWN_IRQ(P7_IRQ_OFF)) {
priv->port_sts[7].link = false;
dsa_port_phylink_mac_change(ds, 7, false);
}
return IRQ_HANDLED;
}
static int bcm_sf2_sw_rst(struct bcm_sf2_priv *priv)
{
unsigned int timeout = 1000;
u32 reg;
int ret;
/* The watchdog reset does not work on 7278, we need to hit the
* "external" reset line through the reset controller.
*/
if (priv->type == BCM7278_DEVICE_ID) {
ret = reset_control_assert(priv->rcdev);
if (ret)
return ret;
return reset_control_deassert(priv->rcdev);
}
reg = core_readl(priv, CORE_WATCHDOG_CTRL);
reg |= SOFTWARE_RESET | EN_CHIP_RST | EN_SW_RESET;
core_writel(priv, reg, CORE_WATCHDOG_CTRL);
do {
reg = core_readl(priv, CORE_WATCHDOG_CTRL);
if (!(reg & SOFTWARE_RESET))
break;
usleep_range(1000, 2000);
} while (timeout-- > 0);
if (timeout == 0)
return -ETIMEDOUT;
return 0;
}
static void bcm_sf2_crossbar_setup(struct bcm_sf2_priv *priv)
{
struct device *dev = priv->dev->ds->dev;
int shift;
u32 mask;
u32 reg;
int i;
mask = BIT(priv->num_crossbar_int_ports) - 1;
reg = reg_readl(priv, REG_CROSSBAR);
switch (priv->type) {
case BCM4908_DEVICE_ID:
shift = CROSSBAR_BCM4908_INT_P7 * priv->num_crossbar_int_ports;
reg &= ~(mask << shift);
if (0) /* FIXME */
reg |= CROSSBAR_BCM4908_EXT_SERDES << shift;
else if (priv->int_phy_mask & BIT(7))
reg |= CROSSBAR_BCM4908_EXT_GPHY4 << shift;
else if (phy_interface_mode_is_rgmii(priv->port_sts[7].mode))
reg |= CROSSBAR_BCM4908_EXT_RGMII << shift;
else if (WARN(1, "Invalid port mode\n"))
return;
break;
default:
return;
}
reg_writel(priv, reg, REG_CROSSBAR);
reg = reg_readl(priv, REG_CROSSBAR);
for (i = 0; i < priv->num_crossbar_int_ports; i++) {
shift = i * priv->num_crossbar_int_ports;
dev_dbg(dev, "crossbar int port #%d - ext port #%d\n", i,
(reg >> shift) & mask);
}
}
static void bcm_sf2_intr_disable(struct bcm_sf2_priv *priv)
{
intrl2_0_mask_set(priv, 0xffffffff);
intrl2_0_writel(priv, 0xffffffff, INTRL2_CPU_CLEAR);
intrl2_1_mask_set(priv, 0xffffffff);
intrl2_1_writel(priv, 0xffffffff, INTRL2_CPU_CLEAR);
}
static void bcm_sf2_identify_ports(struct bcm_sf2_priv *priv,
struct device_node *dn)
{
struct device *dev = priv->dev->ds->dev;
struct bcm_sf2_port_status *port_st;
struct device_node *port;
unsigned int port_num;
struct property *prop;
int err;
priv->moca_port = -1;
for_each_available_child_of_node(dn, port) {
if (of_property_read_u32(port, "reg", &port_num))
continue;
if (port_num >= DSA_MAX_PORTS) {
dev_err(dev, "Invalid port number %d\n", port_num);
continue;
}
port_st = &priv->port_sts[port_num];
/* Internal PHYs get assigned a specific 'phy-mode' property
* value: "internal" to help flag them before MDIO probing
* has completed, since they might be turned off at that
* time
*/
err = of_get_phy_mode(port, &port_st->mode);
if (err)
continue;
if (port_st->mode == PHY_INTERFACE_MODE_INTERNAL)
priv->int_phy_mask |= 1 << port_num;
if (port_st->mode == PHY_INTERFACE_MODE_MOCA)
priv->moca_port = port_num;
if (of_property_read_bool(port, "brcm,use-bcm-hdr"))
priv->brcm_tag_mask |= 1 << port_num;
/* Ensure that port 5 is not picked up as a DSA CPU port
* flavour but a regular port instead. We should be using
* devlink to be able to set the port flavour.
*/
if (port_num == 5 && priv->type == BCM7278_DEVICE_ID) {
prop = of_find_property(port, "ethernet", NULL);
if (prop)
of_remove_property(port, prop);
}
}
}
static int bcm_sf2_mdio_register(struct dsa_switch *ds)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
net: dsa: bcm_sf2: Ensure that MDIO diversion is used Registering our slave MDIO bus outside of the OF infrastructure is necessary in order to avoid creating double references of the same Device Tree nodes, however it is not sufficient to guarantee that the MDIO bus diversion is used because of_phy_connect() will still resolve to a valid PHY phandle and it will connect to the PHY using its parent MDIO bus which is still the SF2 master MDIO bus. The reason for that is because BCM7445 systems were already shipped with a Device Tree blob looking like this (irrelevant parts omitted for simplicity): ports { #address-cells = <1>; #size-cells = <0>; port@1 { phy-mode = "rgmii-txid"; phy-handle = <&phy0>; reg = <1>; label = "rgmii_1"; }; ... mdio@403c0 { ... phy0: ethernet-phy@0 { broken-turn-around; device_type = "ethernet-phy"; max-speed = <0x3e8>; reg = <0>; compatible = "brcm,bcm53125", "ethernet-phy-ieee802.3-c22"; }; }; There is a hardware issue with chip revisions (Dx) that lead to the development of the following commits: 461cd1b03e32 ("net: dsa: bcm_sf2: Register our slave MDIO bus") 536fab5bf582 ("net: dsa: bcm_sf2: Do not register slave MDIO bus with OF") b8c6cd1d316f ("net: dsa: bcm_sf2: do not use indirect reads and writes for 7445E0") There should have been an internal MDIO bus node created for the chip revision (Dx) that suffers from this problem, but it did not happen back then. Had that happen, that we should have correctly parented phy@0 (bcm53125 below) as child node of the internal MDIO bus, but the production Device Tree blob that was shipped with the firmware targeted the fixed version of the chip, despite both the affected and corrected chips being shipped into production. The problem is that of_phy_connect() for port@1 will happily resolve the 'phy-handle' from the mdio@403c0 node, which bypasses the diversion completely. This results in this double programming that the diversion refers to and aims to avoid. In order to force of_phy_connect() to fail, and have DSA call to dsa_slave_phy_connect(), we must deactivate ethernet-phy@0 from mdio@403c0, and the best way to do that is by removing the phandle property completely. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2020-09-04 21:37:30 +00:00
struct device_node *dn, *child;
struct phy_device *phydev;
struct property *prop;
static int index;
net: dsa: bcm_sf2: Ensure that MDIO diversion is used Registering our slave MDIO bus outside of the OF infrastructure is necessary in order to avoid creating double references of the same Device Tree nodes, however it is not sufficient to guarantee that the MDIO bus diversion is used because of_phy_connect() will still resolve to a valid PHY phandle and it will connect to the PHY using its parent MDIO bus which is still the SF2 master MDIO bus. The reason for that is because BCM7445 systems were already shipped with a Device Tree blob looking like this (irrelevant parts omitted for simplicity): ports { #address-cells = <1>; #size-cells = <0>; port@1 { phy-mode = "rgmii-txid"; phy-handle = <&phy0>; reg = <1>; label = "rgmii_1"; }; ... mdio@403c0 { ... phy0: ethernet-phy@0 { broken-turn-around; device_type = "ethernet-phy"; max-speed = <0x3e8>; reg = <0>; compatible = "brcm,bcm53125", "ethernet-phy-ieee802.3-c22"; }; }; There is a hardware issue with chip revisions (Dx) that lead to the development of the following commits: 461cd1b03e32 ("net: dsa: bcm_sf2: Register our slave MDIO bus") 536fab5bf582 ("net: dsa: bcm_sf2: Do not register slave MDIO bus with OF") b8c6cd1d316f ("net: dsa: bcm_sf2: do not use indirect reads and writes for 7445E0") There should have been an internal MDIO bus node created for the chip revision (Dx) that suffers from this problem, but it did not happen back then. Had that happen, that we should have correctly parented phy@0 (bcm53125 below) as child node of the internal MDIO bus, but the production Device Tree blob that was shipped with the firmware targeted the fixed version of the chip, despite both the affected and corrected chips being shipped into production. The problem is that of_phy_connect() for port@1 will happily resolve the 'phy-handle' from the mdio@403c0 node, which bypasses the diversion completely. This results in this double programming that the diversion refers to and aims to avoid. In order to force of_phy_connect() to fail, and have DSA call to dsa_slave_phy_connect(), we must deactivate ethernet-phy@0 from mdio@403c0, and the best way to do that is by removing the phandle property completely. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2020-09-04 21:37:30 +00:00
int err, reg;
/* Find our integrated MDIO bus node */
dn = of_find_compatible_node(NULL, NULL, "brcm,unimac-mdio");
priv->master_mii_bus = of_mdio_find_bus(dn);
if (!priv->master_mii_bus) {
of_node_put(dn);
return -EPROBE_DEFER;
}
get_device(&priv->master_mii_bus->dev);
priv->master_mii_dn = dn;
net: dsa: bcm_sf2: don't use devres for mdiobus As explained in commits: 74b6d7d13307 ("net: dsa: realtek: register the MDIO bus under devres") 5135e96a3dd2 ("net: dsa: don't allocate the slave_mii_bus using devres") mdiobus_free() will panic when called from devm_mdiobus_free() <- devres_release_all() <- __device_release_driver(), and that mdiobus was not previously unregistered. The Starfighter 2 is a platform device, so the initial set of constraints that I thought would cause this (I2C or SPI buses which call ->remove on ->shutdown) do not apply. But there is one more which applies here. If the DSA master itself is on a bus that calls ->remove from ->shutdown (like dpaa2-eth, which is on the fsl-mc bus), there is a device link between the switch and the DSA master, and device_links_unbind_consumers() will unbind the bcm_sf2 switch driver on shutdown. So the same treatment must be applied to all DSA switch drivers, which is: either use devres for both the mdiobus allocation and registration, or don't use devres at all. The bcm_sf2 driver has the code structure in place for orderly mdiobus removal, so just replace devm_mdiobus_alloc() with the non-devres variant, and add manual free where necessary, to ensure that we don't let devres free a still-registered bus. Fixes: ac3a68d56651 ("net: phy: don't abuse devres in devm_mdiobus_register()") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-02-07 16:15:49 +00:00
priv->slave_mii_bus = mdiobus_alloc();
if (!priv->slave_mii_bus) {
of_node_put(dn);
return -ENOMEM;
}
priv->slave_mii_bus->priv = priv;
priv->slave_mii_bus->name = "sf2 slave mii";
priv->slave_mii_bus->read = bcm_sf2_sw_mdio_read;
priv->slave_mii_bus->write = bcm_sf2_sw_mdio_write;
snprintf(priv->slave_mii_bus->id, MII_BUS_ID_SIZE, "sf2-%d",
index++);
priv->slave_mii_bus->dev.of_node = dn;
/* Include the pseudo-PHY address to divert reads towards our
* workaround. This is only required for 7445D0, since 7445E0
* disconnects the internal switch pseudo-PHY such that we can use the
* regular SWITCH_MDIO master controller instead.
*
* Here we flag the pseudo PHY as needing special treatment and would
* otherwise make all other PHY read/writes go to the master MDIO bus
* controller that comes with this switch backed by the "mdio-unimac"
* driver.
*/
if (of_machine_is_compatible("brcm,bcm7445d0"))
priv->indir_phy_mask |= (1 << BRCM_PSEUDO_PHY_ADDR) | (1 << 0);
else
priv->indir_phy_mask = 0;
ds->phys_mii_mask = priv->indir_phy_mask;
ds->slave_mii_bus = priv->slave_mii_bus;
priv->slave_mii_bus->parent = ds->dev->parent;
priv->slave_mii_bus->phy_mask = ~priv->indir_phy_mask;
net: dsa: bcm_sf2: Ensure that MDIO diversion is used Registering our slave MDIO bus outside of the OF infrastructure is necessary in order to avoid creating double references of the same Device Tree nodes, however it is not sufficient to guarantee that the MDIO bus diversion is used because of_phy_connect() will still resolve to a valid PHY phandle and it will connect to the PHY using its parent MDIO bus which is still the SF2 master MDIO bus. The reason for that is because BCM7445 systems were already shipped with a Device Tree blob looking like this (irrelevant parts omitted for simplicity): ports { #address-cells = <1>; #size-cells = <0>; port@1 { phy-mode = "rgmii-txid"; phy-handle = <&phy0>; reg = <1>; label = "rgmii_1"; }; ... mdio@403c0 { ... phy0: ethernet-phy@0 { broken-turn-around; device_type = "ethernet-phy"; max-speed = <0x3e8>; reg = <0>; compatible = "brcm,bcm53125", "ethernet-phy-ieee802.3-c22"; }; }; There is a hardware issue with chip revisions (Dx) that lead to the development of the following commits: 461cd1b03e32 ("net: dsa: bcm_sf2: Register our slave MDIO bus") 536fab5bf582 ("net: dsa: bcm_sf2: Do not register slave MDIO bus with OF") b8c6cd1d316f ("net: dsa: bcm_sf2: do not use indirect reads and writes for 7445E0") There should have been an internal MDIO bus node created for the chip revision (Dx) that suffers from this problem, but it did not happen back then. Had that happen, that we should have correctly parented phy@0 (bcm53125 below) as child node of the internal MDIO bus, but the production Device Tree blob that was shipped with the firmware targeted the fixed version of the chip, despite both the affected and corrected chips being shipped into production. The problem is that of_phy_connect() for port@1 will happily resolve the 'phy-handle' from the mdio@403c0 node, which bypasses the diversion completely. This results in this double programming that the diversion refers to and aims to avoid. In order to force of_phy_connect() to fail, and have DSA call to dsa_slave_phy_connect(), we must deactivate ethernet-phy@0 from mdio@403c0, and the best way to do that is by removing the phandle property completely. Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2020-09-04 21:37:30 +00:00
/* We need to make sure that of_phy_connect() will not work by
* removing the 'phandle' and 'linux,phandle' properties and
* unregister the existing PHY device that was already registered.
*/
for_each_available_child_of_node(dn, child) {
if (of_property_read_u32(child, "reg", &reg) ||
reg >= PHY_MAX_ADDR)
continue;
if (!(priv->indir_phy_mask & BIT(reg)))
continue;
prop = of_find_property(child, "phandle", NULL);
if (prop)
of_remove_property(child, prop);
prop = of_find_property(child, "linux,phandle", NULL);
if (prop)
of_remove_property(child, prop);
phydev = of_phy_find_device(child);
if (phydev)
phy_device_remove(phydev);
}
err = mdiobus_register(priv->slave_mii_bus);
net: dsa: bcm_sf2: don't use devres for mdiobus As explained in commits: 74b6d7d13307 ("net: dsa: realtek: register the MDIO bus under devres") 5135e96a3dd2 ("net: dsa: don't allocate the slave_mii_bus using devres") mdiobus_free() will panic when called from devm_mdiobus_free() <- devres_release_all() <- __device_release_driver(), and that mdiobus was not previously unregistered. The Starfighter 2 is a platform device, so the initial set of constraints that I thought would cause this (I2C or SPI buses which call ->remove on ->shutdown) do not apply. But there is one more which applies here. If the DSA master itself is on a bus that calls ->remove from ->shutdown (like dpaa2-eth, which is on the fsl-mc bus), there is a device link between the switch and the DSA master, and device_links_unbind_consumers() will unbind the bcm_sf2 switch driver on shutdown. So the same treatment must be applied to all DSA switch drivers, which is: either use devres for both the mdiobus allocation and registration, or don't use devres at all. The bcm_sf2 driver has the code structure in place for orderly mdiobus removal, so just replace devm_mdiobus_alloc() with the non-devres variant, and add manual free where necessary, to ensure that we don't let devres free a still-registered bus. Fixes: ac3a68d56651 ("net: phy: don't abuse devres in devm_mdiobus_register()") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-02-07 16:15:49 +00:00
if (err && dn) {
mdiobus_free(priv->slave_mii_bus);
of_node_put(dn);
net: dsa: bcm_sf2: don't use devres for mdiobus As explained in commits: 74b6d7d13307 ("net: dsa: realtek: register the MDIO bus under devres") 5135e96a3dd2 ("net: dsa: don't allocate the slave_mii_bus using devres") mdiobus_free() will panic when called from devm_mdiobus_free() <- devres_release_all() <- __device_release_driver(), and that mdiobus was not previously unregistered. The Starfighter 2 is a platform device, so the initial set of constraints that I thought would cause this (I2C or SPI buses which call ->remove on ->shutdown) do not apply. But there is one more which applies here. If the DSA master itself is on a bus that calls ->remove from ->shutdown (like dpaa2-eth, which is on the fsl-mc bus), there is a device link between the switch and the DSA master, and device_links_unbind_consumers() will unbind the bcm_sf2 switch driver on shutdown. So the same treatment must be applied to all DSA switch drivers, which is: either use devres for both the mdiobus allocation and registration, or don't use devres at all. The bcm_sf2 driver has the code structure in place for orderly mdiobus removal, so just replace devm_mdiobus_alloc() with the non-devres variant, and add manual free where necessary, to ensure that we don't let devres free a still-registered bus. Fixes: ac3a68d56651 ("net: phy: don't abuse devres in devm_mdiobus_register()") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-02-07 16:15:49 +00:00
}
return err;
}
static void bcm_sf2_mdio_unregister(struct bcm_sf2_priv *priv)
{
mdiobus_unregister(priv->slave_mii_bus);
net: dsa: bcm_sf2: don't use devres for mdiobus As explained in commits: 74b6d7d13307 ("net: dsa: realtek: register the MDIO bus under devres") 5135e96a3dd2 ("net: dsa: don't allocate the slave_mii_bus using devres") mdiobus_free() will panic when called from devm_mdiobus_free() <- devres_release_all() <- __device_release_driver(), and that mdiobus was not previously unregistered. The Starfighter 2 is a platform device, so the initial set of constraints that I thought would cause this (I2C or SPI buses which call ->remove on ->shutdown) do not apply. But there is one more which applies here. If the DSA master itself is on a bus that calls ->remove from ->shutdown (like dpaa2-eth, which is on the fsl-mc bus), there is a device link between the switch and the DSA master, and device_links_unbind_consumers() will unbind the bcm_sf2 switch driver on shutdown. So the same treatment must be applied to all DSA switch drivers, which is: either use devres for both the mdiobus allocation and registration, or don't use devres at all. The bcm_sf2 driver has the code structure in place for orderly mdiobus removal, so just replace devm_mdiobus_alloc() with the non-devres variant, and add manual free where necessary, to ensure that we don't let devres free a still-registered bus. Fixes: ac3a68d56651 ("net: phy: don't abuse devres in devm_mdiobus_register()") Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Reviewed-by: Florian Fainelli <f.fainelli@gmail.com> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
2022-02-07 16:15:49 +00:00
mdiobus_free(priv->slave_mii_bus);
of_node_put(priv->master_mii_dn);
}
static u32 bcm_sf2_sw_get_phy_flags(struct dsa_switch *ds, int port)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
/* The BCM7xxx PHY driver expects to find the integrated PHY revision
* in bits 15:8 and the patch level in bits 7:0 which is exactly what
* the REG_PHY_REVISION register layout is.
*/
if (priv->int_phy_mask & BIT(port))
return priv->hw_params.gphy_rev;
else
return PHY_BRCM_AUTO_PWRDWN_ENABLE |
PHY_BRCM_DIS_TXCRXC_NOENRGY |
PHY_BRCM_IDDQ_SUSPEND;
}
static void bcm_sf2_sw_get_caps(struct dsa_switch *ds, int port,
struct phylink_config *config)
{
unsigned long *interfaces = config->supported_interfaces;
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
if (priv->int_phy_mask & BIT(port)) {
__set_bit(PHY_INTERFACE_MODE_INTERNAL, interfaces);
} else if (priv->moca_port == port) {
__set_bit(PHY_INTERFACE_MODE_MOCA, interfaces);
} else {
__set_bit(PHY_INTERFACE_MODE_MII, interfaces);
__set_bit(PHY_INTERFACE_MODE_REVMII, interfaces);
__set_bit(PHY_INTERFACE_MODE_GMII, interfaces);
phy_interface_set_rgmii(interfaces);
}
config->mac_capabilities = MAC_ASYM_PAUSE | MAC_SYM_PAUSE |
MAC_10 | MAC_100 | MAC_1000;
}
static void bcm_sf2_sw_mac_config(struct dsa_switch *ds, int port,
unsigned int mode,
const struct phylink_link_state *state)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
u32 id_mode_dis = 0, port_mode;
u32 reg_rgmii_ctrl;
u32 reg;
if (port == core_readl(priv, CORE_IMP0_PRT_ID))
return;
switch (state->interface) {
case PHY_INTERFACE_MODE_RGMII:
id_mode_dis = 1;
fallthrough;
case PHY_INTERFACE_MODE_RGMII_TXID:
port_mode = EXT_GPHY;
break;
case PHY_INTERFACE_MODE_MII:
port_mode = EXT_EPHY;
break;
case PHY_INTERFACE_MODE_REVMII:
port_mode = EXT_REVMII;
break;
default:
/* Nothing required for all other PHYs: internal and MoCA */
return;
}
reg_rgmii_ctrl = bcm_sf2_reg_rgmii_cntrl(priv, port);
/* Clear id_mode_dis bit, and the existing port mode, let
* RGMII_MODE_EN bet set by mac_link_{up,down}
*/
reg = reg_readl(priv, reg_rgmii_ctrl);
reg &= ~ID_MODE_DIS;
reg &= ~(PORT_MODE_MASK << PORT_MODE_SHIFT);
reg |= port_mode;
if (id_mode_dis)
reg |= ID_MODE_DIS;
reg_writel(priv, reg, reg_rgmii_ctrl);
}
static void bcm_sf2_sw_mac_link_set(struct dsa_switch *ds, int port,
phy_interface_t interface, bool link)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
u32 reg_rgmii_ctrl;
u32 reg;
if (!phy_interface_mode_is_rgmii(interface) &&
interface != PHY_INTERFACE_MODE_MII &&
interface != PHY_INTERFACE_MODE_REVMII)
return;
reg_rgmii_ctrl = bcm_sf2_reg_rgmii_cntrl(priv, port);
/* If the link is down, just disable the interface to conserve power */
reg = reg_readl(priv, reg_rgmii_ctrl);
if (link)
reg |= RGMII_MODE_EN;
else
reg &= ~RGMII_MODE_EN;
reg_writel(priv, reg, reg_rgmii_ctrl);
}
static void bcm_sf2_sw_mac_link_down(struct dsa_switch *ds, int port,
unsigned int mode,
phy_interface_t interface)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
u32 reg, offset;
if (priv->wol_ports_mask & BIT(port))
return;
if (port != core_readl(priv, CORE_IMP0_PRT_ID)) {
if (priv->type == BCM4908_DEVICE_ID ||
priv->type == BCM7445_DEVICE_ID)
offset = CORE_STS_OVERRIDE_GMIIP_PORT(port);
else
offset = CORE_STS_OVERRIDE_GMIIP2_PORT(port);
reg = core_readl(priv, offset);
reg &= ~LINK_STS;
core_writel(priv, reg, offset);
}
bcm_sf2_sw_mac_link_set(ds, port, interface, false);
}
static void bcm_sf2_sw_mac_link_up(struct dsa_switch *ds, int port,
unsigned int mode,
phy_interface_t interface,
struct phy_device *phydev,
int speed, int duplex,
bool tx_pause, bool rx_pause)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
struct ethtool_eee *p = &priv->dev->ports[port].eee;
bcm_sf2_sw_mac_link_set(ds, port, interface, true);
if (port != core_readl(priv, CORE_IMP0_PRT_ID)) {
u32 reg_rgmii_ctrl = 0;
u32 reg, offset;
if (priv->type == BCM4908_DEVICE_ID ||
priv->type == BCM7445_DEVICE_ID)
offset = CORE_STS_OVERRIDE_GMIIP_PORT(port);
else
offset = CORE_STS_OVERRIDE_GMIIP2_PORT(port);
if (interface == PHY_INTERFACE_MODE_RGMII ||
interface == PHY_INTERFACE_MODE_RGMII_TXID ||
interface == PHY_INTERFACE_MODE_MII ||
interface == PHY_INTERFACE_MODE_REVMII) {
reg_rgmii_ctrl = bcm_sf2_reg_rgmii_cntrl(priv, port);
reg = reg_readl(priv, reg_rgmii_ctrl);
reg &= ~(RX_PAUSE_EN | TX_PAUSE_EN);
if (tx_pause)
reg |= TX_PAUSE_EN;
if (rx_pause)
reg |= RX_PAUSE_EN;
reg_writel(priv, reg, reg_rgmii_ctrl);
}
reg = SW_OVERRIDE | LINK_STS;
switch (speed) {
case SPEED_1000:
reg |= SPDSTS_1000 << SPEED_SHIFT;
break;
case SPEED_100:
reg |= SPDSTS_100 << SPEED_SHIFT;
break;
}
if (duplex == DUPLEX_FULL)
reg |= DUPLX_MODE;
if (tx_pause)
reg |= TXFLOW_CNTL;
if (rx_pause)
reg |= RXFLOW_CNTL;
core_writel(priv, reg, offset);
}
if (mode == MLO_AN_PHY && phydev)
p->eee_enabled = b53_eee_init(ds, port, phydev);
}
static void bcm_sf2_sw_fixed_state(struct dsa_switch *ds, int port,
struct phylink_link_state *status)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
status->link = false;
/* MoCA port is special as we do not get link status from CORE_LNKSTS,
* which means that we need to force the link at the port override
* level to get the data to flow. We do use what the interrupt handler
* did determine before.
*
* For the other ports, we just force the link status, since this is
* a fixed PHY device.
*/
if (port == priv->moca_port) {
status->link = priv->port_sts[port].link;
/* For MoCA interfaces, also force a link down notification
* since some version of the user-space daemon (mocad) use
* cmd->autoneg to force the link, which messes up the PHY
* state machine and make it go in PHY_FORCING state instead.
*/
if (!status->link)
netif_carrier_off(dsa_to_port(ds, port)->slave);
status->duplex = DUPLEX_FULL;
} else {
status->link = true;
}
}
static void bcm_sf2_enable_acb(struct dsa_switch *ds)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
u32 reg;
/* Enable ACB globally */
reg = acb_readl(priv, ACB_CONTROL);
reg |= (ACB_FLUSH_MASK << ACB_FLUSH_SHIFT);
acb_writel(priv, reg, ACB_CONTROL);
reg &= ~(ACB_FLUSH_MASK << ACB_FLUSH_SHIFT);
reg |= ACB_EN | ACB_ALGORITHM;
acb_writel(priv, reg, ACB_CONTROL);
}
static int bcm_sf2_sw_suspend(struct dsa_switch *ds)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
unsigned int port;
bcm_sf2_intr_disable(priv);
/* Disable all ports physically present including the IMP
* port, the other ones have already been disabled during
* bcm_sf2_sw_setup
*/
for (port = 0; port < ds->num_ports; port++) {
if (dsa_is_user_port(ds, port) || dsa_is_cpu_port(ds, port))
bcm_sf2_port_disable(ds, port);
}
if (!priv->wol_ports_mask)
clk_disable_unprepare(priv->clk);
return 0;
}
static int bcm_sf2_sw_resume(struct dsa_switch *ds)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
int ret;
if (!priv->wol_ports_mask)
clk_prepare_enable(priv->clk);
ret = bcm_sf2_sw_rst(priv);
if (ret) {
pr_err("%s: failed to software reset switch\n", __func__);
return ret;
}
bcm_sf2_crossbar_setup(priv);
ret = bcm_sf2_cfp_resume(ds);
if (ret)
return ret;
if (priv->hw_params.num_gphy == 1)
bcm_sf2_gphy_enable_set(ds, true);
ds->ops->setup(ds);
return 0;
}
static void bcm_sf2_sw_get_wol(struct dsa_switch *ds, int port,
struct ethtool_wolinfo *wol)
{
struct net_device *p = dsa_to_port(ds, port)->cpu_dp->master;
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
struct ethtool_wolinfo pwol = { };
/* Get the parent device WoL settings */
if (p->ethtool_ops->get_wol)
p->ethtool_ops->get_wol(p, &pwol);
/* Advertise the parent device supported settings */
wol->supported = pwol.supported;
memset(&wol->sopass, 0, sizeof(wol->sopass));
if (pwol.wolopts & WAKE_MAGICSECURE)
memcpy(&wol->sopass, pwol.sopass, sizeof(wol->sopass));
if (priv->wol_ports_mask & (1 << port))
wol->wolopts = pwol.wolopts;
else
wol->wolopts = 0;
}
static int bcm_sf2_sw_set_wol(struct dsa_switch *ds, int port,
struct ethtool_wolinfo *wol)
{
struct net_device *p = dsa_to_port(ds, port)->cpu_dp->master;
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
s8 cpu_port = dsa_to_port(ds, port)->cpu_dp->index;
struct ethtool_wolinfo pwol = { };
if (p->ethtool_ops->get_wol)
p->ethtool_ops->get_wol(p, &pwol);
if (wol->wolopts & ~pwol.supported)
return -EINVAL;
if (wol->wolopts)
priv->wol_ports_mask |= (1 << port);
else
priv->wol_ports_mask &= ~(1 << port);
/* If we have at least one port enabled, make sure the CPU port
* is also enabled. If the CPU port is the last one enabled, we disable
* it since this configuration does not make sense.
*/
if (priv->wol_ports_mask && priv->wol_ports_mask != (1 << cpu_port))
priv->wol_ports_mask |= (1 << cpu_port);
else
priv->wol_ports_mask &= ~(1 << cpu_port);
return p->ethtool_ops->set_wol(p, wol);
}
static int bcm_sf2_sw_setup(struct dsa_switch *ds)
{
struct bcm_sf2_priv *priv = bcm_sf2_to_priv(ds);
unsigned int port;
/* Enable all valid ports and disable those unused */
for (port = 0; port < priv->hw_params.num_ports; port++) {
/* IMP port receives special treatment */
if (dsa_is_user_port(ds, port))
bcm_sf2_port_setup(ds, port, NULL);
else if (dsa_is_cpu_port(ds, port))
bcm_sf2_imp_setup(ds, port);
else
bcm_sf2_port_disable(ds, port);
}
b53_configure_vlan(ds);
bcm_sf2_enable_acb(ds);
return b53_setup_devlink_resources(ds);
}
static void bcm_sf2_sw_teardown(struct dsa_switch *ds)
{
dsa_devlink_resources_unregister(ds);
}
/* The SWITCH_CORE register space is managed by b53 but operates on a page +
* register basis so we need to translate that into an address that the
* bus-glue understands.
*/
#define SF2_PAGE_REG_MKADDR(page, reg) ((page) << 10 | (reg) << 2)
static int bcm_sf2_core_read8(struct b53_device *dev, u8 page, u8 reg,
u8 *val)
{
struct bcm_sf2_priv *priv = dev->priv;
*val = core_readl(priv, SF2_PAGE_REG_MKADDR(page, reg));
return 0;
}
static int bcm_sf2_core_read16(struct b53_device *dev, u8 page, u8 reg,
u16 *val)
{
struct bcm_sf2_priv *priv = dev->priv;
*val = core_readl(priv, SF2_PAGE_REG_MKADDR(page, reg));
return 0;
}
static int bcm_sf2_core_read32(struct b53_device *dev, u8 page, u8 reg,
u32 *val)
{
struct bcm_sf2_priv *priv = dev->priv;
*val = core_readl(priv, SF2_PAGE_REG_MKADDR(page, reg));
return 0;
}
static int bcm_sf2_core_read64(struct b53_device *dev, u8 page, u8 reg,
u64 *val)
{
struct bcm_sf2_priv *priv = dev->priv;
*val = core_readq(priv, SF2_PAGE_REG_MKADDR(page, reg));
return 0;
}
static int bcm_sf2_core_write8(struct b53_device *dev, u8 page, u8 reg,
u8 value)
{
struct bcm_sf2_priv *priv = dev->priv;
core_writel(priv, value, SF2_PAGE_REG_MKADDR(page, reg));
return 0;
}
static int bcm_sf2_core_write16(struct b53_device *dev, u8 page, u8 reg,
u16 value)
{
struct bcm_sf2_priv *priv = dev->priv;
core_writel(priv, value, SF2_PAGE_REG_MKADDR(page, reg));
return 0;
}
static int bcm_sf2_core_write32(struct b53_device *dev, u8 page, u8 reg,
u32 value)
{
struct bcm_sf2_priv *priv = dev->priv;
core_writel(priv, value, SF2_PAGE_REG_MKADDR(page, reg));
return 0;
}
static int bcm_sf2_core_write64(struct b53_device *dev, u8 page, u8 reg,
u64 value)
{
struct bcm_sf2_priv *priv = dev->priv;
core_writeq(priv, value, SF2_PAGE_REG_MKADDR(page, reg));
return 0;
}
static const struct b53_io_ops bcm_sf2_io_ops = {
.read8 = bcm_sf2_core_read8,
.read16 = bcm_sf2_core_read16,
.read32 = bcm_sf2_core_read32,
.read48 = bcm_sf2_core_read64,
.read64 = bcm_sf2_core_read64,
.write8 = bcm_sf2_core_write8,
.write16 = bcm_sf2_core_write16,
.write32 = bcm_sf2_core_write32,
.write48 = bcm_sf2_core_write64,
.write64 = bcm_sf2_core_write64,
};
static void bcm_sf2_sw_get_strings(struct dsa_switch *ds, int port,
u32 stringset, uint8_t *data)
{
int cnt = b53_get_sset_count(ds, port, stringset);
b53_get_strings(ds, port, stringset, data);
bcm_sf2_cfp_get_strings(ds, port, stringset,
data + cnt * ETH_GSTRING_LEN);
}
static void bcm_sf2_sw_get_ethtool_stats(struct dsa_switch *ds, int port,
uint64_t *data)
{
int cnt = b53_get_sset_count(ds, port, ETH_SS_STATS);
b53_get_ethtool_stats(ds, port, data);
bcm_sf2_cfp_get_ethtool_stats(ds, port, data + cnt);
}
static int bcm_sf2_sw_get_sset_count(struct dsa_switch *ds, int port,
int sset)
{
int cnt = b53_get_sset_count(ds, port, sset);
if (cnt < 0)
return cnt;
cnt += bcm_sf2_cfp_get_sset_count(ds, port, sset);
return cnt;
}
static const struct dsa_switch_ops bcm_sf2_ops = {
.get_tag_protocol = b53_get_tag_protocol,
.setup = bcm_sf2_sw_setup,
.teardown = bcm_sf2_sw_teardown,
.get_strings = bcm_sf2_sw_get_strings,
.get_ethtool_stats = bcm_sf2_sw_get_ethtool_stats,
.get_sset_count = bcm_sf2_sw_get_sset_count,
.get_ethtool_phy_stats = b53_get_ethtool_phy_stats,
.get_phy_flags = bcm_sf2_sw_get_phy_flags,
.phylink_get_caps = bcm_sf2_sw_get_caps,
.phylink_mac_config = bcm_sf2_sw_mac_config,
.phylink_mac_link_down = bcm_sf2_sw_mac_link_down,
.phylink_mac_link_up = bcm_sf2_sw_mac_link_up,
.phylink_fixed_state = bcm_sf2_sw_fixed_state,
.suspend = bcm_sf2_sw_suspend,
.resume = bcm_sf2_sw_resume,
.get_wol = bcm_sf2_sw_get_wol,
.set_wol = bcm_sf2_sw_set_wol,
.port_enable = bcm_sf2_port_setup,
.port_disable = bcm_sf2_port_disable,
.get_mac_eee = b53_get_mac_eee,
.set_mac_eee = b53_set_mac_eee,
.port_bridge_join = b53_br_join,
.port_bridge_leave = b53_br_leave,
.port_pre_bridge_flags = b53_br_flags_pre,
.port_bridge_flags = b53_br_flags,
.port_stp_state_set = b53_br_set_stp_state,
.port_fast_age = b53_br_fast_age,
.port_vlan_filtering = b53_vlan_filtering,
.port_vlan_add = b53_vlan_add,
.port_vlan_del = b53_vlan_del,
.port_fdb_dump = b53_fdb_dump,
.port_fdb_add = b53_fdb_add,
.port_fdb_del = b53_fdb_del,
.get_rxnfc = bcm_sf2_get_rxnfc,
.set_rxnfc = bcm_sf2_set_rxnfc,
.port_mirror_add = b53_mirror_add,
.port_mirror_del = b53_mirror_del,
.port_mdb_add = b53_mdb_add,
.port_mdb_del = b53_mdb_del,
};
struct bcm_sf2_of_data {
u32 type;
const u16 *reg_offsets;
unsigned int core_reg_align;
unsigned int num_cfp_rules;
unsigned int num_crossbar_int_ports;
};
static const u16 bcm_sf2_4908_reg_offsets[] = {
[REG_SWITCH_CNTRL] = 0x00,
[REG_SWITCH_STATUS] = 0x04,
[REG_DIR_DATA_WRITE] = 0x08,
[REG_DIR_DATA_READ] = 0x0c,
[REG_SWITCH_REVISION] = 0x10,
[REG_PHY_REVISION] = 0x14,
[REG_SPHY_CNTRL] = 0x24,
[REG_CROSSBAR] = 0xc8,
[REG_RGMII_11_CNTRL] = 0x014c,
[REG_LED_0_CNTRL] = 0x40,
[REG_LED_1_CNTRL] = 0x4c,
[REG_LED_2_CNTRL] = 0x58,
[REG_LED_3_CNTRL] = 0x64,
[REG_LED_4_CNTRL] = 0x88,
[REG_LED_5_CNTRL] = 0xa0,
[REG_LED_AGGREGATE_CTRL] = 0xb8,
};
static const struct bcm_sf2_of_data bcm_sf2_4908_data = {
.type = BCM4908_DEVICE_ID,
.core_reg_align = 0,
.reg_offsets = bcm_sf2_4908_reg_offsets,
.num_cfp_rules = 256,
.num_crossbar_int_ports = 2,
};
/* Register offsets for the SWITCH_REG_* block */
static const u16 bcm_sf2_7445_reg_offsets[] = {
[REG_SWITCH_CNTRL] = 0x00,
[REG_SWITCH_STATUS] = 0x04,
[REG_DIR_DATA_WRITE] = 0x08,
[REG_DIR_DATA_READ] = 0x0C,
[REG_SWITCH_REVISION] = 0x18,
[REG_PHY_REVISION] = 0x1C,
[REG_SPHY_CNTRL] = 0x2C,
[REG_RGMII_0_CNTRL] = 0x34,
[REG_RGMII_1_CNTRL] = 0x40,
[REG_RGMII_2_CNTRL] = 0x4c,
[REG_LED_0_CNTRL] = 0x90,
[REG_LED_1_CNTRL] = 0x94,
[REG_LED_2_CNTRL] = 0x98,
};
static const struct bcm_sf2_of_data bcm_sf2_7445_data = {
.type = BCM7445_DEVICE_ID,
.core_reg_align = 0,
.reg_offsets = bcm_sf2_7445_reg_offsets,
.num_cfp_rules = 256,
};
static const u16 bcm_sf2_7278_reg_offsets[] = {
[REG_SWITCH_CNTRL] = 0x00,
[REG_SWITCH_STATUS] = 0x04,
[REG_DIR_DATA_WRITE] = 0x08,
[REG_DIR_DATA_READ] = 0x0c,
[REG_SWITCH_REVISION] = 0x10,
[REG_PHY_REVISION] = 0x14,
[REG_SPHY_CNTRL] = 0x24,
[REG_RGMII_0_CNTRL] = 0xe0,
[REG_RGMII_1_CNTRL] = 0xec,
[REG_RGMII_2_CNTRL] = 0xf8,
[REG_LED_0_CNTRL] = 0x40,
[REG_LED_1_CNTRL] = 0x4c,
[REG_LED_2_CNTRL] = 0x58,
};
static const struct bcm_sf2_of_data bcm_sf2_7278_data = {
.type = BCM7278_DEVICE_ID,
.core_reg_align = 1,
.reg_offsets = bcm_sf2_7278_reg_offsets,
.num_cfp_rules = 128,
};
static const struct of_device_id bcm_sf2_of_match[] = {
{ .compatible = "brcm,bcm4908-switch",
.data = &bcm_sf2_4908_data
},
{ .compatible = "brcm,bcm7445-switch-v4.0",
.data = &bcm_sf2_7445_data
},
{ .compatible = "brcm,bcm7278-switch-v4.0",
.data = &bcm_sf2_7278_data
},
{ .compatible = "brcm,bcm7278-switch-v4.8",
.data = &bcm_sf2_7278_data
},
{ /* sentinel */ },
};
MODULE_DEVICE_TABLE(of, bcm_sf2_of_match);
static int bcm_sf2_sw_probe(struct platform_device *pdev)
{
const char *reg_names[BCM_SF2_REGS_NUM] = BCM_SF2_REGS_NAME;
struct device_node *dn = pdev->dev.of_node;
const struct of_device_id *of_id = NULL;
const struct bcm_sf2_of_data *data;
struct b53_platform_data *pdata;
struct dsa_switch_ops *ops;
struct device_node *ports;
struct bcm_sf2_priv *priv;
struct b53_device *dev;
struct dsa_switch *ds;
void __iomem **base;
unsigned int i;
u32 reg, rev;
int ret;
priv = devm_kzalloc(&pdev->dev, sizeof(*priv), GFP_KERNEL);
if (!priv)
return -ENOMEM;
ops = devm_kzalloc(&pdev->dev, sizeof(*ops), GFP_KERNEL);
if (!ops)
return -ENOMEM;
dev = b53_switch_alloc(&pdev->dev, &bcm_sf2_io_ops, priv);
if (!dev)
return -ENOMEM;
pdata = devm_kzalloc(&pdev->dev, sizeof(*pdata), GFP_KERNEL);
if (!pdata)
return -ENOMEM;
of_id = of_match_node(bcm_sf2_of_match, dn);
if (!of_id || !of_id->data)
return -EINVAL;
data = of_id->data;
/* Set SWITCH_REG register offsets and SWITCH_CORE align factor */
priv->type = data->type;
priv->reg_offsets = data->reg_offsets;
priv->core_reg_align = data->core_reg_align;
priv->num_cfp_rules = data->num_cfp_rules;
priv->num_crossbar_int_ports = data->num_crossbar_int_ports;
priv->rcdev = devm_reset_control_get_optional_exclusive(&pdev->dev,
"switch");
if (IS_ERR(priv->rcdev))
return PTR_ERR(priv->rcdev);
/* Auto-detection using standard registers will not work, so
* provide an indication of what kind of device we are for
* b53_common to work with
*/
pdata->chip_id = priv->type;
dev->pdata = pdata;
priv->dev = dev;
ds = dev->ds;
ds->ops = &bcm_sf2_ops;
/* Advertise the 8 egress queues */
ds->num_tx_queues = SF2_NUM_EGRESS_QUEUES;
dev_set_drvdata(&pdev->dev, priv);
spin_lock_init(&priv->indir_lock);
mutex_init(&priv->cfp.lock);
INIT_LIST_HEAD(&priv->cfp.rules_list);
/* CFP rule #0 cannot be used for specific classifications, flag it as
* permanently used
*/
set_bit(0, priv->cfp.used);
set_bit(0, priv->cfp.unique);
net: dsa: bcm_sf2: Fix node reference count of_find_node_by_name() will do an of_node_put() on the "from" argument. With CONFIG_OF_DYNAMIC enabled which checks for device_node reference counts, we would be getting a warning like this: [ 6.347230] refcount_t: increment on 0; use-after-free. [ 6.352498] WARNING: CPU: 3 PID: 77 at lib/refcount.c:156 refcount_inc_checked+0x38/0x44 [ 6.360601] Modules linked in: [ 6.363661] CPU: 3 PID: 77 Comm: kworker/3:1 Tainted: G W 5.4.46-gb78b3e9956e6 #13 [ 6.372546] Hardware name: BCM97278SV (DT) [ 6.376649] Workqueue: events deferred_probe_work_func [ 6.381796] pstate: 60000005 (nZCv daif -PAN -UAO) [ 6.386595] pc : refcount_inc_checked+0x38/0x44 [ 6.391133] lr : refcount_inc_checked+0x38/0x44 ... [ 6.478791] Call trace: [ 6.481243] refcount_inc_checked+0x38/0x44 [ 6.485433] kobject_get+0x3c/0x4c [ 6.488840] of_node_get+0x24/0x34 [ 6.492247] of_irq_find_parent+0x3c/0xe0 [ 6.496263] of_irq_parse_one+0xe4/0x1d0 [ 6.500191] irq_of_parse_and_map+0x44/0x84 [ 6.504381] bcm_sf2_sw_probe+0x22c/0x844 [ 6.508397] platform_drv_probe+0x58/0xa8 [ 6.512413] really_probe+0x238/0x3fc [ 6.516081] driver_probe_device+0x11c/0x12c [ 6.520358] __device_attach_driver+0xa8/0x100 [ 6.524808] bus_for_each_drv+0xb4/0xd0 [ 6.528650] __device_attach+0xd0/0x164 [ 6.532493] device_initial_probe+0x24/0x30 [ 6.536682] bus_probe_device+0x38/0x98 [ 6.540524] deferred_probe_work_func+0xa8/0xd4 [ 6.545061] process_one_work+0x178/0x288 [ 6.549078] process_scheduled_works+0x44/0x48 [ 6.553529] worker_thread+0x218/0x270 [ 6.557285] kthread+0xdc/0xe4 [ 6.560344] ret_from_fork+0x10/0x18 [ 6.563925] ---[ end trace 68f65caf69bb152a ]--- Fix this by adding a of_node_get() to increment the reference count prior to the call. Fixes: afa3b592953b ("net: dsa: bcm_sf2: Ensure correct sub-node is parsed") Signed-off-by: Florian Fainelli <f.fainelli@gmail.com> Reviewed-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2020-06-18 03:42:44 +00:00
/* Balance of_node_put() done by of_find_node_by_name() */
of_node_get(dn);
ports = of_find_node_by_name(dn, "ports");
if (ports) {
bcm_sf2_identify_ports(priv, ports);
of_node_put(ports);
}
priv->irq0 = irq_of_parse_and_map(dn, 0);
priv->irq1 = irq_of_parse_and_map(dn, 1);
base = &priv->core;
for (i = 0; i < BCM_SF2_REGS_NUM; i++) {
*base = devm_platform_ioremap_resource(pdev, i);
if (IS_ERR(*base)) {
pr_err("unable to find register: %s\n", reg_names[i]);
return PTR_ERR(*base);
}
base++;
}
priv->clk = devm_clk_get_optional(&pdev->dev, "sw_switch");
if (IS_ERR(priv->clk))
return PTR_ERR(priv->clk);
clk_prepare_enable(priv->clk);
priv->clk_mdiv = devm_clk_get_optional(&pdev->dev, "sw_switch_mdiv");
if (IS_ERR(priv->clk_mdiv)) {
ret = PTR_ERR(priv->clk_mdiv);
goto out_clk;
}
clk_prepare_enable(priv->clk_mdiv);
ret = bcm_sf2_sw_rst(priv);
if (ret) {
pr_err("unable to software reset switch: %d\n", ret);
goto out_clk_mdiv;
}
bcm_sf2_crossbar_setup(priv);
bcm_sf2_gphy_enable_set(priv->dev->ds, true);
ret = bcm_sf2_mdio_register(ds);
if (ret) {
pr_err("failed to register MDIO bus\n");
goto out_clk_mdiv;
}
bcm_sf2_gphy_enable_set(priv->dev->ds, false);
ret = bcm_sf2_cfp_rst(priv);
if (ret) {
pr_err("failed to reset CFP\n");
goto out_mdio;
}
/* Disable all interrupts and request them */
bcm_sf2_intr_disable(priv);
ret = devm_request_irq(&pdev->dev, priv->irq0, bcm_sf2_switch_0_isr, 0,
"switch_0", ds);
if (ret < 0) {
pr_err("failed to request switch_0 IRQ\n");
goto out_mdio;
}
ret = devm_request_irq(&pdev->dev, priv->irq1, bcm_sf2_switch_1_isr, 0,
"switch_1", ds);
if (ret < 0) {
pr_err("failed to request switch_1 IRQ\n");
goto out_mdio;
}
/* Reset the MIB counters */
reg = core_readl(priv, CORE_GMNCFGCFG);
reg |= RST_MIB_CNT;
core_writel(priv, reg, CORE_GMNCFGCFG);
reg &= ~RST_MIB_CNT;
core_writel(priv, reg, CORE_GMNCFGCFG);
/* Get the maximum number of ports for this switch */
priv->hw_params.num_ports = core_readl(priv, CORE_IMP0_PRT_ID) + 1;
if (priv->hw_params.num_ports > DSA_MAX_PORTS)
priv->hw_params.num_ports = DSA_MAX_PORTS;
/* Assume a single GPHY setup if we can't read that property */
if (of_property_read_u32(dn, "brcm,num-gphy",
&priv->hw_params.num_gphy))
priv->hw_params.num_gphy = 1;
rev = reg_readl(priv, REG_SWITCH_REVISION);
priv->hw_params.top_rev = (rev >> SWITCH_TOP_REV_SHIFT) &
SWITCH_TOP_REV_MASK;
priv->hw_params.core_rev = (rev & SF2_REV_MASK);
rev = reg_readl(priv, REG_PHY_REVISION);
priv->hw_params.gphy_rev = rev & PHY_REVISION_MASK;
ret = b53_switch_register(dev);
if (ret)
goto out_mdio;
dev_info(&pdev->dev,
"Starfighter 2 top: %x.%02x, core: %x.%02x, IRQs: %d, %d\n",
priv->hw_params.top_rev >> 8, priv->hw_params.top_rev & 0xff,
priv->hw_params.core_rev >> 8, priv->hw_params.core_rev & 0xff,
priv->irq0, priv->irq1);
return 0;
out_mdio:
bcm_sf2_mdio_unregister(priv);
out_clk_mdiv:
clk_disable_unprepare(priv->clk_mdiv);
out_clk:
clk_disable_unprepare(priv->clk);
return ret;
}
static int bcm_sf2_sw_remove(struct platform_device *pdev)
{
struct bcm_sf2_priv *priv = platform_get_drvdata(pdev);
net: dsa: be compatible with masters which unregister on shutdown Lino reports that on his system with bcmgenet as DSA master and KSZ9897 as a switch, rebooting or shutting down never works properly. What does the bcmgenet driver have special to trigger this, that other DSA masters do not? It has an implementation of ->shutdown which simply calls its ->remove implementation. Otherwise said, it unregisters its network interface on shutdown. This message can be seen in a loop, and it hangs the reboot process there: unregister_netdevice: waiting for eth0 to become free. Usage count = 3 So why 3? A usage count of 1 is normal for a registered network interface, and any virtual interface which links itself as an upper of that will increment it via dev_hold. In the case of DSA, this is the call path: dsa_slave_create -> netdev_upper_dev_link -> __netdev_upper_dev_link -> __netdev_adjacent_dev_insert -> dev_hold So a DSA switch with 3 interfaces will result in a usage count elevated by two, and netdev_wait_allrefs will wait until they have gone away. Other stacked interfaces, like VLAN, watch NETDEV_UNREGISTER events and delete themselves, but DSA cannot just vanish and go poof, at most it can unbind itself from the switch devices, but that must happen strictly earlier compared to when the DSA master unregisters its net_device, so reacting on the NETDEV_UNREGISTER event is way too late. It seems that it is a pretty established pattern to have a driver's ->shutdown hook redirect to its ->remove hook, so the same code is executed regardless of whether the driver is unbound from the device, or the system is just shutting down. As Florian puts it, it is quite a big hammer for bcmgenet to unregister its net_device during shutdown, but having a common code path with the driver unbind helps ensure it is well tested. So DSA, for better or for worse, has to live with that and engage in an arms race of implementing the ->shutdown hook too, from all individual drivers, and do something sane when paired with masters that unregister their net_device there. The only sane thing to do, of course, is to unlink from the master. However, complications arise really quickly. The pattern of redirecting ->shutdown to ->remove is not unique to bcmgenet or even to net_device drivers. In fact, SPI controllers do it too (see dspi_shutdown -> dspi_remove), and presumably, I2C controllers and MDIO controllers do it too (this is something I have not researched too deeply, but even if this is not the case today, it is certainly plausible to happen in the future, and must be taken into consideration). Since DSA switches might be SPI devices, I2C devices, MDIO devices, the insane implication is that for the exact same DSA switch device, we might have both ->shutdown and ->remove getting called. So we need to do something with that insane environment. The pattern I've come up with is "if this, then not that", so if either ->shutdown or ->remove gets called, we set the device's drvdata to NULL, and in the other hook, we check whether the drvdata is NULL and just do nothing. This is probably not necessary for platform devices, just for devices on buses, but I would really insist for consistency among drivers, because when code is copy-pasted, it is not always copy-pasted from the best sources. So depending on whether the DSA switch's ->remove or ->shutdown will get called first, we cannot really guarantee even for the same driver if rebooting will result in the same code path on all platforms. But nonetheless, we need to do something minimally reasonable on ->shutdown too to fix the bug. Of course, the ->remove will do more (a full teardown of the tree, with all data structures freed, and this is why the bug was not caught for so long). The new ->shutdown method is kept separate from dsa_unregister_switch not because we couldn't have unregistered the switch, but simply in the interest of doing something quick and to the point. The big question is: does the DSA switch's ->shutdown get called earlier than the DSA master's ->shutdown? If not, there is still a risk that we might still trigger the WARN_ON in unregister_netdevice that says we are attempting to unregister a net_device which has uppers. That's no good. Although the reference to the master net_device won't physically go away even if DSA's ->shutdown comes afterwards, remember we have a dev_hold on it. The answer to that question lies in this comment above device_link_add: * A side effect of the link creation is re-ordering of dpm_list and the * devices_kset list by moving the consumer device and all devices depending * on it to the ends of these lists (that does not happen to devices that have * not been registered when this function is called). so the fact that DSA uses device_link_add towards its master is not exactly for nothing. device_shutdown() walks devices_kset from the back, so this is our guarantee that DSA's shutdown happens before the master's shutdown. Fixes: 2f1e8ea726e9 ("net: dsa: link interfaces with the DSA master to get rid of lockdep warnings") Link: https://lore.kernel.org/netdev/20210909095324.12978-1-LinoSanfilippo@gmx.de/ Reported-by: Lino Sanfilippo <LinoSanfilippo@gmx.de> Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Tested-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-09-17 13:34:33 +00:00
if (!priv)
return 0;
priv->wol_ports_mask = 0;
/* Disable interrupts */
bcm_sf2_intr_disable(priv);
dsa_unregister_switch(priv->dev->ds);
bcm_sf2_cfp_exit(priv->dev->ds);
bcm_sf2_mdio_unregister(priv);
clk_disable_unprepare(priv->clk_mdiv);
clk_disable_unprepare(priv->clk);
if (priv->type == BCM7278_DEVICE_ID)
reset_control_assert(priv->rcdev);
net: dsa: be compatible with masters which unregister on shutdown Lino reports that on his system with bcmgenet as DSA master and KSZ9897 as a switch, rebooting or shutting down never works properly. What does the bcmgenet driver have special to trigger this, that other DSA masters do not? It has an implementation of ->shutdown which simply calls its ->remove implementation. Otherwise said, it unregisters its network interface on shutdown. This message can be seen in a loop, and it hangs the reboot process there: unregister_netdevice: waiting for eth0 to become free. Usage count = 3 So why 3? A usage count of 1 is normal for a registered network interface, and any virtual interface which links itself as an upper of that will increment it via dev_hold. In the case of DSA, this is the call path: dsa_slave_create -> netdev_upper_dev_link -> __netdev_upper_dev_link -> __netdev_adjacent_dev_insert -> dev_hold So a DSA switch with 3 interfaces will result in a usage count elevated by two, and netdev_wait_allrefs will wait until they have gone away. Other stacked interfaces, like VLAN, watch NETDEV_UNREGISTER events and delete themselves, but DSA cannot just vanish and go poof, at most it can unbind itself from the switch devices, but that must happen strictly earlier compared to when the DSA master unregisters its net_device, so reacting on the NETDEV_UNREGISTER event is way too late. It seems that it is a pretty established pattern to have a driver's ->shutdown hook redirect to its ->remove hook, so the same code is executed regardless of whether the driver is unbound from the device, or the system is just shutting down. As Florian puts it, it is quite a big hammer for bcmgenet to unregister its net_device during shutdown, but having a common code path with the driver unbind helps ensure it is well tested. So DSA, for better or for worse, has to live with that and engage in an arms race of implementing the ->shutdown hook too, from all individual drivers, and do something sane when paired with masters that unregister their net_device there. The only sane thing to do, of course, is to unlink from the master. However, complications arise really quickly. The pattern of redirecting ->shutdown to ->remove is not unique to bcmgenet or even to net_device drivers. In fact, SPI controllers do it too (see dspi_shutdown -> dspi_remove), and presumably, I2C controllers and MDIO controllers do it too (this is something I have not researched too deeply, but even if this is not the case today, it is certainly plausible to happen in the future, and must be taken into consideration). Since DSA switches might be SPI devices, I2C devices, MDIO devices, the insane implication is that for the exact same DSA switch device, we might have both ->shutdown and ->remove getting called. So we need to do something with that insane environment. The pattern I've come up with is "if this, then not that", so if either ->shutdown or ->remove gets called, we set the device's drvdata to NULL, and in the other hook, we check whether the drvdata is NULL and just do nothing. This is probably not necessary for platform devices, just for devices on buses, but I would really insist for consistency among drivers, because when code is copy-pasted, it is not always copy-pasted from the best sources. So depending on whether the DSA switch's ->remove or ->shutdown will get called first, we cannot really guarantee even for the same driver if rebooting will result in the same code path on all platforms. But nonetheless, we need to do something minimally reasonable on ->shutdown too to fix the bug. Of course, the ->remove will do more (a full teardown of the tree, with all data structures freed, and this is why the bug was not caught for so long). The new ->shutdown method is kept separate from dsa_unregister_switch not because we couldn't have unregistered the switch, but simply in the interest of doing something quick and to the point. The big question is: does the DSA switch's ->shutdown get called earlier than the DSA master's ->shutdown? If not, there is still a risk that we might still trigger the WARN_ON in unregister_netdevice that says we are attempting to unregister a net_device which has uppers. That's no good. Although the reference to the master net_device won't physically go away even if DSA's ->shutdown comes afterwards, remember we have a dev_hold on it. The answer to that question lies in this comment above device_link_add: * A side effect of the link creation is re-ordering of dpm_list and the * devices_kset list by moving the consumer device and all devices depending * on it to the ends of these lists (that does not happen to devices that have * not been registered when this function is called). so the fact that DSA uses device_link_add towards its master is not exactly for nothing. device_shutdown() walks devices_kset from the back, so this is our guarantee that DSA's shutdown happens before the master's shutdown. Fixes: 2f1e8ea726e9 ("net: dsa: link interfaces with the DSA master to get rid of lockdep warnings") Link: https://lore.kernel.org/netdev/20210909095324.12978-1-LinoSanfilippo@gmx.de/ Reported-by: Lino Sanfilippo <LinoSanfilippo@gmx.de> Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Tested-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-09-17 13:34:33 +00:00
platform_set_drvdata(pdev, NULL);
return 0;
}
static void bcm_sf2_sw_shutdown(struct platform_device *pdev)
{
struct bcm_sf2_priv *priv = platform_get_drvdata(pdev);
net: dsa: be compatible with masters which unregister on shutdown Lino reports that on his system with bcmgenet as DSA master and KSZ9897 as a switch, rebooting or shutting down never works properly. What does the bcmgenet driver have special to trigger this, that other DSA masters do not? It has an implementation of ->shutdown which simply calls its ->remove implementation. Otherwise said, it unregisters its network interface on shutdown. This message can be seen in a loop, and it hangs the reboot process there: unregister_netdevice: waiting for eth0 to become free. Usage count = 3 So why 3? A usage count of 1 is normal for a registered network interface, and any virtual interface which links itself as an upper of that will increment it via dev_hold. In the case of DSA, this is the call path: dsa_slave_create -> netdev_upper_dev_link -> __netdev_upper_dev_link -> __netdev_adjacent_dev_insert -> dev_hold So a DSA switch with 3 interfaces will result in a usage count elevated by two, and netdev_wait_allrefs will wait until they have gone away. Other stacked interfaces, like VLAN, watch NETDEV_UNREGISTER events and delete themselves, but DSA cannot just vanish and go poof, at most it can unbind itself from the switch devices, but that must happen strictly earlier compared to when the DSA master unregisters its net_device, so reacting on the NETDEV_UNREGISTER event is way too late. It seems that it is a pretty established pattern to have a driver's ->shutdown hook redirect to its ->remove hook, so the same code is executed regardless of whether the driver is unbound from the device, or the system is just shutting down. As Florian puts it, it is quite a big hammer for bcmgenet to unregister its net_device during shutdown, but having a common code path with the driver unbind helps ensure it is well tested. So DSA, for better or for worse, has to live with that and engage in an arms race of implementing the ->shutdown hook too, from all individual drivers, and do something sane when paired with masters that unregister their net_device there. The only sane thing to do, of course, is to unlink from the master. However, complications arise really quickly. The pattern of redirecting ->shutdown to ->remove is not unique to bcmgenet or even to net_device drivers. In fact, SPI controllers do it too (see dspi_shutdown -> dspi_remove), and presumably, I2C controllers and MDIO controllers do it too (this is something I have not researched too deeply, but even if this is not the case today, it is certainly plausible to happen in the future, and must be taken into consideration). Since DSA switches might be SPI devices, I2C devices, MDIO devices, the insane implication is that for the exact same DSA switch device, we might have both ->shutdown and ->remove getting called. So we need to do something with that insane environment. The pattern I've come up with is "if this, then not that", so if either ->shutdown or ->remove gets called, we set the device's drvdata to NULL, and in the other hook, we check whether the drvdata is NULL and just do nothing. This is probably not necessary for platform devices, just for devices on buses, but I would really insist for consistency among drivers, because when code is copy-pasted, it is not always copy-pasted from the best sources. So depending on whether the DSA switch's ->remove or ->shutdown will get called first, we cannot really guarantee even for the same driver if rebooting will result in the same code path on all platforms. But nonetheless, we need to do something minimally reasonable on ->shutdown too to fix the bug. Of course, the ->remove will do more (a full teardown of the tree, with all data structures freed, and this is why the bug was not caught for so long). The new ->shutdown method is kept separate from dsa_unregister_switch not because we couldn't have unregistered the switch, but simply in the interest of doing something quick and to the point. The big question is: does the DSA switch's ->shutdown get called earlier than the DSA master's ->shutdown? If not, there is still a risk that we might still trigger the WARN_ON in unregister_netdevice that says we are attempting to unregister a net_device which has uppers. That's no good. Although the reference to the master net_device won't physically go away even if DSA's ->shutdown comes afterwards, remember we have a dev_hold on it. The answer to that question lies in this comment above device_link_add: * A side effect of the link creation is re-ordering of dpm_list and the * devices_kset list by moving the consumer device and all devices depending * on it to the ends of these lists (that does not happen to devices that have * not been registered when this function is called). so the fact that DSA uses device_link_add towards its master is not exactly for nothing. device_shutdown() walks devices_kset from the back, so this is our guarantee that DSA's shutdown happens before the master's shutdown. Fixes: 2f1e8ea726e9 ("net: dsa: link interfaces with the DSA master to get rid of lockdep warnings") Link: https://lore.kernel.org/netdev/20210909095324.12978-1-LinoSanfilippo@gmx.de/ Reported-by: Lino Sanfilippo <LinoSanfilippo@gmx.de> Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Tested-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-09-17 13:34:33 +00:00
if (!priv)
return;
/* For a kernel about to be kexec'd we want to keep the GPHY on for a
* successful MDIO bus scan to occur. If we did turn off the GPHY
* before (e.g: port_disable), this will also power it back on.
*
* Do not rely on kexec_in_progress, just power the PHY on.
*/
if (priv->hw_params.num_gphy == 1)
bcm_sf2_gphy_enable_set(priv->dev->ds, true);
net: dsa: be compatible with masters which unregister on shutdown Lino reports that on his system with bcmgenet as DSA master and KSZ9897 as a switch, rebooting or shutting down never works properly. What does the bcmgenet driver have special to trigger this, that other DSA masters do not? It has an implementation of ->shutdown which simply calls its ->remove implementation. Otherwise said, it unregisters its network interface on shutdown. This message can be seen in a loop, and it hangs the reboot process there: unregister_netdevice: waiting for eth0 to become free. Usage count = 3 So why 3? A usage count of 1 is normal for a registered network interface, and any virtual interface which links itself as an upper of that will increment it via dev_hold. In the case of DSA, this is the call path: dsa_slave_create -> netdev_upper_dev_link -> __netdev_upper_dev_link -> __netdev_adjacent_dev_insert -> dev_hold So a DSA switch with 3 interfaces will result in a usage count elevated by two, and netdev_wait_allrefs will wait until they have gone away. Other stacked interfaces, like VLAN, watch NETDEV_UNREGISTER events and delete themselves, but DSA cannot just vanish and go poof, at most it can unbind itself from the switch devices, but that must happen strictly earlier compared to when the DSA master unregisters its net_device, so reacting on the NETDEV_UNREGISTER event is way too late. It seems that it is a pretty established pattern to have a driver's ->shutdown hook redirect to its ->remove hook, so the same code is executed regardless of whether the driver is unbound from the device, or the system is just shutting down. As Florian puts it, it is quite a big hammer for bcmgenet to unregister its net_device during shutdown, but having a common code path with the driver unbind helps ensure it is well tested. So DSA, for better or for worse, has to live with that and engage in an arms race of implementing the ->shutdown hook too, from all individual drivers, and do something sane when paired with masters that unregister their net_device there. The only sane thing to do, of course, is to unlink from the master. However, complications arise really quickly. The pattern of redirecting ->shutdown to ->remove is not unique to bcmgenet or even to net_device drivers. In fact, SPI controllers do it too (see dspi_shutdown -> dspi_remove), and presumably, I2C controllers and MDIO controllers do it too (this is something I have not researched too deeply, but even if this is not the case today, it is certainly plausible to happen in the future, and must be taken into consideration). Since DSA switches might be SPI devices, I2C devices, MDIO devices, the insane implication is that for the exact same DSA switch device, we might have both ->shutdown and ->remove getting called. So we need to do something with that insane environment. The pattern I've come up with is "if this, then not that", so if either ->shutdown or ->remove gets called, we set the device's drvdata to NULL, and in the other hook, we check whether the drvdata is NULL and just do nothing. This is probably not necessary for platform devices, just for devices on buses, but I would really insist for consistency among drivers, because when code is copy-pasted, it is not always copy-pasted from the best sources. So depending on whether the DSA switch's ->remove or ->shutdown will get called first, we cannot really guarantee even for the same driver if rebooting will result in the same code path on all platforms. But nonetheless, we need to do something minimally reasonable on ->shutdown too to fix the bug. Of course, the ->remove will do more (a full teardown of the tree, with all data structures freed, and this is why the bug was not caught for so long). The new ->shutdown method is kept separate from dsa_unregister_switch not because we couldn't have unregistered the switch, but simply in the interest of doing something quick and to the point. The big question is: does the DSA switch's ->shutdown get called earlier than the DSA master's ->shutdown? If not, there is still a risk that we might still trigger the WARN_ON in unregister_netdevice that says we are attempting to unregister a net_device which has uppers. That's no good. Although the reference to the master net_device won't physically go away even if DSA's ->shutdown comes afterwards, remember we have a dev_hold on it. The answer to that question lies in this comment above device_link_add: * A side effect of the link creation is re-ordering of dpm_list and the * devices_kset list by moving the consumer device and all devices depending * on it to the ends of these lists (that does not happen to devices that have * not been registered when this function is called). so the fact that DSA uses device_link_add towards its master is not exactly for nothing. device_shutdown() walks devices_kset from the back, so this is our guarantee that DSA's shutdown happens before the master's shutdown. Fixes: 2f1e8ea726e9 ("net: dsa: link interfaces with the DSA master to get rid of lockdep warnings") Link: https://lore.kernel.org/netdev/20210909095324.12978-1-LinoSanfilippo@gmx.de/ Reported-by: Lino Sanfilippo <LinoSanfilippo@gmx.de> Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com> Tested-by: Andrew Lunn <andrew@lunn.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
2021-09-17 13:34:33 +00:00
dsa_switch_shutdown(priv->dev->ds);
platform_set_drvdata(pdev, NULL);
}
#ifdef CONFIG_PM_SLEEP
static int bcm_sf2_suspend(struct device *dev)
{
struct bcm_sf2_priv *priv = dev_get_drvdata(dev);
return dsa_switch_suspend(priv->dev->ds);
}
static int bcm_sf2_resume(struct device *dev)
{
struct bcm_sf2_priv *priv = dev_get_drvdata(dev);
return dsa_switch_resume(priv->dev->ds);
}
#endif /* CONFIG_PM_SLEEP */
static SIMPLE_DEV_PM_OPS(bcm_sf2_pm_ops,
bcm_sf2_suspend, bcm_sf2_resume);
static struct platform_driver bcm_sf2_driver = {
.probe = bcm_sf2_sw_probe,
.remove = bcm_sf2_sw_remove,
.shutdown = bcm_sf2_sw_shutdown,
.driver = {
.name = "brcm-sf2",
.of_match_table = bcm_sf2_of_match,
.pm = &bcm_sf2_pm_ops,
},
};
module_platform_driver(bcm_sf2_driver);
MODULE_AUTHOR("Broadcom Corporation");
MODULE_DESCRIPTION("Driver for Broadcom Starfighter 2 ethernet switch chip");
MODULE_LICENSE("GPL");
MODULE_ALIAS("platform:brcm-sf2");