Summary: This switch is based off of the AR8327/AR8337 external switch/PHY. However unlike the AR8327/AR8337 it itself doesn't have any PHYs; instead an external PHY connects to it using the PSGMII port. Differential Revision: https://reviews.freebsd.org/D34112 Reviewed by: manu This code is inspired by the ar40xx code in openwrt, which itself is based on the Qualcomm QCA-SSDK. Both of these sources are, amusingly, BSD licenced - and thus I have included some of the comments in the hardware workaround paths to document some of the magic numbers.
288 lines
8.0 KiB
C
288 lines
8.0 KiB
C
/*-
|
|
* SPDX-License-Identifier: BSD-2-Clause-FreeBSD
|
|
*
|
|
* Copyright (c) 2022 Adrian Chadd <adrian@FreeBSD.org>.
|
|
*
|
|
* Redistribution and use in source and binary forms, with or without
|
|
* modification, are permitted provided that the following conditions
|
|
* are met:
|
|
* 1. Redistributions of source code must retain the above copyright
|
|
* notice, this list of conditions and the following disclaimer.
|
|
* 2. Redistributions in binary form must reproduce the above copyright
|
|
* notice, this list of conditions and the following disclaimer in the
|
|
* documentation and/or other materials provided with the distribution.
|
|
*
|
|
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
|
|
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
|
|
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
|
|
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
|
|
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
|
|
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
|
|
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
|
|
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
|
|
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
|
|
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
|
|
* SUCH DAMAGE.
|
|
*/
|
|
|
|
#include <sys/param.h>
|
|
#include <sys/bus.h>
|
|
#include <sys/errno.h>
|
|
#include <sys/kernel.h>
|
|
#include <sys/malloc.h>
|
|
#include <sys/module.h>
|
|
#include <sys/socket.h>
|
|
#include <sys/sockio.h>
|
|
#include <sys/sysctl.h>
|
|
#include <sys/systm.h>
|
|
|
|
#include <net/if.h>
|
|
#include <net/if_var.h>
|
|
#include <net/if_arp.h>
|
|
#include <net/ethernet.h>
|
|
#include <net/if_dl.h>
|
|
#include <net/if_media.h>
|
|
#include <net/if_types.h>
|
|
|
|
#include <machine/bus.h>
|
|
#include <dev/iicbus/iic.h>
|
|
#include <dev/iicbus/iiconf.h>
|
|
#include <dev/iicbus/iicbus.h>
|
|
#include <dev/mii/mii.h>
|
|
#include <dev/mii/miivar.h>
|
|
#include <dev/mdio/mdio.h>
|
|
#include <dev/extres/clk/clk.h>
|
|
#include <dev/extres/hwreset/hwreset.h>
|
|
|
|
#include <dev/fdt/fdt_common.h>
|
|
#include <dev/ofw/ofw_bus.h>
|
|
#include <dev/ofw/ofw_bus_subr.h>
|
|
|
|
#include <dev/etherswitch/etherswitch.h>
|
|
|
|
#include <dev/etherswitch/ar40xx/ar40xx_var.h>
|
|
#include <dev/etherswitch/ar40xx/ar40xx_reg.h>
|
|
#include <dev/etherswitch/ar40xx/ar40xx_debug.h>
|
|
#include <dev/etherswitch/ar40xx/ar40xx_hw_port.h>
|
|
|
|
#include "mdio_if.h"
|
|
#include "miibus_if.h"
|
|
#include "etherswitch_if.h"
|
|
|
|
|
|
int
|
|
ar40xx_hw_port_init(struct ar40xx_softc *sc, int port)
|
|
{
|
|
uint32_t reg;
|
|
|
|
AR40XX_DPRINTF(sc, AR40XX_DBG_HW_PORT_INIT,
|
|
"%s: called; port %d\n", __func__, port);
|
|
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_STATUS(port), 0);
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_HEADER(port), 0);
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_VLAN0(port), 0);
|
|
AR40XX_REG_BARRIER_WRITE(sc);
|
|
|
|
DELAY(20);
|
|
|
|
/*
|
|
* Ok! Here is where things get super fun in the AR40xx
|
|
* driver in uboot/linux.
|
|
*
|
|
* The earlier chipset switch drivers enable auto link enable here.
|
|
* The switch will poll the PHYs too, and configure appropriately.
|
|
*
|
|
* The ar40xx code in linux/u-boot instead has a whole workaround
|
|
* path that polls things directly and does some weird hijinx.
|
|
* NOTABLY - they do NOT enable the TX/RX MAC here or autoneg -
|
|
* it's done in the work around path.
|
|
*
|
|
* SO - for now the port is left off until the PHY state changes.
|
|
* And then we flip it on and off based on the PHY state.
|
|
*/
|
|
#if 0
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_STATUS(port),
|
|
AR40XX_PORT_AUTO_LINK_EN);
|
|
#endif
|
|
|
|
/*
|
|
* Configure the VLAN egress mode (don't touch them) and
|
|
* learning state for STP/ATU. This isn't currently
|
|
* configurable so it's just nailed up here and left alone.
|
|
*/
|
|
reg = AR40XX_PORT_VLAN1_OUT_MODE_UNTOUCH
|
|
<< AR40XX_PORT_VLAN1_OUT_MODE_S;
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_VLAN1(port), reg);
|
|
|
|
reg = AR40XX_PORT_LOOKUP_LEARN;
|
|
reg |= AR40XX_PORT_STATE_FORWARD << AR40XX_PORT_LOOKUP_STATE_S;
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_LOOKUP(port), reg);
|
|
AR40XX_REG_BARRIER_WRITE(sc);
|
|
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* Call when the link for a non-CPU port is down.
|
|
*
|
|
* This will turn off the MAC/forwarding path for this port.
|
|
*/
|
|
int
|
|
ar40xx_hw_port_link_down(struct ar40xx_softc *sc, int port)
|
|
{
|
|
|
|
AR40XX_DPRINTF(sc, AR40XX_DBG_HW_PORT_INIT,
|
|
"%s: called; port %d\n", __func__, port);
|
|
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_STATUS(port), 0);
|
|
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* Call when the link for a non-CPU port is up.
|
|
*
|
|
* This will turn on the default auto-link checking and
|
|
* eventually enable the TX/RX MAC.
|
|
*/
|
|
int
|
|
ar40xx_hw_port_link_up(struct ar40xx_softc *sc, int port)
|
|
{
|
|
uint32_t reg;
|
|
|
|
AR40XX_DPRINTF(sc, AR40XX_DBG_HW_PORT_INIT,
|
|
"%s: called; port %d\n", __func__, port);
|
|
|
|
/* Auto-link enable */
|
|
AR40XX_REG_BARRIER_READ(sc);
|
|
reg = AR40XX_REG_READ(sc, AR40XX_REG_PORT_STATUS(port));
|
|
reg |= AR40XX_PORT_AUTO_LINK_EN;
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_STATUS(port), reg);
|
|
AR40XX_REG_BARRIER_WRITE(sc);
|
|
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* Setup the CPU facing port. For this device it'll only
|
|
* be port 0.
|
|
*/
|
|
int
|
|
ar40xx_hw_port_cpuport_setup(struct ar40xx_softc *sc)
|
|
{
|
|
uint32_t reg;
|
|
|
|
AR40XX_DPRINTF(sc, AR40XX_DBG_HW_PORT_INIT, "%s: called\n",
|
|
__func__);
|
|
|
|
reg = AR40XX_PORT_STATUS_TXFLOW
|
|
| AR40XX_PORT_STATUS_RXFLOW
|
|
| AR40XX_PORT_TXHALF_FLOW
|
|
| AR40XX_PORT_DUPLEX
|
|
| AR40XX_PORT_SPEED_1000M;
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_STATUS(0), reg);
|
|
DELAY(20);
|
|
|
|
reg |= AR40XX_PORT_TX_EN | AR40XX_PORT_RX_EN;
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_STATUS(0), reg);
|
|
AR40XX_REG_BARRIER_WRITE(sc);
|
|
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* Fetch the port PVID.
|
|
*
|
|
* For 802.1q mode this is the default VLAN ID for the port.
|
|
* Frames without an 802.1q VLAN will assume this VLAN ID for
|
|
* transmit/receive.
|
|
*/
|
|
int
|
|
ar40xx_hw_get_port_pvid(struct ar40xx_softc *sc, int port, int *pvid)
|
|
{
|
|
uint32_t reg;
|
|
|
|
AR40XX_LOCK_ASSERT(sc);
|
|
|
|
AR40XX_REG_BARRIER_READ(sc);
|
|
reg = AR40XX_REG_READ(sc, AR40XX_REG_PORT_VLAN0(port));
|
|
|
|
reg = reg >> AR40XX_PORT_VLAN0_DEF_CVID_S;
|
|
reg = reg & 0x0fff; /* XXX */
|
|
|
|
*pvid = reg;
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* Set the port PVID.
|
|
*
|
|
* For now, since double-tagged frames aren't currently supported,
|
|
* CVID=SVID here.
|
|
*/
|
|
int
|
|
ar40xx_hw_set_port_pvid(struct ar40xx_softc *sc, int port, int pvid)
|
|
{
|
|
uint32_t reg;
|
|
|
|
AR40XX_LOCK_ASSERT(sc);
|
|
|
|
pvid &= ETHERSWITCH_VID_MASK;
|
|
|
|
reg = pvid << AR40XX_PORT_VLAN0_DEF_SVID_S;
|
|
reg |= pvid << AR40XX_PORT_VLAN0_DEF_CVID_S;
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_VLAN0(port), reg);
|
|
AR40XX_REG_BARRIER_WRITE(sc);
|
|
|
|
return (0);
|
|
}
|
|
|
|
/*
|
|
* Setup the default port membership configuration.
|
|
*
|
|
* This configures the PVID for the port in the sc_vlan config,
|
|
* along with a set of ports that constitute the "membership"
|
|
* of this particular VID.
|
|
*
|
|
* For 802.1q mode the membership can be viewed as the default
|
|
* learning port group, but this can be added to via VLAN membership.
|
|
* (Eg you could in theory split two LAN ports into separate "member"
|
|
* groups and they'd not learn MAC addresses from each other even
|
|
* inside a VLAN; you'd then end up with the traffic being flooded to
|
|
* the CPU port.)
|
|
*/
|
|
int
|
|
ar40xx_hw_port_setup(struct ar40xx_softc *sc, int port, uint32_t members)
|
|
{
|
|
uint32_t egress, ingress, reg;
|
|
uint32_t pvid = sc->sc_vlan.vlan_id[sc->sc_vlan.pvid[port]]
|
|
& ETHERSWITCH_VID_MASK;
|
|
|
|
if (sc->sc_vlan.vlan) {
|
|
egress = AR40XX_PORT_VLAN1_OUT_MODE_UNMOD;
|
|
ingress = AR40XX_IN_SECURE;
|
|
} else {
|
|
egress = AR40XX_PORT_VLAN1_OUT_MODE_UNTOUCH;
|
|
ingress = AR40XX_IN_PORT_ONLY;
|
|
}
|
|
|
|
reg = pvid << AR40XX_PORT_VLAN0_DEF_SVID_S;
|
|
reg |= pvid << AR40XX_PORT_VLAN0_DEF_CVID_S;
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_VLAN0(port), reg);
|
|
AR40XX_REG_BARRIER_WRITE(sc);
|
|
|
|
reg = AR40XX_PORT_VLAN1_PORT_VLAN_PROP;
|
|
reg |= egress << AR40XX_PORT_VLAN1_OUT_MODE_S;
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_VLAN1(port), reg);
|
|
AR40XX_REG_BARRIER_WRITE(sc);
|
|
|
|
reg = members;
|
|
reg |= AR40XX_PORT_LOOKUP_LEARN;
|
|
reg |= ingress << AR40XX_PORT_LOOKUP_IN_MODE_S;
|
|
reg |= AR40XX_PORT_STATE_FORWARD << AR40XX_PORT_LOOKUP_STATE_S;
|
|
AR40XX_REG_WRITE(sc, AR40XX_REG_PORT_LOOKUP(port), reg);
|
|
AR40XX_REG_BARRIER_WRITE(sc);
|
|
|
|
return (0);
|
|
}
|