aboutsummaryrefslogtreecommitdiff
path: root/drivers/net/ethernet/microchip/encx24j600.c
diff options
context:
space:
mode:
authorLibravatar Linus Torvalds <torvalds@linux-foundation.org>2023-02-21 18:24:12 -0800
committerLibravatar Linus Torvalds <torvalds@linux-foundation.org>2023-02-21 18:24:12 -0800
commit5b7c4cabbb65f5c469464da6c5f614cbd7f730f2 (patch)
treecc5c2d0a898769fd59549594fedb3ee6f84e59a0 /drivers/net/ethernet/microchip/encx24j600.c
downloadlinux-5b7c4cabbb65f5c469464da6c5f614cbd7f730f2.tar.gz
linux-5b7c4cabbb65f5c469464da6c5f614cbd7f730f2.zip
Merge tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-nextgrafted
Pull networking updates from Jakub Kicinski: "Core: - Add dedicated kmem_cache for typical/small skb->head, avoid having to access struct page at kfree time, and improve memory use. - Introduce sysctl to set default RPS configuration for new netdevs. - Define Netlink protocol specification format which can be used to describe messages used by each family and auto-generate parsers. Add tools for generating kernel data structures and uAPI headers. - Expose all net/core sysctls inside netns. - Remove 4s sleep in netpoll if carrier is instantly detected on boot. - Add configurable limit of MDB entries per port, and port-vlan. - Continue populating drop reasons throughout the stack. - Retire a handful of legacy Qdiscs and classifiers. Protocols: - Support IPv4 big TCP (TSO frames larger than 64kB). - Add IP_LOCAL_PORT_RANGE socket option, to control local port range on socket by socket basis. - Track and report in procfs number of MPTCP sockets used. - Support mixing IPv4 and IPv6 flows in the in-kernel MPTCP path manager. - IPv6: don't check net.ipv6.route.max_size and rely on garbage collection to free memory (similarly to IPv4). - Support Penultimate Segment Pop (PSP) flavor in SRv6 (RFC8986). - ICMP: add per-rate limit counters. - Add support for user scanning requests in ieee802154. - Remove static WEP support. - Support minimal Wi-Fi 7 Extremely High Throughput (EHT) rate reporting. - WiFi 7 EHT channel puncturing support (client & AP). BPF: - Add a rbtree data structure following the "next-gen data structure" precedent set by recently added linked list, that is, by using kfunc + kptr instead of adding a new BPF map type. - Expose XDP hints via kfuncs with initial support for RX hash and timestamp metadata. - Add BPF_F_NO_TUNNEL_KEY extension to bpf_skb_set_tunnel_key to better support decap on GRE tunnel devices not operating in collect metadata. - Improve x86 JIT's codegen for PROBE_MEM runtime error checks. - Remove the need for trace_printk_lock for bpf_trace_printk and bpf_trace_vprintk helpers. - Extend libbpf's bpf_tracing.h support for tracing arguments of kprobes/uprobes and syscall as a special case. - Significantly reduce the search time for module symbols by livepatch and BPF. - Enable cpumasks to be used as kptrs, which is useful for tracing programs tracking which tasks end up running on which CPUs in different time intervals. - Add support for BPF trampoline on s390x and riscv64. - Add capability to export the XDP features supported by the NIC. - Add __bpf_kfunc tag for marking kernel functions as kfuncs. - Add cgroup.memory=nobpf kernel parameter option to disable BPF memory accounting for container environments. Netfilter: - Remove the CLUSTERIP target. It has been marked as obsolete for years, and we still have WARN splats wrt races of the out-of-band /proc interface installed by this target. - Add 'destroy' commands to nf_tables. They are identical to the existing 'delete' commands, but do not return an error if the referenced object (set, chain, rule...) did not exist. Driver API: - Improve cpumask_local_spread() locality to help NICs set the right IRQ affinity on AMD platforms. - Separate C22 and C45 MDIO bus transactions more clearly. - Introduce new DCB table to control DSCP rewrite on egress. - Support configuration of Physical Layer Collision Avoidance (PLCA) Reconciliation Sublayer (RS) (802.3cg-2019). Modern version of shared medium Ethernet. - Support for MAC Merge layer (IEEE 802.3-2018 clause 99). Allowing preemption of low priority frames by high priority frames. - Add support for controlling MACSec offload using netlink SET. - Rework devlink instance refcounts to allow registration and de-registration under the instance lock. Split the code into multiple files, drop some of the unnecessarily granular locks and factor out common parts of netlink operation handling. - Add TX frame aggregation parameters (for USB drivers). - Add a new attr TCA_EXT_WARN_MSG to report TC (offload) warning messages with notifications for debug. - Allow offloading of UDP NEW connections via act_ct. - Add support for per action HW stats in TC. - Support hardware miss to TC action (continue processing in SW from a specific point in the action chain). - Warn if old Wireless Extension user space interface is used with modern cfg80211/mac80211 drivers. Do not support Wireless Extensions for Wi-Fi 7 devices at all. Everyone should switch to using nl80211 interface instead. - Improve the CAN bit timing configuration. Use extack to return error messages directly to user space, update the SJW handling, including the definition of a new default value that will benefit CAN-FD controllers, by increasing their oscillator tolerance. New hardware / drivers: - Ethernet: - nVidia BlueField-3 support (control traffic driver) - Ethernet support for imx93 SoCs - Motorcomm yt8531 gigabit Ethernet PHY - onsemi NCN26000 10BASE-T1S PHY (with support for PLCA) - Microchip LAN8841 PHY (incl. cable diagnostics and PTP) - Amlogic gxl MDIO mux - WiFi: - RealTek RTL8188EU (rtl8xxxu) - Qualcomm Wi-Fi 7 devices (ath12k) - CAN: - Renesas R-Car V4H Drivers: - Bluetooth: - Set Per Platform Antenna Gain (PPAG) for Intel controllers. - Ethernet NICs: - Intel (1G, igc): - support TSN / Qbv / packet scheduling features of i226 model - Intel (100G, ice): - use GNSS subsystem instead of TTY - multi-buffer XDP support - extend support for GPIO pins to E823 devices - nVidia/Mellanox: - update the shared buffer configuration on PFC commands - implement PTP adjphase function for HW offset control - TC support for Geneve and GRE with VF tunnel offload - more efficient crypto key management method - multi-port eswitch support - Netronome/Corigine: - add DCB IEEE support - support IPsec offloading for NFP3800 - Freescale/NXP (enetc): - support XDP_REDIRECT for XDP non-linear buffers - improve reconfig, avoid link flap and waiting for idle - support MAC Merge layer - Other NICs: - sfc/ef100: add basic devlink support for ef100 - ionic: rx_push mode operation (writing descriptors via MMIO) - bnxt: use the auxiliary bus abstraction for RDMA - r8169: disable ASPM and reset bus in case of tx timeout - cpsw: support QSGMII mode for J721e CPSW9G - cpts: support pulse-per-second output - ngbe: add an mdio bus driver - usbnet: optimize usbnet_bh() by avoiding unnecessary queuing - r8152: handle devices with FW with NCM support - amd-xgbe: support 10Mbps, 2.5GbE speeds and rx-adaptation - virtio-net: support multi buffer XDP - virtio/vsock: replace virtio_vsock_pkt with sk_buff - tsnep: XDP support - Ethernet high-speed switches: - nVidia/Mellanox (mlxsw): - add support for latency TLV (in FW control messages) - Microchip (sparx5): - separate explicit and implicit traffic forwarding rules, make the implicit rules always active - add support for egress DSCP rewrite - IS0 VCAP support (Ingress Classification) - IS2 VCAP filters (protos, L3 addrs, L4 ports, flags, ToS etc.) - ES2 VCAP support (Egress Access Control) - support for Per-Stream Filtering and Policing (802.1Q, 8.6.5.1) - Ethernet embedded switches: - Marvell (mv88e6xxx): - add MAB (port auth) offload support - enable PTP receive for mv88e6390 - NXP (ocelot): - support MAC Merge layer - support for the the vsc7512 internal copper phys - Microchip: - lan9303: convert to PHYLINK - lan966x: support TC flower filter statistics - lan937x: PTP support for KSZ9563/KSZ8563 and LAN937x - lan937x: support Credit Based Shaper configuration - ksz9477: support Energy Efficient Ethernet - other: - qca8k: convert to regmap read/write API, use bulk operations - rswitch: Improve TX timestamp accuracy - Intel WiFi (iwlwifi): - EHT (Wi-Fi 7) rate reporting - STEP equalizer support: transfer some STEP (connection to radio on platforms with integrated wifi) related parameters from the BIOS to the firmware. - Qualcomm 802.11ax WiFi (ath11k): - IPQ5018 support - Fine Timing Measurement (FTM) responder role support - channel 177 support - MediaTek WiFi (mt76): - per-PHY LED support - mt7996: EHT (Wi-Fi 7) support - Wireless Ethernet Dispatch (WED) reset support - switch to using page pool allocator - RealTek WiFi (rtw89): - support new version of Bluetooth co-existance - Mobile: - rmnet: support TX aggregation" * tag 'net-next-6.3' of git://git.kernel.org/pub/scm/linux/kernel/git/netdev/net-next: (1872 commits) page_pool: add a comment explaining the fragment counter usage net: ethtool: fix __ethtool_dev_mm_supported() implementation ethtool: pse-pd: Fix double word in comments xsk: add linux/vmalloc.h to xsk.c sefltests: netdevsim: wait for devlink instance after netns removal selftest: fib_tests: Always cleanup before exit net/mlx5e: Align IPsec ASO result memory to be as required by hardware net/mlx5e: TC, Set CT miss to the specific ct action instance net/mlx5e: Rename CHAIN_TO_REG to MAPPED_OBJ_TO_REG net/mlx5: Refactor tc miss handling to a single function net/mlx5: Kconfig: Make tc offload depend on tc skb extension net/sched: flower: Support hardware miss to tc action net/sched: flower: Move filter handle initialization earlier net/sched: cls_api: Support hardware miss to tc action net/sched: Rename user cookie and act cookie sfc: fix builds without CONFIG_RTC_LIB sfc: clean up some inconsistent indentings net/mlx4_en: Introduce flexible array to silence overflow warning net: lan966x: Fix possible deadlock inside PTP net/ulp: Remove redundant ->clone() test in inet_clone_ulp(). ...
Diffstat (limited to 'drivers/net/ethernet/microchip/encx24j600.c')
-rw-r--r--drivers/net/ethernet/microchip/encx24j600.c1127
1 files changed, 1127 insertions, 0 deletions
diff --git a/drivers/net/ethernet/microchip/encx24j600.c b/drivers/net/ethernet/microchip/encx24j600.c
new file mode 100644
index 000000000..d7c8aa77e
--- /dev/null
+++ b/drivers/net/ethernet/microchip/encx24j600.c
@@ -0,0 +1,1127 @@
+// SPDX-License-Identifier: GPL-2.0-or-later
+/*
+ * Microchip ENCX24J600 ethernet driver
+ *
+ * Copyright (C) 2015 Gridpoint
+ * Author: Jon Ringle <jringle@gridpoint.com>
+ */
+
+#include <linux/device.h>
+#include <linux/errno.h>
+#include <linux/etherdevice.h>
+#include <linux/ethtool.h>
+#include <linux/interrupt.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/netdevice.h>
+#include <linux/regmap.h>
+#include <linux/skbuff.h>
+#include <linux/spi/spi.h>
+
+#include "encx24j600_hw.h"
+
+#define DRV_NAME "encx24j600"
+#define DRV_VERSION "1.0"
+
+#define DEFAULT_MSG_ENABLE (NETIF_MSG_DRV | NETIF_MSG_PROBE | NETIF_MSG_LINK)
+static int debug = -1;
+module_param(debug, int, 0000);
+MODULE_PARM_DESC(debug, "Debug level (0=none,...,16=all)");
+
+/* SRAM memory layout:
+ *
+ * 0x0000-0x05ff TX buffers 1.5KB (1*1536) reside in the GP area in SRAM
+ * 0x0600-0x5fff RX buffers 22.5KB (15*1536) reside in the RX area in SRAM
+ */
+#define ENC_TX_BUF_START 0x0000U
+#define ENC_RX_BUF_START 0x0600U
+#define ENC_RX_BUF_END 0x5fffU
+#define ENC_SRAM_SIZE 0x6000U
+
+enum {
+ RXFILTER_NORMAL,
+ RXFILTER_MULTI,
+ RXFILTER_PROMISC
+};
+
+struct encx24j600_priv {
+ struct net_device *ndev;
+ struct mutex lock; /* device access lock */
+ struct encx24j600_context ctx;
+ struct sk_buff *tx_skb;
+ struct task_struct *kworker_task;
+ struct kthread_worker kworker;
+ struct kthread_work tx_work;
+ struct kthread_work setrx_work;
+ u16 next_packet;
+ bool hw_enabled;
+ bool full_duplex;
+ bool autoneg;
+ u16 speed;
+ int rxfilter;
+ u32 msg_enable;
+};
+
+static void dump_packet(const char *msg, int len, const char *data)
+{
+ pr_debug(DRV_NAME ": %s - packet len:%d\n", msg, len);
+ print_hex_dump_bytes("pk data: ", DUMP_PREFIX_OFFSET, data, len);
+}
+
+static void encx24j600_dump_rsv(struct encx24j600_priv *priv, const char *msg,
+ struct rsv *rsv)
+{
+ struct net_device *dev = priv->ndev;
+
+ netdev_info(dev, "RX packet Len:%d\n", rsv->len);
+ netdev_dbg(dev, "%s - NextPk: 0x%04x\n", msg,
+ rsv->next_packet);
+ netdev_dbg(dev, "RxOK: %d, DribbleNibble: %d\n",
+ RSV_GETBIT(rsv->rxstat, RSV_RXOK),
+ RSV_GETBIT(rsv->rxstat, RSV_DRIBBLENIBBLE));
+ netdev_dbg(dev, "CRCErr:%d, LenChkErr: %d, LenOutOfRange: %d\n",
+ RSV_GETBIT(rsv->rxstat, RSV_CRCERROR),
+ RSV_GETBIT(rsv->rxstat, RSV_LENCHECKERR),
+ RSV_GETBIT(rsv->rxstat, RSV_LENOUTOFRANGE));
+ netdev_dbg(dev, "Multicast: %d, Broadcast: %d, LongDropEvent: %d, CarrierEvent: %d\n",
+ RSV_GETBIT(rsv->rxstat, RSV_RXMULTICAST),
+ RSV_GETBIT(rsv->rxstat, RSV_RXBROADCAST),
+ RSV_GETBIT(rsv->rxstat, RSV_RXLONGEVDROPEV),
+ RSV_GETBIT(rsv->rxstat, RSV_CARRIEREV));
+ netdev_dbg(dev, "ControlFrame: %d, PauseFrame: %d, UnknownOp: %d, VLanTagFrame: %d\n",
+ RSV_GETBIT(rsv->rxstat, RSV_RXCONTROLFRAME),
+ RSV_GETBIT(rsv->rxstat, RSV_RXPAUSEFRAME),
+ RSV_GETBIT(rsv->rxstat, RSV_RXUNKNOWNOPCODE),
+ RSV_GETBIT(rsv->rxstat, RSV_RXTYPEVLAN));
+}
+
+static u16 encx24j600_read_reg(struct encx24j600_priv *priv, u8 reg)
+{
+ struct net_device *dev = priv->ndev;
+ unsigned int val = 0;
+ int ret = regmap_read(priv->ctx.regmap, reg, &val);
+
+ if (unlikely(ret))
+ netif_err(priv, drv, dev, "%s: error %d reading reg %02x\n",
+ __func__, ret, reg);
+ return val;
+}
+
+static void encx24j600_write_reg(struct encx24j600_priv *priv, u8 reg, u16 val)
+{
+ struct net_device *dev = priv->ndev;
+ int ret = regmap_write(priv->ctx.regmap, reg, val);
+
+ if (unlikely(ret))
+ netif_err(priv, drv, dev, "%s: error %d writing reg %02x=%04x\n",
+ __func__, ret, reg, val);
+}
+
+static void encx24j600_update_reg(struct encx24j600_priv *priv, u8 reg,
+ u16 mask, u16 val)
+{
+ struct net_device *dev = priv->ndev;
+ int ret = regmap_update_bits(priv->ctx.regmap, reg, mask, val);
+
+ if (unlikely(ret))
+ netif_err(priv, drv, dev, "%s: error %d updating reg %02x=%04x~%04x\n",
+ __func__, ret, reg, val, mask);
+}
+
+static u16 encx24j600_read_phy(struct encx24j600_priv *priv, u8 reg)
+{
+ struct net_device *dev = priv->ndev;
+ unsigned int val = 0;
+ int ret = regmap_read(priv->ctx.phymap, reg, &val);
+
+ if (unlikely(ret))
+ netif_err(priv, drv, dev, "%s: error %d reading %02x\n",
+ __func__, ret, reg);
+ return val;
+}
+
+static void encx24j600_write_phy(struct encx24j600_priv *priv, u8 reg, u16 val)
+{
+ struct net_device *dev = priv->ndev;
+ int ret = regmap_write(priv->ctx.phymap, reg, val);
+
+ if (unlikely(ret))
+ netif_err(priv, drv, dev, "%s: error %d writing reg %02x=%04x\n",
+ __func__, ret, reg, val);
+}
+
+static void encx24j600_clr_bits(struct encx24j600_priv *priv, u8 reg, u16 mask)
+{
+ encx24j600_update_reg(priv, reg, mask, 0);
+}
+
+static void encx24j600_set_bits(struct encx24j600_priv *priv, u8 reg, u16 mask)
+{
+ encx24j600_update_reg(priv, reg, mask, mask);
+}
+
+static void encx24j600_cmd(struct encx24j600_priv *priv, u8 cmd)
+{
+ struct net_device *dev = priv->ndev;
+ int ret = regmap_write(priv->ctx.regmap, cmd, 0);
+
+ if (unlikely(ret))
+ netif_err(priv, drv, dev, "%s: error %d with cmd %02x\n",
+ __func__, ret, cmd);
+}
+
+static int encx24j600_raw_read(struct encx24j600_priv *priv, u8 reg, u8 *data,
+ size_t count)
+{
+ int ret;
+
+ mutex_lock(&priv->ctx.mutex);
+ ret = regmap_encx24j600_spi_read(&priv->ctx, reg, data, count);
+ mutex_unlock(&priv->ctx.mutex);
+
+ return ret;
+}
+
+static int encx24j600_raw_write(struct encx24j600_priv *priv, u8 reg,
+ const u8 *data, size_t count)
+{
+ int ret;
+
+ mutex_lock(&priv->ctx.mutex);
+ ret = regmap_encx24j600_spi_write(&priv->ctx, reg, data, count);
+ mutex_unlock(&priv->ctx.mutex);
+
+ return ret;
+}
+
+static void encx24j600_update_phcon1(struct encx24j600_priv *priv)
+{
+ u16 phcon1 = encx24j600_read_phy(priv, PHCON1);
+
+ if (priv->autoneg == AUTONEG_ENABLE) {
+ phcon1 |= ANEN | RENEG;
+ } else {
+ phcon1 &= ~ANEN;
+ if (priv->speed == SPEED_100)
+ phcon1 |= SPD100;
+ else
+ phcon1 &= ~SPD100;
+
+ if (priv->full_duplex)
+ phcon1 |= PFULDPX;
+ else
+ phcon1 &= ~PFULDPX;
+ }
+ encx24j600_write_phy(priv, PHCON1, phcon1);
+}
+
+/* Waits for autonegotiation to complete. */
+static int encx24j600_wait_for_autoneg(struct encx24j600_priv *priv)
+{
+ struct net_device *dev = priv->ndev;
+ unsigned long timeout = jiffies + msecs_to_jiffies(2000);
+ u16 phstat1;
+ u16 estat;
+
+ phstat1 = encx24j600_read_phy(priv, PHSTAT1);
+ while ((phstat1 & ANDONE) == 0) {
+ if (time_after(jiffies, timeout)) {
+ u16 phstat3;
+
+ netif_notice(priv, drv, dev, "timeout waiting for autoneg done\n");
+
+ priv->autoneg = AUTONEG_DISABLE;
+ phstat3 = encx24j600_read_phy(priv, PHSTAT3);
+ priv->speed = (phstat3 & PHY3SPD100)
+ ? SPEED_100 : SPEED_10;
+ priv->full_duplex = (phstat3 & PHY3DPX) ? 1 : 0;
+ encx24j600_update_phcon1(priv);
+ netif_notice(priv, drv, dev, "Using parallel detection: %s/%s",
+ priv->speed == SPEED_100 ? "100" : "10",
+ priv->full_duplex ? "Full" : "Half");
+
+ return -ETIMEDOUT;
+ }
+ cpu_relax();
+ phstat1 = encx24j600_read_phy(priv, PHSTAT1);
+ }
+
+ estat = encx24j600_read_reg(priv, ESTAT);
+ if (estat & PHYDPX) {
+ encx24j600_set_bits(priv, MACON2, FULDPX);
+ encx24j600_write_reg(priv, MABBIPG, 0x15);
+ } else {
+ encx24j600_clr_bits(priv, MACON2, FULDPX);
+ encx24j600_write_reg(priv, MABBIPG, 0x12);
+ /* Max retransmittions attempt */
+ encx24j600_write_reg(priv, MACLCON, 0x370f);
+ }
+
+ return 0;
+}
+
+/* Access the PHY to determine link status */
+static void encx24j600_check_link_status(struct encx24j600_priv *priv)
+{
+ struct net_device *dev = priv->ndev;
+ u16 estat;
+
+ estat = encx24j600_read_reg(priv, ESTAT);
+
+ if (estat & PHYLNK) {
+ if (priv->autoneg == AUTONEG_ENABLE)
+ encx24j600_wait_for_autoneg(priv);
+
+ netif_carrier_on(dev);
+ netif_info(priv, ifup, dev, "link up\n");
+ } else {
+ netif_info(priv, ifdown, dev, "link down\n");
+
+ /* Re-enable autoneg since we won't know what we might be
+ * connected to when the link is brought back up again.
+ */
+ priv->autoneg = AUTONEG_ENABLE;
+ priv->full_duplex = true;
+ priv->speed = SPEED_100;
+ netif_carrier_off(dev);
+ }
+}
+
+static void encx24j600_int_link_handler(struct encx24j600_priv *priv)
+{
+ struct net_device *dev = priv->ndev;
+
+ netif_dbg(priv, intr, dev, "%s", __func__);
+ encx24j600_check_link_status(priv);
+ encx24j600_clr_bits(priv, EIR, LINKIF);
+}
+
+static void encx24j600_tx_complete(struct encx24j600_priv *priv, bool err)
+{
+ struct net_device *dev = priv->ndev;
+
+ if (!priv->tx_skb) {
+ BUG();
+ return;
+ }
+
+ mutex_lock(&priv->lock);
+
+ if (err)
+ dev->stats.tx_errors++;
+ else
+ dev->stats.tx_packets++;
+
+ dev->stats.tx_bytes += priv->tx_skb->len;
+
+ encx24j600_clr_bits(priv, EIR, TXIF | TXABTIF);
+
+ netif_dbg(priv, tx_done, dev, "TX Done%s\n", err ? ": Err" : "");
+
+ dev_kfree_skb(priv->tx_skb);
+ priv->tx_skb = NULL;
+
+ netif_wake_queue(dev);
+
+ mutex_unlock(&priv->lock);
+}
+
+static int encx24j600_receive_packet(struct encx24j600_priv *priv,
+ struct rsv *rsv)
+{
+ struct net_device *dev = priv->ndev;
+ struct sk_buff *skb = netdev_alloc_skb(dev, rsv->len + NET_IP_ALIGN);
+
+ if (!skb) {
+ pr_err_ratelimited("RX: OOM: packet dropped\n");
+ dev->stats.rx_dropped++;
+ return -ENOMEM;
+ }
+ skb_reserve(skb, NET_IP_ALIGN);
+ encx24j600_raw_read(priv, RRXDATA, skb_put(skb, rsv->len), rsv->len);
+
+ if (netif_msg_pktdata(priv))
+ dump_packet("RX", skb->len, skb->data);
+
+ skb->dev = dev;
+ skb->protocol = eth_type_trans(skb, dev);
+ skb->ip_summed = CHECKSUM_COMPLETE;
+
+ /* Maintain stats */
+ dev->stats.rx_packets++;
+ dev->stats.rx_bytes += rsv->len;
+
+ netif_rx(skb);
+
+ return 0;
+}
+
+static void encx24j600_rx_packets(struct encx24j600_priv *priv, u8 packet_count)
+{
+ struct net_device *dev = priv->ndev;
+
+ while (packet_count--) {
+ struct rsv rsv;
+ u16 newrxtail;
+
+ encx24j600_write_reg(priv, ERXRDPT, priv->next_packet);
+ encx24j600_raw_read(priv, RRXDATA, (u8 *)&rsv, sizeof(rsv));
+
+ if (netif_msg_rx_status(priv))
+ encx24j600_dump_rsv(priv, __func__, &rsv);
+
+ if (!RSV_GETBIT(rsv.rxstat, RSV_RXOK) ||
+ (rsv.len > MAX_FRAMELEN)) {
+ netif_err(priv, rx_err, dev, "RX Error %04x\n",
+ rsv.rxstat);
+ dev->stats.rx_errors++;
+
+ if (RSV_GETBIT(rsv.rxstat, RSV_CRCERROR))
+ dev->stats.rx_crc_errors++;
+ if (RSV_GETBIT(rsv.rxstat, RSV_LENCHECKERR))
+ dev->stats.rx_frame_errors++;
+ if (rsv.len > MAX_FRAMELEN)
+ dev->stats.rx_over_errors++;
+ } else {
+ encx24j600_receive_packet(priv, &rsv);
+ }
+
+ priv->next_packet = rsv.next_packet;
+
+ newrxtail = priv->next_packet - 2;
+ if (newrxtail == ENC_RX_BUF_START)
+ newrxtail = SRAM_SIZE - 2;
+
+ encx24j600_cmd(priv, SETPKTDEC);
+ encx24j600_write_reg(priv, ERXTAIL, newrxtail);
+ }
+}
+
+static irqreturn_t encx24j600_isr(int irq, void *dev_id)
+{
+ struct encx24j600_priv *priv = dev_id;
+ struct net_device *dev = priv->ndev;
+ int eir;
+
+ /* Clear interrupts */
+ encx24j600_cmd(priv, CLREIE);
+
+ eir = encx24j600_read_reg(priv, EIR);
+
+ if (eir & LINKIF)
+ encx24j600_int_link_handler(priv);
+
+ if (eir & TXIF)
+ encx24j600_tx_complete(priv, false);
+
+ if (eir & TXABTIF)
+ encx24j600_tx_complete(priv, true);
+
+ if (eir & RXABTIF) {
+ if (eir & PCFULIF) {
+ /* Packet counter is full */
+ netif_err(priv, rx_err, dev, "Packet counter full\n");
+ }
+ dev->stats.rx_dropped++;
+ encx24j600_clr_bits(priv, EIR, RXABTIF);
+ }
+
+ if (eir & PKTIF) {
+ u8 packet_count;
+
+ mutex_lock(&priv->lock);
+
+ packet_count = encx24j600_read_reg(priv, ESTAT) & 0xff;
+ while (packet_count) {
+ encx24j600_rx_packets(priv, packet_count);
+ packet_count = encx24j600_read_reg(priv, ESTAT) & 0xff;
+ }
+
+ mutex_unlock(&priv->lock);
+ }
+
+ /* Enable interrupts */
+ encx24j600_cmd(priv, SETEIE);
+
+ return IRQ_HANDLED;
+}
+
+static int encx24j600_soft_reset(struct encx24j600_priv *priv)
+{
+ int ret = 0;
+ int timeout;
+ u16 eudast;
+
+ /* Write and verify a test value to EUDAST */
+ regcache_cache_bypass(priv->ctx.regmap, true);
+ timeout = 10;
+ do {
+ encx24j600_write_reg(priv, EUDAST, EUDAST_TEST_VAL);
+ eudast = encx24j600_read_reg(priv, EUDAST);
+ usleep_range(25, 100);
+ } while ((eudast != EUDAST_TEST_VAL) && --timeout);
+ regcache_cache_bypass(priv->ctx.regmap, false);
+
+ if (timeout == 0) {
+ ret = -ETIMEDOUT;
+ goto err_out;
+ }
+
+ /* Wait for CLKRDY to become set */
+ timeout = 10;
+ while (!(encx24j600_read_reg(priv, ESTAT) & CLKRDY) && --timeout)
+ usleep_range(25, 100);
+
+ if (timeout == 0) {
+ ret = -ETIMEDOUT;
+ goto err_out;
+ }
+
+ /* Issue a System Reset command */
+ encx24j600_cmd(priv, SETETHRST);
+ usleep_range(25, 100);
+
+ /* Confirm that EUDAST has 0000h after system reset */
+ if (encx24j600_read_reg(priv, EUDAST) != 0) {
+ ret = -EINVAL;
+ goto err_out;
+ }
+
+ /* Wait for PHY register and status bits to become available */
+ usleep_range(256, 1000);
+
+err_out:
+ return ret;
+}
+
+static int encx24j600_hw_reset(struct encx24j600_priv *priv)
+{
+ int ret;
+
+ mutex_lock(&priv->lock);
+ ret = encx24j600_soft_reset(priv);
+ mutex_unlock(&priv->lock);
+
+ return ret;
+}
+
+static void encx24j600_reset_hw_tx(struct encx24j600_priv *priv)
+{
+ encx24j600_set_bits(priv, ECON2, TXRST);
+ encx24j600_clr_bits(priv, ECON2, TXRST);
+}
+
+static void encx24j600_hw_init_tx(struct encx24j600_priv *priv)
+{
+ /* Reset TX */
+ encx24j600_reset_hw_tx(priv);
+
+ /* Clear the TXIF flag if were previously set */
+ encx24j600_clr_bits(priv, EIR, TXIF | TXABTIF);
+
+ /* Write the Tx Buffer pointer */
+ encx24j600_write_reg(priv, EGPWRPT, ENC_TX_BUF_START);
+}
+
+static void encx24j600_hw_init_rx(struct encx24j600_priv *priv)
+{
+ encx24j600_cmd(priv, DISABLERX);
+
+ /* Set up RX packet start address in the SRAM */
+ encx24j600_write_reg(priv, ERXST, ENC_RX_BUF_START);
+
+ /* Preload the RX Data pointer to the beginning of the RX area */
+ encx24j600_write_reg(priv, ERXRDPT, ENC_RX_BUF_START);
+
+ priv->next_packet = ENC_RX_BUF_START;
+
+ /* Set up RX end address in the SRAM */
+ encx24j600_write_reg(priv, ERXTAIL, ENC_SRAM_SIZE - 2);
+
+ /* Reset the user data pointers */
+ encx24j600_write_reg(priv, EUDAST, ENC_SRAM_SIZE);
+ encx24j600_write_reg(priv, EUDAND, ENC_SRAM_SIZE + 1);
+
+ /* Set Max Frame length */
+ encx24j600_write_reg(priv, MAMXFL, MAX_FRAMELEN);
+}
+
+static void encx24j600_dump_config(struct encx24j600_priv *priv,
+ const char *msg)
+{
+ pr_info(DRV_NAME ": %s\n", msg);
+
+ /* CHIP configuration */
+ pr_info(DRV_NAME " ECON1: %04X\n", encx24j600_read_reg(priv, ECON1));
+ pr_info(DRV_NAME " ECON2: %04X\n", encx24j600_read_reg(priv, ECON2));
+ pr_info(DRV_NAME " ERXFCON: %04X\n", encx24j600_read_reg(priv,
+ ERXFCON));
+ pr_info(DRV_NAME " ESTAT: %04X\n", encx24j600_read_reg(priv, ESTAT));
+ pr_info(DRV_NAME " EIR: %04X\n", encx24j600_read_reg(priv, EIR));
+ pr_info(DRV_NAME " EIDLED: %04X\n", encx24j600_read_reg(priv, EIDLED));
+
+ /* MAC layer configuration */
+ pr_info(DRV_NAME " MACON1: %04X\n", encx24j600_read_reg(priv, MACON1));
+ pr_info(DRV_NAME " MACON2: %04X\n", encx24j600_read_reg(priv, MACON2));
+ pr_info(DRV_NAME " MAIPG: %04X\n", encx24j600_read_reg(priv, MAIPG));
+ pr_info(DRV_NAME " MACLCON: %04X\n", encx24j600_read_reg(priv,
+ MACLCON));
+ pr_info(DRV_NAME " MABBIPG: %04X\n", encx24j600_read_reg(priv,
+ MABBIPG));
+
+ /* PHY configuation */
+ pr_info(DRV_NAME " PHCON1: %04X\n", encx24j600_read_phy(priv, PHCON1));
+ pr_info(DRV_NAME " PHCON2: %04X\n", encx24j600_read_phy(priv, PHCON2));
+ pr_info(DRV_NAME " PHANA: %04X\n", encx24j600_read_phy(priv, PHANA));
+ pr_info(DRV_NAME " PHANLPA: %04X\n", encx24j600_read_phy(priv,
+ PHANLPA));
+ pr_info(DRV_NAME " PHANE: %04X\n", encx24j600_read_phy(priv, PHANE));
+ pr_info(DRV_NAME " PHSTAT1: %04X\n", encx24j600_read_phy(priv,
+ PHSTAT1));
+ pr_info(DRV_NAME " PHSTAT2: %04X\n", encx24j600_read_phy(priv,
+ PHSTAT2));
+ pr_info(DRV_NAME " PHSTAT3: %04X\n", encx24j600_read_phy(priv,
+ PHSTAT3));
+}
+
+static void encx24j600_set_rxfilter_mode(struct encx24j600_priv *priv)
+{
+ switch (priv->rxfilter) {
+ case RXFILTER_PROMISC:
+ encx24j600_set_bits(priv, MACON1, PASSALL);
+ encx24j600_write_reg(priv, ERXFCON, UCEN | MCEN | NOTMEEN);
+ break;
+ case RXFILTER_MULTI:
+ encx24j600_clr_bits(priv, MACON1, PASSALL);
+ encx24j600_write_reg(priv, ERXFCON, UCEN | CRCEN | BCEN | MCEN);
+ break;
+ case RXFILTER_NORMAL:
+ default:
+ encx24j600_clr_bits(priv, MACON1, PASSALL);
+ encx24j600_write_reg(priv, ERXFCON, UCEN | CRCEN | BCEN);
+ break;
+ }
+}
+
+static void encx24j600_hw_init(struct encx24j600_priv *priv)
+{
+ u16 macon2;
+
+ priv->hw_enabled = false;
+
+ /* PHY Leds: link status,
+ * LEDA: Link State + collision events
+ * LEDB: Link State + transmit/receive events
+ */
+ encx24j600_update_reg(priv, EIDLED, 0xff00, 0xcb00);
+
+ /* Loopback disabled */
+ encx24j600_write_reg(priv, MACON1, 0x9);
+
+ /* interpacket gap value */
+ encx24j600_write_reg(priv, MAIPG, 0x0c12);
+
+ /* Write the auto negotiation pattern */
+ encx24j600_write_phy(priv, PHANA, PHANA_DEFAULT);
+
+ encx24j600_update_phcon1(priv);
+ encx24j600_check_link_status(priv);
+
+ macon2 = MACON2_RSV1 | TXCRCEN | PADCFG0 | PADCFG2 | MACON2_DEFER;
+ if ((priv->autoneg == AUTONEG_DISABLE) && priv->full_duplex)
+ macon2 |= FULDPX;
+
+ encx24j600_set_bits(priv, MACON2, macon2);
+
+ priv->rxfilter = RXFILTER_NORMAL;
+ encx24j600_set_rxfilter_mode(priv);
+
+ /* Program the Maximum frame length */
+ encx24j600_write_reg(priv, MAMXFL, MAX_FRAMELEN);
+
+ /* Init Tx pointers */
+ encx24j600_hw_init_tx(priv);
+
+ /* Init Rx pointers */
+ encx24j600_hw_init_rx(priv);
+
+ if (netif_msg_hw(priv))
+ encx24j600_dump_config(priv, "Hw is initialized");
+}
+
+static void encx24j600_hw_enable(struct encx24j600_priv *priv)
+{
+ /* Clear the interrupt flags in case was set */
+ encx24j600_clr_bits(priv, EIR, (PCFULIF | RXABTIF | TXABTIF | TXIF |
+ PKTIF | LINKIF));
+
+ /* Enable the interrupts */
+ encx24j600_write_reg(priv, EIE, (PCFULIE | RXABTIE | TXABTIE | TXIE |
+ PKTIE | LINKIE | INTIE));
+
+ /* Enable RX */
+ encx24j600_cmd(priv, ENABLERX);
+
+ priv->hw_enabled = true;
+}
+
+static void encx24j600_hw_disable(struct encx24j600_priv *priv)
+{
+ /* Disable all interrupts */
+ encx24j600_write_reg(priv, EIE, 0);
+
+ /* Disable RX */
+ encx24j600_cmd(priv, DISABLERX);
+
+ priv->hw_enabled = false;
+}
+
+static int encx24j600_setlink(struct net_device *dev, u8 autoneg, u16 speed,
+ u8 duplex)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+ int ret = 0;
+
+ if (!priv->hw_enabled) {
+ /* link is in low power mode now; duplex setting
+ * will take effect on next encx24j600_hw_init()
+ */
+ if (speed == SPEED_10 || speed == SPEED_100) {
+ priv->autoneg = (autoneg == AUTONEG_ENABLE);
+ priv->full_duplex = (duplex == DUPLEX_FULL);
+ priv->speed = (speed == SPEED_100);
+ } else {
+ netif_warn(priv, link, dev, "unsupported link speed setting\n");
+ /*speeds other than SPEED_10 and SPEED_100 */
+ /*are not supported by chip */
+ ret = -EOPNOTSUPP;
+ }
+ } else {
+ netif_warn(priv, link, dev, "Warning: hw must be disabled to set link mode\n");
+ ret = -EBUSY;
+ }
+ return ret;
+}
+
+static void encx24j600_hw_get_macaddr(struct encx24j600_priv *priv,
+ unsigned char *ethaddr)
+{
+ unsigned short val;
+
+ val = encx24j600_read_reg(priv, MAADR1);
+
+ ethaddr[0] = val & 0x00ff;
+ ethaddr[1] = (val & 0xff00) >> 8;
+
+ val = encx24j600_read_reg(priv, MAADR2);
+
+ ethaddr[2] = val & 0x00ffU;
+ ethaddr[3] = (val & 0xff00U) >> 8;
+
+ val = encx24j600_read_reg(priv, MAADR3);
+
+ ethaddr[4] = val & 0x00ffU;
+ ethaddr[5] = (val & 0xff00U) >> 8;
+}
+
+/* Program the hardware MAC address from dev->dev_addr.*/
+static int encx24j600_set_hw_macaddr(struct net_device *dev)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+
+ if (priv->hw_enabled) {
+ netif_info(priv, drv, dev, "Hardware must be disabled to set Mac address\n");
+ return -EBUSY;
+ }
+
+ mutex_lock(&priv->lock);
+
+ netif_info(priv, drv, dev, "%s: Setting MAC address to %pM\n",
+ dev->name, dev->dev_addr);
+
+ encx24j600_write_reg(priv, MAADR3, (dev->dev_addr[4] |
+ dev->dev_addr[5] << 8));
+ encx24j600_write_reg(priv, MAADR2, (dev->dev_addr[2] |
+ dev->dev_addr[3] << 8));
+ encx24j600_write_reg(priv, MAADR1, (dev->dev_addr[0] |
+ dev->dev_addr[1] << 8));
+
+ mutex_unlock(&priv->lock);
+
+ return 0;
+}
+
+/* Store the new hardware address in dev->dev_addr, and update the MAC.*/
+static int encx24j600_set_mac_address(struct net_device *dev, void *addr)
+{
+ struct sockaddr *address = addr;
+
+ if (netif_running(dev))
+ return -EBUSY;
+ if (!is_valid_ether_addr(address->sa_data))
+ return -EADDRNOTAVAIL;
+
+ eth_hw_addr_set(dev, address->sa_data);
+ return encx24j600_set_hw_macaddr(dev);
+}
+
+static int encx24j600_open(struct net_device *dev)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+
+ int ret = request_threaded_irq(priv->ctx.spi->irq, NULL, encx24j600_isr,
+ IRQF_TRIGGER_FALLING | IRQF_ONESHOT,
+ DRV_NAME, priv);
+ if (unlikely(ret < 0)) {
+ netdev_err(dev, "request irq %d failed (ret = %d)\n",
+ priv->ctx.spi->irq, ret);
+ return ret;
+ }
+
+ encx24j600_hw_disable(priv);
+ encx24j600_hw_init(priv);
+ encx24j600_hw_enable(priv);
+ netif_start_queue(dev);
+
+ return 0;
+}
+
+static int encx24j600_stop(struct net_device *dev)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+
+ netif_stop_queue(dev);
+ free_irq(priv->ctx.spi->irq, priv);
+ return 0;
+}
+
+static void encx24j600_setrx_proc(struct kthread_work *ws)
+{
+ struct encx24j600_priv *priv =
+ container_of(ws, struct encx24j600_priv, setrx_work);
+
+ mutex_lock(&priv->lock);
+ encx24j600_set_rxfilter_mode(priv);
+ mutex_unlock(&priv->lock);
+}
+
+static void encx24j600_set_multicast_list(struct net_device *dev)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+ int oldfilter = priv->rxfilter;
+
+ if (dev->flags & IFF_PROMISC) {
+ netif_dbg(priv, link, dev, "promiscuous mode\n");
+ priv->rxfilter = RXFILTER_PROMISC;
+ } else if ((dev->flags & IFF_ALLMULTI) || !netdev_mc_empty(dev)) {
+ netif_dbg(priv, link, dev, "%smulticast mode\n",
+ (dev->flags & IFF_ALLMULTI) ? "all-" : "");
+ priv->rxfilter = RXFILTER_MULTI;
+ } else {
+ netif_dbg(priv, link, dev, "normal mode\n");
+ priv->rxfilter = RXFILTER_NORMAL;
+ }
+
+ if (oldfilter != priv->rxfilter)
+ kthread_queue_work(&priv->kworker, &priv->setrx_work);
+}
+
+static void encx24j600_hw_tx(struct encx24j600_priv *priv)
+{
+ struct net_device *dev = priv->ndev;
+
+ netif_info(priv, tx_queued, dev, "TX Packet Len:%d\n",
+ priv->tx_skb->len);
+
+ if (netif_msg_pktdata(priv))
+ dump_packet("TX", priv->tx_skb->len, priv->tx_skb->data);
+
+ if (encx24j600_read_reg(priv, EIR) & TXABTIF)
+ /* Last transmition aborted due to error. Reset TX interface */
+ encx24j600_reset_hw_tx(priv);
+
+ /* Clear the TXIF flag if were previously set */
+ encx24j600_clr_bits(priv, EIR, TXIF);
+
+ /* Set the data pointer to the TX buffer address in the SRAM */
+ encx24j600_write_reg(priv, EGPWRPT, ENC_TX_BUF_START);
+
+ /* Copy the packet into the SRAM */
+ encx24j600_raw_write(priv, WGPDATA, (u8 *)priv->tx_skb->data,
+ priv->tx_skb->len);
+
+ /* Program the Tx buffer start pointer */
+ encx24j600_write_reg(priv, ETXST, ENC_TX_BUF_START);
+
+ /* Program the packet length */
+ encx24j600_write_reg(priv, ETXLEN, priv->tx_skb->len);
+
+ /* Start the transmission */
+ encx24j600_cmd(priv, SETTXRTS);
+}
+
+static void encx24j600_tx_proc(struct kthread_work *ws)
+{
+ struct encx24j600_priv *priv =
+ container_of(ws, struct encx24j600_priv, tx_work);
+
+ mutex_lock(&priv->lock);
+ encx24j600_hw_tx(priv);
+ mutex_unlock(&priv->lock);
+}
+
+static netdev_tx_t encx24j600_tx(struct sk_buff *skb, struct net_device *dev)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+
+ netif_stop_queue(dev);
+
+ /* save the timestamp */
+ netif_trans_update(dev);
+
+ /* Remember the skb for deferred processing */
+ priv->tx_skb = skb;
+
+ kthread_queue_work(&priv->kworker, &priv->tx_work);
+
+ return NETDEV_TX_OK;
+}
+
+/* Deal with a transmit timeout */
+static void encx24j600_tx_timeout(struct net_device *dev, unsigned int txqueue)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+
+ netif_err(priv, tx_err, dev, "TX timeout at %ld, latency %ld\n",
+ jiffies, jiffies - dev_trans_start(dev));
+
+ dev->stats.tx_errors++;
+ netif_wake_queue(dev);
+}
+
+static int encx24j600_get_regs_len(struct net_device *dev)
+{
+ return SFR_REG_COUNT;
+}
+
+static void encx24j600_get_regs(struct net_device *dev,
+ struct ethtool_regs *regs, void *p)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+ u16 *buff = p;
+ u8 reg;
+
+ regs->version = 1;
+ mutex_lock(&priv->lock);
+ for (reg = 0; reg < SFR_REG_COUNT; reg += 2) {
+ unsigned int val = 0;
+ /* ignore errors for unreadable registers */
+ regmap_read(priv->ctx.regmap, reg, &val);
+ buff[reg] = val & 0xffff;
+ }
+ mutex_unlock(&priv->lock);
+}
+
+static void encx24j600_get_drvinfo(struct net_device *dev,
+ struct ethtool_drvinfo *info)
+{
+ strscpy(info->driver, DRV_NAME, sizeof(info->driver));
+ strscpy(info->version, DRV_VERSION, sizeof(info->version));
+ strscpy(info->bus_info, dev_name(dev->dev.parent),
+ sizeof(info->bus_info));
+}
+
+static int encx24j600_get_link_ksettings(struct net_device *dev,
+ struct ethtool_link_ksettings *cmd)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+ u32 supported;
+
+ supported = SUPPORTED_10baseT_Half | SUPPORTED_10baseT_Full |
+ SUPPORTED_100baseT_Half | SUPPORTED_100baseT_Full |
+ SUPPORTED_Autoneg | SUPPORTED_TP;
+
+ ethtool_convert_legacy_u32_to_link_mode(cmd->link_modes.supported,
+ supported);
+
+ cmd->base.speed = priv->speed;
+ cmd->base.duplex = priv->full_duplex ? DUPLEX_FULL : DUPLEX_HALF;
+ cmd->base.port = PORT_TP;
+ cmd->base.autoneg = priv->autoneg ? AUTONEG_ENABLE : AUTONEG_DISABLE;
+
+ return 0;
+}
+
+static int
+encx24j600_set_link_ksettings(struct net_device *dev,
+ const struct ethtool_link_ksettings *cmd)
+{
+ return encx24j600_setlink(dev, cmd->base.autoneg,
+ cmd->base.speed, cmd->base.duplex);
+}
+
+static u32 encx24j600_get_msglevel(struct net_device *dev)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+
+ return priv->msg_enable;
+}
+
+static void encx24j600_set_msglevel(struct net_device *dev, u32 val)
+{
+ struct encx24j600_priv *priv = netdev_priv(dev);
+
+ priv->msg_enable = val;
+}
+
+static const struct ethtool_ops encx24j600_ethtool_ops = {
+ .get_drvinfo = encx24j600_get_drvinfo,
+ .get_msglevel = encx24j600_get_msglevel,
+ .set_msglevel = encx24j600_set_msglevel,
+ .get_regs_len = encx24j600_get_regs_len,
+ .get_regs = encx24j600_get_regs,
+ .get_link_ksettings = encx24j600_get_link_ksettings,
+ .set_link_ksettings = encx24j600_set_link_ksettings,
+};
+
+static const struct net_device_ops encx24j600_netdev_ops = {
+ .ndo_open = encx24j600_open,
+ .ndo_stop = encx24j600_stop,
+ .ndo_start_xmit = encx24j600_tx,
+ .ndo_set_rx_mode = encx24j600_set_multicast_list,
+ .ndo_set_mac_address = encx24j600_set_mac_address,
+ .ndo_tx_timeout = encx24j600_tx_timeout,
+ .ndo_validate_addr = eth_validate_addr,
+};
+
+static int encx24j600_spi_probe(struct spi_device *spi)
+{
+ int ret;
+
+ struct net_device *ndev;
+ struct encx24j600_priv *priv;
+ u16 eidled;
+ u8 addr[ETH_ALEN];
+
+ ndev = alloc_etherdev(sizeof(struct encx24j600_priv));
+
+ if (!ndev) {
+ ret = -ENOMEM;
+ goto error_out;
+ }
+
+ priv = netdev_priv(ndev);
+ spi_set_drvdata(spi, priv);
+ dev_set_drvdata(&spi->dev, priv);
+ SET_NETDEV_DEV(ndev, &spi->dev);
+
+ priv->msg_enable = netif_msg_init(debug, DEFAULT_MSG_ENABLE);
+ priv->ndev = ndev;
+
+ /* Default configuration PHY configuration */
+ priv->full_duplex = true;
+ priv->autoneg = AUTONEG_ENABLE;
+ priv->speed = SPEED_100;
+
+ priv->ctx.spi = spi;
+ ndev->irq = spi->irq;
+ ndev->netdev_ops = &encx24j600_netdev_ops;
+
+ ret = devm_regmap_init_encx24j600(&spi->dev, &priv->ctx);
+ if (ret)
+ goto out_free;
+
+ mutex_init(&priv->lock);
+
+ /* Reset device and check if it is connected */
+ if (encx24j600_hw_reset(priv)) {
+ netif_err(priv, probe, ndev,
+ DRV_NAME ": Chip is not detected\n");
+ ret = -EIO;
+ goto out_free;
+ }
+
+ /* Initialize the device HW to the consistent state */
+ encx24j600_hw_init(priv);
+
+ kthread_init_worker(&priv->kworker);
+ kthread_init_work(&priv->tx_work, encx24j600_tx_proc);
+ kthread_init_work(&priv->setrx_work, encx24j600_setrx_proc);
+
+ priv->kworker_task = kthread_run(kthread_worker_fn, &priv->kworker,
+ "encx24j600");
+
+ if (IS_ERR(priv->kworker_task)) {
+ ret = PTR_ERR(priv->kworker_task);
+ goto out_free;
+ }
+
+ /* Get the MAC address from the chip */
+ encx24j600_hw_get_macaddr(priv, addr);
+ eth_hw_addr_set(ndev, addr);
+
+ ndev->ethtool_ops = &encx24j600_ethtool_ops;
+
+ ret = register_netdev(ndev);
+ if (unlikely(ret)) {
+ netif_err(priv, probe, ndev, "Error %d initializing card encx24j600 card\n",
+ ret);
+ goto out_stop;
+ }
+
+ eidled = encx24j600_read_reg(priv, EIDLED);
+ if (((eidled & DEVID_MASK) >> DEVID_SHIFT) != ENCX24J600_DEV_ID) {
+ ret = -EINVAL;
+ goto out_unregister;
+ }
+
+ netif_info(priv, probe, ndev, "Silicon rev ID: 0x%02x\n",
+ (eidled & REVID_MASK) >> REVID_SHIFT);
+
+ netif_info(priv, drv, priv->ndev, "MAC address %pM\n", ndev->dev_addr);
+
+ return ret;
+
+out_unregister:
+ unregister_netdev(priv->ndev);
+out_stop:
+ kthread_stop(priv->kworker_task);
+out_free:
+ free_netdev(ndev);
+
+error_out:
+ return ret;
+}
+
+static void encx24j600_spi_remove(struct spi_device *spi)
+{
+ struct encx24j600_priv *priv = dev_get_drvdata(&spi->dev);
+
+ unregister_netdev(priv->ndev);
+ kthread_stop(priv->kworker_task);
+
+ free_netdev(priv->ndev);
+}
+
+static const struct spi_device_id encx24j600_spi_id_table[] = {
+ { .name = "encx24j600" },
+ { /* sentinel */ }
+};
+MODULE_DEVICE_TABLE(spi, encx24j600_spi_id_table);
+
+static struct spi_driver encx24j600_spi_net_driver = {
+ .driver = {
+ .name = DRV_NAME,
+ .owner = THIS_MODULE,
+ .bus = &spi_bus_type,
+ },
+ .probe = encx24j600_spi_probe,
+ .remove = encx24j600_spi_remove,
+ .id_table = encx24j600_spi_id_table,
+};
+
+module_spi_driver(encx24j600_spi_net_driver);
+
+MODULE_DESCRIPTION(DRV_NAME " ethernet driver");
+MODULE_AUTHOR("Jon Ringle <jringle@gridpoint.com>");
+MODULE_LICENSE("GPL");