* [PATCH v2 0/2] Adding new TI Common Platform ethernet SWitch driver
@ 2012-02-20 18:37 Mugunthan V N
2012-02-20 18:37 ` [PATCH v2 1/2] netdev: driver: ethernet: add cpsw address lookup engine support Mugunthan V N
2012-02-20 18:37 ` [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver Mugunthan V N
0 siblings, 2 replies; 10+ messages in thread
From: Mugunthan V N @ 2012-02-20 18:37 UTC (permalink / raw)
To: netdev; +Cc: davem, mugunthanvnm
The following series contains driver implementation for TI Common Platform
ethernet SWitch (CPSW) driver.
CPSW is found in following SoC.
* AM335X - http://www.ti.com/litv/pdf/spruh73c
* DM814X - http://www.ti.com/litv/pdf/sprugz8
CPSW:
The three port switch gigabit ethernet subsystem provides ethernet packet
communication and can be configured as an ethernet switch. It
supports 10/100/1000 Mbps. It provides the gigabit media independent
interface (G/MII), reduced gigabit media independent interface (RGMII),
reduced media independent interface (RMII), the management data input
output (MDIO) for physical layer device (PHY) management.
Changes from first posting:
* Made ale control table as const.
* Removed hw_stats sys fs entry and will be adding this as ethtool interface
in ethtool feature implementation patch.
* If skb_padto() fails, we stop queue forever, corrected the sequence by
incrementing tx_dropped and return.
Mugunthan V N (2):
netdev: driver: ethernet: add cpsw address lookup engine support
netdev: driver: ethernet: Add TI CPSW driver
drivers/net/ethernet/ti/Kconfig | 11 +
drivers/net/ethernet/ti/Makefile | 2 +
drivers/net/ethernet/ti/cpsw.c | 1029 ++++++++++++++++++++++++++++++++++++
drivers/net/ethernet/ti/cpsw_ale.c | 505 ++++++++++++++++++
drivers/net/ethernet/ti/cpsw_ale.h | 99 ++++
include/linux/platform_data/cpsw.h | 55 ++
6 files changed, 1701 insertions(+), 0 deletions(-)
create mode 100644 drivers/net/ethernet/ti/cpsw.c
create mode 100644 drivers/net/ethernet/ti/cpsw_ale.c
create mode 100644 drivers/net/ethernet/ti/cpsw_ale.h
create mode 100644 include/linux/platform_data/cpsw.h
^ permalink raw reply [flat|nested] 10+ messages in thread
* [PATCH v2 1/2] netdev: driver: ethernet: add cpsw address lookup engine support
2012-02-20 18:37 [PATCH v2 0/2] Adding new TI Common Platform ethernet SWitch driver Mugunthan V N
@ 2012-02-20 18:37 ` Mugunthan V N
2012-02-20 22:37 ` Joe Perches
2012-02-20 18:37 ` [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver Mugunthan V N
1 sibling, 1 reply; 10+ messages in thread
From: Mugunthan V N @ 2012-02-20 18:37 UTC (permalink / raw)
To: netdev; +Cc: davem, mugunthanvnm
TI CPSW ethernet switch has a built-in address lookup engine. This patch adds
the code necessary for programming this module.
Signed-off-by: Cyril Chemparathy <cyril@ti.com>
Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
---
drivers/net/ethernet/ti/cpsw_ale.c | 505 ++++++++++++++++++++++++++++++++++++
drivers/net/ethernet/ti/cpsw_ale.h | 99 +++++++
2 files changed, 604 insertions(+), 0 deletions(-)
create mode 100644 drivers/net/ethernet/ti/cpsw_ale.c
create mode 100644 drivers/net/ethernet/ti/cpsw_ale.h
diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c
new file mode 100644
index 0000000..7c26130
--- /dev/null
+++ b/drivers/net/ethernet/ti/cpsw_ale.c
@@ -0,0 +1,505 @@
+/*
+ * Texas Instruments 3-Port Ethernet Switch Address Lookup Engine
+ *
+ * Copyright (C) 2012 Texas Instruments
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+#include <linux/kernel.h>
+#include <linux/platform_device.h>
+#include <linux/seq_file.h>
+#include <linux/slab.h>
+#include <linux/err.h>
+#include <linux/io.h>
+#include <linux/stat.h>
+#include <linux/sysfs.h>
+
+#include "cpsw_ale.h"
+
+#define BITMASK(bits) (BIT(bits) - 1)
+#define ALE_ENTRY_BITS 68
+#define ALE_ENTRY_WORDS DIV_ROUND_UP(ALE_ENTRY_BITS, 32)
+
+#define ALE_VERSION_MAJOR(rev) ((rev >> 8) & 0xff)
+#define ALE_VERSION_MINOR(rev) (rev & 0xff)
+
+/* ALE Registers */
+#define ALE_IDVER 0x00
+#define ALE_CONTROL 0x08
+#define ALE_PRESCALE 0x10
+#define ALE_UNKNOWNVLAN 0x18
+#define ALE_TABLE_CONTROL 0x20
+#define ALE_TABLE 0x34
+#define ALE_PORTCTL 0x40
+
+#define ALE_TABLE_WRITE BIT(31)
+
+#define ALE_TYPE_FREE 0
+#define ALE_TYPE_ADDR 1
+#define ALE_TYPE_VLAN 2
+#define ALE_TYPE_VLAN_ADDR 3
+
+#define ALE_UCAST_PERSISTANT 0
+#define ALE_UCAST_UNTOUCHED 1
+#define ALE_UCAST_OUI 2
+#define ALE_UCAST_TOUCHED 3
+
+static inline int cpsw_ale_get_field(u32 *ale_entry, u32 start, u32 bits)
+{
+ int idx;
+
+ idx = start / 32;
+ start -= idx * 32;
+ idx = 2 - idx; /* flip */
+ return (ale_entry[idx] >> start) & BITMASK(bits);
+}
+
+static inline void cpsw_ale_set_field(u32 *ale_entry, u32 start, u32 bits,
+ u32 value)
+{
+ int idx;
+
+ value &= BITMASK(bits);
+ idx = start / 32;
+ start -= idx * 32;
+ idx = 2 - idx; /* flip */
+ ale_entry[idx] &= ~(BITMASK(bits) << start);
+ ale_entry[idx] |= (value << start);
+}
+
+#define DEFINE_ALE_FIELD(name, start, bits) \
+static inline int cpsw_ale_get_##name(u32 *ale_entry) \
+{ \
+ return cpsw_ale_get_field(ale_entry, start, bits); \
+} \
+static inline void cpsw_ale_set_##name(u32 *ale_entry, u32 value) \
+{ \
+ cpsw_ale_set_field(ale_entry, start, bits, value); \
+}
+
+DEFINE_ALE_FIELD(entry_type, 60, 2)
+DEFINE_ALE_FIELD(vlan_id, 48, 12)
+DEFINE_ALE_FIELD(mcast_state, 62, 2)
+DEFINE_ALE_FIELD(port_mask, 66, 3)
+DEFINE_ALE_FIELD(super, 65, 1)
+DEFINE_ALE_FIELD(ucast_type, 62, 2)
+DEFINE_ALE_FIELD(port_num, 66, 2)
+DEFINE_ALE_FIELD(blocked, 65, 1)
+DEFINE_ALE_FIELD(secure, 64, 1)
+DEFINE_ALE_FIELD(vlan_untag_force, 24, 3)
+DEFINE_ALE_FIELD(vlan_reg_mcast, 16, 3)
+DEFINE_ALE_FIELD(vlan_unreg_mcast, 8, 3)
+DEFINE_ALE_FIELD(vlan_member_list, 0, 3)
+DEFINE_ALE_FIELD(mcast, 40, 1)
+
+/* The MAC address field in the ALE entry cannot be macroized as above */
+static inline void cpsw_ale_get_addr(u32 *ale_entry, u8 *addr)
+{
+ int i;
+
+ for (i = 0; i < 6; i++)
+ addr[i] = cpsw_ale_get_field(ale_entry, 40 - 8*i, 8);
+}
+
+static inline void cpsw_ale_set_addr(u32 *ale_entry, u8 *addr)
+{
+ int i;
+
+ for (i = 0; i < 6; i++)
+ cpsw_ale_set_field(ale_entry, 40 - 8*i, 8, addr[i]);
+}
+
+static int cpsw_ale_read(struct cpsw_ale *ale, int idx, u32 *ale_entry)
+{
+ int i;
+
+ WARN_ON(idx > ale->params.ale_entries);
+
+ __raw_writel(idx, ale->params.ale_regs + ALE_TABLE_CONTROL);
+
+ for (i = 0; i < ALE_ENTRY_WORDS; i++)
+ ale_entry[i] = __raw_readl(ale->params.ale_regs +
+ ALE_TABLE + 4 * i);
+
+ return idx;
+}
+
+static int cpsw_ale_write(struct cpsw_ale *ale, int idx, u32 *ale_entry)
+{
+ int i;
+
+ WARN_ON(idx > ale->params.ale_entries);
+
+ for (i = 0; i < ALE_ENTRY_WORDS; i++)
+ __raw_writel(ale_entry[i], ale->params.ale_regs +
+ ALE_TABLE + 4 * i);
+
+ __raw_writel(idx | ALE_TABLE_WRITE, ale->params.ale_regs +
+ ALE_TABLE_CONTROL);
+
+ return idx;
+}
+
+static int cpsw_ale_match_addr(struct cpsw_ale *ale, u8 *addr)
+{
+ u32 ale_entry[ALE_ENTRY_WORDS];
+ int type, idx;
+
+ for (idx = 0; idx < ale->params.ale_entries; idx++) {
+ u8 entry_addr[6];
+
+ cpsw_ale_read(ale, idx, ale_entry);
+ type = cpsw_ale_get_entry_type(ale_entry);
+ if (type != ALE_TYPE_ADDR && type != ALE_TYPE_VLAN_ADDR)
+ continue;
+ cpsw_ale_get_addr(ale_entry, entry_addr);
+ if (memcmp(entry_addr, addr, 6) == 0)
+ return idx;
+ }
+ return -ENOENT;
+}
+
+static int cpsw_ale_match_free(struct cpsw_ale *ale)
+{
+ u32 ale_entry[ALE_ENTRY_WORDS];
+ int type, idx;
+
+ for (idx = 0; idx < ale->params.ale_entries; idx++) {
+ cpsw_ale_read(ale, idx, ale_entry);
+ type = cpsw_ale_get_entry_type(ale_entry);
+ if (type == ALE_TYPE_FREE)
+ return idx;
+ }
+ return -ENOENT;
+}
+
+static int cpsw_ale_find_ageable(struct cpsw_ale *ale)
+{
+ u32 ale_entry[ALE_ENTRY_WORDS];
+ int type, idx;
+
+ for (idx = 0; idx < ale->params.ale_entries; idx++) {
+ cpsw_ale_read(ale, idx, ale_entry);
+ type = cpsw_ale_get_entry_type(ale_entry);
+ if (type != ALE_TYPE_ADDR && type != ALE_TYPE_VLAN_ADDR)
+ continue;
+ if (cpsw_ale_get_mcast(ale_entry))
+ continue;
+ type = cpsw_ale_get_ucast_type(ale_entry);
+ if (type != ALE_UCAST_PERSISTANT &&
+ type != ALE_UCAST_OUI)
+ return idx;
+ }
+ return -ENOENT;
+}
+
+static void cpsw_ale_flush_mcast(struct cpsw_ale *ale, u32 *ale_entry,
+ int port_mask)
+{
+ int mask;
+
+ mask = cpsw_ale_get_port_mask(ale_entry);
+ if ((mask & port_mask) == 0)
+ return; /* ports dont intersect, not interested */
+ mask &= ~port_mask;
+
+ /* free if only remaining port is host port */
+ if (mask == BIT(ale->params.ale_ports))
+ cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_FREE);
+ else
+ cpsw_ale_set_port_mask(ale_entry, mask);
+}
+
+static void cpsw_ale_flush_ucast(struct cpsw_ale *ale, u32 *ale_entry,
+ int port_mask)
+{
+ int port;
+
+ port = cpsw_ale_get_port_num(ale_entry);
+ if ((BIT(port) & port_mask) == 0)
+ return; /* ports dont intersect, not interested */
+ cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_FREE);
+}
+
+int cpsw_ale_flush(struct cpsw_ale *ale, int port_mask)
+{
+ u32 ale_entry[ALE_ENTRY_WORDS];
+ int ret, idx;
+
+ for (idx = 0; idx < ale->params.ale_entries; idx++) {
+ cpsw_ale_read(ale, idx, ale_entry);
+ ret = cpsw_ale_get_entry_type(ale_entry);
+ if (ret != ALE_TYPE_ADDR && ret != ALE_TYPE_VLAN_ADDR)
+ continue;
+
+ if (cpsw_ale_get_mcast(ale_entry))
+ cpsw_ale_flush_mcast(ale, ale_entry, port_mask);
+ else
+ cpsw_ale_flush_ucast(ale, ale_entry, port_mask);
+
+ cpsw_ale_write(ale, idx, ale_entry);
+ }
+ return 0;
+}
+
+int cpsw_ale_add_ucast(struct cpsw_ale *ale, u8 *addr, int port, int flags)
+{
+ u32 ale_entry[ALE_ENTRY_WORDS] = {0, 0, 0};
+ int idx;
+
+ cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_ADDR);
+ cpsw_ale_set_addr(ale_entry, addr);
+ cpsw_ale_set_ucast_type(ale_entry, ALE_UCAST_PERSISTANT);
+ cpsw_ale_set_secure(ale_entry, (flags & ALE_SECURE) ? 1 : 0);
+ cpsw_ale_set_blocked(ale_entry, (flags & ALE_BLOCKED) ? 1 : 0);
+ cpsw_ale_set_port_num(ale_entry, port);
+
+ idx = cpsw_ale_match_addr(ale, addr);
+ if (idx < 0)
+ idx = cpsw_ale_match_free(ale);
+ if (idx < 0)
+ idx = cpsw_ale_find_ageable(ale);
+ if (idx < 0)
+ return -ENOMEM;
+
+ cpsw_ale_write(ale, idx, ale_entry);
+ return 0;
+}
+
+int cpsw_ale_del_ucast(struct cpsw_ale *ale, u8 *addr, int port)
+{
+ u32 ale_entry[ALE_ENTRY_WORDS] = {0, 0, 0};
+ int idx;
+
+ idx = cpsw_ale_match_addr(ale, addr);
+ if (idx < 0)
+ return -ENOENT;
+
+ cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_FREE);
+ cpsw_ale_write(ale, idx, ale_entry);
+ return 0;
+}
+
+int cpsw_ale_add_mcast(struct cpsw_ale *ale, u8 *addr, int port_mask,
+ int super, int mcast_state)
+{
+ u32 ale_entry[ALE_ENTRY_WORDS] = {0, 0, 0};
+ int idx, mask;
+
+ idx = cpsw_ale_match_addr(ale, addr);
+ if (idx >= 0)
+ cpsw_ale_read(ale, idx, ale_entry);
+
+ cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_ADDR);
+ cpsw_ale_set_addr(ale_entry, addr);
+ cpsw_ale_set_super(ale_entry, super);
+ cpsw_ale_set_mcast_state(ale_entry, mcast_state);
+
+ mask = cpsw_ale_get_port_mask(ale_entry);
+ port_mask |= mask;
+ cpsw_ale_set_port_mask(ale_entry, port_mask);
+
+ if (idx < 0)
+ idx = cpsw_ale_match_free(ale);
+ if (idx < 0)
+ idx = cpsw_ale_find_ageable(ale);
+ if (idx < 0)
+ return -ENOMEM;
+
+ cpsw_ale_write(ale, idx, ale_entry);
+ return 0;
+}
+
+int cpsw_ale_del_mcast(struct cpsw_ale *ale, u8 *addr, int port_mask)
+{
+ u32 ale_entry[ALE_ENTRY_WORDS] = {0, 0, 0};
+ int idx;
+
+ idx = cpsw_ale_match_addr(ale, addr);
+ if (idx < 0)
+ return -EINVAL;
+
+ cpsw_ale_read(ale, idx, ale_entry);
+
+ if (port_mask)
+ cpsw_ale_set_port_mask(ale_entry, port_mask);
+ else
+ cpsw_ale_set_entry_type(ale_entry, ALE_TYPE_FREE);
+
+ cpsw_ale_write(ale, idx, ale_entry);
+ return 0;
+}
+
+struct ale_control_info {
+ const char *name;
+ int offset, port_offset;
+ int shift, port_shift;
+ int bits;
+};
+
+#define CTRL_GLOBAL(name, bit) {#name, ALE_CONTROL, 0, bit, 0, 1}
+#define CTRL_UNK(name, shift, bits) {#name, ALE_UNKNOWNVLAN, 0, shift, \
+ 0, bits}
+#define CTRL_PORTCTL(name, start, bits) {#name, ALE_PORTCTL, 4, start, 0, bits}
+
+static const struct ale_control_info ale_controls[] = {
+ [ALE_ENABLE] = CTRL_GLOBAL(enable, 31),
+ [ALE_CLEAR] = CTRL_GLOBAL(clear, 30),
+ [ALE_AGEOUT] = CTRL_GLOBAL(ageout, 29),
+ [ALE_VLAN_NOLEARN] = CTRL_GLOBAL(vlan_nolearn, 7),
+ [ALE_NO_PORT_VLAN] = CTRL_GLOBAL(no_port_vlan, 6),
+ [ALE_OUI_DENY] = CTRL_GLOBAL(oui_deny, 5),
+ [ALE_BYPASS] = CTRL_GLOBAL(bypass, 4),
+ [ALE_RATE_LIMIT_TX] = CTRL_GLOBAL(rate_limit_tx, 3),
+ [ALE_VLAN_AWARE] = CTRL_GLOBAL(vlan_aware, 2),
+ [ALE_AUTH_ENABLE] = CTRL_GLOBAL(auth_enable, 1),
+ [ALE_RATE_LIMIT] = CTRL_GLOBAL(rate_limit, 0),
+
+ [ALE_PORT_STATE] = CTRL_PORTCTL(port_state, 0, 2),
+ [ALE_PORT_DROP_UNTAGGED] = CTRL_PORTCTL(drop_untagged, 2, 1),
+ [ALE_PORT_DROP_UNKNOWN_VLAN] = CTRL_PORTCTL(drop_unknown, 3, 1),
+ [ALE_PORT_NOLEARN] = CTRL_PORTCTL(nolearn, 4, 1),
+ [ALE_PORT_MCAST_LIMIT] = CTRL_PORTCTL(mcast_limit, 16, 8),
+ [ALE_PORT_BCAST_LIMIT] = CTRL_PORTCTL(bcast_limit, 24, 8),
+
+ [ALE_PORT_UNKNOWN_VLAN_MEMBER] =
+ CTRL_UNK(unknown_vlan_member, 0, 6),
+ [ALE_PORT_UNKNOWN_MCAST_FLOOD] =
+ CTRL_UNK(unknown_mcast_flood, 8, 6),
+ [ALE_PORT_UNKNOWN_REG_MCAST_FLOOD] =
+ CTRL_UNK(unknown_reg_flood, 16, 6),
+ [ALE_PORT_UNTAGGED_EGRESS] =
+ CTRL_UNK(untagged_egress, 24, 6),
+};
+
+int cpsw_ale_control_set(struct cpsw_ale *ale, int port, int control,
+ int value)
+{
+ struct ale_control_info *info = &ale_controls[control];
+ int offset, shift;
+ u32 tmp, mask;
+
+ if (control < 0 || control >= ARRAY_SIZE(ale_controls))
+ return -EINVAL;
+
+ if (info->port_offset == 0 && info->port_shift == 0)
+ port = 0; /* global, port is a dont care */
+
+ if (port < 0 || port > ale->params.ale_ports)
+ return -EINVAL;
+
+ mask = BITMASK(info->bits);
+ if (value & ~mask)
+ return -EINVAL;
+
+ offset = info->offset + (port * info->port_offset);
+ shift = info->shift + (port * info->port_shift);
+
+ tmp = __raw_readl(ale->params.ale_regs + offset);
+ tmp = (tmp & ~(mask << shift)) | (value << shift);
+ __raw_writel(tmp, ale->params.ale_regs + offset);
+
+ return 0;
+}
+
+int cpsw_ale_control_get(struct cpsw_ale *ale, int port, int control)
+{
+ struct ale_control_info *info = &ale_controls[control];
+ int offset, shift;
+ u32 tmp;
+
+ if (control < 0 || control >= ARRAY_SIZE(ale_controls))
+ return -EINVAL;
+
+ if (info->port_offset == 0 && info->port_shift == 0)
+ port = 0; /* global, port is a dont care */
+
+ if (port < 0 || port > ale->params.ale_ports)
+ return -EINVAL;
+
+ offset = info->offset + (port * info->port_offset);
+ shift = info->shift + (port * info->port_shift);
+
+ tmp = __raw_readl(ale->params.ale_regs + offset) >> shift;
+ return tmp & BITMASK(info->bits);
+}
+
+static void cpsw_ale_timer(unsigned long arg)
+{
+ struct cpsw_ale *ale = (struct cpsw_ale *)arg;
+
+ cpsw_ale_control_set(ale, 0, ALE_AGEOUT, 1);
+
+ if (ale->ageout) {
+ ale->timer.expires = jiffies + ale->ageout;
+ add_timer(&ale->timer);
+ }
+}
+
+int cpsw_ale_set_ageout(struct cpsw_ale *ale, int ageout)
+{
+ del_timer_sync(&ale->timer);
+ ale->ageout = ageout * HZ;
+ if (ale->ageout) {
+ ale->timer.expires = jiffies + ale->ageout;
+ add_timer(&ale->timer);
+ }
+ return 0;
+}
+
+void cpsw_ale_start(struct cpsw_ale *ale)
+{
+ u32 rev;
+
+ rev = __raw_readl(ale->params.ale_regs + ALE_IDVER);
+ dev_dbg(ale->params.dev, "initialized cpsw ale revision %d.%d\n",
+ ALE_VERSION_MAJOR(rev), ALE_VERSION_MINOR(rev));
+ cpsw_ale_control_set(ale, 0, ALE_ENABLE, 1);
+ cpsw_ale_control_set(ale, 0, ALE_CLEAR, 1);
+
+ init_timer(&ale->timer);
+ ale->timer.data = (unsigned long)ale;
+ ale->timer.function = cpsw_ale_timer;
+ if (ale->ageout) {
+ ale->timer.expires = jiffies + ale->ageout;
+ add_timer(&ale->timer);
+ }
+}
+
+void cpsw_ale_stop(struct cpsw_ale *ale)
+{
+ del_timer_sync(&ale->timer);
+}
+
+struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params)
+{
+ struct cpsw_ale *ale;
+ int ret;
+
+ ret = -ENOMEM;
+ ale = kzalloc(sizeof(*ale), GFP_KERNEL);
+ if (WARN_ON(!ale))
+ return NULL;
+
+ ale->params = *params;
+ ale->ageout = ale->params.ale_ageout * HZ;
+
+ return ale;
+}
+
+int cpsw_ale_destroy(struct cpsw_ale *ale)
+{
+ if (!ale)
+ return -EINVAL;
+ cpsw_ale_stop(ale);
+ cpsw_ale_control_set(ale, 0, ALE_ENABLE, 0);
+ kfree(ale);
+ return 0;
+}
diff --git a/drivers/net/ethernet/ti/cpsw_ale.h b/drivers/net/ethernet/ti/cpsw_ale.h
new file mode 100644
index 0000000..e3f0f35
--- /dev/null
+++ b/drivers/net/ethernet/ti/cpsw_ale.h
@@ -0,0 +1,99 @@
+/*
+ * Texas Instruments 3-Port Ethernet Switch Address Lookup Engine APIs
+ *
+ * Copyright (C) 2012 Texas Instruments
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+#ifndef __TI_CPSW_ALE_H__
+#define __TI_CPSW_ALE_H__
+
+struct cpsw_ale_params {
+ struct device *dev;
+ void __iomem *ale_regs;
+ unsigned long ale_ageout; /* in secs */
+ unsigned long ale_entries;
+ unsigned long ale_ports;
+};
+
+struct cpsw_ale {
+ struct cpsw_ale_params params;
+ struct timer_list timer;
+ unsigned long ageout;
+ struct device_attribute ale_control_attr;
+#define control_attr_to_ale(attr) \
+ container_of(attr, struct cpsw_ale, ale_control_attr);
+ struct device_attribute ale_table_attr;
+#define table_attr_to_ale(attr) \
+ container_of(attr, struct cpsw_ale, ale_table_attr);
+};
+
+enum cpsw_ale_control {
+ /* global */
+ ALE_ENABLE,
+ ALE_CLEAR,
+ ALE_AGEOUT,
+ ALE_VLAN_NOLEARN,
+ ALE_NO_PORT_VLAN,
+ ALE_OUI_DENY,
+ ALE_BYPASS,
+ ALE_RATE_LIMIT_TX,
+ ALE_VLAN_AWARE,
+ ALE_AUTH_ENABLE,
+ ALE_RATE_LIMIT,
+ /* port controls */
+ ALE_PORT_STATE,
+ ALE_PORT_DROP_UNTAGGED,
+ ALE_PORT_DROP_UNKNOWN_VLAN,
+ ALE_PORT_NOLEARN,
+ ALE_PORT_UNKNOWN_VLAN_MEMBER,
+ ALE_PORT_UNKNOWN_MCAST_FLOOD,
+ ALE_PORT_UNKNOWN_REG_MCAST_FLOOD,
+ ALE_PORT_UNTAGGED_EGRESS,
+ ALE_PORT_BCAST_LIMIT,
+ ALE_PORT_MCAST_LIMIT,
+ ALE_NUM_CONTROLS,
+};
+
+enum cpsw_ale_port_state {
+ ALE_PORT_STATE_DISABLE = 0x00,
+ ALE_PORT_STATE_BLOCK = 0x01,
+ ALE_PORT_STATE_LEARN = 0x02,
+ ALE_PORT_STATE_FORWARD = 0x03,
+};
+
+/* ALE unicast entry flags - passed into cpsw_ale_add_ucast() */
+#define ALE_SECURE 1
+#define ALE_BLOCKED 2
+
+#define ALE_MCAST_FWD 0
+#define ALE_MCAST_BLOCK_LEARN_FWD 1
+#define ALE_MCAST_FWD_LEARN 2
+#define ALE_MCAST_FWD_2 3
+
+struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params);
+int cpsw_ale_destroy(struct cpsw_ale *ale);
+
+void cpsw_ale_start(struct cpsw_ale *ale);
+void cpsw_ale_stop(struct cpsw_ale *ale);
+
+int cpsw_ale_set_ageout(struct cpsw_ale *ale, int ageout);
+int cpsw_ale_flush(struct cpsw_ale *ale, int port_mask);
+int cpsw_ale_add_ucast(struct cpsw_ale *ale, u8 *addr, int port, int flags);
+int cpsw_ale_del_ucast(struct cpsw_ale *ale, u8 *addr, int port);
+int cpsw_ale_add_mcast(struct cpsw_ale *ale, u8 *addr, int port_mask,
+ int super, int mcast_state);
+int cpsw_ale_del_mcast(struct cpsw_ale *ale, u8 *addr, int port_mask);
+
+int cpsw_ale_control_get(struct cpsw_ale *ale, int port, int control);
+int cpsw_ale_control_set(struct cpsw_ale *ale, int port,
+ int control, int value);
+
+#endif
--
1.7.0.4
^ permalink raw reply related [flat|nested] 10+ messages in thread
* [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver
2012-02-20 18:37 [PATCH v2 0/2] Adding new TI Common Platform ethernet SWitch driver Mugunthan V N
2012-02-20 18:37 ` [PATCH v2 1/2] netdev: driver: ethernet: add cpsw address lookup engine support Mugunthan V N
@ 2012-02-20 18:37 ` Mugunthan V N
2012-02-20 23:03 ` Joe Perches
1 sibling, 1 reply; 10+ messages in thread
From: Mugunthan V N @ 2012-02-20 18:37 UTC (permalink / raw)
To: netdev; +Cc: davem, mugunthanvnm
This patch adds support for TI's CPSW driver.
The three port switch gigabit ethernet subsystem provides ethernet packet
communication and can be configured as an ethernet switch. Supports
10/100/1000 Mbps.
Signed-off-by: Cyril Chemparathy <cyril@ti.com>
Signed-off-by: Sriramakrishnan A G <srk@ti.com>
Signed-off-by: Mugunthan V N <mugunthanvnm@ti.com>
---
drivers/net/ethernet/ti/Kconfig | 11 +
drivers/net/ethernet/ti/Makefile | 2 +
drivers/net/ethernet/ti/cpsw.c | 1029 ++++++++++++++++++++++++++++++++++++
include/linux/platform_data/cpsw.h | 55 ++
4 files changed, 1097 insertions(+), 0 deletions(-)
create mode 100644 drivers/net/ethernet/ti/cpsw.c
create mode 100644 include/linux/platform_data/cpsw.h
diff --git a/drivers/net/ethernet/ti/Kconfig b/drivers/net/ethernet/ti/Kconfig
index de76c70..b42252c 100644
--- a/drivers/net/ethernet/ti/Kconfig
+++ b/drivers/net/ethernet/ti/Kconfig
@@ -49,6 +49,17 @@ config TI_DAVINCI_CPDMA
To compile this driver as a module, choose M here: the module
will be called davinci_cpdma. This is recommended.
+config TI_CPSW
+ tristate "TI CPSW Switch Support"
+ depends on ARM && (ARCH_DAVINCI || SOC_OMAPAM33XX)
+ select TI_DAVINCI_CPDMA
+ select TI_DAVINCI_MDIO
+ ---help---
+ This driver supports TI's CPSW Ethernet Switch.
+
+ To compile this driver as a module, choose M here: the module
+ will be called cpsw.
+
config TLAN
tristate "TI ThunderLAN support"
depends on (PCI || EISA)
diff --git a/drivers/net/ethernet/ti/Makefile b/drivers/net/ethernet/ti/Makefile
index aedb3af..91bd8bb 100644
--- a/drivers/net/ethernet/ti/Makefile
+++ b/drivers/net/ethernet/ti/Makefile
@@ -7,3 +7,5 @@ obj-$(CONFIG_CPMAC) += cpmac.o
obj-$(CONFIG_TI_DAVINCI_EMAC) += davinci_emac.o
obj-$(CONFIG_TI_DAVINCI_MDIO) += davinci_mdio.o
obj-$(CONFIG_TI_DAVINCI_CPDMA) += davinci_cpdma.o
+obj-$(CONFIG_TI_CPSW) += ti_cpsw.o
+ti_cpsw-y := cpsw_ale.o cpsw.o
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
new file mode 100644
index 0000000..8c66b49
--- /dev/null
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -0,0 +1,1029 @@
+/*
+ * Texas Instruments Ethernet Switch Driver
+ *
+ * Copyright (C) 2012 Texas Instruments
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+#include <linux/kernel.h>
+#include <linux/io.h>
+#include <linux/clk.h>
+#include <linux/timer.h>
+#include <linux/module.h>
+#include <linux/platform_device.h>
+#include <linux/irqreturn.h>
+#include <linux/interrupt.h>
+#include <linux/if_ether.h>
+#include <linux/etherdevice.h>
+#include <linux/netdevice.h>
+#include <linux/phy.h>
+#include <linux/workqueue.h>
+#include <linux/delay.h>
+
+#include <linux/platform_data/cpsw.h>
+
+#include "cpsw_ale.h"
+#include "davinci_cpdma.h"
+
+#define CPSW_DEBUG (NETIF_MSG_HW | NETIF_MSG_WOL | \
+ NETIF_MSG_DRV | NETIF_MSG_LINK | \
+ NETIF_MSG_IFUP | NETIF_MSG_INTR | \
+ NETIF_MSG_PROBE | NETIF_MSG_TIMER | \
+ NETIF_MSG_IFDOWN | NETIF_MSG_RX_ERR | \
+ NETIF_MSG_TX_ERR | NETIF_MSG_TX_DONE | \
+ NETIF_MSG_PKTDATA | NETIF_MSG_TX_QUEUED | \
+ NETIF_MSG_RX_STATUS)
+
+#define msg(level, type, format, ...) \
+do { \
+ if (netif_msg_##type(priv) && net_ratelimit()) \
+ dev_##level(priv->dev, format, ## __VA_ARGS__); \
+} while (0)
+
+#define CPSW_MAJOR_VERSION(reg) (reg >> 8 & 0x7)
+#define CPSW_MINOR_VERSION(reg) (reg & 0xff)
+#define CPSW_RTL_VERSION(reg) ((reg >> 11) & 0x1f)
+
+#define CPDMA_RXTHRESH 0x0c0
+#define CPDMA_RXFREE 0x0e0
+#define CPDMA_TXHDP 0x00
+#define CPDMA_RXHDP 0x20
+#define CPDMA_TXCP 0x40
+#define CPDMA_RXCP 0x60
+
+#define CPSW_POLL_WEIGHT 64
+#define CPSW_MIN_PACKET_SIZE 60
+#define CPSW_MAX_PACKET_SIZE (1500 + 14 + 4 + 4)
+
+#define RX_PRIORITY_MAPPING 0x76543210
+#define TX_PRIORITY_MAPPING 0x33221100
+#define CPDMA_TX_PRIORITY_MAP 0x76543210
+
+#define cpsw_enable_irq(priv) \
+ do { \
+ u32 i; \
+ for (i = 0; i < priv->num_irqs; i++) \
+ enable_irq(priv->irqs_table[i]); \
+ } while (0);
+#define cpsw_disable_irq(priv) \
+ do { \
+ u32 i; \
+ for (i = 0; i < priv->num_irqs; i++) \
+ disable_irq_nosync(priv->irqs_table[i]); \
+ } while (0);
+
+static int debug_level;
+module_param(debug_level, int, 0);
+MODULE_PARM_DESC(debug_level, "cpsw debug level (NETIF_MSG bits)");
+
+static int ale_ageout = 10;
+module_param(ale_ageout, int, 0);
+MODULE_PARM_DESC(ale_ageout, "cpsw ale ageout interval (seconds)");
+
+static int rx_packet_max = CPSW_MAX_PACKET_SIZE;
+module_param(rx_packet_max, int, 0);
+MODULE_PARM_DESC(rx_packet_max, "maximum receive packet size (bytes)");
+
+struct cpsw_ss_regs {
+ u32 id_ver;
+ u32 soft_reset;
+ u32 control;
+ u32 int_control;
+ u32 rx_thresh_en;
+ u32 rx_en;
+ u32 tx_en;
+ u32 misc_en;
+};
+
+struct cpsw_regs {
+ u32 id_ver;
+ u32 control;
+ u32 soft_reset;
+ u32 stat_port_en;
+ u32 ptype;
+};
+
+struct cpsw_slave_regs {
+ u32 max_blks;
+ u32 blk_cnt;
+ u32 flow_thresh;
+ u32 port_vlan;
+ u32 tx_pri_map;
+ u32 ts_ctl;
+ u32 ts_seq_ltype;
+ u32 ts_vlan;
+ u32 sa_lo;
+ u32 sa_hi;
+};
+
+struct cpsw_host_regs {
+ u32 max_blks;
+ u32 blk_cnt;
+ u32 flow_thresh;
+ u32 port_vlan;
+ u32 tx_pri_map;
+ u32 cpdma_tx_pri_map;
+ u32 cpdma_rx_chan_map;
+};
+
+struct cpsw_sliver_regs {
+ u32 id_ver;
+ u32 mac_control;
+ u32 mac_status;
+ u32 soft_reset;
+ u32 rx_maxlen;
+ u32 __reserved_0;
+ u32 rx_pause;
+ u32 tx_pause;
+ u32 __reserved_1;
+ u32 rx_pri_map;
+};
+
+struct cpsw_hw_stats {
+ u32 rxgoodframes;
+ u32 rxbroadcastframes;
+ u32 rxmulticastframes;
+ u32 rxpauseframes;
+ u32 rxcrcerrors;
+ u32 rxaligncodeerrors;
+ u32 rxoversizedframes;
+ u32 rxjabberframes;
+ u32 rxundersizedframes;
+ u32 rxfragments;
+ u32 __pad_0[2];
+ u32 rxoctets;
+ u32 txgoodframes;
+ u32 txbroadcastframes;
+ u32 txmulticastframes;
+ u32 txpauseframes;
+ u32 txdeferredframes;
+ u32 txcollisionframes;
+ u32 txsinglecollframes;
+ u32 txmultcollframes;
+ u32 txexcessivecollisions;
+ u32 txlatecollisions;
+ u32 txunderrun;
+ u32 txcarriersenseerrors;
+ u32 txoctets;
+ u32 octetframes64;
+ u32 octetframes65t127;
+ u32 octetframes128t255;
+ u32 octetframes256t511;
+ u32 octetframes512t1023;
+ u32 octetframes1024tup;
+ u32 netoctets;
+ u32 rxsofoverruns;
+ u32 rxmofoverruns;
+ u32 rxdmaoverruns;
+};
+
+struct cpsw_slave {
+ struct cpsw_slave_regs __iomem *regs;
+ struct cpsw_sliver_regs __iomem *sliver;
+ int slave_num;
+ u32 mac_control;
+ struct cpsw_slave_data *data;
+ struct phy_device *phy;
+};
+
+struct cpsw_priv {
+ spinlock_t lock;
+ struct platform_device *pdev;
+ struct net_device *ndev;
+ struct resource *cpsw_res;
+ struct resource *cpsw_ss_res;
+ struct napi_struct napi;
+#define napi_to_priv(napi) container_of(napi, struct cpsw_priv, napi)
+ struct device *dev;
+ struct cpsw_platform_data data;
+ struct cpsw_regs __iomem *regs;
+ struct cpsw_ss_regs __iomem *ss_regs;
+ struct cpsw_hw_stats __iomem *hw_stats;
+ struct cpsw_host_regs __iomem *host_port_regs;
+ u32 msg_enable;
+ struct net_device_stats stats;
+ int rx_packet_max;
+ int host_port;
+ struct clk *clk;
+ u8 mac_addr[ETH_ALEN];
+ struct cpsw_slave *slaves;
+#define for_each_slave(priv, func, arg...) \
+ do { \
+ int idx; \
+ for (idx = 0; idx < (priv)->data.slaves; idx++) \
+ (func)((priv)->slaves + idx, ##arg); \
+ } while (0)
+
+ struct cpdma_ctlr *dma;
+ struct cpdma_chan *txch, *rxch;
+ struct cpsw_ale *ale;
+ /* snapshot of IRQ numbers */
+ u32 irqs_table[4];
+ u32 num_irqs;
+};
+
+static void cpsw_intr_enable(struct cpsw_priv *priv)
+{
+ __raw_writel(0xFF, &priv->ss_regs->tx_en);
+ __raw_writel(0xFF, &priv->ss_regs->rx_en);
+
+ cpdma_ctlr_int_ctrl(priv->dma, true);
+ return;
+}
+
+static void cpsw_intr_disable(struct cpsw_priv *priv)
+{
+ __raw_writel(0, &priv->ss_regs->tx_en);
+ __raw_writel(0, &priv->ss_regs->rx_en);
+
+ cpdma_ctlr_int_ctrl(priv->dma, false);
+ return;
+}
+
+void cpsw_tx_handler(void *token, int len, int status)
+{
+ struct sk_buff *skb = token;
+ struct net_device *ndev = skb->dev;
+ struct cpsw_priv *priv = netdev_priv(ndev);
+
+ if (unlikely(netif_queue_stopped(ndev)))
+ netif_start_queue(ndev);
+ priv->stats.tx_packets++;
+ priv->stats.tx_bytes += len;
+ dev_kfree_skb_any(skb);
+}
+
+void cpsw_rx_handler(void *token, int len, int status)
+{
+ struct sk_buff *skb = token;
+ struct net_device *ndev = skb->dev;
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ int ret = 0;
+
+ /* free and bail if we are shutting down */
+ if (unlikely(!netif_running(ndev)) ||
+ unlikely(!netif_carrier_ok(ndev))) {
+ dev_kfree_skb_any(skb);
+ return;
+ }
+ if (likely(status >= 0)) {
+ skb_put(skb, len);
+ skb->protocol = eth_type_trans(skb, ndev);
+ netif_receive_skb(skb);
+ priv->stats.rx_bytes += len;
+ priv->stats.rx_packets++;
+ skb = NULL;
+ }
+
+ if (unlikely(!netif_running(ndev))) {
+ if (skb)
+ dev_kfree_skb_any(skb);
+ return;
+ }
+
+ if (likely(!skb)) {
+ skb = netdev_alloc_skb_ip_align(ndev, priv->rx_packet_max);
+ if (WARN_ON(!skb))
+ return;
+
+ ret = cpdma_chan_submit(priv->rxch, skb, skb->data,
+ skb_tailroom(skb), GFP_KERNEL);
+ }
+ WARN_ON(ret < 0);
+}
+
+static irqreturn_t cpsw_interrupt(int irq, void *dev_id)
+{
+ struct cpsw_priv *priv = dev_id;
+
+ if (likely(netif_running(priv->ndev))) {
+ cpsw_intr_disable(priv);
+ cpsw_disable_irq(priv);
+ napi_schedule(&priv->napi);
+ }
+ return IRQ_HANDLED;
+}
+
+static inline int cpsw_get_slave_port(struct cpsw_priv *priv, u32 slave_num)
+{
+ if (priv->host_port == 0)
+ return slave_num + 1;
+ else
+ return slave_num;
+}
+
+static int cpsw_poll(struct napi_struct *napi, int budget)
+{
+ struct cpsw_priv *priv = napi_to_priv(napi);
+ int num_tx, num_rx;
+
+ num_tx = cpdma_chan_process(priv->txch, 128);
+ num_rx = cpdma_chan_process(priv->rxch, budget);
+
+ if (num_rx || num_tx)
+ msg(dbg, intr, "poll %d rx, %d tx pkts\n", num_rx, num_tx);
+
+ if (num_rx < budget) {
+ napi_complete(napi);
+ cpsw_intr_enable(priv);
+ cpdma_ctlr_eoi(priv->dma);
+ cpsw_enable_irq(priv);
+ }
+
+ return num_rx;
+}
+
+static inline void soft_reset(const char *module, void __iomem *reg)
+{
+ unsigned long timeout = jiffies + HZ;
+
+ __raw_writel(1, reg);
+ do {
+ cpu_relax();
+ } while ((__raw_readl(reg) & 1) && time_after(timeout, jiffies));
+
+ WARN(__raw_readl(reg) & 1, "failed to soft-reset %s\n", module);
+}
+
+#define mac_hi(mac) (((mac)[0] << 0) | ((mac)[1] << 8) | \
+ ((mac)[2] << 16) | ((mac)[3] << 24))
+#define mac_lo(mac) (((mac)[4] << 0) | ((mac)[5] << 8))
+
+static void cpsw_set_slave_mac(struct cpsw_slave *slave,
+ struct cpsw_priv *priv)
+{
+ __raw_writel(mac_hi(priv->mac_addr), &slave->regs->sa_hi);
+ __raw_writel(mac_lo(priv->mac_addr), &slave->regs->sa_lo);
+}
+
+static void _cpsw_adjust_link(struct cpsw_slave *slave,
+ struct cpsw_priv *priv, bool *link)
+{
+ struct phy_device *phy = slave->phy;
+ u32 mac_control = 0;
+ u32 slave_port;
+
+ if (!phy)
+ return;
+
+ slave_port = cpsw_get_slave_port(priv, slave->slave_num);
+
+ if (phy->link) {
+ mac_control = priv->data.mac_control;
+
+ /* enable forwarding */
+ cpsw_ale_control_set(priv->ale, slave_port,
+ ALE_PORT_STATE, ALE_PORT_STATE_FORWARD);
+
+ if (phy->speed == 1000)
+ mac_control |= BIT(7); /* GIGABITEN */
+ if (phy->duplex)
+ mac_control |= BIT(0); /* FULLDUPLEXEN */
+ *link = true;
+ } else {
+ mac_control = 0;
+ /* diable forwarding */
+ cpsw_ale_control_set(priv->ale, slave_port,
+ ALE_PORT_STATE, ALE_PORT_STATE_DISABLE);
+ }
+
+ if (mac_control != slave->mac_control) {
+ phy_print_status(phy);
+ __raw_writel(mac_control, &slave->sliver->mac_control);
+ }
+
+ slave->mac_control = mac_control;
+}
+
+static void cpsw_adjust_link(struct net_device *ndev)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ bool link = false;
+
+ for_each_slave(priv, _cpsw_adjust_link, priv, &link);
+
+ if (link) {
+ netif_carrier_on(ndev);
+ if (netif_running(ndev))
+ netif_wake_queue(ndev);
+ } else {
+ netif_carrier_off(ndev);
+ netif_stop_queue(ndev);
+ }
+}
+
+static inline int __show_stat(char *buf, int maxlen, const char *name, u32 val)
+{
+ static char *leader = "........................................";
+
+ if (!val)
+ return 0;
+ else
+ return snprintf(buf, maxlen, "%s %s %10d\n", name,
+ leader + strlen(name), val);
+}
+
+static void cpsw_slave_open(struct cpsw_slave *slave, struct cpsw_priv *priv)
+{
+ char name[32];
+ u32 slave_port;
+
+ sprintf(name, "slave-%d", slave->slave_num);
+
+ soft_reset(name, &slave->sliver->soft_reset);
+
+ /* setup priority mapping */
+ __raw_writel(RX_PRIORITY_MAPPING, &slave->sliver->rx_pri_map);
+ __raw_writel(TX_PRIORITY_MAPPING, &slave->regs->tx_pri_map);
+
+ /* setup max packet size, and mac address */
+ __raw_writel(priv->rx_packet_max, &slave->sliver->rx_maxlen);
+ cpsw_set_slave_mac(slave, priv);
+
+ slave->mac_control = 0; /* no link yet */
+
+ slave_port = cpsw_get_slave_port(priv, slave->slave_num);
+
+ cpsw_ale_add_mcast(priv->ale, priv->ndev->broadcast,
+ 1 << slave_port, 0, ALE_MCAST_FWD_2);
+
+ slave->phy = phy_connect(priv->ndev, slave->data->phy_id,
+ &cpsw_adjust_link, 0, slave->data->phy_if);
+ if (IS_ERR(slave->phy)) {
+ msg(err, ifup, "phy %s not found on slave %d\n",
+ slave->data->phy_id, slave->slave_num);
+ slave->phy = NULL;
+ } else {
+ printk(KERN_ERR "\nCPSW phy found : id is : 0x%x\n",
+ slave->phy->phy_id);
+ phy_start(slave->phy);
+ }
+}
+
+static void cpsw_init_host_port(struct cpsw_priv *priv)
+{
+ /* soft reset the controller and initialize ale */
+ soft_reset("cpsw", &priv->regs->soft_reset);
+ cpsw_ale_start(priv->ale);
+
+ /* switch to vlan unaware mode */
+ cpsw_ale_control_set(priv->ale, 0, ALE_VLAN_AWARE, 0);
+
+ /* setup host port priority mapping */
+ __raw_writel(CPDMA_TX_PRIORITY_MAP,
+ &priv->host_port_regs->cpdma_tx_pri_map);
+ __raw_writel(0, &priv->host_port_regs->cpdma_rx_chan_map);
+
+ cpsw_ale_control_set(priv->ale, priv->host_port,
+ ALE_PORT_STATE, ALE_PORT_STATE_FORWARD);
+
+ cpsw_ale_add_ucast(priv->ale, priv->mac_addr, priv->host_port, 0);
+ cpsw_ale_add_mcast(priv->ale, priv->ndev->broadcast,
+ 1 << priv->host_port, 0, ALE_MCAST_FWD_2);
+}
+
+static int cpsw_ndo_open(struct net_device *ndev)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ int i, ret;
+ u32 reg;
+
+ cpsw_intr_disable(priv);
+ netif_carrier_off(ndev);
+
+ ret = clk_enable(priv->clk);
+ if (ret < 0) {
+ dev_err(priv->dev, "unable to turn on device clock\n");
+ return ret;
+ }
+
+ reg = __raw_readl(&priv->regs->id_ver);
+
+ msg(info, ifup, "initializing cpsw version %d.%d (%d)\n",
+ CPSW_MAJOR_VERSION(reg), CPSW_MINOR_VERSION(reg),
+ CPSW_RTL_VERSION(reg));
+
+ /* initialize host and slave ports */
+ cpsw_init_host_port(priv);
+ for_each_slave(priv, cpsw_slave_open, priv);
+
+ /* setup tx dma to fixed prio and zero offset */
+ cpdma_control_set(priv->dma, CPDMA_TX_PRIO_FIXED, 1);
+ cpdma_control_set(priv->dma, CPDMA_RX_BUFFER_OFFSET, 0);
+
+ /* disable priority elevation and enable statistics on all ports */
+ __raw_writel(0, &priv->regs->ptype);
+
+ /* enable statistics collection only on the host port */
+ __raw_writel(0x7, &priv->regs->stat_port_en);
+
+ if (WARN_ON(!priv->data.rx_descs))
+ priv->data.rx_descs = 128;
+
+ for (i = 0; i < priv->data.rx_descs; i++) {
+ struct sk_buff *skb;
+
+ ret = -ENOMEM;
+ skb = netdev_alloc_skb_ip_align(priv->ndev,
+ priv->rx_packet_max);
+ if (!skb)
+ break;
+ ret = cpdma_chan_submit(priv->rxch, skb, skb->data,
+ skb_tailroom(skb), GFP_KERNEL);
+ if (WARN_ON(ret < 0))
+ break;
+ }
+ /* continue even if we didn't manage to submit all receive descs */
+ msg(info, ifup, "submitted %d rx descriptors\n", i);
+
+ cpdma_ctlr_start(priv->dma);
+ cpsw_intr_enable(priv);
+ napi_enable(&priv->napi);
+ cpdma_ctlr_eoi(priv->dma);
+
+ return 0;
+}
+
+static void cpsw_slave_stop(struct cpsw_slave *slave, struct cpsw_priv *priv)
+{
+ if (!slave->phy)
+ return;
+ phy_stop(slave->phy);
+ phy_disconnect(slave->phy);
+ slave->phy = NULL;
+}
+
+static int cpsw_ndo_stop(struct net_device *ndev)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+
+ msg(info, ifdown, "shutting down cpsw device\n");
+ cpsw_intr_disable(priv);
+ cpdma_ctlr_int_ctrl(priv->dma, false);
+ cpdma_ctlr_stop(priv->dma);
+ netif_stop_queue(priv->ndev);
+ napi_disable(&priv->napi);
+ netif_carrier_off(priv->ndev);
+ cpsw_ale_stop(priv->ale);
+ for_each_slave(priv, cpsw_slave_stop, priv);
+ clk_disable(priv->clk);
+ return 0;
+}
+
+static netdev_tx_t cpsw_ndo_start_xmit(struct sk_buff *skb,
+ struct net_device *ndev)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ int ret;
+
+ ndev->trans_start = jiffies;
+
+ if (skb_padto(skb, CPSW_MIN_PACKET_SIZE)) {
+ msg(err, tx_err, "packet pad failed");
+ priv->stats.tx_dropped++;
+ return NETDEV_TX_OK;
+ }
+
+ ret = cpdma_chan_submit(priv->txch, skb, skb->data,
+ skb->len, GFP_KERNEL);
+ if (unlikely(ret != 0)) {
+ msg(err, tx_err, "desc submit failed");
+ goto fail;
+ }
+
+ return NETDEV_TX_OK;
+fail:
+ priv->stats.tx_dropped++;
+ netif_stop_queue(ndev);
+ return NETDEV_TX_BUSY;
+}
+
+static void cpsw_ndo_change_rx_flags(struct net_device *ndev, int flags)
+{
+ /*
+ * The switch cannot operate in promiscuous mode without substantial
+ * headache. For promiscuous mode to work, we would need to put the
+ * ALE in bypass mode and route all traffic to the host port.
+ * Subsequently, the host will need to operate as a "bridge", learn,
+ * and flood as needed. For now, we simply complain here and
+ * do nothing about it :-)
+ */
+ if ((flags & IFF_PROMISC) && (ndev->flags & IFF_PROMISC))
+ dev_err(&ndev->dev, "promiscuity ignored!\n");
+
+ /*
+ * The switch cannot filter multicast traffic unless it is configured
+ * in "VLAN Aware" mode. Unfortunately, VLAN awareness requires a
+ * whole bunch of additional logic that this driver does not implement
+ * at present.
+ */
+ if ((flags & IFF_ALLMULTI) && !(ndev->flags & IFF_ALLMULTI))
+ dev_err(&ndev->dev, "multicast traffic cannot be filtered!\n");
+}
+
+static void cpsw_ndo_tx_timeout(struct net_device *ndev)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+
+ msg(err, tx_err, "transmit timeout, restarting dma");
+ priv->stats.tx_errors++;
+ cpsw_intr_disable(priv);
+ cpdma_ctlr_int_ctrl(priv->dma, false);
+ cpdma_chan_stop(priv->txch);
+ cpdma_chan_start(priv->txch);
+ cpdma_ctlr_int_ctrl(priv->dma, true);
+ cpsw_intr_enable(priv);
+ cpdma_ctlr_eoi(priv->dma);
+}
+
+static struct net_device_stats *cpsw_ndo_get_stats(struct net_device *ndev)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ return &priv->stats;
+}
+
+#ifdef CONFIG_NET_POLL_CONTROLLER
+static void cpsw_ndo_poll_controller(struct net_device *ndev)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+
+ cpsw_intr_disable(priv);
+ cpdma_ctlr_int_ctrl(priv->dma, false);
+ cpsw_interrupt(ndev->irq, priv);
+ cpdma_ctlr_int_ctrl(priv->dma, true);
+ cpsw_intr_enable(priv);
+ cpdma_ctlr_eoi(priv->dma);
+}
+#endif
+
+static const struct net_device_ops cpsw_netdev_ops = {
+ .ndo_open = cpsw_ndo_open,
+ .ndo_stop = cpsw_ndo_stop,
+ .ndo_start_xmit = cpsw_ndo_start_xmit,
+ .ndo_change_rx_flags = cpsw_ndo_change_rx_flags,
+ .ndo_validate_addr = eth_validate_addr,
+ .ndo_tx_timeout = cpsw_ndo_tx_timeout,
+ .ndo_get_stats = cpsw_ndo_get_stats,
+#ifdef CONFIG_NET_POLL_CONTROLLER
+ .ndo_poll_controller = cpsw_ndo_poll_controller,
+#endif
+};
+
+static void cpsw_get_drvinfo(struct net_device *ndev,
+ struct ethtool_drvinfo *info)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ strcpy(info->driver, "TI CPSW Driver v1.0");
+ strcpy(info->version, "1.0");
+ strcpy(info->bus_info, priv->pdev->name);
+}
+
+static u32 cpsw_get_msglevel(struct net_device *ndev)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ return priv->msg_enable;
+}
+
+static void cpsw_set_msglevel(struct net_device *ndev, u32 value)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ priv->msg_enable = value;
+}
+
+static const struct ethtool_ops cpsw_ethtool_ops = {
+ .get_drvinfo = cpsw_get_drvinfo,
+ .get_msglevel = cpsw_get_msglevel,
+ .set_msglevel = cpsw_set_msglevel,
+ .get_link = ethtool_op_get_link,
+};
+
+static void cpsw_slave_init(struct cpsw_slave *slave, struct cpsw_priv *priv)
+{
+ void __iomem *regs = priv->regs;
+ int slave_num = slave->slave_num;
+ struct cpsw_slave_data *data = priv->data.slave_data + slave_num;
+
+ slave->data = data;
+ slave->regs = regs + data->slave_reg_ofs;
+ slave->sliver = regs + data->sliver_reg_ofs;
+}
+
+static int __devinit cpsw_probe(struct platform_device *pdev)
+{
+ struct cpsw_platform_data *data = pdev->dev.platform_data;
+ struct net_device *ndev;
+ struct cpsw_priv *priv;
+ struct cpdma_params dma_params;
+ struct cpsw_ale_params ale_params;
+ void __iomem *regs;
+ struct resource *res;
+ int ret = 0, i, k = 0;
+
+ if (!data) {
+ pr_err("cpsw: platform data missing\n");
+ return -ENODEV;
+ }
+
+ ndev = alloc_etherdev(sizeof(struct cpsw_priv));
+ if (!ndev) {
+ pr_err("cpsw: error allocating net_device\n");
+ return -ENOMEM;
+ }
+
+ platform_set_drvdata(pdev, ndev);
+ priv = netdev_priv(ndev);
+ spin_lock_init(&priv->lock);
+ priv->data = *data;
+ priv->pdev = pdev;
+ priv->ndev = ndev;
+ priv->dev = &ndev->dev;
+ priv->msg_enable = netif_msg_init(debug_level, CPSW_DEBUG);
+ priv->rx_packet_max = max(rx_packet_max, 128);
+
+ if (is_valid_ether_addr(data->slave_data[0].mac_addr)) {
+ memcpy(priv->mac_addr, data->slave_data[0].mac_addr, ETH_ALEN);
+ printk(KERN_INFO"Detected MACID=%x:%x:%x:%x:%x:%x\n",
+ priv->mac_addr[0], priv->mac_addr[1],
+ priv->mac_addr[2], priv->mac_addr[3],
+ priv->mac_addr[4], priv->mac_addr[5]);
+ } else {
+ random_ether_addr(priv->mac_addr);
+ printk(KERN_INFO"Random MACID=%x:%x:%x:%x:%x:%x\n",
+ priv->mac_addr[0], priv->mac_addr[1],
+ priv->mac_addr[2], priv->mac_addr[3],
+ priv->mac_addr[4], priv->mac_addr[5]);
+ }
+
+ memcpy(ndev->dev_addr, priv->mac_addr, ETH_ALEN);
+
+ priv->slaves = kzalloc(sizeof(struct cpsw_slave) * data->slaves,
+ GFP_KERNEL);
+ if (!priv->slaves) {
+ dev_err(priv->dev, "failed to allocate slave ports\n");
+ ret = -EBUSY;
+ goto clean_ndev_ret;
+ }
+ for (i = 0; i < data->slaves; i++)
+ priv->slaves[i].slave_num = i;
+
+ priv->clk = clk_get(&pdev->dev, NULL);
+ if (IS_ERR(priv->clk)) {
+ dev_err(priv->dev, "failed to get device clock\n");
+ ret = -EBUSY;
+ }
+
+ priv->cpsw_res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+ if (!priv->cpsw_res) {
+ dev_err(priv->dev, "error getting i/o resource\n");
+ ret = -ENOENT;
+ goto clean_clk_ret;
+ }
+
+ if (!request_mem_region(priv->cpsw_res->start,
+ resource_size(priv->cpsw_res), ndev->name)) {
+ dev_err(priv->dev, "failed request i/o region\n");
+ ret = -ENXIO;
+ goto clean_clk_ret;
+ }
+
+ regs = ioremap(priv->cpsw_res->start, resource_size(priv->cpsw_res));
+ if (!regs) {
+ dev_err(priv->dev, "unable to map i/o region\n");
+ goto clean_cpsw_iores_ret;
+ }
+ priv->regs = regs;
+ priv->host_port = data->host_port_num;
+ priv->host_port_regs = regs + data->host_port_reg_ofs;
+ priv->hw_stats = regs + data->hw_stats_reg_ofs;
+
+ priv->cpsw_ss_res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
+ if (!priv->cpsw_ss_res) {
+ dev_err(priv->dev, "error getting i/o resource\n");
+ ret = -ENOENT;
+ goto clean_clk_ret;
+ }
+
+ if (!request_mem_region(priv->cpsw_ss_res->start,
+ resource_size(priv->cpsw_ss_res), ndev->name)) {
+ dev_err(priv->dev, "failed request i/o region\n");
+ ret = -ENXIO;
+ goto clean_clk_ret;
+ }
+
+ regs = ioremap(priv->cpsw_ss_res->start,
+ resource_size(priv->cpsw_ss_res));
+ if (!regs) {
+ dev_err(priv->dev, "unable to map i/o region\n");
+ goto clean_cpsw_ss_iores_ret;
+ }
+ priv->ss_regs = regs;
+
+ for_each_slave(priv, cpsw_slave_init, priv);
+
+ memset(&dma_params, 0, sizeof(dma_params));
+ dma_params.dev = &pdev->dev;
+ dma_params.dmaregs = (void __iomem *)(((u32)priv->regs) +
+ data->cpdma_reg_ofs);
+ dma_params.rxthresh = (void __iomem *)(((u32)priv->regs) +
+ data->cpdma_reg_ofs + CPDMA_RXTHRESH);
+ dma_params.rxfree = (void __iomem *)(((u32)priv->regs) +
+ data->cpdma_reg_ofs + CPDMA_RXFREE);
+
+ dma_params.txhdp = (void __iomem *)(((u32)priv->regs) +
+ data->cpdma_sram_ofs + CPDMA_TXHDP);
+ dma_params.rxhdp = (void __iomem *)(((u32)priv->regs) +
+ data->cpdma_sram_ofs + CPDMA_RXHDP);
+ dma_params.txcp = (void __iomem *)(((u32)priv->regs) +
+ data->cpdma_sram_ofs + CPDMA_TXCP);
+ dma_params.rxcp = (void __iomem *)(((u32)priv->regs) +
+ data->cpdma_sram_ofs + CPDMA_RXCP);
+
+ dma_params.num_chan = data->channels;
+ dma_params.has_soft_reset = true;
+ dma_params.min_packet_size = CPSW_MIN_PACKET_SIZE;
+ dma_params.desc_mem_size = data->bd_ram_size;
+ dma_params.desc_align = 16;
+ dma_params.has_ext_regs = true;
+ dma_params.desc_mem_phys = data->no_bd_ram ? 0 :
+ (u32 __force)priv->cpsw_res->start + data->bd_ram_ofs;
+ dma_params.desc_hw_addr = data->hw_ram_addr ?
+ data->hw_ram_addr : dma_params.desc_mem_phys ;
+
+ priv->dma = cpdma_ctlr_create(&dma_params);
+ if (!priv->dma) {
+ dev_err(priv->dev, "error initializing dma\n");
+ ret = -ENOMEM;
+ goto clean_iomap_ret;
+ }
+
+ priv->txch = cpdma_chan_create(priv->dma, tx_chan_num(0),
+ cpsw_tx_handler);
+ priv->rxch = cpdma_chan_create(priv->dma, rx_chan_num(0),
+ cpsw_rx_handler);
+
+ if (WARN_ON(!priv->txch || !priv->rxch)) {
+ dev_err(priv->dev, "error initializing dma channels\n");
+ ret = -ENOMEM;
+ goto clean_dma_ret;
+ }
+
+ memset(&ale_params, 0, sizeof(ale_params));
+ ale_params.dev = &ndev->dev;
+ ale_params.ale_regs = (void *)((u32)priv->regs) +
+ ((u32)data->ale_reg_ofs);
+ ale_params.ale_ageout = ale_ageout;
+ ale_params.ale_entries = data->ale_entries;
+ ale_params.ale_ports = data->slaves;
+
+ priv->ale = cpsw_ale_create(&ale_params);
+ if (!priv->ale) {
+ dev_err(priv->dev, "error initializing ale engine\n");
+ ret = -ENODEV;
+ goto clean_dma_ret;
+ }
+
+ ndev->irq = platform_get_irq(pdev, 0);
+ if (ndev->irq < 0) {
+ dev_err(priv->dev, "error getting irq resource\n");
+ ret = -ENOENT;
+ goto clean_ale_ret;
+ }
+
+ while ((res = platform_get_resource(priv->pdev, IORESOURCE_IRQ, k))) {
+ for (i = res->start; i <= res->end; i++) {
+ if (request_irq(i, cpsw_interrupt, IRQF_DISABLED,
+ dev_name(&pdev->dev), priv)) {
+ dev_err(priv->dev, "error attaching irq\n");
+ goto clean_ale_ret;
+ }
+ priv->irqs_table[k] = i;
+ priv->num_irqs = k;
+ }
+ k++;
+ }
+
+ ndev->flags |= IFF_ALLMULTI; /* see cpsw_ndo_change_rx_flags() */
+
+ ndev->netdev_ops = &cpsw_netdev_ops;
+ SET_ETHTOOL_OPS(ndev, &cpsw_ethtool_ops);
+ netif_napi_add(ndev, &priv->napi, cpsw_poll, CPSW_POLL_WEIGHT);
+
+ /* register the network device */
+ SET_NETDEV_DEV(ndev, &pdev->dev);
+ ret = register_netdev(ndev);
+ if (ret) {
+ dev_err(priv->dev, "error registering net device\n");
+ ret = -ENODEV;
+ goto clean_irq_ret;
+ }
+
+ msg(notice, probe, "initialized device (regs %x, irq %d)\n",
+ priv->cpsw_res->start, ndev->irq);
+
+ return 0;
+
+clean_irq_ret:
+ free_irq(ndev->irq, priv);
+clean_ale_ret:
+ cpsw_ale_destroy(priv->ale);
+clean_dma_ret:
+ cpdma_chan_destroy(priv->txch);
+ cpdma_chan_destroy(priv->rxch);
+ cpdma_ctlr_destroy(priv->dma);
+clean_iomap_ret:
+ iounmap(priv->regs);
+clean_cpsw_ss_iores_ret:
+ release_mem_region(priv->cpsw_ss_res->start,
+ resource_size(priv->cpsw_ss_res));
+clean_cpsw_iores_ret:
+ release_mem_region(priv->cpsw_res->start,
+ resource_size(priv->cpsw_res));
+clean_clk_ret:
+ clk_put(priv->clk);
+ kfree(priv->slaves);
+clean_ndev_ret:
+ free_netdev(ndev);
+ return ret;
+}
+
+static int __devexit cpsw_remove(struct platform_device *pdev)
+{
+ struct net_device *ndev = platform_get_drvdata(pdev);
+ struct cpsw_priv *priv = netdev_priv(ndev);
+
+ msg(notice, probe, "removing device\n");
+ platform_set_drvdata(pdev, NULL);
+
+ free_irq(ndev->irq, priv);
+ cpsw_ale_destroy(priv->ale);
+ cpdma_chan_destroy(priv->txch);
+ cpdma_chan_destroy(priv->rxch);
+ cpdma_ctlr_destroy(priv->dma);
+ iounmap(priv->regs);
+ release_mem_region(priv->cpsw_res->start,
+ resource_size(priv->cpsw_res));
+ release_mem_region(priv->cpsw_ss_res->start,
+ resource_size(priv->cpsw_ss_res));
+ clk_put(priv->clk);
+ kfree(priv->slaves);
+ free_netdev(ndev);
+
+ return 0;
+}
+
+static int cpsw_suspend(struct device *dev)
+{
+ struct platform_device *pdev = to_platform_device(dev);
+ struct net_device *ndev = platform_get_drvdata(pdev);
+
+ if (netif_running(ndev))
+ cpsw_ndo_stop(ndev);
+ return 0;
+}
+
+static int cpsw_resume(struct device *dev)
+{
+ struct platform_device *pdev = to_platform_device(dev);
+ struct net_device *ndev = platform_get_drvdata(pdev);
+
+ if (netif_running(ndev))
+ cpsw_ndo_open(ndev);
+ return 0;
+}
+
+static const struct dev_pm_ops cpsw_pm_ops = {
+ .suspend = cpsw_suspend,
+ .resume = cpsw_resume,
+};
+
+static struct platform_driver cpsw_driver = {
+ .driver = {
+ .name = "cpsw",
+ .owner = THIS_MODULE,
+ .pm = &cpsw_pm_ops,
+ },
+ .probe = cpsw_probe,
+ .remove = __devexit_p(cpsw_remove),
+};
+
+static int __init cpsw_init(void)
+{
+ return platform_driver_register(&cpsw_driver);
+}
+late_initcall(cpsw_init);
+
+static void __exit cpsw_exit(void)
+{
+ platform_driver_unregister(&cpsw_driver);
+}
+module_exit(cpsw_exit);
+
+MODULE_LICENSE("GPL");
+MODULE_DESCRIPTION("TI CPSW Ethernet driver");
diff --git a/include/linux/platform_data/cpsw.h b/include/linux/platform_data/cpsw.h
new file mode 100644
index 0000000..c4e23d0
--- /dev/null
+++ b/include/linux/platform_data/cpsw.h
@@ -0,0 +1,55 @@
+/*
+ * Texas Instruments Ethernet Switch Driver
+ *
+ * Copyright (C) 2012 Texas Instruments
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
+ * GNU General Public License for more details.
+ */
+#ifndef __CPSW_H__
+#define __CPSW_H__
+
+#include <linux/if_ether.h>
+
+struct cpsw_slave_data {
+ u32 slave_reg_ofs;
+ u32 sliver_reg_ofs;
+ const char *phy_id;
+ int phy_if;
+ u8 mac_addr[ETH_ALEN];
+};
+
+struct cpsw_platform_data {
+ u32 ss_reg_ofs; /* Subsystem control register offset */
+ u32 channels; /* number of cpdma channels (symmetric) */
+ u32 cpdma_reg_ofs; /* cpdma register offset */
+ u32 cpdma_sram_ofs; /* cpdma sram offset */
+
+ u32 slaves; /* number of slave cpgmac ports */
+ struct cpsw_slave_data *slave_data;
+
+ u32 ale_reg_ofs; /* address lookup engine reg offset */
+ u32 ale_entries; /* ale table size */
+
+ u32 host_port_reg_ofs; /* cpsw cpdma host port registers */
+ u32 host_port_num; /* The port number for the host port */
+
+ u32 hw_stats_reg_ofs; /* cpsw hardware statistics counters */
+
+ u32 bd_ram_ofs; /* embedded buffer descriptor RAM offset*/
+ u32 bd_ram_size; /*buffer descriptor ram size */
+ u32 hw_ram_addr; /*if the HW address for BD RAM is different */
+ bool no_bd_ram; /* no embedded BD ram*/
+
+ u32 rx_descs; /* Number of Rx Descriptios */
+
+ u32 mac_control; /* Mac control register */
+};
+
+#endif /* __CPSW_H__ */
--
1.7.0.4
^ permalink raw reply related [flat|nested] 10+ messages in thread
* Re: [PATCH v2 1/2] netdev: driver: ethernet: add cpsw address lookup engine support
2012-02-20 18:37 ` [PATCH v2 1/2] netdev: driver: ethernet: add cpsw address lookup engine support Mugunthan V N
@ 2012-02-20 22:37 ` Joe Perches
2012-02-22 16:43 ` N, Mugunthan V
0 siblings, 1 reply; 10+ messages in thread
From: Joe Perches @ 2012-02-20 22:37 UTC (permalink / raw)
To: Mugunthan V N; +Cc: netdev, davem
On Tue, 2012-02-21 at 00:07 +0530, Mugunthan V N wrote:
> TI CPSW ethernet switch has a built-in address lookup engine. This patch adds
> the code necessary for programming this module.
just some style notes.
> diff --git a/drivers/net/ethernet/ti/cpsw_ale.c b/drivers/net/ethernet/ti/cpsw_ale.c
[]
> +struct ale_control_info {
> + const char *name;
> + int offset, port_offset;
> + int shift, port_shift;
> + int bits;
> +};
> +
> +#define CTRL_GLOBAL(name, bit) {#name, ALE_CONTROL, 0, bit, 0, 1}
> +#define CTRL_UNK(name, shift, bits) {#name, ALE_UNKNOWNVLAN, 0, shift, \
> + 0, bits}
> +#define CTRL_PORTCTL(name, start, bits) {#name, ALE_PORTCTL, 4, start, 0, bits}
named initializers might be more readable,
embedding the index might be too.
[]
> +struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params)
> +{
> + struct cpsw_ale *ale;
> + int ret;
> +
> + ret = -ENOMEM;
> + ale = kzalloc(sizeof(*ale), GFP_KERNEL);
> + if (WARN_ON(!ale))
> + return NULL;
The WARN_ON is unnecessary. alloc failures get a dump_stack.
> diff --git a/drivers/net/ethernet/ti/cpsw_ale.h b/drivers/net/ethernet/ti/cpsw_ale.h
[]
> +struct cpsw_ale {
> + struct cpsw_ale_params params;
> + struct timer_list timer;
> + unsigned long ageout;
> + struct device_attribute ale_control_attr;
> +#define control_attr_to_ale(attr) \
> + container_of(attr, struct cpsw_ale, ale_control_attr);
> + struct device_attribute ale_table_attr;
> +#define table_attr_to_ale(attr) \
> + container_of(attr, struct cpsw_ale, ale_table_attr);
> +};
Embedding the #defines in a struct makes it difficult to read.
Perhaps this is simpler:
struct cpsw_ale {
struct cpsw_ale_params params;
struct timer_list timer;
unsigned long ageout;
struct device_attribute ale_control_attr;
struct device_attribute ale_table_attr;
};
#define control_attr_to_ale(attr) \
container_of(attr, struct cpsw_ale, ale_control_attr);
#define table_attr_to_ale(attr) \
container_of(attr, struct cpsw_ale, ale_table_attr);
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver
2012-02-20 18:37 ` [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver Mugunthan V N
@ 2012-02-20 23:03 ` Joe Perches
2012-02-22 16:43 ` N, Mugunthan V
2012-02-23 12:47 ` N, Mugunthan V
0 siblings, 2 replies; 10+ messages in thread
From: Joe Perches @ 2012-02-20 23:03 UTC (permalink / raw)
To: Mugunthan V N; +Cc: netdev, davem
On Tue, 2012-02-21 at 00:07 +0530, Mugunthan V N wrote:
> This patch adds support for TI's CPSW driver.
>
> The three port switch gigabit ethernet subsystem provides ethernet packet
> communication and can be configured as an ethernet switch. Supports
> 10/100/1000 Mbps.
some more notes.
> diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
[]
> +#define msg(level, type, format, ...) \
> +do { \
> + if (netif_msg_##type(priv) && net_ratelimit()) \
> + dev_##level(priv->dev, format, ## __VA_ARGS__); \
> +} while (0)
This hides an always used priv variable.
I suggest you expand this to multiple #defines like:
#define cpsw_info(priv, type, format, ...) \
do { \
if (netif_msg_##type(priv) && net_ratelimit()) \
dev_info(priv->dev, format, ##__VA_ARGS__); \
} while (0)
and change the uses from
msg(info,...)
to
cpsw_info(priv, type, format, ...)
[]
> +struct cpsw_priv {
> + spinlock_t lock;
> + struct platform_device *pdev;
> + struct net_device *ndev;
> + struct resource *cpsw_res;
> + struct resource *cpsw_ss_res;
> + struct napi_struct napi;
> +#define napi_to_priv(napi) container_of(napi, struct cpsw_priv, napi)
I still think embedded #defines in structs are hard to read.
[]
> +static int cpsw_poll(struct napi_struct *napi, int budget)
> +{
> + struct cpsw_priv *priv = napi_to_priv(napi);
> + int num_tx, num_rx;
> +
> + num_tx = cpdma_chan_process(priv->txch, 128);
> + num_rx = cpdma_chan_process(priv->rxch, budget);
> +
> + if (num_rx || num_tx)
> + msg(dbg, intr, "poll %d rx, %d tx pkts\n", num_rx, num_tx);
cpsw_dbg(priv, intr, etc...)
[]
> + *link = true;
> + } else {
> + mac_control = 0;
> + /* diable forwarding */
disable tyop
> + if (IS_ERR(slave->phy)) {
> + msg(err, ifup, "phy %s not found on slave %d\n",
> + slave->data->phy_id, slave->slave_num);
> + slave->phy = NULL;
> + } else {
> + printk(KERN_ERR "\nCPSW phy found : id is : 0x%x\n",
> + slave->phy->phy_id);
Using leading newlines doesn't help much.
I suggest dev_err.
[]
> + ndev = alloc_etherdev(sizeof(struct cpsw_priv));
> + if (!ndev) {
> + pr_err("cpsw: error allocating net_device\n");
Add
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
before all #includes and remove the embedded prefix
from all pr_<level> uses.
> + if (is_valid_ether_addr(data->slave_data[0].mac_addr)) {
> + memcpy(priv->mac_addr, data->slave_data[0].mac_addr, ETH_ALEN);
> + printk(KERN_INFO"Detected MACID=%x:%x:%x:%x:%x:%x\n",
> + priv->mac_addr[0], priv->mac_addr[1],
> + priv->mac_addr[2], priv->mac_addr[3],
> + priv->mac_addr[4], priv->mac_addr[5]);
Use printf extension %pM to print the MAC address
pr_info("Detected MAC=%pM\n", priv->mac_addr);
> + priv->slaves = kzalloc(sizeof(struct cpsw_slave) * data->slaves,
> + GFP_KERNEL);
> + if (!priv->slaves) {
> + dev_err(priv->dev, "failed to allocate slave ports\n");
No real need for this dev_err
[]
> + memset(&dma_params, 0, sizeof(dma_params));
> + dma_params.dev = &pdev->dev;
> + dma_params.dmaregs = (void __iomem *)(((u32)priv->regs) +
> + data->cpdma_reg_ofs);
> + dma_params.rxthresh = (void __iomem *)(((u32)priv->regs) +
> + data->cpdma_reg_ofs + CPDMA_RXTHRESH);
> + dma_params.rxfree = (void __iomem *)(((u32)priv->regs) +
> + data->cpdma_reg_ofs + CPDMA_RXFREE);
> +
> + dma_params.txhdp = (void __iomem *)(((u32)priv->regs) +
> + data->cpdma_sram_ofs + CPDMA_TXHDP);
> + dma_params.rxhdp = (void __iomem *)(((u32)priv->regs) +
> + data->cpdma_sram_ofs + CPDMA_RXHDP);
> + dma_params.txcp = (void __iomem *)(((u32)priv->regs) +
> + data->cpdma_sram_ofs + CPDMA_TXCP);
> + dma_params.rxcp = (void __iomem *)(((u32)priv->regs) +
> + data->cpdma_sram_ofs + CPDMA_RXCP);
These might be more readable if you didn't try to align it
or maybe add macros or inline functions like:
#define cpsw_sram_addr(priv, data, offset) \
(void __iomem *)(((u32)(priv)->regs) + (data)->cpdma_sram_ofs + offset)
^ permalink raw reply [flat|nested] 10+ messages in thread
* RE: [PATCH v2 1/2] netdev: driver: ethernet: add cpsw address lookup engine support
2012-02-20 22:37 ` Joe Perches
@ 2012-02-22 16:43 ` N, Mugunthan V
0 siblings, 0 replies; 10+ messages in thread
From: N, Mugunthan V @ 2012-02-22 16:43 UTC (permalink / raw)
To: Joe Perches; +Cc: netdev, davem
Joe
> -----Original Message-----
> From: Joe Perches [mailto:joe@perches.com]
> Sent: Tuesday, February 21, 2012 4:08 AM
> To: N, Mugunthan V
> Cc: netdev@vger.kernel.org; davem@davemloft.net
> Subject: Re: [PATCH v2 1/2] netdev: driver: ethernet: add cpsw address
> lookup engine support
>
> On Tue, 2012-02-21 at 00:07 +0530, Mugunthan V N wrote:
> > TI CPSW ethernet switch has a built-in address lookup engine. This
> patch adds
> > the code necessary for programming this module.
>
> just some style notes.
>
> > diff --git a/drivers/net/ethernet/ti/cpsw_ale.c
> b/drivers/net/ethernet/ti/cpsw_ale.c
> []
> > +struct ale_control_info {
> > + const char *name;
> > + int offset, port_offset;
> > + int shift, port_shift;
> > + int bits;
> > +};
> > +
> > +#define CTRL_GLOBAL(name, bit) {#name, ALE_CONTROL, 0, bit, 0,
> 1}
> > +#define CTRL_UNK(name, shift, bits) {#name, ALE_UNKNOWNVLAN, 0,
> shift, \
> > + 0, bits}
> > +#define CTRL_PORTCTL(name, start, bits) {#name, ALE_PORTCTL, 4,
> start, 0, bits}
>
> named initializers might be more readable,
> embedding the index might be too.
Agreed, will used named initializers also with embedding the index.
> []
>
> > +struct cpsw_ale *cpsw_ale_create(struct cpsw_ale_params *params)
> > +{
> > + struct cpsw_ale *ale;
> > + int ret;
> > +
> > + ret = -ENOMEM;
> > + ale = kzalloc(sizeof(*ale), GFP_KERNEL);
> > + if (WARN_ON(!ale))
> > + return NULL;
>
> The WARN_ON is unnecessary. alloc failures get a dump_stack.
You mean kzalloc will provide a dump stack during allocation failure, if so
I will remove the WARN_ON
>
> > diff --git a/drivers/net/ethernet/ti/cpsw_ale.h
> b/drivers/net/ethernet/ti/cpsw_ale.h
> []
> > +struct cpsw_ale {
> > + struct cpsw_ale_params params;
> > + struct timer_list timer;
> > + unsigned long ageout;
> > + struct device_attribute ale_control_attr;
> > +#define control_attr_to_ale(attr) \
> > + container_of(attr, struct cpsw_ale, ale_control_attr);
> > + struct device_attribute ale_table_attr;
> > +#define table_attr_to_ale(attr) \
> > + container_of(attr, struct cpsw_ale, ale_table_attr);
> > +};
>
> Embedding the #defines in a struct makes it difficult to read.
> Perhaps this is simpler:
>
> struct cpsw_ale {
> struct cpsw_ale_params params;
> struct timer_list timer;
> unsigned long ageout;
> struct device_attribute ale_control_attr;
> struct device_attribute ale_table_attr;
> };
>
> #define control_attr_to_ale(attr) \
> container_of(attr, struct cpsw_ale, ale_control_attr);
> #define table_attr_to_ale(attr) \
> container_of(attr, struct cpsw_ale, ale_table_attr);
>
Yes, I will move the define outside struct definition in next set of
patch set.
>
With regards
Mugunthan V N
^ permalink raw reply [flat|nested] 10+ messages in thread
* RE: [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver
2012-02-20 23:03 ` Joe Perches
@ 2012-02-22 16:43 ` N, Mugunthan V
2012-02-23 12:47 ` N, Mugunthan V
1 sibling, 0 replies; 10+ messages in thread
From: N, Mugunthan V @ 2012-02-22 16:43 UTC (permalink / raw)
To: Joe Perches; +Cc: netdev, davem
Joe
> -----Original Message-----
> From: Joe Perches [mailto:joe@perches.com]
> Sent: Tuesday, February 21, 2012 4:34 AM
> To: N, Mugunthan V
> Cc: netdev@vger.kernel.org; davem@davemloft.net
> Subject: Re: [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver
>
> On Tue, 2012-02-21 at 00:07 +0530, Mugunthan V N wrote:
> > This patch adds support for TI's CPSW driver.
> >
> > The three port switch gigabit ethernet subsystem provides ethernet
> packet
> > communication and can be configured as an ethernet switch. Supports
> > 10/100/1000 Mbps.
>
> some more notes.
>
> > diff --git a/drivers/net/ethernet/ti/cpsw.c
> b/drivers/net/ethernet/ti/cpsw.c
> []
> > +#define msg(level, type, format, ...) \
> > +do { \
> > + if (netif_msg_##type(priv) && net_ratelimit()) \
> > + dev_##level(priv->dev, format, ## __VA_ARGS__); \
> > +} while (0)
>
> This hides an always used priv variable.
>
> I suggest you expand this to multiple #defines like:
>
> #define cpsw_info(priv, type, format, ...) \
> do { \
> if (netif_msg_##type(priv) && net_ratelimit()) \
> dev_info(priv->dev, format, ##__VA_ARGS__); \
> } while (0)
>
> and change the uses from
> msg(info,...)
> to
> cpsw_info(priv, type, format, ...)
Agreed, will change the define.
>
> []
>
> > +struct cpsw_priv {
> > + spinlock_t lock;
> > + struct platform_device *pdev;
> > + struct net_device *ndev;
> > + struct resource *cpsw_res;
> > + struct resource *cpsw_ss_res;
> > + struct napi_struct napi;
> > +#define napi_to_priv(napi) container_of(napi, struct cpsw_priv, napi)
>
> I still think embedded #defines in structs are hard to read.
Agreed, will move defines outside the struct definition.
>
> []
> > +static int cpsw_poll(struct napi_struct *napi, int budget)
> > +{
> > + struct cpsw_priv *priv = napi_to_priv(napi);
> > + int num_tx, num_rx;
> > +
> > + num_tx = cpdma_chan_process(priv->txch, 128);
> > + num_rx = cpdma_chan_process(priv->rxch, budget);
> > +
> > + if (num_rx || num_tx)
> > + msg(dbg, intr, "poll %d rx, %d tx pkts\n", num_rx, num_tx);
>
> cpsw_dbg(priv, intr, etc...)
Will change this debug.
> []
> > + *link = true;
> > + } else {
> > + mac_control = 0;
> > + /* diable forwarding */
>
> disable tyop
>
> > + if (IS_ERR(slave->phy)) {
> > + msg(err, ifup, "phy %s not found on slave %d\n",
> > + slave->data->phy_id, slave->slave_num);
> > + slave->phy = NULL;
> > + } else {
> > + printk(KERN_ERR "\nCPSW phy found : id is : 0x%x\n",
> > + slave->phy->phy_id);
>
> Using leading newlines doesn't help much.
> I suggest dev_err.
Agreed, will change this to dev_err
> []
>
> > + ndev = alloc_etherdev(sizeof(struct cpsw_priv));
> > + if (!ndev) {
> > + pr_err("cpsw: error allocating net_device\n");
>
> Add
> #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> before all #includes and remove the embedded prefix
> from all pr_<level> uses.
Agreed, will make use of pr_fmt and remove embedded prefix in pr_<level>
uses
>
> > + if (is_valid_ether_addr(data->slave_data[0].mac_addr)) {
> > + memcpy(priv->mac_addr, data->slave_data[0].mac_addr,
> ETH_ALEN);
> > + printk(KERN_INFO"Detected MACID=%x:%x:%x:%x:%x:%x\n",
> > + priv->mac_addr[0], priv->mac_addr[1],
> > + priv->mac_addr[2], priv->mac_addr[3],
> > + priv->mac_addr[4], priv->mac_addr[5]);
>
> Use printf extension %pM to print the MAC address
>
> pr_info("Detected MAC=%pM\n", priv->mac_addr);
Will use %pM to MAC address
>
> > + priv->slaves = kzalloc(sizeof(struct cpsw_slave) * data->slaves,
> > + GFP_KERNEL);
> > + if (!priv->slaves) {
> > + dev_err(priv->dev, "failed to allocate slave ports\n");
>
> No real need for this dev_err
Agreed, will remove this dev_err
> []
> > + memset(&dma_params, 0, sizeof(dma_params));
> > + dma_params.dev = &pdev->dev;
> > + dma_params.dmaregs = (void __iomem *)(((u32)priv->regs) +
> > + data->cpdma_reg_ofs);
> > + dma_params.rxthresh = (void __iomem *)(((u32)priv->regs) +
> > + data->cpdma_reg_ofs + CPDMA_RXTHRESH);
> > + dma_params.rxfree = (void __iomem *)(((u32)priv->regs) +
> > + data->cpdma_reg_ofs + CPDMA_RXFREE);
> > +
> > + dma_params.txhdp = (void __iomem *)(((u32)priv->regs) +
> > + data->cpdma_sram_ofs + CPDMA_TXHDP);
> > + dma_params.rxhdp = (void __iomem *)(((u32)priv->regs) +
> > + data->cpdma_sram_ofs + CPDMA_RXHDP);
> > + dma_params.txcp = (void __iomem *)(((u32)priv->regs) +
> > + data->cpdma_sram_ofs + CPDMA_TXCP);
> > + dma_params.rxcp = (void __iomem *)(((u32)priv->regs) +
> > + data->cpdma_sram_ofs + CPDMA_RXCP);
>
> These might be more readable if you didn't try to align it
> or maybe add macros or inline functions like:
>
> #define cpsw_sram_addr(priv, data, offset) \
> (void __iomem *)(((u32)(priv)->regs) + (data)->cpdma_sram_ofs +
> offset)
Agreed, will make it as macros
With regards
Mugunthan V N
^ permalink raw reply [flat|nested] 10+ messages in thread
* RE: [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver
2012-02-20 23:03 ` Joe Perches
2012-02-22 16:43 ` N, Mugunthan V
@ 2012-02-23 12:47 ` N, Mugunthan V
2012-02-23 15:46 ` Joe Perches
1 sibling, 1 reply; 10+ messages in thread
From: N, Mugunthan V @ 2012-02-23 12:47 UTC (permalink / raw)
To: Joe Perches, davem; +Cc: netdev
Joe/David
> -----Original Message-----
> From: N, Mugunthan V
> Sent: Wednesday, February 22, 2012 10:13 PM
> To: 'Joe Perches'
> Cc: netdev@vger.kernel.org; davem@davemloft.net
> Subject: RE: [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver
>
> Joe
>
> > -----Original Message-----
> > From: Joe Perches [mailto:joe@perches.com]
> > Sent: Tuesday, February 21, 2012 4:34 AM
[...]
> > Add
> > #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> > before all #includes and remove the embedded prefix
> > from all pr_<level> uses.
>
> Agreed, will make use of pr_fmt and remove embedded prefix in pr_<level>
> uses
Adding
#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
in the driver gives a checkpatch error as follows
#77: FILE: drivers/net/ethernet/ti/cpsw.c:16:
+#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
total: 1 errors, 0 warnings, 1137 lines checked
Is it ok to ignore the checkpatch error as most of the drivers use it.
[...]
With regards
Mugunthan V N
^ permalink raw reply [flat|nested] 10+ messages in thread
* RE: [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver
2012-02-23 12:47 ` N, Mugunthan V
@ 2012-02-23 15:46 ` Joe Perches
2012-02-23 16:38 ` Andy Whitcroft
0 siblings, 1 reply; 10+ messages in thread
From: Joe Perches @ 2012-02-23 15:46 UTC (permalink / raw)
To: N, Mugunthan V, Andy Whitcroft; +Cc: davem, netdev
On Thu, 2012-02-23 at 12:47 +0000, N, Mugunthan V wrote:
> Joe/David
> > > -----Original Message-----
> > > From: Joe Perches [mailto:joe@perches.com]
> > > Sent: Tuesday, February 21, 2012 4:34 AM
> [...]
> > > Add
> > > #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> > > before all #includes and remove the embedded prefix
> > > from all pr_<level> uses.
> > Agreed, will make use of pr_fmt and remove embedded prefix in pr_<level>
> > uses
> Adding
> #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> in the driver gives a checkpatch error as follows
> #77: FILE: drivers/net/ethernet/ti/cpsw.c:16:
> +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> total: 1 errors, 0 warnings, 1137 lines checked
> Is it ok to ignore the checkpatch error as most of the drivers use it.
Yes.
Andy Whitcroft has a pending checkpatch fix
for this. I hope he pushes it out soon.
^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver
2012-02-23 15:46 ` Joe Perches
@ 2012-02-23 16:38 ` Andy Whitcroft
0 siblings, 0 replies; 10+ messages in thread
From: Andy Whitcroft @ 2012-02-23 16:38 UTC (permalink / raw)
To: Joe Perches; +Cc: N, Mugunthan V, davem, netdev
On Thu, Feb 23, 2012 at 07:46:02AM -0800, Joe Perches wrote:
> On Thu, 2012-02-23 at 12:47 +0000, N, Mugunthan V wrote:
> > Joe/David
> > > > -----Original Message-----
> > > > From: Joe Perches [mailto:joe@perches.com]
> > > > Sent: Tuesday, February 21, 2012 4:34 AM
> > [...]
> > > > Add
> > > > #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> > > > before all #includes and remove the embedded prefix
> > > > from all pr_<level> uses.
> > > Agreed, will make use of pr_fmt and remove embedded prefix in pr_<level>
> > > uses
> > Adding
> > #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> > in the driver gives a checkpatch error as follows
> > #77: FILE: drivers/net/ethernet/ti/cpsw.c:16:
> > +#define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
> > total: 1 errors, 0 warnings, 1137 lines checked
> > Is it ok to ignore the checkpatch error as most of the drivers use it.
>
> Yes.
>
> Andy Whitcroft has a pending checkpatch fix
> for this. I hope he pushes it out soon.
Bah good reminder. Will get those re-tested and out as soon as I can
...
-apw
^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2012-02-23 17:12 UTC | newest]
Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2012-02-20 18:37 [PATCH v2 0/2] Adding new TI Common Platform ethernet SWitch driver Mugunthan V N
2012-02-20 18:37 ` [PATCH v2 1/2] netdev: driver: ethernet: add cpsw address lookup engine support Mugunthan V N
2012-02-20 22:37 ` Joe Perches
2012-02-22 16:43 ` N, Mugunthan V
2012-02-20 18:37 ` [PATCH v2 2/2] netdev: driver: ethernet: Add TI CPSW driver Mugunthan V N
2012-02-20 23:03 ` Joe Perches
2012-02-22 16:43 ` N, Mugunthan V
2012-02-23 12:47 ` N, Mugunthan V
2012-02-23 15:46 ` Joe Perches
2012-02-23 16:38 ` Andy Whitcroft
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.