* [PATCH net 2/3] of_mdio: fix device reference leak in of_phy_find_device
From: Johan Hovold @ 2016-11-16 14:20 UTC (permalink / raw)
To: Florian Fainelli
Cc: Rob Herring, Frank Rowand, netdev, linux-kernel, Johan Hovold
In-Reply-To: <1479306038-27211-1-git-send-email-johan@kernel.org>
Make sure to drop the reference taken by bus_find_device() before
returning NULL from of_phy_find_device() when the found device is not a
PHY.
Fixes: 6ed742363b9c ("of: of_mdio: Ensure mdio device is a PHY")
Signed-off-by: Johan Hovold <johan@kernel.org>
---
drivers/of/of_mdio.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/of/of_mdio.c b/drivers/of/of_mdio.c
index 8f4648383fb2..5a3145a02547 100644
--- a/drivers/of/of_mdio.c
+++ b/drivers/of/of_mdio.c
@@ -292,6 +292,7 @@ struct phy_device *of_phy_find_device(struct device_node *phy_np)
mdiodev = to_mdio_device(d);
if (mdiodev->flags & MDIO_DEVICE_FLAG_PHY)
return to_phy_device(d);
+ put_device(d);
}
return NULL;
--
2.7.3
^ permalink raw reply related
* [PATCH net 3/3] net: phy: fixed_phy: fix of_node leak in fixed_phy_unregister
From: Johan Hovold @ 2016-11-16 14:20 UTC (permalink / raw)
To: Florian Fainelli
Cc: Rob Herring, Frank Rowand, netdev, linux-kernel, Johan Hovold
In-Reply-To: <1479306038-27211-1-git-send-email-johan@kernel.org>
Make sure to drop the of_node reference taken in fixed_phy_register()
when deregistering a PHY.
Fixes: a75951217472 ("net: phy: extend fixed driver with
fixed_phy_register()")
Signed-off-by: Johan Hovold <johan@kernel.org>
---
drivers/net/phy/fixed_phy.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/phy/fixed_phy.c b/drivers/net/phy/fixed_phy.c
index c649c101bbab..eb5167210681 100644
--- a/drivers/net/phy/fixed_phy.c
+++ b/drivers/net/phy/fixed_phy.c
@@ -279,7 +279,7 @@ EXPORT_SYMBOL_GPL(fixed_phy_register);
void fixed_phy_unregister(struct phy_device *phy)
{
phy_device_remove(phy);
-
+ of_node_put(phy->mdio.dev.of_node);
fixed_phy_del(phy->mdio.addr);
}
EXPORT_SYMBOL_GPL(fixed_phy_unregister);
--
2.7.3
^ permalink raw reply related
* [patch net-next 0/4] mlxsw: Introduce support for I2C bus
From: Jiri Pirko @ 2016-11-16 14:20 UTC (permalink / raw)
To: netdev; +Cc: davem, vadimp, idosch, eladr, yotamg, nogahf, arkadis, ogerlitz
From: Jiri Pirko <jiri@mellanox.com>
Vadim says:
This patchset adds I2C access support for SwitchX, SwitchX2, SwitchIB,
SwitchIB2 and Spectrum silicones.
It contains:
- Small changes in mlxsw core code, needed for I2C bus support;
- I2C driver, which obtains I2C input/output mailboxes setting and
provides command interface implementation.
- Minimal driver, which works on top of I2C driver and allows running
of mlxsw command interface over I2C bus;
Use case:
On system, which does not have PCI to ASIC (BMC), hwmon functionality
(sensors, pwm, tacho) will be available through I2C.
Usage (manual probing):
echo mlxsw_minimal 0x48 > /sys/bus/i2c/devices/i2c-2/new_device
Sysfs interface:
/sys/bus/i2c/devices/2-0048/hwmon/hwmon5/pwm1
/sys/bus/i2c/devices/2-0048/hwmon/hwmon5/temp1_input
Vadim Pasternak (4):
mlxsw: Add bus capability flag
mlxsw: Introduce support for I2C bus
mlxsw: Invoke driver's init/fini methods only if defined
mlxsw: minimal: Add I2C support for Mellanox ASICs
drivers/net/ethernet/mellanox/mlxsw/Kconfig | 21 +
drivers/net/ethernet/mellanox/mlxsw/Makefile | 4 +
drivers/net/ethernet/mellanox/mlxsw/core.c | 20 +-
drivers/net/ethernet/mellanox/mlxsw/core.h | 3 +
drivers/net/ethernet/mellanox/mlxsw/i2c.c | 582 ++++++++++++++++++++++++++
drivers/net/ethernet/mellanox/mlxsw/i2c.h | 60 +++
drivers/net/ethernet/mellanox/mlxsw/minimal.c | 97 +++++
drivers/net/ethernet/mellanox/mlxsw/pci.c | 1 +
8 files changed, 783 insertions(+), 5 deletions(-)
create mode 100644 drivers/net/ethernet/mellanox/mlxsw/i2c.c
create mode 100644 drivers/net/ethernet/mellanox/mlxsw/i2c.h
create mode 100644 drivers/net/ethernet/mellanox/mlxsw/minimal.c
--
2.7.4
^ permalink raw reply
* [patch net-next 1/4] mlxsw: Add bus capability flag
From: Jiri Pirko @ 2016-11-16 14:20 UTC (permalink / raw)
To: netdev; +Cc: davem, vadimp, idosch, eladr, yotamg, nogahf, arkadis, ogerlitz
In-Reply-To: <1479306046-13324-1-git-send-email-jiri@resnulli.us>
From: Vadim Pasternak <vadimp@mellanox.com>
The mlxsw core infrastructure currently assumes that communication with
the ASIC is always possible using Ethernet management datagrams (EMADs),
but this is only possible when the PCI bus is used.
The bus capability flag is added to indicate EMAD support and make core
initialize EMAD communication only when it's set. Otherwise, register
access is done using command interface.
Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
drivers/net/ethernet/mellanox/mlxsw/core.c | 6 ++++++
drivers/net/ethernet/mellanox/mlxsw/core.h | 3 +++
drivers/net/ethernet/mellanox/mlxsw/pci.c | 1 +
3 files changed, 10 insertions(+)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c
index 7874e30..33b0cd4 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.c
@@ -598,6 +598,9 @@ static int mlxsw_emad_init(struct mlxsw_core *mlxsw_core)
u64 tid;
int err;
+ if (!(mlxsw_core->bus->features & MLXSW_BUS_F_TXRX))
+ return 0;
+
/* Set the upper 32 bits of the transaction ID field to a random
* number. This allows us to discard EMADs addressed to other
* devices.
@@ -634,6 +637,9 @@ static void mlxsw_emad_fini(struct mlxsw_core *mlxsw_core)
{
char hpkt_pl[MLXSW_REG_HPKT_LEN];
+ if (!(mlxsw_core->bus->features & MLXSW_BUS_F_TXRX))
+ return;
+
mlxsw_core->emad.use_emad = false;
mlxsw_reg_hpkt_pack(hpkt_pl, MLXSW_REG_HPKT_ACTION_DISCARD,
MLXSW_TRAP_ID_ETHEMAD);
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.h b/drivers/net/ethernet/mellanox/mlxsw/core.h
index e382ed0..a2bd6c6a 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.h
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.h
@@ -269,6 +269,8 @@ u64 mlxsw_core_res_get(struct mlxsw_core *mlxsw_core,
#define MLXSW_CORE_RES_GET(res, short_res_id) \
mlxsw_core_res_get(res, MLXSW_RES_ID_##short_res_id)
+#define MLXSW_BUS_F_TXRX BIT(0)
+
struct mlxsw_bus {
const char *kind;
int (*init)(void *bus_priv, struct mlxsw_core *mlxsw_core,
@@ -284,6 +286,7 @@ struct mlxsw_bus {
char *in_mbox, size_t in_mbox_size,
char *out_mbox, size_t out_mbox_size,
u8 *p_status);
+ u8 features;
};
struct mlxsw_bus_info {
diff --git a/drivers/net/ethernet/mellanox/mlxsw/pci.c b/drivers/net/ethernet/mellanox/mlxsw/pci.c
index d5cf1ea..a223c85 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/pci.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/pci.c
@@ -1755,6 +1755,7 @@ static const struct mlxsw_bus mlxsw_pci_bus = {
.skb_transmit_busy = mlxsw_pci_skb_transmit_busy,
.skb_transmit = mlxsw_pci_skb_transmit,
.cmd_exec = mlxsw_pci_cmd_exec,
+ .features = MLXSW_BUS_F_TXRX,
};
static int mlxsw_pci_sw_reset(struct mlxsw_pci *mlxsw_pci,
--
2.7.4
^ permalink raw reply related
* [patch net-next 2/4] mlxsw: Introduce support for I2C bus
From: Jiri Pirko @ 2016-11-16 14:20 UTC (permalink / raw)
To: netdev; +Cc: davem, vadimp, idosch, eladr, yotamg, nogahf, arkadis, ogerlitz
In-Reply-To: <1479306046-13324-1-git-send-email-jiri@resnulli.us>
From: Vadim Pasternak <vadimp@mellanox.com>
Add I2C bus implementation for Mellanox Technologies Switch ASICs.
This includes command interface implementation using input / out mailboxes,
whose location is retrieved from the firmware during probe time.
Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
drivers/net/ethernet/mellanox/mlxsw/Kconfig | 10 +
drivers/net/ethernet/mellanox/mlxsw/Makefile | 2 +
drivers/net/ethernet/mellanox/mlxsw/i2c.c | 582 +++++++++++++++++++++++++++
drivers/net/ethernet/mellanox/mlxsw/i2c.h | 60 +++
4 files changed, 654 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlxsw/i2c.c
create mode 100644 drivers/net/ethernet/mellanox/mlxsw/i2c.h
diff --git a/drivers/net/ethernet/mellanox/mlxsw/Kconfig b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
index bac2e5e..f0bc7b3 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
@@ -29,6 +29,16 @@ config MLXSW_PCI
To compile this driver as a module, choose M here: the
module will be called mlxsw_pci.
+config MLXSW_I2C
+ tristate "I2C bus implementation for Mellanox Technologies Switch ASICs"
+ depends on I2C && MLXSW_CORE
+ default m
+ ---help---
+ This is I2C bus implementation for Mellanox Technologies Switch ASICs.
+
+ To compile this driver as a module, choose M here: the
+ module will be called mlxsw_i2c.
+
config MLXSW_SWITCHIB
tristate "Mellanox Technologies SwitchIB and SwitchIB-2 support"
depends on MLXSW_CORE && NET_SWITCHDEV
diff --git a/drivers/net/ethernet/mellanox/mlxsw/Makefile b/drivers/net/ethernet/mellanox/mlxsw/Makefile
index badef87..4623ede 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/Makefile
+++ b/drivers/net/ethernet/mellanox/mlxsw/Makefile
@@ -3,6 +3,8 @@ mlxsw_core-objs := core.o
mlxsw_core-$(CONFIG_MLXSW_CORE_HWMON) += core_hwmon.o
obj-$(CONFIG_MLXSW_PCI) += mlxsw_pci.o
mlxsw_pci-objs := pci.o
+obj-$(CONFIG_MLXSW_I2C) += mlxsw_i2c.o
+mlxsw_i2c-objs := i2c.o
obj-$(CONFIG_MLXSW_SWITCHIB) += mlxsw_switchib.o
mlxsw_switchib-objs := switchib.o
obj-$(CONFIG_MLXSW_SWITCHX2) += mlxsw_switchx2.o
diff --git a/drivers/net/ethernet/mellanox/mlxsw/i2c.c b/drivers/net/ethernet/mellanox/mlxsw/i2c.c
new file mode 100644
index 0000000..e50c8db
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlxsw/i2c.c
@@ -0,0 +1,582 @@
+/*
+ * drivers/net/ethernet/mellanox/mlxsw/i2c.c
+ * Copyright (c) 2016 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2016 Vadim Pasternak <vadimp@mellanox.com>
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * 3. Neither the names of the copyright holders nor the names of its
+ * contributors may be used to endorse or promote products derived from
+ * this software without specific prior written permission.
+ *
+ * Alternatively, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") version 2 as published by the Free
+ * Software Foundation.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/err.h>
+#include <linux/i2c.h>
+#include <linux/init.h>
+#include <linux/jiffies.h>
+#include <linux/kernel.h>
+#include <linux/mutex.h>
+#include <linux/module.h>
+#include <linux/mod_devicetable.h>
+#include <linux/slab.h>
+
+#include "cmd.h"
+#include "core.h"
+#include "i2c.h"
+
+static const char mlxsw_i2c_driver_name[] = "mlxsw_i2c";
+
+#define MLXSW_I2C_CIR2_BASE 0x72000
+#define MLXSW_I2C_CIR_STATUS_OFF 0x18
+#define MLXSW_I2C_CIR2_OFF_STATUS (MLXSW_I2C_CIR2_BASE + \
+ MLXSW_I2C_CIR_STATUS_OFF)
+#define MLXSW_I2C_OPMOD_SHIFT 12
+#define MLXSW_I2C_GO_BIT_SHIFT 23
+#define MLXSW_I2C_CIR_CTRL_STATUS_SHIFT 24
+#define MLXSW_I2C_GO_BIT BIT(MLXSW_I2C_GO_BIT_SHIFT)
+#define MLXSW_I2C_GO_OPMODE BIT(MLXSW_I2C_OPMOD_SHIFT)
+#define MLXSW_I2C_SET_IMM_CMD (MLXSW_I2C_GO_OPMODE | \
+ MLXSW_CMD_OPCODE_QUERY_FW)
+#define MLXSW_I2C_PUSH_IMM_CMD (MLXSW_I2C_GO_BIT | \
+ MLXSW_I2C_SET_IMM_CMD)
+#define MLXSW_I2C_SET_CMD (MLXSW_CMD_OPCODE_ACCESS_REG)
+#define MLXSW_I2C_PUSH_CMD (MLXSW_I2C_GO_BIT | MLXSW_I2C_SET_CMD)
+#define MLXSW_I2C_TLV_HDR_SIZE 0x10
+#define MLXSW_I2C_ADDR_WIDTH 4
+#define MLXSW_I2C_PUSH_CMD_SIZE (MLXSW_I2C_ADDR_WIDTH + 4)
+#define MLXSW_I2C_READ_SEMA_SIZE 4
+#define MLXSW_I2C_PREP_SIZE (MLXSW_I2C_ADDR_WIDTH + 28)
+#define MLXSW_I2C_MBOX_SIZE 20
+#define MLXSW_I2C_MBOX_OUT_PARAM_OFF 12
+#define MLXSW_I2C_MAX_BUFF_SIZE 32
+#define MLXSW_I2C_MBOX_OFFSET_BITS 20
+#define MLXSW_I2C_MBOX_SIZE_BITS 12
+#define MLXSW_I2C_ADDR_BUF_SIZE 4
+#define MLXSW_I2C_BLK_MAX 32
+#define MLXSW_I2C_RETRY 5
+#define MLXSW_I2C_TIMEOUT_MSECS 5000
+
+/**
+ * struct mlxsw_i2c - device private data:
+ * @cmd.mb_size_in: input mailbox size;
+ * @cmd.mb_off_in: input mailbox offset in register space;
+ * @cmd.mb_size_out: output mailbox size;
+ * @cmd.mb_off_out: output mailbox offset in register space;
+ * @cmd.lock: command execution lock;
+ * @dev: I2C device;
+ * @core: switch core pointer;
+ * @bus_info: bus info block;
+ */
+struct mlxsw_i2c {
+ struct {
+ u32 mb_size_in;
+ u32 mb_off_in;
+ u32 mb_size_out;
+ u32 mb_off_out;
+ struct mutex lock;
+ } cmd;
+ struct device *dev;
+ struct mlxsw_core *core;
+ struct mlxsw_bus_info bus_info;
+};
+
+#define MLXSW_I2C_READ_MSG(_client, _addr_buf, _buf, _len) { \
+ { .addr = (_client)->addr, \
+ .buf = (_addr_buf), \
+ .len = MLXSW_I2C_ADDR_BUF_SIZE, \
+ .flags = 0 }, \
+ { .addr = (_client)->addr, \
+ .buf = (_buf), \
+ .len = (_len), \
+ .flags = I2C_M_RD } }
+
+#define MLXSW_I2C_WRITE_MSG(_client, _buf, _len) \
+ { .addr = (_client)->addr, \
+ .buf = (u8 *)(_buf), \
+ .len = (_len), \
+ .flags = 0 }
+
+/* Routine converts in and out mail boxes offset and size. */
+static inline void
+mlxsw_i2c_convert_mbox(struct mlxsw_i2c *mlxsw_i2c, u8 *buf)
+{
+ u32 tmp;
+
+ /* Local in/out mailboxes: 20 bits for offset, 12 for size */
+ tmp = be32_to_cpup((__be32 *) buf);
+ mlxsw_i2c->cmd.mb_off_in = tmp &
+ GENMASK(MLXSW_I2C_MBOX_OFFSET_BITS - 1, 0);
+ mlxsw_i2c->cmd.mb_size_in = (tmp & GENMASK(31,
+ MLXSW_I2C_MBOX_OFFSET_BITS)) >>
+ MLXSW_I2C_MBOX_OFFSET_BITS;
+
+ tmp = be32_to_cpup((__be32 *) (buf + MLXSW_I2C_ADDR_WIDTH));
+ mlxsw_i2c->cmd.mb_off_out = tmp &
+ GENMASK(MLXSW_I2C_MBOX_OFFSET_BITS - 1, 0);
+ mlxsw_i2c->cmd.mb_size_out = (tmp & GENMASK(31,
+ MLXSW_I2C_MBOX_OFFSET_BITS)) >>
+ MLXSW_I2C_MBOX_OFFSET_BITS;
+}
+
+/* Routine obtains register size from mail box buffer. */
+static inline int mlxsw_i2c_get_reg_size(u8 *in_mbox)
+{
+ u16 tmp = be16_to_cpup((__be16 *) (in_mbox + MLXSW_I2C_TLV_HDR_SIZE));
+
+ return (tmp & 0x7ff) * 4 + MLXSW_I2C_TLV_HDR_SIZE;
+}
+
+/* Routine sets I2C device internal offset in the transaction buffer. */
+static inline void mlxsw_i2c_set_slave_addr(u8 *buf, u32 off)
+{
+ __be32 *val = (__be32 *) buf;
+
+ *val = htonl(off);
+}
+
+/* Routine waits until go bit is cleared. */
+static int mlxsw_i2c_wait_go_bit(struct i2c_client *client,
+ struct mlxsw_i2c *mlxsw_i2c, u8 *p_status)
+{
+ u8 addr_buf[MLXSW_I2C_ADDR_BUF_SIZE];
+ u8 buf[MLXSW_I2C_READ_SEMA_SIZE];
+ int len = MLXSW_I2C_READ_SEMA_SIZE;
+ struct i2c_msg read_sema[] =
+ MLXSW_I2C_READ_MSG(client, addr_buf, buf, len);
+ bool wait_done = false;
+ unsigned long end;
+ int i = 0, err;
+
+ mlxsw_i2c_set_slave_addr(addr_buf, MLXSW_I2C_CIR2_OFF_STATUS);
+
+ end = jiffies + msecs_to_jiffies(MLXSW_I2C_TIMEOUT_MSECS);
+ do {
+ u32 ctrl;
+
+ err = i2c_transfer(client->adapter, read_sema,
+ ARRAY_SIZE(read_sema));
+
+ ctrl = be32_to_cpu(*(__be32 *) buf);
+ if (err == ARRAY_SIZE(read_sema)) {
+ if (!(ctrl & MLXSW_I2C_GO_BIT)) {
+ wait_done = true;
+ *p_status = ctrl >>
+ MLXSW_I2C_CIR_CTRL_STATUS_SHIFT;
+ break;
+ }
+ }
+ cond_resched();
+ } while ((time_before(jiffies, end)) || (i++ < MLXSW_I2C_RETRY));
+
+ if (wait_done) {
+ if (*p_status)
+ err = -EIO;
+ } else {
+ return -ETIMEDOUT;
+ }
+
+ return err > 0 ? 0 : err;
+}
+
+/* Routine posts a command to ASIC though mail box. */
+static int mlxsw_i2c_write_cmd(struct i2c_client *client,
+ struct mlxsw_i2c *mlxsw_i2c,
+ int immediate)
+{
+ __be32 push_cmd_buf[MLXSW_I2C_PUSH_CMD_SIZE / 4] = {
+ 0, cpu_to_be32(MLXSW_I2C_PUSH_IMM_CMD)
+ };
+ __be32 prep_cmd_buf[MLXSW_I2C_PREP_SIZE / 4] = {
+ 0, 0, 0, 0, 0, 0,
+ cpu_to_be32(client->adapter->nr & 0xffff),
+ cpu_to_be32(MLXSW_I2C_SET_IMM_CMD)
+ };
+ struct i2c_msg push_cmd =
+ MLXSW_I2C_WRITE_MSG(client, push_cmd_buf,
+ MLXSW_I2C_PUSH_CMD_SIZE);
+ struct i2c_msg prep_cmd =
+ MLXSW_I2C_WRITE_MSG(client, prep_cmd_buf, MLXSW_I2C_PREP_SIZE);
+ int err;
+
+ if (!immediate) {
+ push_cmd_buf[1] = cpu_to_be32(MLXSW_I2C_PUSH_CMD);
+ prep_cmd_buf[7] = cpu_to_be32(MLXSW_I2C_SET_CMD);
+ }
+ mlxsw_i2c_set_slave_addr((u8 *)prep_cmd_buf,
+ MLXSW_I2C_CIR2_BASE);
+ mlxsw_i2c_set_slave_addr((u8 *)push_cmd_buf,
+ MLXSW_I2C_CIR2_OFF_STATUS);
+
+ /* Prepare Command Interface Register for transaction */
+ err = i2c_transfer(client->adapter, &prep_cmd, 1);
+ if (err < 0)
+ return err;
+ else if (err != 1)
+ return -EIO;
+
+ /* Write out Command Interface Register GO bit to push transaction */
+ err = i2c_transfer(client->adapter, &push_cmd, 1);
+ if (err < 0)
+ return err;
+ else if (err != 1)
+ return -EIO;
+
+ return 0;
+}
+
+/* Routine obtains mail box offsets from ASIC register space. */
+static int mlxsw_i2c_get_mbox(struct i2c_client *client,
+ struct mlxsw_i2c *mlxsw_i2c)
+{
+ u8 addr_buf[MLXSW_I2C_ADDR_BUF_SIZE];
+ u8 buf[MLXSW_I2C_MBOX_SIZE];
+ struct i2c_msg mbox_cmd[] =
+ MLXSW_I2C_READ_MSG(client, addr_buf, buf, MLXSW_I2C_MBOX_SIZE);
+ int err;
+
+ /* Read mail boxes offsets. */
+ mlxsw_i2c_set_slave_addr(addr_buf, MLXSW_I2C_CIR2_BASE);
+ err = i2c_transfer(client->adapter, mbox_cmd, 2);
+ if (err != 2) {
+ dev_err(&client->dev, "Could not obtain mail boxes\n");
+ if (!err)
+ return -EIO;
+ else
+ return err;
+ }
+
+ /* Convert mail boxes. */
+ mlxsw_i2c_convert_mbox(mlxsw_i2c, &buf[MLXSW_I2C_MBOX_OUT_PARAM_OFF]);
+
+ return err;
+}
+
+/* Routine sends I2C write transaction to ASIC device. */
+static int
+mlxsw_i2c_write(struct device *dev, size_t in_mbox_size, u8 *in_mbox, int num,
+ u8 *p_status)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct mlxsw_i2c *mlxsw_i2c = i2c_get_clientdata(client);
+ unsigned long timeout = msecs_to_jiffies(MLXSW_I2C_TIMEOUT_MSECS);
+ u8 tran_buf[MLXSW_I2C_MAX_BUFF_SIZE + MLXSW_I2C_ADDR_BUF_SIZE];
+ int off = mlxsw_i2c->cmd.mb_off_in, chunk_size, i, j;
+ unsigned long end;
+ struct i2c_msg write_tran =
+ MLXSW_I2C_WRITE_MSG(client, tran_buf, MLXSW_I2C_PUSH_CMD_SIZE);
+ int err;
+
+ for (i = 0; i < num; i++) {
+ chunk_size = (in_mbox_size > MLXSW_I2C_BLK_MAX) ?
+ MLXSW_I2C_BLK_MAX : in_mbox_size;
+ write_tran.len = MLXSW_I2C_ADDR_WIDTH + chunk_size;
+ mlxsw_i2c_set_slave_addr(tran_buf, off);
+ memcpy(&tran_buf[MLXSW_I2C_ADDR_BUF_SIZE], in_mbox +
+ chunk_size * i, chunk_size);
+
+ j = 0;
+ end = jiffies + timeout;
+ do {
+ err = i2c_transfer(client->adapter, &write_tran, 1);
+ if (err == 1)
+ break;
+
+ cond_resched();
+ } while ((time_before(jiffies, end)) ||
+ (j++ < MLXSW_I2C_RETRY));
+
+ if (err != 1) {
+ if (!err)
+ err = -EIO;
+ return err;
+ }
+
+ off += chunk_size;
+ in_mbox_size -= chunk_size;
+ }
+
+ /* Prepare and write out Command Interface Register for transaction. */
+ err = mlxsw_i2c_write_cmd(client, mlxsw_i2c, 0);
+ if (err) {
+ dev_err(&client->dev, "Could not start transaction");
+ return -EIO;
+ }
+
+ /* Wait until go bit is cleared. */
+ err = mlxsw_i2c_wait_go_bit(client, mlxsw_i2c, p_status);
+ if (err) {
+ dev_err(&client->dev, "HW semaphore is not released");
+ return err;
+ }
+
+ /* Validate transaction completion status. */
+ if (*p_status) {
+ dev_err(&client->dev, "Bad transaction completion status %x\n",
+ *p_status);
+ return -EIO;
+ }
+
+ return err > 0 ? 0 : err;
+}
+
+/* Routine executes I2C command. */
+static int
+mlxsw_i2c_cmd(struct device *dev, size_t in_mbox_size, u8 *in_mbox,
+ size_t out_mbox_size, u8 *out_mbox, u8 *status)
+{
+ struct i2c_client *client = to_i2c_client(dev);
+ struct mlxsw_i2c *mlxsw_i2c = i2c_get_clientdata(client);
+ unsigned long timeout = msecs_to_jiffies(MLXSW_I2C_TIMEOUT_MSECS);
+ u8 tran_buf[MLXSW_I2C_ADDR_BUF_SIZE];
+ int num, chunk_size, reg_size, i, j;
+ int off = mlxsw_i2c->cmd.mb_off_out;
+ unsigned long end;
+ struct i2c_msg read_tran[] =
+ MLXSW_I2C_READ_MSG(client, tran_buf, NULL, 0);
+ int err;
+
+ WARN_ON(in_mbox_size % sizeof(u32) || out_mbox_size % sizeof(u32));
+
+ reg_size = mlxsw_i2c_get_reg_size(in_mbox);
+ num = reg_size / MLXSW_I2C_BLK_MAX;
+ if (reg_size % MLXSW_I2C_BLK_MAX)
+ num++;
+
+ if (mutex_lock_interruptible(&mlxsw_i2c->cmd.lock) < 0) {
+ dev_err(&client->dev, "Could not acquire lock");
+ return -EINVAL;
+ }
+
+ err = mlxsw_i2c_write(dev, reg_size, in_mbox, num, status);
+ if (err)
+ goto cmd_fail;
+
+ /* No out mailbox is case of write transaction. */
+ if (!out_mbox) {
+ mutex_unlock(&mlxsw_i2c->cmd.lock);
+ return 0;
+ }
+
+ /* Send read transaction to get output mailbox content. */
+ read_tran[1].buf = out_mbox;
+ for (i = 0; i < num; i++) {
+ chunk_size = (reg_size > MLXSW_I2C_BLK_MAX) ?
+ MLXSW_I2C_BLK_MAX : reg_size;
+ read_tran[1].len = chunk_size;
+ mlxsw_i2c_set_slave_addr(tran_buf, off);
+
+ j = 0;
+ end = jiffies + timeout;
+ do {
+ err = i2c_transfer(client->adapter, read_tran,
+ ARRAY_SIZE(read_tran));
+ if (err == ARRAY_SIZE(read_tran))
+ break;
+
+ cond_resched();
+ } while ((time_before(jiffies, end)) ||
+ (j++ < MLXSW_I2C_RETRY));
+
+ if (err != ARRAY_SIZE(read_tran)) {
+ if (!err)
+ err = -EIO;
+
+ goto cmd_fail;
+ }
+
+ off += chunk_size;
+ reg_size -= chunk_size;
+ read_tran[1].buf += chunk_size;
+ }
+
+ mutex_unlock(&mlxsw_i2c->cmd.lock);
+
+ return 0;
+
+cmd_fail:
+ mutex_unlock(&mlxsw_i2c->cmd.lock);
+ return err;
+}
+
+static int mlxsw_i2c_cmd_exec(void *bus_priv, u16 opcode, u8 opcode_mod,
+ u32 in_mod, bool out_mbox_direct,
+ char *in_mbox, size_t in_mbox_size,
+ char *out_mbox, size_t out_mbox_size,
+ u8 *status)
+{
+ struct mlxsw_i2c *mlxsw_i2c = bus_priv;
+
+ return mlxsw_i2c_cmd(mlxsw_i2c->dev, in_mbox_size, in_mbox,
+ out_mbox_size, out_mbox, status);
+}
+
+static bool mlxsw_i2c_skb_transmit_busy(void *bus_priv,
+ const struct mlxsw_tx_info *tx_info)
+{
+ return false;
+}
+
+static int mlxsw_i2c_skb_transmit(void *bus_priv, struct sk_buff *skb,
+ const struct mlxsw_tx_info *tx_info)
+{
+ return 0;
+}
+
+static int
+mlxsw_i2c_init(void *bus_priv, struct mlxsw_core *mlxsw_core,
+ const struct mlxsw_config_profile *profile,
+ struct mlxsw_res *resources)
+{
+ struct mlxsw_i2c *mlxsw_i2c = bus_priv;
+
+ mlxsw_i2c->core = mlxsw_core;
+
+ return 0;
+}
+
+static void mlxsw_i2c_fini(void *bus_priv)
+{
+ struct mlxsw_i2c *mlxsw_i2c = bus_priv;
+
+ mlxsw_i2c->core = NULL;
+}
+
+static const struct mlxsw_bus mlxsw_i2c_bus = {
+ .kind = "i2c",
+ .init = mlxsw_i2c_init,
+ .fini = mlxsw_i2c_fini,
+ .skb_transmit_busy = mlxsw_i2c_skb_transmit_busy,
+ .skb_transmit = mlxsw_i2c_skb_transmit,
+ .cmd_exec = mlxsw_i2c_cmd_exec,
+};
+
+static int mlxsw_i2c_probe(struct i2c_client *client,
+ const struct i2c_device_id *id)
+{
+ struct mlxsw_i2c *mlxsw_i2c;
+ u8 status;
+ int err;
+
+ mlxsw_i2c = devm_kzalloc(&client->dev, sizeof(*mlxsw_i2c), GFP_KERNEL);
+ if (!mlxsw_i2c)
+ return -ENOMEM;
+
+ i2c_set_clientdata(client, mlxsw_i2c);
+ mutex_init(&mlxsw_i2c->cmd.lock);
+
+ /* In order to use mailboxes through the i2c, special area is reserved
+ * on the i2c address space that can be used for input and output
+ * mailboxes. Such mailboxes are called local mailboxes. When using a
+ * local mailbox, software should specify 0 as the Input/Output
+ * parameters. The location of the Local Mailbox addresses on the i2c
+ * space can be retrieved through the QUERY_FW command.
+ * For this purpose QUERY_FW is to be issued with opcode modifier equal
+ * 0x01. For such command the output parameter is an immediate value.
+ * Here QUERY_FW command is invoked for ASIC probing and for getting
+ * local mailboxes addresses from immedate output parameters.
+ */
+
+ /* Prepare and write out Command Interface Register for transaction */
+ err = mlxsw_i2c_write_cmd(client, mlxsw_i2c, 1);
+ if (err) {
+ dev_err(&client->dev, "Could not start transaction");
+ goto errout;
+ }
+
+ /* Wait until go bit is cleared. */
+ err = mlxsw_i2c_wait_go_bit(client, mlxsw_i2c, &status);
+ if (err) {
+ dev_err(&client->dev, "HW semaphore is not released");
+ goto errout;
+ }
+
+ /* Validate transaction completion status. */
+ if (status) {
+ dev_err(&client->dev, "Bad transaction completion status %x\n",
+ status);
+ err = -EIO;
+ goto errout;
+ }
+
+ /* Get mailbox offsets. */
+ err = mlxsw_i2c_get_mbox(client, mlxsw_i2c);
+ if (err < 0) {
+ dev_err(&client->dev, "Fail to get mailboxes\n");
+ goto errout;
+ }
+
+ dev_info(&client->dev, "%s mb size=%x off=0x%08x out mb size=%x off=0x%08x\n",
+ id->name, mlxsw_i2c->cmd.mb_size_in,
+ mlxsw_i2c->cmd.mb_off_in, mlxsw_i2c->cmd.mb_size_out,
+ mlxsw_i2c->cmd.mb_off_out);
+
+ /* Register device bus. */
+ mlxsw_i2c->bus_info.device_kind = id->name;
+ mlxsw_i2c->bus_info.device_name = client->name;
+ mlxsw_i2c->bus_info.dev = &client->dev;
+ mlxsw_i2c->dev = &client->dev;
+
+ err = mlxsw_core_bus_device_register(&mlxsw_i2c->bus_info,
+ &mlxsw_i2c_bus, mlxsw_i2c);
+ if (err) {
+ dev_err(&client->dev, "Fail to register core bus\n");
+ return err;
+ }
+
+ return 0;
+
+errout:
+ i2c_set_clientdata(client, NULL);
+
+ return err;
+}
+
+static int mlxsw_i2c_remove(struct i2c_client *client)
+{
+ struct mlxsw_i2c *mlxsw_i2c = i2c_get_clientdata(client);
+
+ mlxsw_core_bus_device_unregister(mlxsw_i2c->core);
+ mutex_destroy(&mlxsw_i2c->cmd.lock);
+
+ return 0;
+}
+
+int mlxsw_i2c_driver_register(struct i2c_driver *i2c_driver)
+{
+ i2c_driver->probe = mlxsw_i2c_probe;
+ i2c_driver->remove = mlxsw_i2c_remove;
+ return i2c_add_driver(i2c_driver);
+}
+EXPORT_SYMBOL(mlxsw_i2c_driver_register);
+
+void mlxsw_i2c_driver_unregister(struct i2c_driver *i2c_driver)
+{
+ i2c_del_driver(i2c_driver);
+}
+EXPORT_SYMBOL(mlxsw_i2c_driver_unregister);
+
+MODULE_AUTHOR("Vadim Pasternak <vadimp@mellanox.com>");
+MODULE_DESCRIPTION("Mellanox switch I2C interface driver");
+MODULE_LICENSE("Dual BSD/GPL");
diff --git a/drivers/net/ethernet/mellanox/mlxsw/i2c.h b/drivers/net/ethernet/mellanox/mlxsw/i2c.h
new file mode 100644
index 0000000..daa24b2
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlxsw/i2c.h
@@ -0,0 +1,60 @@
+/*
+ * drivers/net/ethernet/mellanox/mlxsw/i2c.h
+ * Copyright (c) 2016 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2016 Vadim Pasternak <vadimp@mellanox.com>
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * 3. Neither the names of the copyright holders nor the names of its
+ * contributors may be used to endorse or promote products derived from
+ * this software without specific prior written permission.
+ *
+ * Alternatively, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") version 2 as published by the Free
+ * Software Foundation.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#ifndef _MLXSW_I2C_H
+#define _MLXSW_I2C_H
+
+#include <linux/i2c.h>
+
+#if IS_ENABLED(CONFIG_MLXSW_I2C)
+
+int mlxsw_i2c_driver_register(struct i2c_driver *i2c_driver);
+void mlxsw_i2c_driver_unregister(struct i2c_driver *i2c_driver);
+
+#else
+
+static inline int
+mlxsw_i2c_driver_register(struct i2c_driver *i2c_driver)
+{
+ return -ENODEV;
+}
+
+static inline void
+mlxsw_i2c_driver_unregister(struct i2c_driver *i2c_driver)
+{
+}
+
+#endif
+
+#endif
--
2.7.4
^ permalink raw reply related
* [patch net-next 3/4] mlxsw: Invoke driver's init/fini methods only if defined
From: Jiri Pirko @ 2016-11-16 14:20 UTC (permalink / raw)
To: netdev; +Cc: davem, vadimp, idosch, eladr, yotamg, nogahf, arkadis, ogerlitz
In-Reply-To: <1479306046-13324-1-git-send-email-jiri@resnulli.us>
From: Vadim Pasternak <vadimp@mellanox.com>
We are going to add a minimal driver on top of the mlxsw core
infrastructure, which will be mainly used for hardware monitoring in
Baseboard management controller (BMC) installations.
Unlike the switch drivers (e.g., spectrum, switchx2), this driver does not
initialize the ASIC and therefore doesn't need to implement the init() and
fini() methods in its 'mlxsw_driver' struct.
Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
drivers/net/ethernet/mellanox/mlxsw/core.c | 14 +++++++++-----
1 file changed, 9 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlxsw/core.c b/drivers/net/ethernet/mellanox/mlxsw/core.c
index 33b0cd4..9272ef1 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/core.c
+++ b/drivers/net/ethernet/mellanox/mlxsw/core.c
@@ -1162,9 +1162,11 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
if (err)
goto err_hwmon_init;
- err = mlxsw_driver->init(mlxsw_core, mlxsw_bus_info);
- if (err)
- goto err_driver_init;
+ if (mlxsw_driver->init) {
+ err = mlxsw_driver->init(mlxsw_core, mlxsw_bus_info);
+ if (err)
+ goto err_driver_init;
+ }
err = mlxsw_core_debugfs_init(mlxsw_core);
if (err)
@@ -1173,7 +1175,8 @@ int mlxsw_core_bus_device_register(const struct mlxsw_bus_info *mlxsw_bus_info,
return 0;
err_debugfs_init:
- mlxsw_core->driver->fini(mlxsw_core);
+ if (mlxsw_core->driver->fini)
+ mlxsw_core->driver->fini(mlxsw_core);
err_driver_init:
err_hwmon_init:
devlink_unregister(devlink);
@@ -1199,7 +1202,8 @@ void mlxsw_core_bus_device_unregister(struct mlxsw_core *mlxsw_core)
struct devlink *devlink = priv_to_devlink(mlxsw_core);
mlxsw_core_debugfs_fini(mlxsw_core);
- mlxsw_core->driver->fini(mlxsw_core);
+ if (mlxsw_core->driver->fini)
+ mlxsw_core->driver->fini(mlxsw_core);
devlink_unregister(devlink);
mlxsw_emad_fini(mlxsw_core);
mlxsw_core->bus->fini(mlxsw_core->bus_priv);
--
2.7.4
^ permalink raw reply related
* [patch net-next 4/4] mlxsw: minimal: Add I2C support for Mellanox ASICs
From: Jiri Pirko @ 2016-11-16 14:20 UTC (permalink / raw)
To: netdev; +Cc: davem, vadimp, idosch, eladr, yotamg, nogahf, arkadis, ogerlitz
In-Reply-To: <1479306046-13324-1-git-send-email-jiri@resnulli.us>
From: Vadim Pasternak <vadimp@mellanox.com>
Add I2C access support for Mellanox ASICs:
- Virtual Protocol Interconnect switches SwitchX, SwitchX2,
providing InfiniBand, Ethernet and Fibre Channel connectivity;
- Infiniband switches SwitchIB, SwitchIB2:
- Ethernet switch Spectrum.
Example of probing activation:
echo mlxsw_minimal 0x48 > /sys/bus/i2c/devices/i2c-2/new_device
Signed-off-by: Vadim Pasternak <vadimp@mellanox.com>
Reviewed-by: Ido Schimmel <idosch@mellanox.com>
Signed-off-by: Jiri Pirko <jiri@mellanox.com>
---
drivers/net/ethernet/mellanox/mlxsw/Kconfig | 11 +++
drivers/net/ethernet/mellanox/mlxsw/Makefile | 2 +
drivers/net/ethernet/mellanox/mlxsw/minimal.c | 97 +++++++++++++++++++++++++++
3 files changed, 110 insertions(+)
create mode 100644 drivers/net/ethernet/mellanox/mlxsw/minimal.c
diff --git a/drivers/net/ethernet/mellanox/mlxsw/Kconfig b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
index f0bc7b3..c9822e6 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/Kconfig
+++ b/drivers/net/ethernet/mellanox/mlxsw/Kconfig
@@ -79,3 +79,14 @@ config MLXSW_SPECTRUM_DCB
---help---
Say Y here if you want to use Data Center Bridging (DCB) in the
driver.
+
+config MLXSW_MINIMAL
+ tristate "Mellanox Technologies minimal I2C support"
+ depends on MLXSW_CORE && MLXSW_I2C
+ default m
+ ---help---
+ This driver supports I2C access for Mellanox Technologies Switch
+ ASICs.
+
+ To compile this driver as a module, choose M here: the
+ module will be called mlxsw_minimal.
diff --git a/drivers/net/ethernet/mellanox/mlxsw/Makefile b/drivers/net/ethernet/mellanox/mlxsw/Makefile
index 4623ede..2722942 100644
--- a/drivers/net/ethernet/mellanox/mlxsw/Makefile
+++ b/drivers/net/ethernet/mellanox/mlxsw/Makefile
@@ -14,3 +14,5 @@ mlxsw_spectrum-objs := spectrum.o spectrum_buffers.o \
spectrum_switchdev.o spectrum_router.o \
spectrum_kvdl.o
mlxsw_spectrum-$(CONFIG_MLXSW_SPECTRUM_DCB) += spectrum_dcb.o
+obj-$(CONFIG_MLXSW_MINIMAL) += mlxsw_minimal.o
+mlxsw_minimal-objs := minimal.o
diff --git a/drivers/net/ethernet/mellanox/mlxsw/minimal.c b/drivers/net/ethernet/mellanox/mlxsw/minimal.c
new file mode 100644
index 0000000..3dd1626
--- /dev/null
+++ b/drivers/net/ethernet/mellanox/mlxsw/minimal.c
@@ -0,0 +1,97 @@
+/*
+ * drivers/net/ethernet/mellanox/mlxsw/minimal.c
+ * Copyright (c) 2016 Mellanox Technologies. All rights reserved.
+ * Copyright (c) 2016 Vadim Pasternak <vadimp@mellanox.com>
+ *
+ * Redistribution and use in source and binary forms, with or without
+ * modification, are permitted provided that the following conditions are met:
+ *
+ * 1. Redistributions of source code must retain the above copyright
+ * notice, this list of conditions and the following disclaimer.
+ * 2. Redistributions in binary form must reproduce the above copyright
+ * notice, this list of conditions and the following disclaimer in the
+ * documentation and/or other materials provided with the distribution.
+ * 3. Neither the names of the copyright holders nor the names of its
+ * contributors may be used to endorse or promote products derived from
+ * this software without specific prior written permission.
+ *
+ * Alternatively, this software may be distributed under the terms of the
+ * GNU General Public License ("GPL") version 2 as published by the Free
+ * Software Foundation.
+ *
+ * THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+ * AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+ * IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
+ * ARE DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT OWNER OR CONTRIBUTORS BE
+ * LIABLE FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR
+ * CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF
+ * SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS
+ * INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN
+ * CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
+ * ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE
+ * POSSIBILITY OF SUCH DAMAGE.
+ */
+
+#include <linux/i2c.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+#include <linux/mod_devicetable.h>
+#include <linux/types.h>
+
+#include "core.h"
+#include "i2c.h"
+
+static const char mlxsw_minimal_driver_name[] = "mlxsw_minimal";
+
+static const struct mlxsw_config_profile mlxsw_minimal_config_profile;
+
+static struct mlxsw_driver mlxsw_minimal_driver = {
+ .kind = mlxsw_minimal_driver_name,
+ .priv_size = 1,
+ .profile = &mlxsw_minimal_config_profile,
+};
+
+static const struct i2c_device_id mlxsw_minimal_i2c_id[] = {
+ { "mlxsw_minimal", 0},
+ { },
+};
+
+static struct i2c_driver mlxsw_minimal_i2c_driver = {
+ .driver.name = "mlxsw_minimal",
+ .class = I2C_CLASS_HWMON,
+ .id_table = mlxsw_minimal_i2c_id,
+};
+
+static int __init mlxsw_minimal_module_init(void)
+{
+ int err;
+
+ err = mlxsw_core_driver_register(&mlxsw_minimal_driver);
+ if (err)
+ return err;
+
+ err = mlxsw_i2c_driver_register(&mlxsw_minimal_i2c_driver);
+ if (err)
+ goto err_i2c_driver_register;
+
+ return 0;
+
+err_i2c_driver_register:
+ mlxsw_core_driver_unregister(&mlxsw_minimal_driver);
+
+ return err;
+}
+
+static void __exit mlxsw_minimal_module_exit(void)
+{
+ mlxsw_i2c_driver_unregister(&mlxsw_minimal_i2c_driver);
+ mlxsw_core_driver_unregister(&mlxsw_minimal_driver);
+}
+
+module_init(mlxsw_minimal_module_init);
+module_exit(mlxsw_minimal_module_exit);
+
+MODULE_LICENSE("Dual BSD/GPL");
+MODULE_AUTHOR("Vadim Pasternak <vadimp@mellanox.com>");
+MODULE_DESCRIPTION("Mellanox minimal driver");
+MODULE_DEVICE_TABLE(i2c, mlxsw_minimal_i2c_id);
--
2.7.4
^ permalink raw reply related
* [PATCH v2 net-next] net/mlx5e: remove napi_hash_del() calls
From: Eric Dumazet @ 2016-11-16 14:21 UTC (permalink / raw)
To: David Miller; +Cc: Saeed Mahameed, netdev
In-Reply-To: <1479304537.8455.167.camel@edumazet-glaptop3.roam.corp.google.com>
From: Eric Dumazet <edumazet@google.com>
Calling napi_hash_del() after netif_napi_del() is pointless.
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Saeed Mahameed <saeedm@mellanox.com>
---
drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 4 ----
1 file changed, 4 deletions(-)
diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
index 52b4fd53b1a74d17b7b40efc279305fdb9ed7521..00579ce0e96fb8f992d93ddfa0e5fd58df25813a 100644
--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -1533,7 +1533,6 @@ static int mlx5e_open_channel(struct mlx5e_priv *priv, int ix,
err_napi_del:
netif_napi_del(&c->napi);
- napi_hash_del(&c->napi);
kfree(c);
return err;
@@ -1554,9 +1553,6 @@ static void mlx5e_close_channel(struct mlx5e_channel *c)
mlx5e_close_cq(&c->icosq.cq);
netif_napi_del(&c->napi);
- napi_hash_del(&c->napi);
- synchronize_rcu();
-
kfree(c);
}
^ permalink raw reply related
* Re: [patch net-next 0/4] mlxsw: Introduce support for I2C bus
From: Jiri Pirko @ 2016-11-16 14:23 UTC (permalink / raw)
To: netdev; +Cc: davem, vadimp, idosch, eladr, yotamg, nogahf, arkadis, ogerlitz
In-Reply-To: <1479306046-13324-1-git-send-email-jiri@resnulli.us>
Dave, I made sure that the patchsets in flight which touch mlxsw code do
not depend on each other and could be applied in any order you please.
Thanks.
^ permalink raw reply
* Re: [PATCH] net: ethernet: faraday: To support device tree usage.
From: Greentime Hu @ 2016-11-16 14:26 UTC (permalink / raw)
To: Andrew Lunn; +Cc: netdev, linux-kernel
In-Reply-To: <20161116134715.GE19962@lunn.ch>
On Wed, Nov 16, 2016 at 9:47 PM, Andrew Lunn <andrew@lunn.ch> wrote:
> On Wed, Nov 16, 2016 at 04:43:15PM +0800, Greentime Hu wrote:
>> To support device tree usage for ftmac100.
>>
>> Signed-off-by: Greentime Hu <green.hu@gmail.com>
>> ---
>> drivers/net/ethernet/faraday/ftmac100.c | 7 +++++++
>> 1 file changed, 7 insertions(+)
>>
>> diff --git a/drivers/net/ethernet/faraday/ftmac100.c b/drivers/net/ethernet/faraday/ftmac100.c
>> index dce5f7b..81dd9e1 100644
>> --- a/drivers/net/ethernet/faraday/ftmac100.c
>> +++ b/drivers/net/ethernet/faraday/ftmac100.c
>> @@ -1172,11 +1172,17 @@ static int __exit ftmac100_remove(struct platform_device *pdev)
>> return 0;
>> }
>>
>> +static const struct of_device_id mac_of_ids[] = {
>> + { .compatible = "andestech,atmac100" },
>> + { }
>
> andestech is not in
> Documentation/devicetree/bindings/vendor-prefixes.txt Please provide a
> separate patch adding it.
OK. I will provide another patch to add andestech.
> Humm, why andestech? Why not something based around faraday
> technology?
It is because we use the same ftmac100 IP provided from faraday
technology but I am now using it in andestech SoC.
^ permalink raw reply
* Re: [PATCH net-next] sfc: remove napi_hash_del() call
From: Bert Kenward @ 2016-11-16 14:15 UTC (permalink / raw)
To: Eric Dumazet, David Miller; +Cc: netdev, Edward Cree
In-Reply-To: <1479304907.8455.171.camel@edumazet-glaptop3.roam.corp.google.com>
On 16/11/16 14:01, Eric Dumazet wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> Calling napi_hash_del() after netif_napi_del() is pointless.
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Cc: Edward Cree <ecree@solarflare.com>
> Cc: Bert Kenward <bkenward@solarflare.com>
> ---
> drivers/net/ethernet/sfc/efx.c | 5 ++---
> 1 file changed, 2 insertions(+), 3 deletions(-)
Acked-by: Bert Kenward <bkenward@solarflare.com>
^ permalink raw reply
* Re: [PATCH net-next v2 2/4] bpf, mlx5: fix various refcount issues in mlx5e_xdp_set
From: Saeed Mahameed @ 2016-11-16 14:30 UTC (permalink / raw)
To: Daniel Borkmann
Cc: David S. Miller, Alexei Starovoitov, Brenden Blanco, zhiyisun,
Rana Shahout, Saeed Mahameed, Linux Netdev List
In-Reply-To: <582C64FC.4010307@iogearbox.net>
On Wed, Nov 16, 2016 at 3:54 PM, Daniel Borkmann <daniel@iogearbox.net> wrote:
> On 11/16/2016 01:25 PM, Saeed Mahameed wrote:
>>
>> On Wed, Nov 16, 2016 at 2:04 AM, Daniel Borkmann <daniel@iogearbox.net>
>> wrote:
>>>
>>> There are multiple issues in mlx5e_xdp_set():
>>>
>>> 1) The batched bpf_prog_add() is currently not checked for errors! When
>>> doing so, it should be done at an earlier point in time to makes sure
>>> that we cannot fail anymore at the time we want to set the program
>>> for
>>> each channel. This only means that we have to undo the bpf_prog_add()
>>> in case we return due to required reset.
>>>
>>> 2) When swapping the priv->xdp_prog, then no extra reference count must
>>> be
>>> taken since we got that from call path via dev_change_xdp_fd()
>>> already.
>>> Otherwise, we'd never be able to free the program. Also,
>>> bpf_prog_add()
>>> without checking the return code could fail.
>>>
>>> Fixes: 86994156c736 ("net/mlx5e: XDP fast RX drop bpf programs support")
>>> Signed-off-by: Daniel Borkmann <daniel@iogearbox.net>
>>> ---
>>> drivers/net/ethernet/mellanox/mlx5/core/en_main.c | 25
>>> ++++++++++++++++++-----
>>> 1 file changed, 20 insertions(+), 5 deletions(-)
>>>
>>> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
>>> b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
>>> index ab0c336..cf26672 100644
>>> --- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
>>> +++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
>>> @@ -3142,6 +3142,17 @@ static int mlx5e_xdp_set(struct net_device
>>> *netdev, struct bpf_prog *prog)
>>> goto unlock;
>>> }
>>>
>>> + if (prog) {
>>> + /* num_channels is invariant here, so we can take the
>>> + * batched reference right upfront.
>>> + */
>>> + prog = bpf_prog_add(prog, priv->params.num_channels);
>>> + if (IS_ERR(prog)) {
>>> + err = PTR_ERR(prog);
>>> + goto unlock;
>>> + }
>>> + }
>>> +
>>
>>
>> With this approach you will end up taking a ref count twice per RQ! on
>> the first time xdp_set is called i.e (old_prog = NULL, prog != NULL).
>> One ref will be taken per RQ/Channel from the above code, and since
>> reset will be TRUE mlx5e_open_locked will be called and another ref
>> count will be taken on mlx5e_create_rq.
>>
>> The issue here is that we have two places for ref count accounting,
>> xdp_set and mlx5e_create_rq. Having ref-count updates in
>> mlx5e_create_rq is essential for num_channels configuration changes
>> (mlx5e_set_ringparam).
>>
>> Our previous approach made sure that only one path will do the ref
>> counting (mlx5e_open_locked vs. mlx5e_xdp_set batch ref update when
>> reset == false).
>
>
> That is correct, for a short time bpf_prog_add() was charged also when
> we reset. When we reset, we will then jump to unlock_put and safely undo
> this since we took ref from mlx5e_create_rq() already in that case, and
> return successfully. That was intentional, so that eventually we end up
> just taking a /single/ ref per RQ when we exit mlx5e_xdp_set(), but more
> below ...
>
> [...]
>>
>> 2. Keep the current approach and make sure to not update the ref count
>> twice, you can batch update only if (!reset && was_open) otherwise you
>> can rely on mlx5e_open_locked to take the num_channels ref count for
>> you.
>>
>> Personally I prefer option 2, since we will keep the current logic
>> which will allow configuration updates such as (mlx5e_set_ringparam)
>> to not worry about ref counting since it will be done in the reset
>> flow.
>
>
> ... agree on keeping current approach. I actually like the idea, so we'd
> end up with this simpler version for the batched ref then.
>
Great :).
So let's do the batched update only if we are not going to reset (we
already know that in advance), instead of the current patch where you
batch update unconditionally and then
unlock_put in case reset was performed (which is just redundant and confusing).
Please note that if open_locked fails you need to goto unlock_put.
> Note, your "bpf_prog_add(prog, 1) // one ref for the device." is not needed
> since this we already got that one through dev_change_xdp_fd() as mentioned.
>
If it is not needed then why we need bpf_prog_put on mlx5e_nic_cleanup
in your next patch? this doesn't look symmetric (right) !
you shouldn't release a reference you didn't take.
Overall with this series the driver can take num_channels refs and
will release num_channels refs on mlx5e_close. we shouldn't release
one extra ref on NIC cleanup.
^ permalink raw reply
* [PATCH net] : add a missing rcu synchronization
From: Eric Dumazet @ 2016-11-16 14:31 UTC (permalink / raw)
To: David Miller; +Cc: netdev, Michael Chan
From: Eric Dumazet <edumazet@google.com>
Add a missing synchronize_net() call to avoid potential use after free,
since we explicitly call napi_hash_del() to factorize the RCU grace
period.
Fixes: c0c050c58d84 ("bnxt_en: New Broadcom ethernet driver.")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Cc: Michael Chan <michael.chan@broadcom.com>
---
drivers/net/ethernet/broadcom/bnxt/bnxt.c | 4 ++++
1 file changed, 4 insertions(+)
diff --git a/drivers/net/ethernet/broadcom/bnxt/bnxt.c b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
index c6909660e097..e18635b2a002 100644
--- a/drivers/net/ethernet/broadcom/bnxt/bnxt.c
+++ b/drivers/net/ethernet/broadcom/bnxt/bnxt.c
@@ -4934,6 +4934,10 @@ static void bnxt_del_napi(struct bnxt *bp)
napi_hash_del(&bnapi->napi);
netif_napi_del(&bnapi->napi);
}
+ /* We called napi_hash_del() before netif_napi_del(), we need
+ * to respect an RCU grace period before freeing napi structures.
+ */
+ synchronize_net();
}
static void bnxt_init_napi(struct bnxt *bp)
^ permalink raw reply related
* [PATCH net 0/7] net: cpsw: fix leaks and probe deferral
From: Johan Hovold @ 2016-11-16 14:35 UTC (permalink / raw)
To: Mugunthan V N
Cc: Grygorii Strashko, linux-omap, netdev, linux-kernel, Johan Hovold
This series fixes as number of leaks and issues in the cpsw probe-error
and driver-unbind paths, some which specifically prevented deferred
probing.
Johan
Johan Hovold (7):
net: ethernet: ti: cpsw: fix bad register access in probe error path
net: ethernet: ti: cpsw: fix mdio device reference leak
net: ethernet: ti: cpsw: fix deferred probe
net: ethernet: ti: cpsw: fix of_node and phydev leaks
net: ethernet: ti: cpsw: fix secondary-emac probe error path
net: ethernet: ti: cpsw: add missing sanity check
net: ethernet: ti: cpsw: fix fixed-link phy probe deferral
drivers/net/ethernet/ti/cpsw.c | 88 ++++++++++++++++++++++++++++++++++--------
1 file changed, 71 insertions(+), 17 deletions(-)
--
2.7.3
^ permalink raw reply
* [PATCH net 2/7] net: ethernet: ti: cpsw: fix mdio device reference leak
From: Johan Hovold @ 2016-11-16 14:35 UTC (permalink / raw)
To: Mugunthan V N
Cc: Grygorii Strashko, linux-omap, netdev, linux-kernel, Johan Hovold
In-Reply-To: <1479306916-27673-1-git-send-email-johan@kernel.org>
Make sure to drop the reference taken by of_find_device_by_node() when
looking up an mdio device from a phy_id property during probe.
Fixes: 549985ee9c72 ("cpsw: simplify the setup of the register
pointers")
Signed-off-by: Johan Hovold <johan@kernel.org>
---
drivers/net/ethernet/ti/cpsw.c | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 5bc5e6189661..ee1fae914abc 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -2397,6 +2397,7 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
}
snprintf(slave_data->phy_id, sizeof(slave_data->phy_id),
PHY_ID_FMT, mdio->name, phyid);
+ put_device(&mdio->dev);
} else {
dev_err(&pdev->dev,
"No slave[%d] phy_id, phy-handle, or fixed-link property\n",
--
2.7.3
^ permalink raw reply related
* Re: [PATCH v2 net-next] net/mlx5e: remove napi_hash_del() calls
From: Saeed Mahameed @ 2016-11-16 14:34 UTC (permalink / raw)
To: Eric Dumazet; +Cc: David Miller, Saeed Mahameed, netdev
In-Reply-To: <1479306094.8455.181.camel@edumazet-glaptop3.roam.corp.google.com>
On Wed, Nov 16, 2016 at 4:21 PM, Eric Dumazet <eric.dumazet@gmail.com> wrote:
> From: Eric Dumazet <edumazet@google.com>
>
> Calling napi_hash_del() after netif_napi_del() is pointless.
>
> Signed-off-by: Eric Dumazet <edumazet@google.com>
> Cc: Saeed Mahameed <saeedm@mellanox.com>
Acked-by: Saeed Mahameed <saeedm@mellanox.com>
^ permalink raw reply
* [PATCH net 3/7] net: ethernet: ti: cpsw: fix deferred probe
From: Johan Hovold @ 2016-11-16 14:35 UTC (permalink / raw)
To: Mugunthan V N
Cc: Grygorii Strashko, linux-omap, netdev, linux-kernel, Johan Hovold
In-Reply-To: <1479306916-27673-1-git-send-email-johan@kernel.org>
Make sure to deregister all child devices also on probe errors to avoid
leaks and to fix probe deferral.
cpsw 4a100000.ethernet: omap_device: omap_device_enable() called from invalid state 1
cpsw 4a100000.ethernet: use pm_runtime_put_sync_suspend() in driver?
cpsw: probe of 4a100000.ethernet failed with error -22
Add generic helper to undo the effects of cpsw_probe_dt(), which will
also be used in a follow-on patch to fix further leaks that have been
introduced more recently.
Fixes: 1fb19aa730e4 ("net: cpsw: Add parent<->child relation support
between cpsw and mdio")
Signed-off-by: Johan Hovold <johan@kernel.org>
---
drivers/net/ethernet/ti/cpsw.c | 26 ++++++++++++++++----------
1 file changed, 16 insertions(+), 10 deletions(-)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index ee1fae914abc..4374ba05610b 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -2441,6 +2441,11 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
return 0;
}
+static void cpsw_remove_dt(struct platform_device *pdev)
+{
+ of_platform_depopulate(&pdev->dev);
+}
+
static int cpsw_probe_dual_emac(struct cpsw_priv *priv)
{
struct cpsw_common *cpsw = priv->cpsw;
@@ -2588,7 +2593,7 @@ static int cpsw_probe(struct platform_device *pdev)
if (cpsw_probe_dt(&cpsw->data, pdev)) {
dev_err(&pdev->dev, "cpsw: platform data missing\n");
ret = -ENODEV;
- goto clean_runtime_disable_ret;
+ goto clean_dt_ret;
}
data = &cpsw->data;
cpsw->rx_ch_num = 1;
@@ -2609,7 +2614,7 @@ static int cpsw_probe(struct platform_device *pdev)
GFP_KERNEL);
if (!cpsw->slaves) {
ret = -ENOMEM;
- goto clean_runtime_disable_ret;
+ goto clean_dt_ret;
}
for (i = 0; i < data->slaves; i++)
cpsw->slaves[i].slave_num = i;
@@ -2621,7 +2626,7 @@ static int cpsw_probe(struct platform_device *pdev)
if (IS_ERR(clk)) {
dev_err(priv->dev, "fck is not found\n");
ret = -ENODEV;
- goto clean_runtime_disable_ret;
+ goto clean_dt_ret;
}
cpsw->bus_freq_mhz = clk_get_rate(clk) / 1000000;
@@ -2629,7 +2634,7 @@ static int cpsw_probe(struct platform_device *pdev)
ss_regs = devm_ioremap_resource(&pdev->dev, ss_res);
if (IS_ERR(ss_regs)) {
ret = PTR_ERR(ss_regs);
- goto clean_runtime_disable_ret;
+ goto clean_dt_ret;
}
cpsw->regs = ss_regs;
@@ -2639,7 +2644,7 @@ static int cpsw_probe(struct platform_device *pdev)
ret = pm_runtime_get_sync(&pdev->dev);
if (ret < 0) {
pm_runtime_put_noidle(&pdev->dev);
- goto clean_runtime_disable_ret;
+ goto clean_dt_ret;
}
cpsw->version = readl(&cpsw->regs->id_ver);
pm_runtime_put_sync(&pdev->dev);
@@ -2648,7 +2653,7 @@ static int cpsw_probe(struct platform_device *pdev)
cpsw->wr_regs = devm_ioremap_resource(&pdev->dev, res);
if (IS_ERR(cpsw->wr_regs)) {
ret = PTR_ERR(cpsw->wr_regs);
- goto clean_runtime_disable_ret;
+ goto clean_dt_ret;
}
memset(&dma_params, 0, sizeof(dma_params));
@@ -2685,7 +2690,7 @@ static int cpsw_probe(struct platform_device *pdev)
default:
dev_err(priv->dev, "unknown version 0x%08x\n", cpsw->version);
ret = -ENODEV;
- goto clean_runtime_disable_ret;
+ goto clean_dt_ret;
}
for (i = 0; i < cpsw->data.slaves; i++) {
struct cpsw_slave *slave = &cpsw->slaves[i];
@@ -2714,7 +2719,7 @@ static int cpsw_probe(struct platform_device *pdev)
if (!cpsw->dma) {
dev_err(priv->dev, "error initializing dma\n");
ret = -ENOMEM;
- goto clean_runtime_disable_ret;
+ goto clean_dt_ret;
}
cpsw->txch[0] = cpdma_chan_create(cpsw->dma, 0, cpsw_tx_handler, 0);
@@ -2827,7 +2832,8 @@ static int cpsw_probe(struct platform_device *pdev)
}
clean_dma_ret:
cpdma_ctlr_destroy(cpsw->dma);
-clean_runtime_disable_ret:
+clean_dt_ret:
+ cpsw_remove_dt(pdev);
pm_runtime_disable(&pdev->dev);
clean_ndev_ret:
free_netdev(priv->ndev);
@@ -2852,7 +2858,7 @@ static int cpsw_remove(struct platform_device *pdev)
cpsw_ale_destroy(cpsw->ale);
cpdma_ctlr_destroy(cpsw->dma);
- of_platform_depopulate(&pdev->dev);
+ cpsw_remove_dt(pdev);
pm_runtime_put_sync(&pdev->dev);
pm_runtime_disable(&pdev->dev);
if (cpsw->data.dual_emac)
--
2.7.3
^ permalink raw reply related
* [PATCH net 4/7] net: ethernet: ti: cpsw: fix of_node and phydev leaks
From: Johan Hovold @ 2016-11-16 14:35 UTC (permalink / raw)
To: Mugunthan V N
Cc: Grygorii Strashko, linux-omap, netdev, linux-kernel, Johan Hovold
In-Reply-To: <1479306916-27673-1-git-send-email-johan@kernel.org>
Make sure to drop references taken and deregister devices registered
during probe on probe errors (including deferred probe) and driver
unbind.
Specifically, PHY of-node references were never released and fixed-link
PHY devices were never deregistered.
Fixes: 9e42f715264f ("drivers: net: cpsw: add phy-handle parsing")
Fixes: 1f71e8c96fc6 ("drivers: net: cpsw: Add support for fixed-link
PHY")
Signed-off-by: Johan Hovold <johan@kernel.org>
---
drivers/net/ethernet/ti/cpsw.c | 35 +++++++++++++++++++++++++++++++++++
1 file changed, 35 insertions(+)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 4374ba05610b..4bd96732291c 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -2443,6 +2443,41 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
static void cpsw_remove_dt(struct platform_device *pdev)
{
+ struct net_device *ndev = platform_get_drvdata(pdev);
+ struct cpsw_common *cpsw = ndev_to_cpsw(ndev);
+ struct cpsw_platform_data *data = &cpsw->data;
+ struct device_node *node = pdev->dev.of_node;
+ struct device_node *slave_node;
+ int i = 0;
+
+ for_each_available_child_of_node(node, slave_node) {
+ struct cpsw_slave_data *slave_data = &data->slave_data[i];
+
+ if (strcmp(slave_node->name, "slave"))
+ continue;
+
+ if (of_phy_is_fixed_link(slave_node)) {
+ struct phy_device *phydev;
+
+ phydev = of_phy_find_device(slave_node);
+ if (phydev) {
+ fixed_phy_unregister(phydev);
+ /* Put references taken by
+ * of_phy_find_device() and
+ * of_phy_register_fixed_link().
+ */
+ phy_device_free(phydev);
+ phy_device_free(phydev);
+ }
+ }
+
+ of_node_put(slave_data->phy_node);
+
+ i++;
+ if (i == data->slaves)
+ break;
+ }
+
of_platform_depopulate(&pdev->dev);
}
--
2.7.3
^ permalink raw reply related
* [PATCH net 5/7] net: ethernet: ti: cpsw: fix secondary-emac probe error path
From: Johan Hovold @ 2016-11-16 14:35 UTC (permalink / raw)
To: Mugunthan V N
Cc: Grygorii Strashko, linux-omap, netdev, linux-kernel, Johan Hovold
In-Reply-To: <1479306916-27673-1-git-send-email-johan@kernel.org>
Make sure to deregister the primary device in case the secondary emac
fails to probe.
kernel BUG at /home/johan/work/omicron/src/linux/net/core/dev.c:7743!
...
[<c05b3dec>] (free_netdev) from [<c04fe6c0>] (cpsw_probe+0x9cc/0xe50)
[<c04fe6c0>] (cpsw_probe) from [<c047b28c>] (platform_drv_probe+0x5c/0xc0)
Fixes: d9ba8f9e6298 ("driver: net: ethernet: cpsw: dual emac interface
implementation")
Signed-off-by: Johan Hovold <johan@kernel.org>
---
drivers/net/ethernet/ti/cpsw.c | 4 +++-
1 file changed, 3 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 4bd96732291c..421da97a5be6 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -2852,12 +2852,14 @@ static int cpsw_probe(struct platform_device *pdev)
ret = cpsw_probe_dual_emac(priv);
if (ret) {
cpsw_err(priv, probe, "error probe slave 2 emac interface\n");
- goto clean_ale_ret;
+ goto clean_unregister_netdev_ret;
}
}
return 0;
+clean_unregister_netdev_ret:
+ unregister_netdev(ndev);
clean_ale_ret:
if (pm_runtime_get_sync(&pdev->dev) < 0) {
pm_runtime_put_noidle(&pdev->dev);
--
2.7.3
^ permalink raw reply related
* [PATCH net 6/7] net: ethernet: ti: cpsw: add missing sanity check
From: Johan Hovold @ 2016-11-16 14:35 UTC (permalink / raw)
To: Mugunthan V N
Cc: Grygorii Strashko, linux-omap, netdev, linux-kernel, Johan Hovold
In-Reply-To: <1479306916-27673-1-git-send-email-johan@kernel.org>
Make sure to check for allocation failures before dereferencing a
NULL-pointer during probe.
Fixes: 649a1688c960 ("net: ethernet: ti: cpsw: create common struct to
hold shared driver data")
Signed-off-by: Johan Hovold <johan@kernel.org>
---
drivers/net/ethernet/ti/cpsw.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 421da97a5be6..411be060fee6 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -2588,6 +2588,9 @@ static int cpsw_probe(struct platform_device *pdev)
int irq;
cpsw = devm_kzalloc(&pdev->dev, sizeof(struct cpsw_common), GFP_KERNEL);
+ if (!cpsw)
+ return -ENOMEM;
+
cpsw->dev = &pdev->dev;
ndev = alloc_etherdev_mq(sizeof(struct cpsw_priv), CPSW_MAX_QUEUES);
--
2.7.3
^ permalink raw reply related
* [PATCH net 7/7] net: ethernet: ti: cpsw: fix fixed-link phy probe deferral
From: Johan Hovold @ 2016-11-16 14:35 UTC (permalink / raw)
To: Mugunthan V N
Cc: Grygorii Strashko, linux-omap, netdev, linux-kernel, Johan Hovold
In-Reply-To: <1479306916-27673-1-git-send-email-johan@kernel.org>
Make sure to propagate errors from of_phy_register_fixed_link() which
can fail with -EPROBE_DEFER.
Fixes: 1f71e8c96fc6 ("drivers: net: cpsw: Add support for fixed-link
PHY")
Signed-off-by: Johan Hovold <johan@kernel.org>
---
drivers/net/ethernet/ti/cpsw.c | 12 +++++++-----
1 file changed, 7 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 411be060fee6..cf8f1afa46bf 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -2375,8 +2375,11 @@ static int cpsw_probe_dt(struct cpsw_platform_data *data,
* to the PHY is the Ethernet MAC DT node.
*/
ret = of_phy_register_fixed_link(slave_node);
- if (ret)
+ if (ret) {
+ if (ret != -EPROBE_DEFER)
+ dev_err(&pdev->dev, "failed to register fixed-link phy: %d\n", ret);
return ret;
+ }
slave_data->phy_node = of_node_get(slave_node);
} else if (parp) {
u32 phyid;
@@ -2628,11 +2631,10 @@ static int cpsw_probe(struct platform_device *pdev)
/* Select default pin state */
pinctrl_pm_select_default_state(&pdev->dev);
- if (cpsw_probe_dt(&cpsw->data, pdev)) {
- dev_err(&pdev->dev, "cpsw: platform data missing\n");
- ret = -ENODEV;
+ ret = cpsw_probe_dt(&cpsw->data, pdev);
+ if (ret)
goto clean_dt_ret;
- }
+
data = &cpsw->data;
cpsw->rx_ch_num = 1;
cpsw->tx_ch_num = 1;
--
2.7.3
^ permalink raw reply related
* [PATCH net 1/7] net: ethernet: ti: cpsw: fix bad register access in probe error path
From: Johan Hovold @ 2016-11-16 14:35 UTC (permalink / raw)
To: Mugunthan V N
Cc: Grygorii Strashko, linux-omap, netdev, linux-kernel, Johan Hovold
In-Reply-To: <1479306916-27673-1-git-send-email-johan@kernel.org>
Make sure to resume the platform device to enable clocks before
accessing the CPSW registers in the probe error path (e.g. for deferred
probe).
Unhandled fault: external abort on non-linefetch (0x1008) at 0xd0872d08
...
[<c04fabcc>] (cpsw_ale_control_set) from [<c04fb8b4>] (cpsw_ale_destroy+0x2c/0x44)
[<c04fb8b4>] (cpsw_ale_destroy) from [<c04fea58>] (cpsw_probe+0xbd0/0x10c4)
[<c04fea58>] (cpsw_probe) from [<c047b2a0>] (platform_drv_probe+0x5c/0xc0)
Note that in the unlikely event of a runtime-resume failure, we'll leak
the ale struct.
Fixes: df828598a755 ("netdev: driver: ethernet: Add TI CPSW driver")
Signed-off-by: Johan Hovold <johan@kernel.org>
---
drivers/net/ethernet/ti/cpsw.c | 7 ++++++-
1 file changed, 6 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index c6cff3d2ff05..5bc5e6189661 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -2818,7 +2818,12 @@ static int cpsw_probe(struct platform_device *pdev)
return 0;
clean_ale_ret:
- cpsw_ale_destroy(cpsw->ale);
+ if (pm_runtime_get_sync(&pdev->dev) < 0) {
+ pm_runtime_put_noidle(&pdev->dev);
+ } else {
+ cpsw_ale_destroy(cpsw->ale);
+ pm_runtime_put_sync(&pdev->dev);
+ }
clean_dma_ret:
cpdma_ctlr_destroy(cpsw->dma);
clean_runtime_disable_ret:
--
2.7.3
^ permalink raw reply related
* Re: [PATCH] net: ethernet: faraday: To support device tree usage.
From: Andrew Lunn @ 2016-11-16 14:37 UTC (permalink / raw)
To: Greentime Hu; +Cc: netdev, devicetree
In-Reply-To: <CAEbi=3esyRi4cnUNf4uPKCQtyRekBRo81Frd=BSJy8qf=3uc+A@mail.gmail.com>
On Wed, Nov 16, 2016 at 10:26:52PM +0800, Greentime Hu wrote:
> On Wed, Nov 16, 2016 at 9:47 PM, Andrew Lunn <andrew@lunn.ch> wrote:
> > On Wed, Nov 16, 2016 at 04:43:15PM +0800, Greentime Hu wrote:
> >> To support device tree usage for ftmac100.
> >>
> >> Signed-off-by: Greentime Hu <green.hu@gmail.com>
> >> ---
> >> drivers/net/ethernet/faraday/ftmac100.c | 7 +++++++
> >> 1 file changed, 7 insertions(+)
> >>
> >> diff --git a/drivers/net/ethernet/faraday/ftmac100.c b/drivers/net/ethernet/faraday/ftmac100.c
> >> index dce5f7b..81dd9e1 100644
> >> --- a/drivers/net/ethernet/faraday/ftmac100.c
> >> +++ b/drivers/net/ethernet/faraday/ftmac100.c
> >> @@ -1172,11 +1172,17 @@ static int __exit ftmac100_remove(struct platform_device *pdev)
> >> return 0;
> >> }
> >>
> >> +static const struct of_device_id mac_of_ids[] = {
> >> + { .compatible = "andestech,atmac100" },
> >> + { }
> >
> > andestech is not in
> > Documentation/devicetree/bindings/vendor-prefixes.txt Please provide a
> > separate patch adding it.
> OK. I will provide another patch to add andestech.
>
> > Humm, why andestech? Why not something based around faraday
> > technology?
> It is because we use the same ftmac100 IP provided from faraday
> technology but I am now using it in andestech SoC.
Please make sure you get an acked-by: from the device tree
maintainers. They might want you to use faraday, since that is the
original IP provider. For example, all Synopsys licensed IP uses
"snps,XXX", not the SoC vendor with the license.
Andrew
^ permalink raw reply
* [PATCH] bpf: fix possible uninitialized access in inactive rotation
From: Arnd Bergmann @ 2016-11-16 14:38 UTC (permalink / raw)
To: Alexei Starovoitov
Cc: Arnd Bergmann, Martin KaFai Lau, David S. Miller, netdev,
linux-kernel
This newly added code causes a build warning:
kernel/bpf/bpf_lru_list.c: In function '__bpf_lru_list_rotate_inactive':
kernel/bpf/bpf_lru_list.c:201:28: error: 'next' may be used uninitialized in this function [-Werror=maybe-uninitialized]
The warning is plausible from looking at the code, though there might
be non-obvious external constraints that ensure it always works.
Moving the assignment of ->next_inactive_rotation inside of the
loop makes it obvious to the reader and the compiler when we
actually want to update ->next.
Fixes: 3a08c2fd7634 ("bpf: LRU List")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
---
kernel/bpf/bpf_lru_list.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/kernel/bpf/bpf_lru_list.c b/kernel/bpf/bpf_lru_list.c
index bfebff010ba9..f462e5f09703 100644
--- a/kernel/bpf/bpf_lru_list.c
+++ b/kernel/bpf/bpf_lru_list.c
@@ -192,13 +192,14 @@ static void __bpf_lru_list_rotate_inactive(struct bpf_lru *lru,
next = cur->prev;
if (bpf_lru_node_is_ref(node))
__bpf_lru_node_move(l, node, BPF_LRU_LIST_T_ACTIVE);
- if (cur == last)
+ if (cur == last) {
+ l->next_inactive_rotation = next;
break;
+ }
cur = next;
i++;
}
- l->next_inactive_rotation = next;
}
/* Shrink the inactive list. It starts from the tail of the
--
2.9.0
^ permalink raw reply related
* [PATCH] net: dsa: fix fixed-link-phy device leaks
From: Johan Hovold @ 2016-11-16 14:47 UTC (permalink / raw)
To: Andrew Lunn
Cc: Vivien Didelot, Florian Fainelli, David S. Miller, netdev,
linux-kernel, Johan Hovold
Make sure to drop the reference taken by of_phy_find_device() when
registering and deregistering the fixed-link PHY-device.
Note that we need to put both references held at deregistration.
Fixes: 39b0c705195e ("net: dsa: Allow configuration of CPU & DSA port
speeds/duplex")
Signed-off-by: Johan Hovold <johan@kernel.org>
---
Hi,
This is one has been compile tested only, but fixes a couple of leaks
similar to one that was found in the cpsw driver for which I just posted
a patch.
It turns out all drivers but DSA fail to deregister the fixed-link PHYs
registered by of_phy_register_fixed_link(). Due to the way this
interface was designed, deregistering such a PHY is a bit cumbersome and
looks like it would benefit from a common helper.
However, perhaps the interface should instead be changed so that the PHY
device is returned so that drivers do not need to use
of_phy_find_device() when they need to access properties of the fixed
link (e.g. as in dsu_cpu_dsa_setup below).
Thoughts?
Thanks,
Johan
net/dsa/dsa.c | 8 +++++++-
1 file changed, 7 insertions(+), 1 deletion(-)
diff --git a/net/dsa/dsa.c b/net/dsa/dsa.c
index a6902c1e2f28..798a6a776a5f 100644
--- a/net/dsa/dsa.c
+++ b/net/dsa/dsa.c
@@ -233,6 +233,8 @@ int dsa_cpu_dsa_setup(struct dsa_switch *ds, struct device *dev,
genphy_read_status(phydev);
if (ds->ops->adjust_link)
ds->ops->adjust_link(ds, port, phydev);
+
+ phy_device_free(phydev); /* of_phy_find_device */
}
return 0;
@@ -509,8 +511,12 @@ void dsa_cpu_dsa_destroy(struct device_node *port_dn)
if (of_phy_is_fixed_link(port_dn)) {
phydev = of_phy_find_device(port_dn);
if (phydev) {
- phy_device_free(phydev);
fixed_phy_unregister(phydev);
+ /* Put references taken by of_phy_find_device() and
+ * of_phy_register_fixed_link().
+ */
+ phy_device_free(phydev);
+ phy_device_free(phydev);
}
}
}
--
2.7.3
^ permalink raw reply related
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox