* [PATCH v8 0/8] fsi: Add IBM I2C Responder virtual FSI master
@ 2023-03-06 16:10 Eddie James
2023-03-06 16:10 ` [PATCH v8 1/8] fsi: Move fsi_slave structure definition to header Eddie James
` (7 more replies)
0 siblings, 8 replies; 9+ messages in thread
From: Eddie James @ 2023-03-06 16:10 UTC (permalink / raw)
To: linux-fsi
Cc: rostedt, linux-trace-kernel, devicetree, linux-kernel, mhiramat,
alistair, joel, jk, andrew, robh+dt, krzysztof.kozlowski+dt,
eajames
The I2C Responder (I2CR) is an I2C device that translates I2C commands
to CFAM or SCOM operations, effectively implementing an FSI master and
bus.
Changes since v7:
- Add FSI master device naming patch
- Use I2C adapter number as FSI master index with "i2cr" name
- Export the fsi_master_i2cr_read/write functions for the SCOM driver
Changes since v6:
- Drop conversions to BE except in the FSI core interface
- Emulate the CFAM config table since it doesn't exist off the I2CR
- Drop status check during probe
- Add release function for device
- Update SCOM driver with newer interfaces and drop BE conversion
- New sbefifo change to not check state during probe
Changes since v5:
- Make I2CR scom driver depend on I2CR FSI master
Changes since v4:
- Add I2CR scom driver and associated patches
- Use compatible strings for FSI drivers if specified
- Include aliased device numbering patch
- Restructure the trace events to eliminate holes
Changes since v3:
- Rework the endian-ness in i2cr_write
- Rework the tracing to include the i2c bus and device address
Changes since v2:
- Fix the bindings again, sorry for the spam
Changes since v1:
- Fix the binding document
- Change the binding name
- Clean up the size argument checking
- Reduce __force by using packed struct for the command
Eddie James (8):
fsi: Move fsi_slave structure definition to header
fsi: Make master device indexing and naming optional
dt-bindings: fsi: Document the IBM I2C Responder virtual FSI master
fsi: Add IBM I2C Responder virtual FSI master
fsi: Add I2C Responder SCOM driver
fsi: Add aliased device numbering
fsi: Use of_match_table for bus matching if specified
fsi: sbefifo: Don't check status during probe
.../bindings/fsi/ibm,i2cr-fsi-master.yaml | 41 +++
drivers/fsi/Kconfig | 17 +
drivers/fsi/Makefile | 2 +
drivers/fsi/fsi-core.c | 73 ++--
drivers/fsi/fsi-master-aspeed.c | 2 +-
drivers/fsi/fsi-master-ast-cf.c | 2 +-
drivers/fsi/fsi-master-gpio.c | 2 +-
drivers/fsi/fsi-master-hub.c | 2 +-
drivers/fsi/fsi-master-i2cr.c | 318 ++++++++++++++++++
drivers/fsi/fsi-master-i2cr.h | 24 ++
drivers/fsi/fsi-master.h | 3 +-
drivers/fsi/fsi-sbefifo.c | 8 -
drivers/fsi/fsi-scom.c | 8 +
drivers/fsi/fsi-slave.h | 28 ++
drivers/fsi/i2cr-scom.c | 154 +++++++++
include/trace/events/fsi_master_i2cr.h | 107 ++++++
16 files changed, 752 insertions(+), 39 deletions(-)
create mode 100644 Documentation/devicetree/bindings/fsi/ibm,i2cr-fsi-master.yaml
create mode 100644 drivers/fsi/fsi-master-i2cr.c
create mode 100644 drivers/fsi/fsi-master-i2cr.h
create mode 100644 drivers/fsi/fsi-slave.h
create mode 100644 drivers/fsi/i2cr-scom.c
create mode 100644 include/trace/events/fsi_master_i2cr.h
--
2.31.1
^ permalink raw reply [flat|nested] 9+ messages in thread
* [PATCH v8 1/8] fsi: Move fsi_slave structure definition to header
2023-03-06 16:10 [PATCH v8 0/8] fsi: Add IBM I2C Responder virtual FSI master Eddie James
@ 2023-03-06 16:10 ` Eddie James
2023-03-06 16:10 ` [PATCH v8 2/8] fsi: Make master device indexing and naming optional Eddie James
` (6 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Eddie James @ 2023-03-06 16:10 UTC (permalink / raw)
To: linux-fsi
Cc: rostedt, linux-trace-kernel, devicetree, linux-kernel, mhiramat,
alistair, joel, jk, andrew, robh+dt, krzysztof.kozlowski+dt,
eajames
Some FSI drivers may have need of the slave definition, so
move it to a header file. Also use one macro for obtaining a
pointer to the fsi_master structure.
Signed-off-by: Eddie James <eajames@linux.ibm.com>
---
drivers/fsi/fsi-core.c | 24 ++++--------------------
drivers/fsi/fsi-master-aspeed.c | 2 +-
drivers/fsi/fsi-master-ast-cf.c | 2 +-
drivers/fsi/fsi-master-gpio.c | 2 +-
drivers/fsi/fsi-master-hub.c | 2 +-
drivers/fsi/fsi-master.h | 2 +-
drivers/fsi/fsi-slave.h | 28 ++++++++++++++++++++++++++++
7 files changed, 37 insertions(+), 25 deletions(-)
create mode 100644 drivers/fsi/fsi-slave.h
diff --git a/drivers/fsi/fsi-core.c b/drivers/fsi/fsi-core.c
index 0b927c9f4267..d591e68afd11 100644
--- a/drivers/fsi/fsi-core.c
+++ b/drivers/fsi/fsi-core.c
@@ -23,6 +23,10 @@
#include <linux/uaccess.h>
#include "fsi-master.h"
+#include "fsi-slave.h"
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/fsi.h>
#define FSI_SLAVE_CONF_NEXT_MASK GENMASK(31, 31)
#define FSI_SLAVE_CONF_SLOTS_MASK GENMASK(23, 16)
@@ -78,26 +82,6 @@ static const int engine_page_size = 0x400;
static DEFINE_IDA(master_ida);
-struct fsi_slave {
- struct device dev;
- struct fsi_master *master;
- struct cdev cdev;
- int cdev_idx;
- int id; /* FSI address */
- int link; /* FSI link# */
- u32 cfam_id;
- int chip_id;
- uint32_t size; /* size of slave address space */
- u8 t_send_delay;
- u8 t_echo_delay;
-};
-
-#define CREATE_TRACE_POINTS
-#include <trace/events/fsi.h>
-
-#define to_fsi_master(d) container_of(d, struct fsi_master, dev)
-#define to_fsi_slave(d) container_of(d, struct fsi_slave, dev)
-
static const int slave_retries = 2;
static int discard_errors;
diff --git a/drivers/fsi/fsi-master-aspeed.c b/drivers/fsi/fsi-master-aspeed.c
index 7cec1772820d..437f87b4a6a3 100644
--- a/drivers/fsi/fsi-master-aspeed.c
+++ b/drivers/fsi/fsi-master-aspeed.c
@@ -376,7 +376,7 @@ static int aspeed_master_break(struct fsi_master *master, int link)
static void aspeed_master_release(struct device *dev)
{
struct fsi_master_aspeed *aspeed =
- to_fsi_master_aspeed(dev_to_fsi_master(dev));
+ to_fsi_master_aspeed(to_fsi_master(dev));
kfree(aspeed);
}
diff --git a/drivers/fsi/fsi-master-ast-cf.c b/drivers/fsi/fsi-master-ast-cf.c
index 5f608ef8b53c..6124978305eb 100644
--- a/drivers/fsi/fsi-master-ast-cf.c
+++ b/drivers/fsi/fsi-master-ast-cf.c
@@ -1190,7 +1190,7 @@ static int fsi_master_acf_gpio_release(void *data)
static void fsi_master_acf_release(struct device *dev)
{
- struct fsi_master_acf *master = to_fsi_master_acf(dev_to_fsi_master(dev));
+ struct fsi_master_acf *master = to_fsi_master_acf(to_fsi_master(dev));
/* Cleanup, stop coprocessor */
mutex_lock(&master->lock);
diff --git a/drivers/fsi/fsi-master-gpio.c b/drivers/fsi/fsi-master-gpio.c
index 7d5f29b4b595..ed03da4f2447 100644
--- a/drivers/fsi/fsi-master-gpio.c
+++ b/drivers/fsi/fsi-master-gpio.c
@@ -761,7 +761,7 @@ static DEVICE_ATTR(external_mode, 0664,
static void fsi_master_gpio_release(struct device *dev)
{
- struct fsi_master_gpio *master = to_fsi_master_gpio(dev_to_fsi_master(dev));
+ struct fsi_master_gpio *master = to_fsi_master_gpio(to_fsi_master(dev));
of_node_put(dev_of_node(master->dev));
diff --git a/drivers/fsi/fsi-master-hub.c b/drivers/fsi/fsi-master-hub.c
index 01f0a796111e..6d8b6e8854e5 100644
--- a/drivers/fsi/fsi-master-hub.c
+++ b/drivers/fsi/fsi-master-hub.c
@@ -105,7 +105,7 @@ static int hub_master_link_enable(struct fsi_master *master, int link,
static void hub_master_release(struct device *dev)
{
- struct fsi_master_hub *hub = to_fsi_master_hub(dev_to_fsi_master(dev));
+ struct fsi_master_hub *hub = to_fsi_master_hub(to_fsi_master(dev));
kfree(hub);
}
diff --git a/drivers/fsi/fsi-master.h b/drivers/fsi/fsi-master.h
index 4762315a46ba..967622c1cabf 100644
--- a/drivers/fsi/fsi-master.h
+++ b/drivers/fsi/fsi-master.h
@@ -136,7 +136,7 @@ struct fsi_master {
u8 t_send_delay, u8 t_echo_delay);
};
-#define dev_to_fsi_master(d) container_of(d, struct fsi_master, dev)
+#define to_fsi_master(d) container_of(d, struct fsi_master, dev)
/**
* fsi_master registration & lifetime: the fsi_master_register() and
diff --git a/drivers/fsi/fsi-slave.h b/drivers/fsi/fsi-slave.h
new file mode 100644
index 000000000000..1d63a585829d
--- /dev/null
+++ b/drivers/fsi/fsi-slave.h
@@ -0,0 +1,28 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (C) IBM Corporation 2023 */
+
+#ifndef DRIVERS_FSI_SLAVE_H
+#define DRIVERS_FSI_SLAVE_H
+
+#include <linux/cdev.h>
+#include <linux/device.h>
+
+struct fsi_master;
+
+struct fsi_slave {
+ struct device dev;
+ struct fsi_master *master;
+ struct cdev cdev;
+ int cdev_idx;
+ int id; /* FSI address */
+ int link; /* FSI link# */
+ u32 cfam_id;
+ int chip_id;
+ uint32_t size; /* size of slave address space */
+ u8 t_send_delay;
+ u8 t_echo_delay;
+};
+
+#define to_fsi_slave(d) container_of(d, struct fsi_slave, dev)
+
+#endif /* DRIVERS_FSI_SLAVE_H */
--
2.31.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH v8 2/8] fsi: Make master device indexing and naming optional
2023-03-06 16:10 [PATCH v8 0/8] fsi: Add IBM I2C Responder virtual FSI master Eddie James
2023-03-06 16:10 ` [PATCH v8 1/8] fsi: Move fsi_slave structure definition to header Eddie James
@ 2023-03-06 16:10 ` Eddie James
2023-03-06 16:10 ` [PATCH v8 3/8] dt-bindings: fsi: Document the IBM I2C Responder virtual FSI master Eddie James
` (5 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Eddie James @ 2023-03-06 16:10 UTC (permalink / raw)
To: linux-fsi
Cc: rostedt, linux-trace-kernel, devicetree, linux-kernel, mhiramat,
alistair, joel, jk, andrew, robh+dt, krzysztof.kozlowski+dt,
eajames
Some FSI masters need different device naming, so make the indexing and naming
an optional part of FSI master registration.
Signed-off-by: Eddie James <eajames@linux.ibm.com>
---
drivers/fsi/fsi-core.c | 13 +++++++++----
1 file changed, 9 insertions(+), 4 deletions(-)
diff --git a/drivers/fsi/fsi-core.c b/drivers/fsi/fsi-core.c
index d591e68afd11..3271e40702ec 100644
--- a/drivers/fsi/fsi-core.c
+++ b/drivers/fsi/fsi-core.c
@@ -1297,11 +1297,16 @@ int fsi_master_register(struct fsi_master *master)
struct device_node *np;
mutex_init(&master->scan_lock);
- master->idx = ida_simple_get(&master_ida, 0, INT_MAX, GFP_KERNEL);
- if (master->idx < 0)
- return master->idx;
- dev_set_name(&master->dev, "fsi%d", master->idx);
+ if (!master->idx) {
+ master->idx = ida_simple_get(&master_ida, 0, INT_MAX,
+ GFP_KERNEL);
+ if (master->idx < 0)
+ return master->idx;
+
+ dev_set_name(&master->dev, "fsi%d", master->idx);
+ }
+
master->dev.class = &fsi_master_class;
rc = device_register(&master->dev);
--
2.31.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH v8 3/8] dt-bindings: fsi: Document the IBM I2C Responder virtual FSI master
2023-03-06 16:10 [PATCH v8 0/8] fsi: Add IBM I2C Responder virtual FSI master Eddie James
2023-03-06 16:10 ` [PATCH v8 1/8] fsi: Move fsi_slave structure definition to header Eddie James
2023-03-06 16:10 ` [PATCH v8 2/8] fsi: Make master device indexing and naming optional Eddie James
@ 2023-03-06 16:10 ` Eddie James
2023-03-06 16:10 ` [PATCH v8 4/8] fsi: Add " Eddie James
` (4 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Eddie James @ 2023-03-06 16:10 UTC (permalink / raw)
To: linux-fsi
Cc: rostedt, linux-trace-kernel, devicetree, linux-kernel, mhiramat,
alistair, joel, jk, andrew, robh+dt, krzysztof.kozlowski+dt,
eajames
The I2C Responder translates I2C commands to CFAM or SCOM operations,
effectively implementing an FSI master.
Signed-off-by: Eddie James <eajames@linux.ibm.com>
Reviewed-by: Andrew Jeffery <andrew@aj.id.au>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
---
.../bindings/fsi/ibm,i2cr-fsi-master.yaml | 41 +++++++++++++++++++
1 file changed, 41 insertions(+)
create mode 100644 Documentation/devicetree/bindings/fsi/ibm,i2cr-fsi-master.yaml
diff --git a/Documentation/devicetree/bindings/fsi/ibm,i2cr-fsi-master.yaml b/Documentation/devicetree/bindings/fsi/ibm,i2cr-fsi-master.yaml
new file mode 100644
index 000000000000..442cecdc57cb
--- /dev/null
+++ b/Documentation/devicetree/bindings/fsi/ibm,i2cr-fsi-master.yaml
@@ -0,0 +1,41 @@
+# SPDX-License-Identifier: (GPL-2.0-only OR BSD-2-Clause)
+%YAML 1.2
+---
+$id: http://devicetree.org/schemas/fsi/ibm,i2cr-fsi-master.yaml#
+$schema: http://devicetree.org/meta-schemas/core.yaml#
+
+title: IBM I2C Responder virtual FSI master
+
+maintainers:
+ - Eddie James <eajames@linux.ibm.com>
+
+description: |
+ The I2C Responder (I2CR) is a an I2C device that's connected to an FSI CFAM
+ (see fsi.txt). The I2CR translates I2C bus operations to FSI CFAM reads and
+ writes or SCOM operations, thereby acting as an FSI master.
+
+properties:
+ compatible:
+ enum:
+ - ibm,i2cr-fsi-master
+
+ reg:
+ maxItems: 1
+
+required:
+ - compatible
+ - reg
+
+additionalProperties: false
+
+examples:
+ - |
+ i2c {
+ #address-cells = <1>;
+ #size-cells = <0>;
+
+ i2cr@20 {
+ compatible = "ibm,i2cr-fsi-master";
+ reg = <0x20>;
+ };
+ };
--
2.31.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH v8 4/8] fsi: Add IBM I2C Responder virtual FSI master
2023-03-06 16:10 [PATCH v8 0/8] fsi: Add IBM I2C Responder virtual FSI master Eddie James
` (2 preceding siblings ...)
2023-03-06 16:10 ` [PATCH v8 3/8] dt-bindings: fsi: Document the IBM I2C Responder virtual FSI master Eddie James
@ 2023-03-06 16:10 ` Eddie James
2023-03-06 16:10 ` [PATCH v8 5/8] fsi: Add I2C Responder SCOM driver Eddie James
` (3 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Eddie James @ 2023-03-06 16:10 UTC (permalink / raw)
To: linux-fsi
Cc: rostedt, linux-trace-kernel, devicetree, linux-kernel, mhiramat,
alistair, joel, jk, andrew, robh+dt, krzysztof.kozlowski+dt,
eajames
The I2C Responder (I2CR) is an I2C device that translates I2C commands
to CFAM or SCOM operations, effectively implementing an FSI master and
bus.
Signed-off-by: Eddie James <eajames@linux.ibm.com>
---
drivers/fsi/Kconfig | 9 +
drivers/fsi/Makefile | 1 +
drivers/fsi/fsi-master-i2cr.c | 318 +++++++++++++++++++++++++
drivers/fsi/fsi-master-i2cr.h | 24 ++
drivers/fsi/fsi-master.h | 1 +
include/trace/events/fsi_master_i2cr.h | 107 +++++++++
6 files changed, 460 insertions(+)
create mode 100644 drivers/fsi/fsi-master-i2cr.c
create mode 100644 drivers/fsi/fsi-master-i2cr.h
create mode 100644 include/trace/events/fsi_master_i2cr.h
diff --git a/drivers/fsi/Kconfig b/drivers/fsi/Kconfig
index e6668a869913..999be82720c5 100644
--- a/drivers/fsi/Kconfig
+++ b/drivers/fsi/Kconfig
@@ -62,6 +62,15 @@ config FSI_MASTER_ASPEED
Enable it for your BMC kernel in an OpenPower or IBM Power system.
+config FSI_MASTER_I2CR
+ tristate "IBM I2C Responder virtual FSI master"
+ depends on I2C
+ help
+ This option enables a virtual FSI master in order to access a CFAM
+ behind an IBM I2C Responder (I2CR) chip. The I2CR is an I2C device
+ that translates I2C commands to CFAM or SCOM operations, effectively
+ implementing an FSI master and bus.
+
config FSI_SCOM
tristate "SCOM FSI client device driver"
help
diff --git a/drivers/fsi/Makefile b/drivers/fsi/Makefile
index da218a1ad8e1..34dbaa1c452e 100644
--- a/drivers/fsi/Makefile
+++ b/drivers/fsi/Makefile
@@ -4,6 +4,7 @@ obj-$(CONFIG_FSI) += fsi-core.o
obj-$(CONFIG_FSI_MASTER_HUB) += fsi-master-hub.o
obj-$(CONFIG_FSI_MASTER_ASPEED) += fsi-master-aspeed.o
obj-$(CONFIG_FSI_MASTER_GPIO) += fsi-master-gpio.o
+obj-$(CONFIG_FSI_MASTER_I2CR) += fsi-master-i2cr.o
obj-$(CONFIG_FSI_MASTER_AST_CF) += fsi-master-ast-cf.o
obj-$(CONFIG_FSI_SCOM) += fsi-scom.o
obj-$(CONFIG_FSI_SBEFIFO) += fsi-sbefifo.o
diff --git a/drivers/fsi/fsi-master-i2cr.c b/drivers/fsi/fsi-master-i2cr.c
new file mode 100644
index 000000000000..e4e8a5931ca3
--- /dev/null
+++ b/drivers/fsi/fsi-master-i2cr.c
@@ -0,0 +1,318 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (C) IBM Corporation 2023 */
+
+#include <linux/device.h>
+#include <linux/fsi.h>
+#include <linux/i2c.h>
+#include <linux/module.h>
+#include <linux/mod_devicetable.h>
+#include <linux/mutex.h>
+
+#include "fsi-master-i2cr.h"
+
+#define CREATE_TRACE_POINTS
+#include <trace/events/fsi_master_i2cr.h>
+
+#define I2CR_ADDRESS_CFAM(a) ((a) >> 2)
+#define I2CR_INITIAL_PARITY true
+
+#define I2CR_STATUS_CMD 0x60002
+#define I2CR_STATUS_ERR BIT_ULL(61)
+#define I2CR_ERROR_CMD 0x60004
+#define I2CR_LOG_CMD 0x60008
+
+static const u8 i2cr_cfam[] = {
+ 0xc0, 0x02, 0x0d, 0xa6,
+ 0x80, 0x01, 0x10, 0x02,
+ 0x80, 0x01, 0x10, 0x02,
+ 0x80, 0x01, 0x10, 0x02,
+ 0x80, 0x01, 0x80, 0x52,
+ 0x80, 0x01, 0x10, 0x02,
+ 0x80, 0x01, 0x10, 0x02,
+ 0x80, 0x01, 0x10, 0x02,
+ 0x80, 0x01, 0x10, 0x02,
+ 0x80, 0x01, 0x22, 0x2d,
+ 0x00, 0x00, 0x00, 0x00,
+ 0xde, 0xad, 0xc0, 0xde
+};
+
+static bool i2cr_check_parity32(u32 v, bool parity)
+{
+ u32 i;
+
+ for (i = 0; i < 32; ++i) {
+ if (v & (1u << i))
+ parity = !parity;
+ }
+
+ return parity;
+}
+
+static bool i2cr_check_parity64(u64 v)
+{
+ u32 i;
+ bool parity = I2CR_INITIAL_PARITY;
+
+ for (i = 0; i < 64; ++i) {
+ if (v & (1llu << i))
+ parity = !parity;
+ }
+
+ return parity;
+}
+
+static u32 i2cr_get_command(u32 address, bool parity)
+{
+ address <<= 1;
+
+ if (i2cr_check_parity32(address, parity))
+ address |= 1;
+
+ return address;
+}
+
+static int i2cr_transfer(struct i2c_client *client, u32 command, u64 *data)
+{
+ struct i2c_msg msgs[2];
+ int ret;
+
+ msgs[0].addr = client->addr;
+ msgs[0].flags = 0;
+ msgs[0].len = sizeof(command);
+ msgs[0].buf = (__u8 *)&command;
+ msgs[1].addr = client->addr;
+ msgs[1].flags = I2C_M_RD;
+ msgs[1].len = sizeof(*data);
+ msgs[1].buf = (__u8 *)data;
+
+ ret = i2c_transfer(client->adapter, msgs, 2);
+ if (ret == 2)
+ return 0;
+
+ trace_i2cr_i2c_error(client, command, ret);
+
+ if (ret < 0)
+ return ret;
+
+ return -EIO;
+}
+
+static int i2cr_check_status(struct i2c_client *client)
+{
+ u64 status;
+ int ret;
+
+ ret = i2cr_transfer(client, I2CR_STATUS_CMD, &status);
+ if (ret)
+ return ret;
+
+ if (status & I2CR_STATUS_ERR) {
+ u32 buf[3] = { 0, 0, 0 };
+ u64 error;
+ u64 log;
+
+ i2cr_transfer(client, I2CR_ERROR_CMD, &error);
+ i2cr_transfer(client, I2CR_LOG_CMD, &log);
+
+ trace_i2cr_status_error(client, status, error, log);
+
+ buf[0] = I2CR_STATUS_CMD;
+ i2c_master_send(client, (const char *)buf, sizeof(buf));
+
+ buf[0] = I2CR_ERROR_CMD;
+ i2c_master_send(client, (const char *)buf, sizeof(buf));
+
+ buf[0] = I2CR_LOG_CMD;
+ i2c_master_send(client, (const char *)buf, sizeof(buf));
+
+ dev_err(&client->dev, "status:%016llx error:%016llx log:%016llx\n", status, error,
+ log);
+ return -EREMOTEIO;
+ }
+
+ trace_i2cr_status(client, status);
+ return 0;
+}
+
+int fsi_master_i2cr_read(struct fsi_master_i2cr *i2cr, u32 addr, u64 *data)
+{
+ u32 command = i2cr_get_command(addr, I2CR_INITIAL_PARITY);
+ int ret;
+
+ mutex_lock(&i2cr->lock);
+
+ ret = i2cr_transfer(i2cr->client, command, data);
+ if (ret)
+ goto unlock;
+
+ ret = i2cr_check_status(i2cr->client);
+ if (ret)
+ goto unlock;
+
+ trace_i2cr_read(i2cr->client, command, data);
+
+unlock:
+ mutex_unlock(&i2cr->lock);
+ return ret;
+}
+EXPORT_SYMBOL_GPL(fsi_master_i2cr_read);
+
+int fsi_master_i2cr_write(struct fsi_master_i2cr *i2cr, u32 addr, u64 data)
+{
+ u32 buf[3] = { 0 };
+ int ret;
+
+ buf[0] = i2cr_get_command(addr, i2cr_check_parity64(data));
+ memcpy(&buf[1], &data, sizeof(data));
+
+ mutex_lock(&i2cr->lock);
+
+ ret = i2c_master_send(i2cr->client, (const char *)buf, sizeof(buf));
+ if (ret == sizeof(buf)) {
+ ret = i2cr_check_status(i2cr->client);
+ if (!ret)
+ trace_i2cr_write(i2cr->client, buf[0], data);
+ } else {
+ trace_i2cr_i2c_error(i2cr->client, buf[0], ret);
+
+ if (ret >= 0)
+ ret = -EIO;
+ }
+
+ mutex_unlock(&i2cr->lock);
+ return ret;
+}
+EXPORT_SYMBOL_GPL(fsi_master_i2cr_write);
+
+static int i2cr_read(struct fsi_master *master, int link, uint8_t id, uint32_t addr, void *val,
+ size_t size)
+{
+ struct fsi_master_i2cr *i2cr = container_of(master, struct fsi_master_i2cr, master);
+ u64 data;
+ size_t i;
+ int ret;
+
+ if (link || id || (addr & 0xffff0000) || !(size == 1 || size == 2 || size == 4))
+ return -EINVAL;
+
+ /*
+ * The I2CR doesn't have CFAM or FSI slave address space - only the
+ * engines. In order for this to work with the FSI core, we need to
+ * emulate at minimum the CFAM config table so that the appropriate
+ * engines are discovered.
+ */
+ if (addr < 0xc00) {
+ if (addr > sizeof(i2cr_cfam) - 4)
+ addr = (addr & 0x3) + (sizeof(i2cr_cfam) - 4);
+
+ memcpy(val, &i2cr_cfam[addr], size);
+ return 0;
+ }
+
+ ret = fsi_master_i2cr_read(i2cr, I2CR_ADDRESS_CFAM(addr), &data);
+ if (ret)
+ return ret;
+
+ /*
+ * FSI core expects up to 4 bytes BE back, while I2CR replied with LE
+ * bytes on the wire.
+ */
+ for (i = 0; i < size; ++i)
+ ((u8 *)val)[i] = ((u8 *)&data)[7 - i];
+
+ return 0;
+}
+
+static int i2cr_write(struct fsi_master *master, int link, uint8_t id, uint32_t addr,
+ const void *val, size_t size)
+{
+ struct fsi_master_i2cr *i2cr = container_of(master, struct fsi_master_i2cr, master);
+ u64 data = 0;
+ size_t i;
+
+ if (link || id || (addr & 0xffff0000) || !(size == 1 || size == 2 || size == 4))
+ return -EINVAL;
+
+ /* I2CR writes to CFAM or FSI slave address are a successful no-op. */
+ if (addr < 0xc00)
+ return 0;
+
+ /*
+ * FSI core passes up to 4 bytes BE, while the I2CR expects LE bytes on
+ * the wire.
+ */
+ for (i = 0; i < size; ++i)
+ ((u8 *)&data)[7 - i] = ((u8 *)val)[i];
+
+ return fsi_master_i2cr_write(i2cr, I2CR_ADDRESS_CFAM(addr), data);
+}
+
+static void i2cr_release(struct device *dev)
+{
+ struct fsi_master_i2cr *i2cr = to_fsi_master_i2cr(to_fsi_master(dev));
+
+ of_node_put(dev->of_node);
+
+ kfree(i2cr);
+}
+
+static int i2cr_probe(struct i2c_client *client)
+{
+ struct fsi_master_i2cr *i2cr;
+ int ret;
+
+ i2cr = kzalloc(sizeof(*i2cr), GFP_KERNEL);
+ if (!i2cr)
+ return -ENOMEM;
+
+ /* Only one I2CR on any given I2C bus (fixed I2C device address) */
+ i2cr->master.idx = client->adapter->nr;
+ dev_set_name(&i2cr->master.dev, "i2cr%d",i2cr->master.idx);
+ i2cr->master.dev.parent = &client->dev;
+ i2cr->master.dev.of_node = of_node_get(dev_of_node(&client->dev));
+ i2cr->master.dev.release = i2cr_release;
+
+ i2cr->master.n_links = 1;
+ i2cr->master.flags = FSI_MASTER_FLAG_I2CR;
+ i2cr->master.read = i2cr_read;
+ i2cr->master.write = i2cr_write;
+
+ mutex_init(&i2cr->lock);
+ i2cr->client = client;
+
+ ret = fsi_master_register(&i2cr->master);
+ if (ret)
+ return ret;
+
+ i2c_set_clientdata(client, i2cr);
+ return 0;
+}
+
+static void i2cr_remove(struct i2c_client *client)
+{
+ struct fsi_master_i2cr *i2cr = i2c_get_clientdata(client);
+
+ i2cr->master.idx = -1;
+ fsi_master_unregister(&i2cr->master);
+}
+
+static const struct of_device_id i2cr_ids[] = {
+ { .compatible = "ibm,i2cr-fsi-master" },
+ { }
+};
+MODULE_DEVICE_TABLE(of, i2cr_ids);
+
+static struct i2c_driver i2cr_driver = {
+ .probe_new = i2cr_probe,
+ .remove = i2cr_remove,
+ .driver = {
+ .name = "fsi-master-i2cr",
+ .of_match_table = i2cr_ids,
+ },
+};
+
+module_i2c_driver(i2cr_driver)
+
+MODULE_AUTHOR("Eddie James <eajames@linux.ibm.com>");
+MODULE_DESCRIPTION("IBM I2C Responder virtual FSI master driver");
+MODULE_LICENSE("GPL");
diff --git a/drivers/fsi/fsi-master-i2cr.h b/drivers/fsi/fsi-master-i2cr.h
new file mode 100644
index 000000000000..929d63995c7b
--- /dev/null
+++ b/drivers/fsi/fsi-master-i2cr.h
@@ -0,0 +1,24 @@
+/* SPDX-License-Identifier: GPL-2.0-only */
+/* Copyright (C) IBM Corporation 2023 */
+
+#ifndef DRIVERS_FSI_MASTER_I2CR_H
+#define DRIVERS_FSI_MASTER_I2CR_H
+
+#include <linux/mutex.h>
+
+#include "fsi-master.h"
+
+struct i2c_client;
+
+struct fsi_master_i2cr {
+ struct fsi_master master;
+ struct mutex lock; /* protect HW access */
+ struct i2c_client *client;
+};
+
+#define to_fsi_master_i2cr(m) container_of(m, struct fsi_master_i2cr, master)
+
+int fsi_master_i2cr_read(struct fsi_master_i2cr *i2cr, u32 addr, u64 *data);
+int fsi_master_i2cr_write(struct fsi_master_i2cr *i2cr, u32 addr, u64 data);
+
+#endif /* DRIVERS_FSI_MASTER_I2CR_H */
diff --git a/drivers/fsi/fsi-master.h b/drivers/fsi/fsi-master.h
index 967622c1cabf..a1fa315849d2 100644
--- a/drivers/fsi/fsi-master.h
+++ b/drivers/fsi/fsi-master.h
@@ -111,6 +111,7 @@
/* fsi-master definition and flags */
#define FSI_MASTER_FLAG_SWCLOCK 0x1
+#define FSI_MASTER_FLAG_I2CR 0x2
/*
* Structures and function prototypes
diff --git a/include/trace/events/fsi_master_i2cr.h b/include/trace/events/fsi_master_i2cr.h
new file mode 100644
index 000000000000..c33eba130049
--- /dev/null
+++ b/include/trace/events/fsi_master_i2cr.h
@@ -0,0 +1,107 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+
+#undef TRACE_SYSTEM
+#define TRACE_SYSTEM fsi_master_i2cr
+
+#if !defined(_TRACE_FSI_MASTER_I2CR_H) || defined(TRACE_HEADER_MULTI_READ)
+#define _TRACE_FSI_MASTER_I2CR_H
+
+#include <linux/tracepoint.h>
+
+TRACE_EVENT(i2cr_i2c_error,
+ TP_PROTO(const struct i2c_client *client, uint32_t command, int rc),
+ TP_ARGS(client, command, rc),
+ TP_STRUCT__entry(
+ __field(int, bus)
+ __field(int, rc)
+ __array(unsigned char, command, sizeof(uint32_t))
+ __field(unsigned short, addr)
+ ),
+ TP_fast_assign(
+ __entry->bus = client->adapter->nr;
+ __entry->rc = rc;
+ memcpy(__entry->command, &command, sizeof(uint32_t));
+ __entry->addr = client->addr;
+ ),
+ TP_printk("%d-%02x command:{ %*ph } rc:%d", __entry->bus, __entry->addr,
+ (int)sizeof(uint32_t), __entry->command, __entry->rc)
+);
+
+TRACE_EVENT(i2cr_read,
+ TP_PROTO(const struct i2c_client *client, uint32_t command, uint64_t *data),
+ TP_ARGS(client, command, data),
+ TP_STRUCT__entry(
+ __field(int, bus)
+ __array(unsigned char, data, sizeof(uint64_t))
+ __array(unsigned char, command, sizeof(uint32_t))
+ __field(unsigned short, addr)
+ ),
+ TP_fast_assign(
+ __entry->bus = client->adapter->nr;
+ memcpy(__entry->data, data, sizeof(uint64_t));
+ memcpy(__entry->command, &command, sizeof(uint32_t));
+ __entry->addr = client->addr;
+ ),
+ TP_printk("%d-%02x command:{ %*ph } { %*ph }", __entry->bus, __entry->addr,
+ (int)sizeof(uint32_t), __entry->command, (int)sizeof(uint64_t), __entry->data)
+);
+
+TRACE_EVENT(i2cr_status,
+ TP_PROTO(const struct i2c_client *client, uint64_t status),
+ TP_ARGS(client, status),
+ TP_STRUCT__entry(
+ __field(uint64_t, status)
+ __field(int, bus)
+ __field(unsigned short, addr)
+ ),
+ TP_fast_assign(
+ __entry->status = status;
+ __entry->bus = client->adapter->nr;
+ __entry->addr = client->addr;
+ ),
+ TP_printk("%d-%02x %016llx", __entry->bus, __entry->addr, __entry->status)
+);
+
+TRACE_EVENT(i2cr_status_error,
+ TP_PROTO(const struct i2c_client *client, uint64_t status, uint64_t error, uint64_t log),
+ TP_ARGS(client, status, error, log),
+ TP_STRUCT__entry(
+ __field(uint64_t, error)
+ __field(uint64_t, log)
+ __field(uint64_t, status)
+ __field(int, bus)
+ __field(unsigned short, addr)
+ ),
+ TP_fast_assign(
+ __entry->error = error;
+ __entry->log = log;
+ __entry->status = status;
+ __entry->bus = client->adapter->nr;
+ __entry->addr = client->addr;
+ ),
+ TP_printk("%d-%02x status:%016llx error:%016llx log:%016llx", __entry->bus, __entry->addr,
+ __entry->status, __entry->error, __entry->log)
+);
+
+TRACE_EVENT(i2cr_write,
+ TP_PROTO(const struct i2c_client *client, uint32_t command, uint64_t data),
+ TP_ARGS(client, command, data),
+ TP_STRUCT__entry(
+ __field(int, bus)
+ __array(unsigned char, data, sizeof(uint64_t))
+ __array(unsigned char, command, sizeof(uint32_t))
+ __field(unsigned short, addr)
+ ),
+ TP_fast_assign(
+ __entry->bus = client->adapter->nr;
+ memcpy(__entry->data, &data, sizeof(uint64_t));
+ memcpy(__entry->command, &command, sizeof(uint32_t));
+ __entry->addr = client->addr;
+ ),
+ TP_printk("%d-%02x command:{ %*ph } { %*ph }", __entry->bus, __entry->addr,
+ (int)sizeof(uint32_t), __entry->command, (int)sizeof(uint64_t), __entry->data)
+);
+
+#endif
+
+#include <trace/define_trace.h>
--
2.31.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH v8 5/8] fsi: Add I2C Responder SCOM driver
2023-03-06 16:10 [PATCH v8 0/8] fsi: Add IBM I2C Responder virtual FSI master Eddie James
` (3 preceding siblings ...)
2023-03-06 16:10 ` [PATCH v8 4/8] fsi: Add " Eddie James
@ 2023-03-06 16:10 ` Eddie James
2023-03-06 16:10 ` [PATCH v8 6/8] fsi: Add aliased device numbering Eddie James
` (2 subsequent siblings)
7 siblings, 0 replies; 9+ messages in thread
From: Eddie James @ 2023-03-06 16:10 UTC (permalink / raw)
To: linux-fsi
Cc: rostedt, linux-trace-kernel, devicetree, linux-kernel, mhiramat,
alistair, joel, jk, andrew, robh+dt, krzysztof.kozlowski+dt,
eajames
The I2CR has the capability to directly perform SCOM operations,
circumventing the need to drive the FSI2PIB engine. Add a new
driver to perform SCOM operations through the I2CR.
Signed-off-by: Eddie James <eajames@linux.ibm.com>
---
drivers/fsi/Kconfig | 8 +++
drivers/fsi/Makefile | 1 +
drivers/fsi/i2cr-scom.c | 154 ++++++++++++++++++++++++++++++++++++++++
3 files changed, 163 insertions(+)
create mode 100644 drivers/fsi/i2cr-scom.c
diff --git a/drivers/fsi/Kconfig b/drivers/fsi/Kconfig
index 999be82720c5..79a31593618a 100644
--- a/drivers/fsi/Kconfig
+++ b/drivers/fsi/Kconfig
@@ -94,4 +94,12 @@ config FSI_OCC
provide the raw sensor data as well as perform thermal and power
management on the system.
+config I2CR_SCOM
+ tristate "IBM I2C Responder SCOM driver"
+ depends on FSI_MASTER_I2CR
+ help
+ This option enables an I2C Responder based SCOM device driver. The
+ I2CR has the capability to directly perform SCOM operations instead
+ of using the FSI2PIB engine.
+
endif
diff --git a/drivers/fsi/Makefile b/drivers/fsi/Makefile
index 34dbaa1c452e..5550aa15e0b1 100644
--- a/drivers/fsi/Makefile
+++ b/drivers/fsi/Makefile
@@ -9,3 +9,4 @@ obj-$(CONFIG_FSI_MASTER_AST_CF) += fsi-master-ast-cf.o
obj-$(CONFIG_FSI_SCOM) += fsi-scom.o
obj-$(CONFIG_FSI_SBEFIFO) += fsi-sbefifo.o
obj-$(CONFIG_FSI_OCC) += fsi-occ.o
+obj-$(CONFIG_I2CR_SCOM) += i2cr-scom.o
diff --git a/drivers/fsi/i2cr-scom.c b/drivers/fsi/i2cr-scom.c
new file mode 100644
index 000000000000..63b548bdef3e
--- /dev/null
+++ b/drivers/fsi/i2cr-scom.c
@@ -0,0 +1,154 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Copyright (C) IBM Corporation 2023 */
+
+#include <linux/cdev.h>
+#include <linux/device.h>
+#include <linux/fs.h>
+#include <linux/fsi.h>
+#include <linux/module.h>
+#include <linux/mod_devicetable.h>
+
+#include "fsi-master-i2cr.h"
+#include "fsi-slave.h"
+
+struct i2cr_scom {
+ struct device dev;
+ struct cdev cdev;
+ struct fsi_master_i2cr *i2cr;
+};
+
+static loff_t i2cr_scom_llseek(struct file *file, loff_t offset, int whence)
+{
+ switch (whence) {
+ case SEEK_CUR:
+ break;
+ case SEEK_SET:
+ file->f_pos = offset;
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return offset;
+}
+
+static ssize_t i2cr_scom_read(struct file *filep, char __user *buf, size_t len, loff_t *offset)
+{
+ struct i2cr_scom *scom = filep->private_data;
+ u64 data;
+ int ret;
+
+ if (len != sizeof(data))
+ return -EINVAL;
+
+ ret = fsi_master_i2cr_read(scom->i2cr, (u32)*offset, &data);
+ if (ret)
+ return ret;
+
+ ret = copy_to_user(buf, &data, len);
+ if (ret)
+ return ret;
+
+ return len;
+}
+
+static ssize_t i2cr_scom_write(struct file *filep, const char __user *buf, size_t len,
+ loff_t *offset)
+{
+ struct i2cr_scom *scom = filep->private_data;
+ u64 data;
+ int ret;
+
+ if (len != sizeof(data))
+ return -EINVAL;
+
+ ret = copy_from_user(&data, buf, len);
+ if (ret)
+ return ret;
+
+ ret = fsi_master_i2cr_write(scom->i2cr, (u32)*offset, data);
+ if (ret)
+ return ret;
+
+ return len;
+}
+
+static const struct file_operations i2cr_scom_fops = {
+ .owner = THIS_MODULE,
+ .open = simple_open,
+ .llseek = i2cr_scom_llseek,
+ .read = i2cr_scom_read,
+ .write = i2cr_scom_write,
+};
+
+static int i2cr_scom_probe(struct device *dev)
+{
+ struct fsi_device *fsi_dev = to_fsi_dev(dev);
+ struct i2cr_scom *scom;
+ int didx;
+ int ret;
+
+ if (!(fsi_dev->slave->master->flags & FSI_MASTER_FLAG_I2CR))
+ return -ENODEV;
+
+ scom = devm_kzalloc(dev, sizeof(*scom), GFP_KERNEL);
+ if (!scom)
+ return -ENOMEM;
+
+ scom->i2cr = to_fsi_master_i2cr(fsi_dev->slave->master);
+ dev_set_drvdata(dev, scom);
+
+ scom->dev.type = &fsi_cdev_type;
+ scom->dev.parent = dev;
+ device_initialize(&scom->dev);
+
+ ret = fsi_get_new_minor(fsi_dev, fsi_dev_scom, &scom->dev.devt, &didx);
+ if (ret)
+ return ret;
+
+ dev_set_name(&scom->dev, "scom%d", didx);
+ cdev_init(&scom->cdev, &i2cr_scom_fops);
+ ret = cdev_device_add(&scom->cdev, &scom->dev);
+ if (ret)
+ fsi_free_minor(scom->dev.devt);
+
+ return ret;
+}
+
+static int i2cr_scom_remove(struct device *dev)
+{
+ struct i2cr_scom *scom = dev_get_drvdata(dev);
+
+ cdev_device_del(&scom->cdev, &scom->dev);
+ fsi_free_minor(scom->dev.devt);
+
+ return 0;
+}
+
+static const struct of_device_id i2cr_scom_of_ids[] = {
+ { .compatible = "ibm,i2cr-scom" },
+ { }
+};
+MODULE_DEVICE_TABLE(of, i2cr_scom_of_ids);
+
+static const struct fsi_device_id i2cr_scom_ids[] = {
+ { 0x5, FSI_VERSION_ANY },
+ { }
+};
+
+static struct fsi_driver i2cr_scom_driver = {
+ .id_table = i2cr_scom_ids,
+ .drv = {
+ .name = "i2cr_scom",
+ .bus = &fsi_bus_type,
+ .of_match_table = i2cr_scom_of_ids,
+ .probe = i2cr_scom_probe,
+ .remove = i2cr_scom_remove,
+ }
+};
+
+module_fsi_driver(i2cr_scom_driver);
+
+MODULE_AUTHOR("Eddie James <eajames@linux.ibm.com>");
+MODULE_DESCRIPTION("IBM I2C Responder SCOM driver");
+MODULE_LICENSE("GPL");
--
2.31.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH v8 6/8] fsi: Add aliased device numbering
2023-03-06 16:10 [PATCH v8 0/8] fsi: Add IBM I2C Responder virtual FSI master Eddie James
` (4 preceding siblings ...)
2023-03-06 16:10 ` [PATCH v8 5/8] fsi: Add I2C Responder SCOM driver Eddie James
@ 2023-03-06 16:10 ` Eddie James
2023-03-06 16:10 ` [PATCH v8 7/8] fsi: Use of_match_table for bus matching if specified Eddie James
2023-03-06 16:10 ` [PATCH v8 8/8] fsi: sbefifo: Don't check status during probe Eddie James
7 siblings, 0 replies; 9+ messages in thread
From: Eddie James @ 2023-03-06 16:10 UTC (permalink / raw)
To: linux-fsi
Cc: rostedt, linux-trace-kernel, devicetree, linux-kernel, mhiramat,
alistair, joel, jk, andrew, robh+dt, krzysztof.kozlowski+dt,
eajames
The I2C and SPI subsystems can use an aliased name to number the device.
Add similar support to the FSI subsystem for any device type.
Signed-off-by: Eddie James <eajames@linux.ibm.com>
---
drivers/fsi/fsi-core.c | 25 +++++++++++++++++++++++++
1 file changed, 25 insertions(+)
diff --git a/drivers/fsi/fsi-core.c b/drivers/fsi/fsi-core.c
index 3271e40702ec..48f25dc0fad7 100644
--- a/drivers/fsi/fsi-core.c
+++ b/drivers/fsi/fsi-core.c
@@ -955,9 +955,34 @@ static int __fsi_get_new_minor(struct fsi_slave *slave, enum fsi_dev_type type,
return 0;
}
+static const char *const fsi_dev_type_names[] = {
+ "cfam",
+ "sbefifo",
+ "scom",
+ "occ",
+};
+
int fsi_get_new_minor(struct fsi_device *fdev, enum fsi_dev_type type,
dev_t *out_dev, int *out_index)
{
+ if (fdev->dev.of_node) {
+ int aid = of_alias_get_id(fdev->dev.of_node, fsi_dev_type_names[type]);
+
+ if (aid >= 0) {
+ int id = (aid << 4) | type;
+
+ id = ida_simple_get(&fsi_minor_ida, id, id + 1, GFP_KERNEL);
+ if (id >= 0) {
+ *out_index = aid;
+ *out_dev = fsi_base_dev + id;
+ return 0;
+ }
+
+ if (id != -ENOSPC)
+ return id;
+ }
+ }
+
return __fsi_get_new_minor(fdev->slave, type, out_dev, out_index);
}
EXPORT_SYMBOL_GPL(fsi_get_new_minor);
--
2.31.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH v8 7/8] fsi: Use of_match_table for bus matching if specified
2023-03-06 16:10 [PATCH v8 0/8] fsi: Add IBM I2C Responder virtual FSI master Eddie James
` (5 preceding siblings ...)
2023-03-06 16:10 ` [PATCH v8 6/8] fsi: Add aliased device numbering Eddie James
@ 2023-03-06 16:10 ` Eddie James
2023-03-06 16:10 ` [PATCH v8 8/8] fsi: sbefifo: Don't check status during probe Eddie James
7 siblings, 0 replies; 9+ messages in thread
From: Eddie James @ 2023-03-06 16:10 UTC (permalink / raw)
To: linux-fsi
Cc: rostedt, linux-trace-kernel, devicetree, linux-kernel, mhiramat,
alistair, joel, jk, andrew, robh+dt, krzysztof.kozlowski+dt,
eajames
Since we have two scom drivers, use the standard of matching if
the driver specifies a table so that the right devices go to the
right driver.
Signed-off-by: Eddie James <eajames@linux.ibm.com>
---
drivers/fsi/fsi-core.c | 11 +++++++++--
drivers/fsi/fsi-scom.c | 8 ++++++++
2 files changed, 17 insertions(+), 2 deletions(-)
diff --git a/drivers/fsi/fsi-core.c b/drivers/fsi/fsi-core.c
index 48f25dc0fad7..5191f6979346 100644
--- a/drivers/fsi/fsi-core.c
+++ b/drivers/fsi/fsi-core.c
@@ -16,6 +16,7 @@
#include <linux/idr.h>
#include <linux/module.h>
#include <linux/of.h>
+#include <linux/of_device.h>
#include <linux/slab.h>
#include <linux/bitops.h>
#include <linux/cdev.h>
@@ -1380,8 +1381,14 @@ static int fsi_bus_match(struct device *dev, struct device_driver *drv)
if (id->engine_type != fsi_dev->engine_type)
continue;
if (id->version == FSI_VERSION_ANY ||
- id->version == fsi_dev->version)
- return 1;
+ id->version == fsi_dev->version) {
+ if (drv->of_match_table) {
+ if (of_driver_match_device(dev, drv))
+ return 1;
+ } else {
+ return 1;
+ }
+ }
}
return 0;
diff --git a/drivers/fsi/fsi-scom.c b/drivers/fsi/fsi-scom.c
index bcb756dc9866..61dbda9dbe2b 100644
--- a/drivers/fsi/fsi-scom.c
+++ b/drivers/fsi/fsi-scom.c
@@ -10,6 +10,7 @@
#include <linux/cdev.h>
#include <linux/delay.h>
#include <linux/fs.h>
+#include <linux/mod_devicetable.h>
#include <linux/uaccess.h>
#include <linux/slab.h>
#include <linux/list.h>
@@ -587,6 +588,12 @@ static int scom_remove(struct device *dev)
return 0;
}
+static const struct of_device_id scom_of_ids[] = {
+ { .compatible = "ibm,fsi2pib" },
+ { }
+};
+MODULE_DEVICE_TABLE(of, scom_of_ids);
+
static const struct fsi_device_id scom_ids[] = {
{
.engine_type = FSI_ENGID_SCOM,
@@ -600,6 +607,7 @@ static struct fsi_driver scom_drv = {
.drv = {
.name = "scom",
.bus = &fsi_bus_type,
+ .of_match_table = scom_of_ids,
.probe = scom_probe,
.remove = scom_remove,
}
--
2.31.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
* [PATCH v8 8/8] fsi: sbefifo: Don't check status during probe
2023-03-06 16:10 [PATCH v8 0/8] fsi: Add IBM I2C Responder virtual FSI master Eddie James
` (6 preceding siblings ...)
2023-03-06 16:10 ` [PATCH v8 7/8] fsi: Use of_match_table for bus matching if specified Eddie James
@ 2023-03-06 16:10 ` Eddie James
7 siblings, 0 replies; 9+ messages in thread
From: Eddie James @ 2023-03-06 16:10 UTC (permalink / raw)
To: linux-fsi
Cc: rostedt, linux-trace-kernel, devicetree, linux-kernel, mhiramat,
alistair, joel, jk, andrew, robh+dt, krzysztof.kozlowski+dt,
eajames
The status check during probe doesn't serve any purpose. Any attempt
to use the SBEFIFO will result in the same check and cleanup.
Signed-off-by: Eddie James <eajames@linux.ibm.com>
---
drivers/fsi/fsi-sbefifo.c | 8 --------
1 file changed, 8 deletions(-)
diff --git a/drivers/fsi/fsi-sbefifo.c b/drivers/fsi/fsi-sbefifo.c
index 9912b7a6a4b9..42d7c95528d1 100644
--- a/drivers/fsi/fsi-sbefifo.c
+++ b/drivers/fsi/fsi-sbefifo.c
@@ -1027,14 +1027,6 @@ static int sbefifo_probe(struct device *dev)
mutex_init(&sbefifo->lock);
sbefifo->timeout_start_rsp_ms = SBEFIFO_TIMEOUT_START_RSP;
- /*
- * Try cleaning up the FIFO. If this fails, we still register the
- * driver and will try cleaning things up again on the next access.
- */
- rc = sbefifo_cleanup_hw(sbefifo);
- if (rc && rc != -ESHUTDOWN)
- dev_err(dev, "Initial HW cleanup failed, will retry later\n");
-
/* Create chardev for userspace access */
sbefifo->dev.type = &fsi_cdev_type;
sbefifo->dev.parent = dev;
--
2.31.1
^ permalink raw reply related [flat|nested] 9+ messages in thread
end of thread, other threads:[~2023-03-06 17:09 UTC | newest]
Thread overview: 9+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-03-06 16:10 [PATCH v8 0/8] fsi: Add IBM I2C Responder virtual FSI master Eddie James
2023-03-06 16:10 ` [PATCH v8 1/8] fsi: Move fsi_slave structure definition to header Eddie James
2023-03-06 16:10 ` [PATCH v8 2/8] fsi: Make master device indexing and naming optional Eddie James
2023-03-06 16:10 ` [PATCH v8 3/8] dt-bindings: fsi: Document the IBM I2C Responder virtual FSI master Eddie James
2023-03-06 16:10 ` [PATCH v8 4/8] fsi: Add " Eddie James
2023-03-06 16:10 ` [PATCH v8 5/8] fsi: Add I2C Responder SCOM driver Eddie James
2023-03-06 16:10 ` [PATCH v8 6/8] fsi: Add aliased device numbering Eddie James
2023-03-06 16:10 ` [PATCH v8 7/8] fsi: Use of_match_table for bus matching if specified Eddie James
2023-03-06 16:10 ` [PATCH v8 8/8] fsi: sbefifo: Don't check status during probe Eddie James
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.