tmp_suning_uos_patched/drivers/greybus/svc.c
Greg Kroah-Hartman 8465def499 staging: greybus: move the greybus core to drivers/greybus
The Greybus core code has been stable for a long time, and has been
shipping for many years in millions of phones.  With the advent of a
recent Google Summer of Code project, and a number of new devices in the
works from various companies, it is time to get the core greybus code
out of staging as it really is going to be with us for a while.

Cc: Johan Hovold <johan@kernel.org>
Cc: linux-kernel@vger.kernel.org
Cc: greybus-dev@lists.linaro.org
Acked-by: Viresh Kumar <viresh.kumar@linaro.org>
Acked-by: Alex Elder <elder@kernel.org>
Link: https://lore.kernel.org/r/20190825055429.18547-9-gregkh@linuxfoundation.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
2019-08-27 19:03:04 +02:00

1398 lines
35 KiB
C

// SPDX-License-Identifier: GPL-2.0
/*
* SVC Greybus driver.
*
* Copyright 2015 Google Inc.
* Copyright 2015 Linaro Ltd.
*/
#include <linux/debugfs.h>
#include <linux/workqueue.h>
#include <linux/greybus.h>
#define SVC_INTF_EJECT_TIMEOUT 9000
#define SVC_INTF_ACTIVATE_TIMEOUT 6000
#define SVC_INTF_RESUME_TIMEOUT 3000
struct gb_svc_deferred_request {
struct work_struct work;
struct gb_operation *operation;
};
static int gb_svc_queue_deferred_request(struct gb_operation *operation);
static ssize_t endo_id_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct gb_svc *svc = to_gb_svc(dev);
return sprintf(buf, "0x%04x\n", svc->endo_id);
}
static DEVICE_ATTR_RO(endo_id);
static ssize_t ap_intf_id_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct gb_svc *svc = to_gb_svc(dev);
return sprintf(buf, "%u\n", svc->ap_intf_id);
}
static DEVICE_ATTR_RO(ap_intf_id);
// FIXME
// This is a hack, we need to do this "right" and clean the interface up
// properly, not just forcibly yank the thing out of the system and hope for the
// best. But for now, people want their modules to come out without having to
// throw the thing to the ground or get out a screwdriver.
static ssize_t intf_eject_store(struct device *dev,
struct device_attribute *attr, const char *buf,
size_t len)
{
struct gb_svc *svc = to_gb_svc(dev);
unsigned short intf_id;
int ret;
ret = kstrtou16(buf, 10, &intf_id);
if (ret < 0)
return ret;
dev_warn(dev, "Forcibly trying to eject interface %d\n", intf_id);
ret = gb_svc_intf_eject(svc, intf_id);
if (ret < 0)
return ret;
return len;
}
static DEVICE_ATTR_WO(intf_eject);
static ssize_t watchdog_show(struct device *dev, struct device_attribute *attr,
char *buf)
{
struct gb_svc *svc = to_gb_svc(dev);
return sprintf(buf, "%s\n",
gb_svc_watchdog_enabled(svc) ? "enabled" : "disabled");
}
static ssize_t watchdog_store(struct device *dev,
struct device_attribute *attr, const char *buf,
size_t len)
{
struct gb_svc *svc = to_gb_svc(dev);
int retval;
bool user_request;
retval = strtobool(buf, &user_request);
if (retval)
return retval;
if (user_request)
retval = gb_svc_watchdog_enable(svc);
else
retval = gb_svc_watchdog_disable(svc);
if (retval)
return retval;
return len;
}
static DEVICE_ATTR_RW(watchdog);
static ssize_t watchdog_action_show(struct device *dev,
struct device_attribute *attr, char *buf)
{
struct gb_svc *svc = to_gb_svc(dev);
if (svc->action == GB_SVC_WATCHDOG_BITE_PANIC_KERNEL)
return sprintf(buf, "panic\n");
else if (svc->action == GB_SVC_WATCHDOG_BITE_RESET_UNIPRO)
return sprintf(buf, "reset\n");
return -EINVAL;
}
static ssize_t watchdog_action_store(struct device *dev,
struct device_attribute *attr,
const char *buf, size_t len)
{
struct gb_svc *svc = to_gb_svc(dev);
if (sysfs_streq(buf, "panic"))
svc->action = GB_SVC_WATCHDOG_BITE_PANIC_KERNEL;
else if (sysfs_streq(buf, "reset"))
svc->action = GB_SVC_WATCHDOG_BITE_RESET_UNIPRO;
else
return -EINVAL;
return len;
}
static DEVICE_ATTR_RW(watchdog_action);
static int gb_svc_pwrmon_rail_count_get(struct gb_svc *svc, u8 *value)
{
struct gb_svc_pwrmon_rail_count_get_response response;
int ret;
ret = gb_operation_sync(svc->connection,
GB_SVC_TYPE_PWRMON_RAIL_COUNT_GET, NULL, 0,
&response, sizeof(response));
if (ret) {
dev_err(&svc->dev, "failed to get rail count: %d\n", ret);
return ret;
}
*value = response.rail_count;
return 0;
}
static int gb_svc_pwrmon_rail_names_get(struct gb_svc *svc,
struct gb_svc_pwrmon_rail_names_get_response *response,
size_t bufsize)
{
int ret;
ret = gb_operation_sync(svc->connection,
GB_SVC_TYPE_PWRMON_RAIL_NAMES_GET, NULL, 0,
response, bufsize);
if (ret) {
dev_err(&svc->dev, "failed to get rail names: %d\n", ret);
return ret;
}
if (response->status != GB_SVC_OP_SUCCESS) {
dev_err(&svc->dev,
"SVC error while getting rail names: %u\n",
response->status);
return -EREMOTEIO;
}
return 0;
}
static int gb_svc_pwrmon_sample_get(struct gb_svc *svc, u8 rail_id,
u8 measurement_type, u32 *value)
{
struct gb_svc_pwrmon_sample_get_request request;
struct gb_svc_pwrmon_sample_get_response response;
int ret;
request.rail_id = rail_id;
request.measurement_type = measurement_type;
ret = gb_operation_sync(svc->connection, GB_SVC_TYPE_PWRMON_SAMPLE_GET,
&request, sizeof(request),
&response, sizeof(response));
if (ret) {
dev_err(&svc->dev, "failed to get rail sample: %d\n", ret);
return ret;
}
if (response.result) {
dev_err(&svc->dev,
"UniPro error while getting rail power sample (%d %d): %d\n",
rail_id, measurement_type, response.result);
switch (response.result) {
case GB_SVC_PWRMON_GET_SAMPLE_INVAL:
return -EINVAL;
case GB_SVC_PWRMON_GET_SAMPLE_NOSUPP:
return -ENOMSG;
default:
return -EREMOTEIO;
}
}
*value = le32_to_cpu(response.measurement);
return 0;
}
int gb_svc_pwrmon_intf_sample_get(struct gb_svc *svc, u8 intf_id,
u8 measurement_type, u32 *value)
{
struct gb_svc_pwrmon_intf_sample_get_request request;
struct gb_svc_pwrmon_intf_sample_get_response response;
int ret;
request.intf_id = intf_id;
request.measurement_type = measurement_type;
ret = gb_operation_sync(svc->connection,
GB_SVC_TYPE_PWRMON_INTF_SAMPLE_GET,
&request, sizeof(request),
&response, sizeof(response));
if (ret) {
dev_err(&svc->dev, "failed to get intf sample: %d\n", ret);
return ret;
}
if (response.result) {
dev_err(&svc->dev,
"UniPro error while getting intf power sample (%d %d): %d\n",
intf_id, measurement_type, response.result);
switch (response.result) {
case GB_SVC_PWRMON_GET_SAMPLE_INVAL:
return -EINVAL;
case GB_SVC_PWRMON_GET_SAMPLE_NOSUPP:
return -ENOMSG;
default:
return -EREMOTEIO;
}
}
*value = le32_to_cpu(response.measurement);
return 0;
}
static struct attribute *svc_attrs[] = {
&dev_attr_endo_id.attr,
&dev_attr_ap_intf_id.attr,
&dev_attr_intf_eject.attr,
&dev_attr_watchdog.attr,
&dev_attr_watchdog_action.attr,
NULL,
};
ATTRIBUTE_GROUPS(svc);
int gb_svc_intf_device_id(struct gb_svc *svc, u8 intf_id, u8 device_id)
{
struct gb_svc_intf_device_id_request request;
request.intf_id = intf_id;
request.device_id = device_id;
return gb_operation_sync(svc->connection, GB_SVC_TYPE_INTF_DEVICE_ID,
&request, sizeof(request), NULL, 0);
}
int gb_svc_intf_eject(struct gb_svc *svc, u8 intf_id)
{
struct gb_svc_intf_eject_request request;
int ret;
request.intf_id = intf_id;
/*
* The pulse width for module release in svc is long so we need to
* increase the timeout so the operation will not return to soon.
*/
ret = gb_operation_sync_timeout(svc->connection,
GB_SVC_TYPE_INTF_EJECT, &request,
sizeof(request), NULL, 0,
SVC_INTF_EJECT_TIMEOUT);
if (ret) {
dev_err(&svc->dev, "failed to eject interface %u\n", intf_id);
return ret;
}
return 0;
}
int gb_svc_intf_vsys_set(struct gb_svc *svc, u8 intf_id, bool enable)
{
struct gb_svc_intf_vsys_request request;
struct gb_svc_intf_vsys_response response;
int type, ret;
request.intf_id = intf_id;
if (enable)
type = GB_SVC_TYPE_INTF_VSYS_ENABLE;
else
type = GB_SVC_TYPE_INTF_VSYS_DISABLE;
ret = gb_operation_sync(svc->connection, type,
&request, sizeof(request),
&response, sizeof(response));
if (ret < 0)
return ret;
if (response.result_code != GB_SVC_INTF_VSYS_OK)
return -EREMOTEIO;
return 0;
}
int gb_svc_intf_refclk_set(struct gb_svc *svc, u8 intf_id, bool enable)
{
struct gb_svc_intf_refclk_request request;
struct gb_svc_intf_refclk_response response;
int type, ret;
request.intf_id = intf_id;
if (enable)
type = GB_SVC_TYPE_INTF_REFCLK_ENABLE;
else
type = GB_SVC_TYPE_INTF_REFCLK_DISABLE;
ret = gb_operation_sync(svc->connection, type,
&request, sizeof(request),
&response, sizeof(response));
if (ret < 0)
return ret;
if (response.result_code != GB_SVC_INTF_REFCLK_OK)
return -EREMOTEIO;
return 0;
}
int gb_svc_intf_unipro_set(struct gb_svc *svc, u8 intf_id, bool enable)
{
struct gb_svc_intf_unipro_request request;
struct gb_svc_intf_unipro_response response;
int type, ret;
request.intf_id = intf_id;
if (enable)
type = GB_SVC_TYPE_INTF_UNIPRO_ENABLE;
else
type = GB_SVC_TYPE_INTF_UNIPRO_DISABLE;
ret = gb_operation_sync(svc->connection, type,
&request, sizeof(request),
&response, sizeof(response));
if (ret < 0)
return ret;
if (response.result_code != GB_SVC_INTF_UNIPRO_OK)
return -EREMOTEIO;
return 0;
}
int gb_svc_intf_activate(struct gb_svc *svc, u8 intf_id, u8 *intf_type)
{
struct gb_svc_intf_activate_request request;
struct gb_svc_intf_activate_response response;
int ret;
request.intf_id = intf_id;
ret = gb_operation_sync_timeout(svc->connection,
GB_SVC_TYPE_INTF_ACTIVATE,
&request, sizeof(request),
&response, sizeof(response),
SVC_INTF_ACTIVATE_TIMEOUT);
if (ret < 0)
return ret;
if (response.status != GB_SVC_OP_SUCCESS) {
dev_err(&svc->dev, "failed to activate interface %u: %u\n",
intf_id, response.status);
return -EREMOTEIO;
}
*intf_type = response.intf_type;
return 0;
}
int gb_svc_intf_resume(struct gb_svc *svc, u8 intf_id)
{
struct gb_svc_intf_resume_request request;
struct gb_svc_intf_resume_response response;
int ret;
request.intf_id = intf_id;
ret = gb_operation_sync_timeout(svc->connection,
GB_SVC_TYPE_INTF_RESUME,
&request, sizeof(request),
&response, sizeof(response),
SVC_INTF_RESUME_TIMEOUT);
if (ret < 0) {
dev_err(&svc->dev, "failed to send interface resume %u: %d\n",
intf_id, ret);
return ret;
}
if (response.status != GB_SVC_OP_SUCCESS) {
dev_err(&svc->dev, "failed to resume interface %u: %u\n",
intf_id, response.status);
return -EREMOTEIO;
}
return 0;
}
int gb_svc_dme_peer_get(struct gb_svc *svc, u8 intf_id, u16 attr, u16 selector,
u32 *value)
{
struct gb_svc_dme_peer_get_request request;
struct gb_svc_dme_peer_get_response response;
u16 result;
int ret;
request.intf_id = intf_id;
request.attr = cpu_to_le16(attr);
request.selector = cpu_to_le16(selector);
ret = gb_operation_sync(svc->connection, GB_SVC_TYPE_DME_PEER_GET,
&request, sizeof(request),
&response, sizeof(response));
if (ret) {
dev_err(&svc->dev, "failed to get DME attribute (%u 0x%04x %u): %d\n",
intf_id, attr, selector, ret);
return ret;
}
result = le16_to_cpu(response.result_code);
if (result) {
dev_err(&svc->dev, "UniPro error while getting DME attribute (%u 0x%04x %u): %u\n",
intf_id, attr, selector, result);
return -EREMOTEIO;
}
if (value)
*value = le32_to_cpu(response.attr_value);
return 0;
}
int gb_svc_dme_peer_set(struct gb_svc *svc, u8 intf_id, u16 attr, u16 selector,
u32 value)
{
struct gb_svc_dme_peer_set_request request;
struct gb_svc_dme_peer_set_response response;
u16 result;
int ret;
request.intf_id = intf_id;
request.attr = cpu_to_le16(attr);
request.selector = cpu_to_le16(selector);
request.value = cpu_to_le32(value);
ret = gb_operation_sync(svc->connection, GB_SVC_TYPE_DME_PEER_SET,
&request, sizeof(request),
&response, sizeof(response));
if (ret) {
dev_err(&svc->dev, "failed to set DME attribute (%u 0x%04x %u %u): %d\n",
intf_id, attr, selector, value, ret);
return ret;
}
result = le16_to_cpu(response.result_code);
if (result) {
dev_err(&svc->dev, "UniPro error while setting DME attribute (%u 0x%04x %u %u): %u\n",
intf_id, attr, selector, value, result);
return -EREMOTEIO;
}
return 0;
}
int gb_svc_connection_create(struct gb_svc *svc,
u8 intf1_id, u16 cport1_id,
u8 intf2_id, u16 cport2_id,
u8 cport_flags)
{
struct gb_svc_conn_create_request request;
request.intf1_id = intf1_id;
request.cport1_id = cpu_to_le16(cport1_id);
request.intf2_id = intf2_id;
request.cport2_id = cpu_to_le16(cport2_id);
request.tc = 0; /* TC0 */
request.flags = cport_flags;
return gb_operation_sync(svc->connection, GB_SVC_TYPE_CONN_CREATE,
&request, sizeof(request), NULL, 0);
}
void gb_svc_connection_destroy(struct gb_svc *svc, u8 intf1_id, u16 cport1_id,
u8 intf2_id, u16 cport2_id)
{
struct gb_svc_conn_destroy_request request;
struct gb_connection *connection = svc->connection;
int ret;
request.intf1_id = intf1_id;
request.cport1_id = cpu_to_le16(cport1_id);
request.intf2_id = intf2_id;
request.cport2_id = cpu_to_le16(cport2_id);
ret = gb_operation_sync(connection, GB_SVC_TYPE_CONN_DESTROY,
&request, sizeof(request), NULL, 0);
if (ret) {
dev_err(&svc->dev, "failed to destroy connection (%u:%u %u:%u): %d\n",
intf1_id, cport1_id, intf2_id, cport2_id, ret);
}
}
/* Creates bi-directional routes between the devices */
int gb_svc_route_create(struct gb_svc *svc, u8 intf1_id, u8 dev1_id,
u8 intf2_id, u8 dev2_id)
{
struct gb_svc_route_create_request request;
request.intf1_id = intf1_id;
request.dev1_id = dev1_id;
request.intf2_id = intf2_id;
request.dev2_id = dev2_id;
return gb_operation_sync(svc->connection, GB_SVC_TYPE_ROUTE_CREATE,
&request, sizeof(request), NULL, 0);
}
/* Destroys bi-directional routes between the devices */
void gb_svc_route_destroy(struct gb_svc *svc, u8 intf1_id, u8 intf2_id)
{
struct gb_svc_route_destroy_request request;
int ret;
request.intf1_id = intf1_id;
request.intf2_id = intf2_id;
ret = gb_operation_sync(svc->connection, GB_SVC_TYPE_ROUTE_DESTROY,
&request, sizeof(request), NULL, 0);
if (ret) {
dev_err(&svc->dev, "failed to destroy route (%u %u): %d\n",
intf1_id, intf2_id, ret);
}
}
int gb_svc_intf_set_power_mode(struct gb_svc *svc, u8 intf_id, u8 hs_series,
u8 tx_mode, u8 tx_gear, u8 tx_nlanes,
u8 tx_amplitude, u8 tx_hs_equalizer,
u8 rx_mode, u8 rx_gear, u8 rx_nlanes,
u8 flags, u32 quirks,
struct gb_svc_l2_timer_cfg *local,
struct gb_svc_l2_timer_cfg *remote)
{
struct gb_svc_intf_set_pwrm_request request;
struct gb_svc_intf_set_pwrm_response response;
int ret;
u16 result_code;
memset(&request, 0, sizeof(request));
request.intf_id = intf_id;
request.hs_series = hs_series;
request.tx_mode = tx_mode;
request.tx_gear = tx_gear;
request.tx_nlanes = tx_nlanes;
request.tx_amplitude = tx_amplitude;
request.tx_hs_equalizer = tx_hs_equalizer;
request.rx_mode = rx_mode;
request.rx_gear = rx_gear;
request.rx_nlanes = rx_nlanes;
request.flags = flags;
request.quirks = cpu_to_le32(quirks);
if (local)
request.local_l2timerdata = *local;
if (remote)
request.remote_l2timerdata = *remote;
ret = gb_operation_sync(svc->connection, GB_SVC_TYPE_INTF_SET_PWRM,
&request, sizeof(request),
&response, sizeof(response));
if (ret < 0)
return ret;
result_code = response.result_code;
if (result_code != GB_SVC_SETPWRM_PWR_LOCAL) {
dev_err(&svc->dev, "set power mode = %d\n", result_code);
return -EIO;
}
return 0;
}
EXPORT_SYMBOL_GPL(gb_svc_intf_set_power_mode);
int gb_svc_intf_set_power_mode_hibernate(struct gb_svc *svc, u8 intf_id)
{
struct gb_svc_intf_set_pwrm_request request;
struct gb_svc_intf_set_pwrm_response response;
int ret;
u16 result_code;
memset(&request, 0, sizeof(request));
request.intf_id = intf_id;
request.hs_series = GB_SVC_UNIPRO_HS_SERIES_A;
request.tx_mode = GB_SVC_UNIPRO_HIBERNATE_MODE;
request.rx_mode = GB_SVC_UNIPRO_HIBERNATE_MODE;
ret = gb_operation_sync(svc->connection, GB_SVC_TYPE_INTF_SET_PWRM,
&request, sizeof(request),
&response, sizeof(response));
if (ret < 0) {
dev_err(&svc->dev,
"failed to send set power mode operation to interface %u: %d\n",
intf_id, ret);
return ret;
}
result_code = response.result_code;
if (result_code != GB_SVC_SETPWRM_PWR_OK) {
dev_err(&svc->dev,
"failed to hibernate the link for interface %u: %u\n",
intf_id, result_code);
return -EIO;
}
return 0;
}
int gb_svc_ping(struct gb_svc *svc)
{
return gb_operation_sync_timeout(svc->connection, GB_SVC_TYPE_PING,
NULL, 0, NULL, 0,
GB_OPERATION_TIMEOUT_DEFAULT * 2);
}
static int gb_svc_version_request(struct gb_operation *op)
{
struct gb_connection *connection = op->connection;
struct gb_svc *svc = gb_connection_get_data(connection);
struct gb_svc_version_request *request;
struct gb_svc_version_response *response;
if (op->request->payload_size < sizeof(*request)) {
dev_err(&svc->dev, "short version request (%zu < %zu)\n",
op->request->payload_size,
sizeof(*request));
return -EINVAL;
}
request = op->request->payload;
if (request->major > GB_SVC_VERSION_MAJOR) {
dev_warn(&svc->dev, "unsupported major version (%u > %u)\n",
request->major, GB_SVC_VERSION_MAJOR);
return -ENOTSUPP;
}
svc->protocol_major = request->major;
svc->protocol_minor = request->minor;
if (!gb_operation_response_alloc(op, sizeof(*response), GFP_KERNEL))
return -ENOMEM;
response = op->response->payload;
response->major = svc->protocol_major;
response->minor = svc->protocol_minor;
return 0;
}
static ssize_t pwr_debugfs_voltage_read(struct file *file, char __user *buf,
size_t len, loff_t *offset)
{
struct svc_debugfs_pwrmon_rail *pwrmon_rails =
file_inode(file)->i_private;
struct gb_svc *svc = pwrmon_rails->svc;
int ret, desc;
u32 value;
char buff[16];
ret = gb_svc_pwrmon_sample_get(svc, pwrmon_rails->id,
GB_SVC_PWRMON_TYPE_VOL, &value);
if (ret) {
dev_err(&svc->dev,
"failed to get voltage sample %u: %d\n",
pwrmon_rails->id, ret);
return ret;
}
desc = scnprintf(buff, sizeof(buff), "%u\n", value);
return simple_read_from_buffer(buf, len, offset, buff, desc);
}
static ssize_t pwr_debugfs_current_read(struct file *file, char __user *buf,
size_t len, loff_t *offset)
{
struct svc_debugfs_pwrmon_rail *pwrmon_rails =
file_inode(file)->i_private;
struct gb_svc *svc = pwrmon_rails->svc;
int ret, desc;
u32 value;
char buff[16];
ret = gb_svc_pwrmon_sample_get(svc, pwrmon_rails->id,
GB_SVC_PWRMON_TYPE_CURR, &value);
if (ret) {
dev_err(&svc->dev,
"failed to get current sample %u: %d\n",
pwrmon_rails->id, ret);
return ret;
}
desc = scnprintf(buff, sizeof(buff), "%u\n", value);
return simple_read_from_buffer(buf, len, offset, buff, desc);
}
static ssize_t pwr_debugfs_power_read(struct file *file, char __user *buf,
size_t len, loff_t *offset)
{
struct svc_debugfs_pwrmon_rail *pwrmon_rails =
file_inode(file)->i_private;
struct gb_svc *svc = pwrmon_rails->svc;
int ret, desc;
u32 value;
char buff[16];
ret = gb_svc_pwrmon_sample_get(svc, pwrmon_rails->id,
GB_SVC_PWRMON_TYPE_PWR, &value);
if (ret) {
dev_err(&svc->dev, "failed to get power sample %u: %d\n",
pwrmon_rails->id, ret);
return ret;
}
desc = scnprintf(buff, sizeof(buff), "%u\n", value);
return simple_read_from_buffer(buf, len, offset, buff, desc);
}
static const struct file_operations pwrmon_debugfs_voltage_fops = {
.read = pwr_debugfs_voltage_read,
};
static const struct file_operations pwrmon_debugfs_current_fops = {
.read = pwr_debugfs_current_read,
};
static const struct file_operations pwrmon_debugfs_power_fops = {
.read = pwr_debugfs_power_read,
};
static void gb_svc_pwrmon_debugfs_init(struct gb_svc *svc)
{
int i;
size_t bufsize;
struct dentry *dent;
struct gb_svc_pwrmon_rail_names_get_response *rail_names;
u8 rail_count;
dent = debugfs_create_dir("pwrmon", svc->debugfs_dentry);
if (IS_ERR_OR_NULL(dent))
return;
if (gb_svc_pwrmon_rail_count_get(svc, &rail_count))
goto err_pwrmon_debugfs;
if (!rail_count || rail_count > GB_SVC_PWRMON_MAX_RAIL_COUNT)
goto err_pwrmon_debugfs;
bufsize = sizeof(*rail_names) +
GB_SVC_PWRMON_RAIL_NAME_BUFSIZE * rail_count;
rail_names = kzalloc(bufsize, GFP_KERNEL);
if (!rail_names)
goto err_pwrmon_debugfs;
svc->pwrmon_rails = kcalloc(rail_count, sizeof(*svc->pwrmon_rails),
GFP_KERNEL);
if (!svc->pwrmon_rails)
goto err_pwrmon_debugfs_free;
if (gb_svc_pwrmon_rail_names_get(svc, rail_names, bufsize))
goto err_pwrmon_debugfs_free;
for (i = 0; i < rail_count; i++) {
struct dentry *dir;
struct svc_debugfs_pwrmon_rail *rail = &svc->pwrmon_rails[i];
char fname[GB_SVC_PWRMON_RAIL_NAME_BUFSIZE];
snprintf(fname, sizeof(fname), "%s",
(char *)&rail_names->name[i]);
rail->id = i;
rail->svc = svc;
dir = debugfs_create_dir(fname, dent);
debugfs_create_file("voltage_now", 0444, dir, rail,
&pwrmon_debugfs_voltage_fops);
debugfs_create_file("current_now", 0444, dir, rail,
&pwrmon_debugfs_current_fops);
debugfs_create_file("power_now", 0444, dir, rail,
&pwrmon_debugfs_power_fops);
}
kfree(rail_names);
return;
err_pwrmon_debugfs_free:
kfree(rail_names);
kfree(svc->pwrmon_rails);
svc->pwrmon_rails = NULL;
err_pwrmon_debugfs:
debugfs_remove(dent);
}
static void gb_svc_debugfs_init(struct gb_svc *svc)
{
svc->debugfs_dentry = debugfs_create_dir(dev_name(&svc->dev),
gb_debugfs_get());
gb_svc_pwrmon_debugfs_init(svc);
}
static void gb_svc_debugfs_exit(struct gb_svc *svc)
{
debugfs_remove_recursive(svc->debugfs_dentry);
kfree(svc->pwrmon_rails);
svc->pwrmon_rails = NULL;
}
static int gb_svc_hello(struct gb_operation *op)
{
struct gb_connection *connection = op->connection;
struct gb_svc *svc = gb_connection_get_data(connection);
struct gb_svc_hello_request *hello_request;
int ret;
if (op->request->payload_size < sizeof(*hello_request)) {
dev_warn(&svc->dev, "short hello request (%zu < %zu)\n",
op->request->payload_size,
sizeof(*hello_request));
return -EINVAL;
}
hello_request = op->request->payload;
svc->endo_id = le16_to_cpu(hello_request->endo_id);
svc->ap_intf_id = hello_request->interface_id;
ret = device_add(&svc->dev);
if (ret) {
dev_err(&svc->dev, "failed to register svc device: %d\n", ret);
return ret;
}
ret = gb_svc_watchdog_create(svc);
if (ret) {
dev_err(&svc->dev, "failed to create watchdog: %d\n", ret);
goto err_unregister_device;
}
gb_svc_debugfs_init(svc);
return gb_svc_queue_deferred_request(op);
err_unregister_device:
gb_svc_watchdog_destroy(svc);
device_del(&svc->dev);
return ret;
}
static struct gb_interface *gb_svc_interface_lookup(struct gb_svc *svc,
u8 intf_id)
{
struct gb_host_device *hd = svc->hd;
struct gb_module *module;
size_t num_interfaces;
u8 module_id;
list_for_each_entry(module, &hd->modules, hd_node) {
module_id = module->module_id;
num_interfaces = module->num_interfaces;
if (intf_id >= module_id &&
intf_id < module_id + num_interfaces) {
return module->interfaces[intf_id - module_id];
}
}
return NULL;
}
static struct gb_module *gb_svc_module_lookup(struct gb_svc *svc, u8 module_id)
{
struct gb_host_device *hd = svc->hd;
struct gb_module *module;
list_for_each_entry(module, &hd->modules, hd_node) {
if (module->module_id == module_id)
return module;
}
return NULL;
}
static void gb_svc_process_hello_deferred(struct gb_operation *operation)
{
struct gb_connection *connection = operation->connection;
struct gb_svc *svc = gb_connection_get_data(connection);
int ret;
/*
* XXX This is a hack/work-around to reconfigure the APBridgeA-Switch
* link to PWM G2, 1 Lane, Slow Auto, so that it has sufficient
* bandwidth for 3 audio streams plus boot-over-UniPro of a hot-plugged
* module.
*
* The code should be removed once SW-2217, Heuristic for UniPro
* Power Mode Changes is resolved.
*/
ret = gb_svc_intf_set_power_mode(svc, svc->ap_intf_id,
GB_SVC_UNIPRO_HS_SERIES_A,
GB_SVC_UNIPRO_SLOW_AUTO_MODE,
2, 1,
GB_SVC_SMALL_AMPLITUDE,
GB_SVC_NO_DE_EMPHASIS,
GB_SVC_UNIPRO_SLOW_AUTO_MODE,
2, 1,
0, 0,
NULL, NULL);
if (ret)
dev_warn(&svc->dev,
"power mode change failed on AP to switch link: %d\n",
ret);
}
static void gb_svc_process_module_inserted(struct gb_operation *operation)
{
struct gb_svc_module_inserted_request *request;
struct gb_connection *connection = operation->connection;
struct gb_svc *svc = gb_connection_get_data(connection);
struct gb_host_device *hd = svc->hd;
struct gb_module *module;
size_t num_interfaces;
u8 module_id;
u16 flags;
int ret;
/* The request message size has already been verified. */
request = operation->request->payload;
module_id = request->primary_intf_id;
num_interfaces = request->intf_count;
flags = le16_to_cpu(request->flags);
dev_dbg(&svc->dev, "%s - id = %u, num_interfaces = %zu, flags = 0x%04x\n",
__func__, module_id, num_interfaces, flags);
if (flags & GB_SVC_MODULE_INSERTED_FLAG_NO_PRIMARY) {
dev_warn(&svc->dev, "no primary interface detected on module %u\n",
module_id);
}
module = gb_svc_module_lookup(svc, module_id);
if (module) {
dev_warn(&svc->dev, "unexpected module-inserted event %u\n",
module_id);
return;
}
module = gb_module_create(hd, module_id, num_interfaces);
if (!module) {
dev_err(&svc->dev, "failed to create module\n");
return;
}
ret = gb_module_add(module);
if (ret) {
gb_module_put(module);
return;
}
list_add(&module->hd_node, &hd->modules);
}
static void gb_svc_process_module_removed(struct gb_operation *operation)
{
struct gb_svc_module_removed_request *request;
struct gb_connection *connection = operation->connection;
struct gb_svc *svc = gb_connection_get_data(connection);
struct gb_module *module;
u8 module_id;
/* The request message size has already been verified. */
request = operation->request->payload;
module_id = request->primary_intf_id;
dev_dbg(&svc->dev, "%s - id = %u\n", __func__, module_id);
module = gb_svc_module_lookup(svc, module_id);
if (!module) {
dev_warn(&svc->dev, "unexpected module-removed event %u\n",
module_id);
return;
}
module->disconnected = true;
gb_module_del(module);
list_del(&module->hd_node);
gb_module_put(module);
}
static void gb_svc_process_intf_oops(struct gb_operation *operation)
{
struct gb_svc_intf_oops_request *request;
struct gb_connection *connection = operation->connection;
struct gb_svc *svc = gb_connection_get_data(connection);
struct gb_interface *intf;
u8 intf_id;
u8 reason;
/* The request message size has already been verified. */
request = operation->request->payload;
intf_id = request->intf_id;
reason = request->reason;
intf = gb_svc_interface_lookup(svc, intf_id);
if (!intf) {
dev_warn(&svc->dev, "unexpected interface-oops event %u\n",
intf_id);
return;
}
dev_info(&svc->dev, "Deactivating interface %u, interface oops reason = %u\n",
intf_id, reason);
mutex_lock(&intf->mutex);
intf->disconnected = true;
gb_interface_disable(intf);
gb_interface_deactivate(intf);
mutex_unlock(&intf->mutex);
}
static void gb_svc_process_intf_mailbox_event(struct gb_operation *operation)
{
struct gb_svc_intf_mailbox_event_request *request;
struct gb_connection *connection = operation->connection;
struct gb_svc *svc = gb_connection_get_data(connection);
struct gb_interface *intf;
u8 intf_id;
u16 result_code;
u32 mailbox;
/* The request message size has already been verified. */
request = operation->request->payload;
intf_id = request->intf_id;
result_code = le16_to_cpu(request->result_code);
mailbox = le32_to_cpu(request->mailbox);
dev_dbg(&svc->dev, "%s - id = %u, result = 0x%04x, mailbox = 0x%08x\n",
__func__, intf_id, result_code, mailbox);
intf = gb_svc_interface_lookup(svc, intf_id);
if (!intf) {
dev_warn(&svc->dev, "unexpected mailbox event %u\n", intf_id);
return;
}
gb_interface_mailbox_event(intf, result_code, mailbox);
}
static void gb_svc_process_deferred_request(struct work_struct *work)
{
struct gb_svc_deferred_request *dr;
struct gb_operation *operation;
struct gb_svc *svc;
u8 type;
dr = container_of(work, struct gb_svc_deferred_request, work);
operation = dr->operation;
svc = gb_connection_get_data(operation->connection);
type = operation->request->header->type;
switch (type) {
case GB_SVC_TYPE_SVC_HELLO:
gb_svc_process_hello_deferred(operation);
break;
case GB_SVC_TYPE_MODULE_INSERTED:
gb_svc_process_module_inserted(operation);
break;
case GB_SVC_TYPE_MODULE_REMOVED:
gb_svc_process_module_removed(operation);
break;
case GB_SVC_TYPE_INTF_MAILBOX_EVENT:
gb_svc_process_intf_mailbox_event(operation);
break;
case GB_SVC_TYPE_INTF_OOPS:
gb_svc_process_intf_oops(operation);
break;
default:
dev_err(&svc->dev, "bad deferred request type: 0x%02x\n", type);
}
gb_operation_put(operation);
kfree(dr);
}
static int gb_svc_queue_deferred_request(struct gb_operation *operation)
{
struct gb_svc *svc = gb_connection_get_data(operation->connection);
struct gb_svc_deferred_request *dr;
dr = kmalloc(sizeof(*dr), GFP_KERNEL);
if (!dr)
return -ENOMEM;
gb_operation_get(operation);
dr->operation = operation;
INIT_WORK(&dr->work, gb_svc_process_deferred_request);
queue_work(svc->wq, &dr->work);
return 0;
}
static int gb_svc_intf_reset_recv(struct gb_operation *op)
{
struct gb_svc *svc = gb_connection_get_data(op->connection);
struct gb_message *request = op->request;
struct gb_svc_intf_reset_request *reset;
if (request->payload_size < sizeof(*reset)) {
dev_warn(&svc->dev, "short reset request received (%zu < %zu)\n",
request->payload_size, sizeof(*reset));
return -EINVAL;
}
reset = request->payload;
/* FIXME Reset the interface here */
return 0;
}
static int gb_svc_module_inserted_recv(struct gb_operation *op)
{
struct gb_svc *svc = gb_connection_get_data(op->connection);
struct gb_svc_module_inserted_request *request;
if (op->request->payload_size < sizeof(*request)) {
dev_warn(&svc->dev, "short module-inserted request received (%zu < %zu)\n",
op->request->payload_size, sizeof(*request));
return -EINVAL;
}
request = op->request->payload;
dev_dbg(&svc->dev, "%s - id = %u\n", __func__,
request->primary_intf_id);
return gb_svc_queue_deferred_request(op);
}
static int gb_svc_module_removed_recv(struct gb_operation *op)
{
struct gb_svc *svc = gb_connection_get_data(op->connection);
struct gb_svc_module_removed_request *request;
if (op->request->payload_size < sizeof(*request)) {
dev_warn(&svc->dev, "short module-removed request received (%zu < %zu)\n",
op->request->payload_size, sizeof(*request));
return -EINVAL;
}
request = op->request->payload;
dev_dbg(&svc->dev, "%s - id = %u\n", __func__,
request->primary_intf_id);
return gb_svc_queue_deferred_request(op);
}
static int gb_svc_intf_oops_recv(struct gb_operation *op)
{
struct gb_svc *svc = gb_connection_get_data(op->connection);
struct gb_svc_intf_oops_request *request;
if (op->request->payload_size < sizeof(*request)) {
dev_warn(&svc->dev, "short intf-oops request received (%zu < %zu)\n",
op->request->payload_size, sizeof(*request));
return -EINVAL;
}
return gb_svc_queue_deferred_request(op);
}
static int gb_svc_intf_mailbox_event_recv(struct gb_operation *op)
{
struct gb_svc *svc = gb_connection_get_data(op->connection);
struct gb_svc_intf_mailbox_event_request *request;
if (op->request->payload_size < sizeof(*request)) {
dev_warn(&svc->dev, "short mailbox request received (%zu < %zu)\n",
op->request->payload_size, sizeof(*request));
return -EINVAL;
}
request = op->request->payload;
dev_dbg(&svc->dev, "%s - id = %u\n", __func__, request->intf_id);
return gb_svc_queue_deferred_request(op);
}
static int gb_svc_request_handler(struct gb_operation *op)
{
struct gb_connection *connection = op->connection;
struct gb_svc *svc = gb_connection_get_data(connection);
u8 type = op->type;
int ret = 0;
/*
* SVC requests need to follow a specific order (at least initially) and
* below code takes care of enforcing that. The expected order is:
* - PROTOCOL_VERSION
* - SVC_HELLO
* - Any other request, but the earlier two.
*
* Incoming requests are guaranteed to be serialized and so we don't
* need to protect 'state' for any races.
*/
switch (type) {
case GB_SVC_TYPE_PROTOCOL_VERSION:
if (svc->state != GB_SVC_STATE_RESET)
ret = -EINVAL;
break;
case GB_SVC_TYPE_SVC_HELLO:
if (svc->state != GB_SVC_STATE_PROTOCOL_VERSION)
ret = -EINVAL;
break;
default:
if (svc->state != GB_SVC_STATE_SVC_HELLO)
ret = -EINVAL;
break;
}
if (ret) {
dev_warn(&svc->dev, "unexpected request 0x%02x received (state %u)\n",
type, svc->state);
return ret;
}
switch (type) {
case GB_SVC_TYPE_PROTOCOL_VERSION:
ret = gb_svc_version_request(op);
if (!ret)
svc->state = GB_SVC_STATE_PROTOCOL_VERSION;
return ret;
case GB_SVC_TYPE_SVC_HELLO:
ret = gb_svc_hello(op);
if (!ret)
svc->state = GB_SVC_STATE_SVC_HELLO;
return ret;
case GB_SVC_TYPE_INTF_RESET:
return gb_svc_intf_reset_recv(op);
case GB_SVC_TYPE_MODULE_INSERTED:
return gb_svc_module_inserted_recv(op);
case GB_SVC_TYPE_MODULE_REMOVED:
return gb_svc_module_removed_recv(op);
case GB_SVC_TYPE_INTF_MAILBOX_EVENT:
return gb_svc_intf_mailbox_event_recv(op);
case GB_SVC_TYPE_INTF_OOPS:
return gb_svc_intf_oops_recv(op);
default:
dev_warn(&svc->dev, "unsupported request 0x%02x\n", type);
return -EINVAL;
}
}
static void gb_svc_release(struct device *dev)
{
struct gb_svc *svc = to_gb_svc(dev);
if (svc->connection)
gb_connection_destroy(svc->connection);
ida_destroy(&svc->device_id_map);
destroy_workqueue(svc->wq);
kfree(svc);
}
struct device_type greybus_svc_type = {
.name = "greybus_svc",
.release = gb_svc_release,
};
struct gb_svc *gb_svc_create(struct gb_host_device *hd)
{
struct gb_svc *svc;
svc = kzalloc(sizeof(*svc), GFP_KERNEL);
if (!svc)
return NULL;
svc->wq = alloc_workqueue("%s:svc", WQ_UNBOUND, 1, dev_name(&hd->dev));
if (!svc->wq) {
kfree(svc);
return NULL;
}
svc->dev.parent = &hd->dev;
svc->dev.bus = &greybus_bus_type;
svc->dev.type = &greybus_svc_type;
svc->dev.groups = svc_groups;
svc->dev.dma_mask = svc->dev.parent->dma_mask;
device_initialize(&svc->dev);
dev_set_name(&svc->dev, "%d-svc", hd->bus_id);
ida_init(&svc->device_id_map);
svc->state = GB_SVC_STATE_RESET;
svc->hd = hd;
svc->connection = gb_connection_create_static(hd, GB_SVC_CPORT_ID,
gb_svc_request_handler);
if (IS_ERR(svc->connection)) {
dev_err(&svc->dev, "failed to create connection: %ld\n",
PTR_ERR(svc->connection));
goto err_put_device;
}
gb_connection_set_data(svc->connection, svc);
return svc;
err_put_device:
put_device(&svc->dev);
return NULL;
}
int gb_svc_add(struct gb_svc *svc)
{
int ret;
/*
* The SVC protocol is currently driven by the SVC, so the SVC device
* is added from the connection request handler when enough
* information has been received.
*/
ret = gb_connection_enable(svc->connection);
if (ret)
return ret;
return 0;
}
static void gb_svc_remove_modules(struct gb_svc *svc)
{
struct gb_host_device *hd = svc->hd;
struct gb_module *module, *tmp;
list_for_each_entry_safe(module, tmp, &hd->modules, hd_node) {
gb_module_del(module);
list_del(&module->hd_node);
gb_module_put(module);
}
}
void gb_svc_del(struct gb_svc *svc)
{
gb_connection_disable_rx(svc->connection);
/*
* The SVC device may have been registered from the request handler.
*/
if (device_is_registered(&svc->dev)) {
gb_svc_debugfs_exit(svc);
gb_svc_watchdog_destroy(svc);
device_del(&svc->dev);
}
flush_workqueue(svc->wq);
gb_svc_remove_modules(svc);
gb_connection_disable(svc->connection);
}
void gb_svc_put(struct gb_svc *svc)
{
put_device(&svc->dev);
}