--- /dev/null
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright 2017 NXP
+ */
+
+#include <assert.h>
+#include <stdio.h>
+#include <stdbool.h>
+#include <errno.h>
+#include <stdint.h>
+#include <inttypes.h>
+#include <string.h>
+
+#include <rte_byteorder.h>
+#include <rte_common.h>
+#include <rte_debug.h>
+#include <rte_dev.h>
+#include <rte_eal.h>
+#include <rte_kvargs.h>
+#include <rte_log.h>
+#include <rte_malloc.h>
+#include <rte_memory.h>
+#include <rte_memcpy.h>
+#include <rte_lcore.h>
+#include <rte_bus_vdev.h>
+
+#include <rte_rawdev.h>
+#include <rte_rawdev_pmd.h>
+
+#include "skeleton_rawdev.h"
+
+/* Dynamic log type identifier */
+int skeleton_pmd_logtype;
+
+/* Count of instances */
+uint16_t skeldev_init_once;
+
+/**< Rawdev Skeleton dummy driver name */
+#define SKELETON_PMD_RAWDEV_NAME rawdev_skeleton
+
+/**< Skeleton rawdev driver object */
+static struct rte_vdev_driver skeleton_pmd_drv;
+
+struct queue_buffers {
+ void *bufs[SKELETON_QUEUE_MAX_DEPTH];
+};
+
+static struct queue_buffers queue_buf[SKELETON_MAX_QUEUES] = {};
+static void clear_queue_bufs(int queue_id);
+
+static void skeleton_rawdev_info_get(struct rte_rawdev *dev,
+ rte_rawdev_obj_t dev_info)
+{
+ struct skeleton_rawdev *skeldev;
+ struct skeleton_rawdev_conf *skeldev_conf;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ if (!dev_info) {
+ SKELETON_PMD_ERR("Invalid request");
+ return;
+ }
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ skeldev_conf = dev_info;
+
+ skeldev_conf->num_queues = skeldev->num_queues;
+ skeldev_conf->capabilities = skeldev->capabilities;
+ skeldev_conf->device_state = skeldev->device_state;
+ skeldev_conf->firmware_state = skeldev->fw.firmware_state;
+}
+
+static int skeleton_rawdev_configure(const struct rte_rawdev *dev,
+ rte_rawdev_obj_t config)
+{
+ struct skeleton_rawdev *skeldev;
+ struct skeleton_rawdev_conf *skeldev_conf;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ RTE_FUNC_PTR_OR_ERR_RET(dev, -EINVAL);
+
+ if (!config) {
+ SKELETON_PMD_ERR("Invalid configuration");
+ return -EINVAL;
+ }
+
+ skeldev_conf = config;
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ if (skeldev_conf->num_queues <= SKELETON_MAX_QUEUES)
+ skeldev->num_queues = skeldev_conf->num_queues;
+ else
+ return -EINVAL;
+
+ skeldev->capabilities = skeldev_conf->capabilities;
+ skeldev->num_queues = skeldev_conf->num_queues;
+
+ return 0;
+}
+
+static int skeleton_rawdev_start(struct rte_rawdev *dev)
+{
+ int ret = 0;
+ struct skeleton_rawdev *skeldev;
+ enum skeleton_firmware_state fw_state;
+ enum skeleton_device_state device_state;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ RTE_FUNC_PTR_OR_ERR_RET(dev, -EINVAL);
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ fw_state = skeldev->fw.firmware_state;
+ device_state = skeldev->device_state;
+
+ if (fw_state == SKELETON_FW_LOADED &&
+ device_state == SKELETON_DEV_STOPPED) {
+ skeldev->device_state = SKELETON_DEV_RUNNING;
+ } else {
+ SKELETON_PMD_ERR("Device not ready for starting");
+ ret = -EINVAL;
+ }
+
+ return ret;
+}
+
+static void skeleton_rawdev_stop(struct rte_rawdev *dev)
+{
+ struct skeleton_rawdev *skeldev;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ if (dev) {
+ skeldev = skeleton_rawdev_get_priv(dev);
+ skeldev->device_state = SKELETON_DEV_STOPPED;
+ }
+}
+
+static void
+reset_queues(struct skeleton_rawdev *skeldev)
+{
+ int i;
+
+ for (i = 0; i < SKELETON_MAX_QUEUES; i++) {
+ skeldev->queues[i].depth = SKELETON_QUEUE_DEF_DEPTH;
+ skeldev->queues[i].state = SKELETON_QUEUE_DETACH;
+ }
+}
+
+static void
+reset_attribute_table(struct skeleton_rawdev *skeldev)
+{
+ int i;
+
+ for (i = 0; i < SKELETON_MAX_ATTRIBUTES; i++) {
+ if (skeldev->attr[i].name) {
+ free(skeldev->attr[i].name);
+ skeldev->attr[i].name = NULL;
+ }
+ }
+}
+
+static int skeleton_rawdev_close(struct rte_rawdev *dev)
+{
+ int ret = 0, i;
+ struct skeleton_rawdev *skeldev;
+ enum skeleton_firmware_state fw_state;
+ enum skeleton_device_state device_state;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ RTE_FUNC_PTR_OR_ERR_RET(dev, -EINVAL);
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ fw_state = skeldev->fw.firmware_state;
+ device_state = skeldev->device_state;
+
+ reset_queues(skeldev);
+ reset_attribute_table(skeldev);
+
+ switch (fw_state) {
+ case SKELETON_FW_LOADED:
+ if (device_state == SKELETON_DEV_RUNNING) {
+ SKELETON_PMD_ERR("Cannot close running device");
+ ret = -EINVAL;
+ } else {
+ /* Probably call fw reset here */
+ skeldev->fw.firmware_state = SKELETON_FW_READY;
+ }
+ break;
+ case SKELETON_FW_READY:
+ case SKELETON_FW_ERROR:
+ default:
+ SKELETON_PMD_DEBUG("Device already in stopped state");
+ ret = -EINVAL;
+ break;
+ }
+
+ /* Clear all allocated queues */
+ for (i = 0; i < SKELETON_MAX_QUEUES; i++)
+ clear_queue_bufs(i);
+
+ return ret;
+}
+
+static int skeleton_rawdev_reset(struct rte_rawdev *dev)
+{
+ struct skeleton_rawdev *skeldev;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ RTE_FUNC_PTR_OR_ERR_RET(dev, -EINVAL);
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ SKELETON_PMD_DEBUG("Resetting device");
+ skeldev->fw.firmware_state = SKELETON_FW_READY;
+
+ return 0;
+}
+
+static void skeleton_rawdev_queue_def_conf(struct rte_rawdev *dev,
+ uint16_t queue_id,
+ rte_rawdev_obj_t queue_conf)
+{
+ struct skeleton_rawdev *skeldev;
+ struct skeleton_rawdev_queue *skelq;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ if (!dev || !queue_conf)
+ return;
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+ skelq = &skeldev->queues[queue_id];
+
+ if (queue_id < SKELETON_MAX_QUEUES)
+ rte_memcpy(queue_conf, skelq,
+ sizeof(struct skeleton_rawdev_queue));
+}
+
+static void
+clear_queue_bufs(int queue_id)
+{
+ int i;
+
+ /* Clear buffers for queue_id */
+ for (i = 0; i < SKELETON_QUEUE_MAX_DEPTH; i++)
+ queue_buf[queue_id].bufs[i] = NULL;
+}
+
+static int skeleton_rawdev_queue_setup(struct rte_rawdev *dev,
+ uint16_t queue_id,
+ rte_rawdev_obj_t queue_conf)
+{
+ int ret = 0;
+ struct skeleton_rawdev *skeldev;
+ struct skeleton_rawdev_queue *q;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ if (!dev || !queue_conf)
+ return -EINVAL;
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+ q = &skeldev->queues[queue_id];
+
+ if (skeldev->num_queues > queue_id &&
+ q->depth < SKELETON_QUEUE_MAX_DEPTH) {
+ rte_memcpy(q, queue_conf,
+ sizeof(struct skeleton_rawdev_queue));
+ clear_queue_bufs(queue_id);
+ } else {
+ SKELETON_PMD_ERR("Invalid queue configuration");
+ ret = -EINVAL;
+ }
+
+ return ret;
+}
+
+static int skeleton_rawdev_queue_release(struct rte_rawdev *dev,
+ uint16_t queue_id)
+{
+ int ret = 0;
+ struct skeleton_rawdev *skeldev;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ RTE_FUNC_PTR_OR_ERR_RET(dev, -EINVAL);
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ if (skeldev->num_queues > queue_id) {
+ skeldev->queues[queue_id].state = SKELETON_QUEUE_DETACH;
+ skeldev->queues[queue_id].depth = SKELETON_QUEUE_DEF_DEPTH;
+ clear_queue_bufs(queue_id);
+ } else {
+ SKELETON_PMD_ERR("Invalid queue configuration");
+ ret = -EINVAL;
+ }
+
+ return ret;
+}
+
+static int skeleton_rawdev_get_attr(struct rte_rawdev *dev,
+ const char *attr_name,
+ uint64_t *attr_value)
+{
+ int i;
+ uint8_t done = 0;
+ struct skeleton_rawdev *skeldev;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ if (!dev || !attr_name || !attr_value) {
+ SKELETON_PMD_ERR("Invalid arguments for getting attributes");
+ return -EINVAL;
+ }
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ for (i = 0; i < SKELETON_MAX_ATTRIBUTES; i++) {
+ if (!skeldev->attr[i].name)
+ continue;
+
+ if (!strncmp(skeldev->attr[i].name, attr_name,
+ SKELETON_ATTRIBUTE_NAME_MAX)) {
+ *attr_value = skeldev->attr[i].value;
+ done = 1;
+ SKELETON_PMD_DEBUG("Attribute (%s) Value (%" PRIu64 ")",
+ attr_name, *attr_value);
+ break;
+ }
+ }
+
+ if (done)
+ return 0;
+
+ /* Attribute not found */
+ return -EINVAL;
+}
+
+static int skeleton_rawdev_set_attr(struct rte_rawdev *dev,
+ const char *attr_name,
+ const uint64_t attr_value)
+{
+ int i;
+ uint8_t done = 0;
+ struct skeleton_rawdev *skeldev;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ if (!dev || !attr_name) {
+ SKELETON_PMD_ERR("Invalid arguments for setting attributes");
+ return -EINVAL;
+ }
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ /* Check if attribute already exists */
+ for (i = 0; i < SKELETON_MAX_ATTRIBUTES; i++) {
+ if (!skeldev->attr[i].name)
+ break;
+
+ if (!strncmp(skeldev->attr[i].name, attr_name,
+ SKELETON_ATTRIBUTE_NAME_MAX)) {
+ /* Update value */
+ skeldev->attr[i].value = attr_value;
+ done = 1;
+ break;
+ }
+ }
+
+ if (!done) {
+ if (i < (SKELETON_MAX_ATTRIBUTES - 1)) {
+ /* There is still space to insert one more */
+ skeldev->attr[i].name = strdup(attr_name);
+ if (!skeldev->attr[i].name)
+ return -ENOMEM;
+
+ skeldev->attr[i].value = attr_value;
+ return 0;
+ }
+ }
+
+ return -EINVAL;
+}
+
+static int skeleton_rawdev_enqueue_bufs(struct rte_rawdev *dev,
+ struct rte_rawdev_buf **buffers,
+ unsigned int count,
+ rte_rawdev_obj_t context)
+{
+ unsigned int i;
+ uint16_t q_id;
+ RTE_SET_USED(dev);
+
+ /* context is essentially the queue_id which is
+ * transferred as opaque object through the library layer. This can
+ * help in complex implementation which require more information than
+ * just an integer - for example, a queue-pair.
+ */
+ q_id = *((int *)context);
+
+ for (i = 0; i < count; i++)
+ queue_buf[q_id].bufs[i] = buffers[i]->buf_addr;
+
+ return i;
+}
+
+static int skeleton_rawdev_dequeue_bufs(struct rte_rawdev *dev,
+ struct rte_rawdev_buf **buffers,
+ unsigned int count,
+ rte_rawdev_obj_t context)
+{
+ unsigned int i;
+ uint16_t q_id;
+ RTE_SET_USED(dev);
+
+ /* context is essentially the queue_id which is
+ * transferred as opaque object through the library layer. This can
+ * help in complex implementation which require more information than
+ * just an integer - for example, a queue-pair.
+ */
+ q_id = *((int *)context);
+
+ for (i = 0; i < count; i++)
+ buffers[i]->buf_addr = queue_buf[q_id].bufs[i];
+
+ return i;
+}
+
+static int skeleton_rawdev_dump(struct rte_rawdev *dev, FILE *f)
+{
+ RTE_SET_USED(dev);
+ RTE_SET_USED(f);
+
+ return 0;
+}
+
+static int skeleton_rawdev_firmware_status_get(struct rte_rawdev *dev,
+ rte_rawdev_obj_t status_info)
+{
+ struct skeleton_rawdev *skeldev;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ RTE_FUNC_PTR_OR_ERR_RET(dev, -EINVAL);
+
+ if (status_info)
+ memcpy(status_info, &skeldev->fw.firmware_state,
+ sizeof(enum skeleton_firmware_state));
+
+ return 0;
+}
+
+
+static int skeleton_rawdev_firmware_version_get(
+ struct rte_rawdev *dev,
+ rte_rawdev_obj_t version_info)
+{
+ struct skeleton_rawdev *skeldev;
+ struct skeleton_firmware_version_info *vi;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+ vi = version_info;
+
+ vi->major = skeldev->fw.firmware_version.major;
+ vi->minor = skeldev->fw.firmware_version.minor;
+ vi->subrel = skeldev->fw.firmware_version.subrel;
+
+ return 0;
+}
+
+static int skeleton_rawdev_firmware_load(struct rte_rawdev *dev,
+ rte_rawdev_obj_t firmware_buf)
+{
+ struct skeleton_rawdev *skeldev;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ /* firmware_buf is a mmaped, possibly DMA'able area, buffer. Being
+ * dummy, all this does is check if firmware_buf is not NULL and
+ * sets the state of the firmware.
+ */
+ if (!firmware_buf)
+ return -EINVAL;
+
+ skeldev->fw.firmware_state = SKELETON_FW_LOADED;
+
+ return 0;
+}
+
+static int skeleton_rawdev_firmware_unload(struct rte_rawdev *dev)
+{
+ struct skeleton_rawdev *skeldev;
+
+ SKELETON_PMD_FUNC_TRACE();
+
+ skeldev = skeleton_rawdev_get_priv(dev);
+
+ skeldev->fw.firmware_state = SKELETON_FW_READY;
+
+ return 0;
+}
+
+static const struct rte_rawdev_ops skeleton_rawdev_ops = {
+ .dev_info_get = skeleton_rawdev_info_get,
+ .dev_configure = skeleton_rawdev_configure,
+ .dev_start = skeleton_rawdev_start,
+ .dev_stop = skeleton_rawdev_stop,
+ .dev_close = skeleton_rawdev_close,
+ .dev_reset = skeleton_rawdev_reset,
+
+ .queue_def_conf = skeleton_rawdev_queue_def_conf,
+ .queue_setup = skeleton_rawdev_queue_setup,
+ .queue_release = skeleton_rawdev_queue_release,
+
+ .attr_get = skeleton_rawdev_get_attr,
+ .attr_set = skeleton_rawdev_set_attr,
+
+ .enqueue_bufs = skeleton_rawdev_enqueue_bufs,
+ .dequeue_bufs = skeleton_rawdev_dequeue_bufs,
+
+ .dump = skeleton_rawdev_dump,
+
+ .xstats_get = NULL,
+ .xstats_get_names = NULL,
+ .xstats_get_by_name = NULL,
+ .xstats_reset = NULL,
+
+ .firmware_status_get = skeleton_rawdev_firmware_status_get,
+ .firmware_version_get = skeleton_rawdev_firmware_version_get,
+ .firmware_load = skeleton_rawdev_firmware_load,
+ .firmware_unload = skeleton_rawdev_firmware_unload,
+};
+
+static int
+skeleton_rawdev_create(const char *name,
+ struct rte_vdev_device *vdev,
+ int socket_id)
+{
+ int ret = 0, i;
+ struct rte_rawdev *rawdev = NULL;
+ struct skeleton_rawdev *skeldev = NULL;
+
+ if (!name) {
+ SKELETON_PMD_ERR("Invalid name of the device!");
+ ret = -EINVAL;
+ goto cleanup;
+ }
+
+ /* Allocate device structure */
+ rawdev = rte_rawdev_pmd_allocate(name, sizeof(struct skeleton_rawdev),
+ socket_id);
+ if (rawdev == NULL) {
+ SKELETON_PMD_ERR("Unable to allocate rawdevice");
+ ret = -EINVAL;
+ goto cleanup;
+ }
+
+ rawdev->dev_ops = &skeleton_rawdev_ops;
+ rawdev->device = &vdev->device;
+ rawdev->driver_name = vdev->device.driver->name;
+
+ skeldev = skeleton_rawdev_get_priv(rawdev);
+
+ skeldev->device_id = SKELETON_DEVICE_ID;
+ skeldev->vendor_id = SKELETON_VENDOR_ID;
+ skeldev->capabilities = SKELETON_DEFAULT_CAPA;
+
+ memset(&skeldev->fw, 0, sizeof(struct skeleton_firmware));
+
+ skeldev->fw.firmware_state = SKELETON_FW_READY;
+ skeldev->fw.firmware_version.major = SKELETON_MAJOR_VER;
+ skeldev->fw.firmware_version.minor = SKELETON_MINOR_VER;
+ skeldev->fw.firmware_version.subrel = SKELETON_SUB_VER;
+
+ skeldev->device_state = SKELETON_DEV_STOPPED;
+
+ /* Reset/set to default queue configuration for this device */
+ for (i = 0; i < SKELETON_MAX_QUEUES; i++) {
+ skeldev->queues[i].state = SKELETON_QUEUE_DETACH;
+ skeldev->queues[i].depth = SKELETON_QUEUE_DEF_DEPTH;
+ }
+
+ /* Clear all allocated queue buffers */
+ for (i = 0; i < SKELETON_MAX_QUEUES; i++)
+ clear_queue_bufs(i);
+
+ return ret;
+
+cleanup:
+ if (rawdev)
+ rte_rawdev_pmd_release(rawdev);
+
+ return ret;
+}
+
+static int
+skeleton_rawdev_destroy(const char *name)
+{
+ int ret;
+ struct rte_rawdev *rdev;
+
+ if (!name) {
+ SKELETON_PMD_ERR("Invalid device name");
+ return -EINVAL;
+ }
+
+ rdev = rte_rawdev_pmd_get_named_dev(name);
+ if (!rdev) {
+ SKELETON_PMD_ERR("Invalid device name (%s)", name);
+ return -EINVAL;
+ }
+
+ /* rte_rawdev_close is called by pmd_release */
+ ret = rte_rawdev_pmd_release(rdev);
+ if (ret)
+ SKELETON_PMD_DEBUG("Device cleanup failed");
+
+ return 0;
+}
+
+static int
+skeleton_rawdev_probe(struct rte_vdev_device *vdev)
+{
+ const char *name;
+ int ret = 0;
+
+
+ name = rte_vdev_device_name(vdev);
+ /* More than one instance is not supported */
+ if (skeldev_init_once) {
+ SKELETON_PMD_ERR("Multiple instance not supported for %s",
+ name);
+ return -EINVAL;
+ }
+
+ SKELETON_PMD_INFO("Init %s on NUMA node %d", name, rte_socket_id());
+
+ ret = skeleton_rawdev_create(name, vdev, rte_socket_id());
+
+ /* Device instance created; Second instance not possible */
+ skeldev_init_once = 1;
+
+ return ret;
+}
+
+static int
+skeleton_rawdev_remove(struct rte_vdev_device *vdev)
+{
+ const char *name;
+ int ret;
+
+ name = rte_vdev_device_name(vdev);
+
+ SKELETON_PMD_INFO("Closing %s on NUMA node %d", name, rte_socket_id());
+
+ ret = skeleton_rawdev_destroy(name);
+ if (!ret)
+ skeldev_init_once = 0;
+
+ return ret;
+}
+
+static struct rte_vdev_driver skeleton_pmd_drv = {
+ .probe = skeleton_rawdev_probe,
+ .remove = skeleton_rawdev_remove
+};
+
+RTE_PMD_REGISTER_VDEV(SKELETON_PMD_RAWDEV_NAME, skeleton_pmd_drv);
+
+RTE_INIT(skeleton_pmd_init_log);
+
+static void
+skeleton_pmd_init_log(void)
+{
+ skeleton_pmd_logtype = rte_log_register("rawdev.skeleton");
+ if (skeleton_pmd_logtype >= 0)
+ rte_log_set_level(skeleton_pmd_logtype, RTE_LOG_INFO);
+}
--- /dev/null
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright 2017 NXP
+ */
+
+#ifndef __SKELETON_RAWDEV_H__
+#define __SKELETON_RAWDEV_H__
+
+#include <rte_rawdev.h>
+
+extern int skeleton_pmd_logtype;
+
+#define SKELETON_PMD_LOG(level, fmt, args...) \
+ rte_log(RTE_LOG_ ## level, skeleton_pmd_logtype, "%s(): " fmt "\n", \
+ __func__, ##args)
+
+#define SKELETON_PMD_FUNC_TRACE() SKELETON_PMD_LOG(DEBUG, ">>")
+
+#define SKELETON_PMD_DEBUG(fmt, args...) \
+ SKELETON_PMD_LOG(DEBUG, fmt, ## args)
+#define SKELETON_PMD_INFO(fmt, args...) \
+ SKELETON_PMD_LOG(INFO, fmt, ## args)
+#define SKELETON_PMD_ERR(fmt, args...) \
+ SKELETON_PMD_LOG(ERR, fmt, ## args)
+#define SKELETON_PMD_WARN(fmt, args...) \
+ SKELETON_PMD_LOG(WARNING, fmt, ## args)
+/* Macros for self test application */
+#define SKELETON_TEST_INFO SKELETON_PMD_INFO
+#define SKELETON_TEST_DEBUG SKELETON_PMD_DEBUG
+#define SKELETON_TEST_ERR SKELETON_PMD_ERR
+#define SKELETON_TEST_WARN SKELETON_PMD_WARN
+
+#define SKELETON_SELFTEST_ARG ("selftest")
+
+#define SKELETON_VENDOR_ID 0x10
+#define SKELETON_DEVICE_ID 0x01
+
+#define SKELETON_MAJOR_VER 1
+#define SKELETON_MINOR_VER 0
+#define SKELETON_SUB_VER 0
+
+#define SKELETON_MAX_QUEUES 1
+
+enum skeleton_firmware_state {
+ SKELETON_FW_READY,
+ SKELETON_FW_LOADED,
+ SKELETON_FW_ERROR
+};
+
+enum skeleton_device_state {
+ SKELETON_DEV_RUNNING,
+ SKELETON_DEV_STOPPED
+};
+
+enum skeleton_queue_state {
+ SKELETON_QUEUE_DETACH,
+ SKELETON_QUEUE_ATTACH
+};
+
+#define SKELETON_QUEUE_DEF_DEPTH 10
+#define SKELETON_QUEUE_MAX_DEPTH 25
+
+struct skeleton_firmware_version_info {
+ uint8_t major;
+ uint8_t minor;
+ uint8_t subrel;
+};
+
+struct skeleton_firmware {
+ /**< Device firmware information */
+ struct skeleton_firmware_version_info firmware_version;
+ /**< Device state */
+ enum skeleton_firmware_state firmware_state;
+
+};
+
+#define SKELETON_MAX_ATTRIBUTES 10
+#define SKELETON_ATTRIBUTE_NAME_MAX 20
+
+struct skeleton_rawdev_attributes {
+ /**< Name of the attribute */
+ char *name;
+ /**< Value or reference of value of attribute */
+ uint64_t value;
+};
+
+/**< Device supports firmware loading/unloading */
+#define SKELETON_CAPA_FW_LOAD 0x0001
+/**< Device supports firmware reset */
+#define SKELETON_CAPA_FW_RESET 0x0002
+/**< Device support queue based communication */
+#define SKELETON_CAPA_QUEUES 0x0004
+/**< Default Capabilities: FW_LOAD, FW_RESET, QUEUES */
+#define SKELETON_DEFAULT_CAPA 0x7
+
+struct skeleton_rawdev_queue {
+ uint8_t state;
+ uint32_t depth;
+};
+
+struct skeleton_rawdev {
+ uint16_t device_id;
+ uint16_t vendor_id;
+ uint16_t num_queues;
+ /**< One of SKELETON_CAPA_* */
+ uint16_t capabilities;
+ /**< State of device; linked to firmware state */
+ enum skeleton_device_state device_state;
+ /**< Firmware configuration */
+ struct skeleton_firmware fw;
+ /**< Collection of all communication channels - which can be referred
+ * to as queues.
+ */
+ struct skeleton_rawdev_queue queues[SKELETON_MAX_QUEUES];
+ /**< Global table containing various pre-defined and user-defined
+ * attributes.
+ */
+ struct skeleton_rawdev_attributes attr[SKELETON_MAX_ATTRIBUTES];
+ struct rte_device *device;
+};
+
+struct skeleton_rawdev_conf {
+ uint16_t num_queues;
+ unsigned int capabilities;
+ enum skeleton_device_state device_state;
+ enum skeleton_firmware_state firmware_state;
+};
+
+static inline struct skeleton_rawdev *
+skeleton_rawdev_get_priv(const struct rte_rawdev *rawdev)
+{
+ return rawdev->dev_private;
+}
+
+int test_rawdev_skeldev(void);
+
+#endif /* __SKELETON_RAWDEV_H__ */
git.droids-corp.org / dpdk.git / commitdiff