+#include "eal_private.h"
+
+#ifdef VFIO_PRESENT
+
+/* per-process VFIO config */
+static struct vfio_config vfio_cfg;
+
+static int vfio_type1_dma_map(int);
+static int vfio_spapr_dma_map(int);
+static int vfio_noiommu_dma_map(int);
+
+/* IOMMU types we support */
+static const struct vfio_iommu_type iommu_types[] = {
+ /* x86 IOMMU, otherwise known as type 1 */
+ { RTE_VFIO_TYPE1, "Type 1", &vfio_type1_dma_map},
+ /* ppc64 IOMMU, otherwise known as spapr */
+ { RTE_VFIO_SPAPR, "sPAPR", &vfio_spapr_dma_map},
+ /* IOMMU-less mode */
+ { RTE_VFIO_NOIOMMU, "No-IOMMU", &vfio_noiommu_dma_map},
+};
+
+int
+vfio_get_group_fd(int iommu_group_no)
+{
+ int i;
+ int vfio_group_fd;
+ int group_idx = -1;
+ char filename[PATH_MAX];
+
+ /* check if we already have the group descriptor open */
+ for (i = 0; i < VFIO_MAX_GROUPS; i++)
+ if (vfio_cfg.vfio_groups[i].group_no == iommu_group_no)
+ return vfio_cfg.vfio_groups[i].fd;
+
+ /* Lets see first if there is room for a new group */
+ if (vfio_cfg.vfio_active_groups == VFIO_MAX_GROUPS) {
+ RTE_LOG(ERR, EAL, "Maximum number of VFIO groups reached!\n");
+ return -1;
+ }
+
+ /* Now lets get an index for the new group */
+ for (i = 0; i < VFIO_MAX_GROUPS; i++)
+ if (vfio_cfg.vfio_groups[i].group_no == -1) {
+ group_idx = i;
+ break;
+ }
+
+ /* This should not happen */
+ if (group_idx == -1) {
+ RTE_LOG(ERR, EAL, "No VFIO group free slot found\n");
+ return -1;
+ }
+ /* if primary, try to open the group */
+ if (internal_config.process_type == RTE_PROC_PRIMARY) {
+ /* try regular group format */
+ snprintf(filename, sizeof(filename),
+ VFIO_GROUP_FMT, iommu_group_no);
+ vfio_group_fd = open(filename, O_RDWR);
+ if (vfio_group_fd < 0) {
+ /* if file not found, it's not an error */
+ if (errno != ENOENT) {
+ RTE_LOG(ERR, EAL, "Cannot open %s: %s\n", filename,
+ strerror(errno));
+ return -1;
+ }
+
+ /* special case: try no-IOMMU path as well */
+ snprintf(filename, sizeof(filename),
+ VFIO_NOIOMMU_GROUP_FMT, iommu_group_no);
+ vfio_group_fd = open(filename, O_RDWR);
+ if (vfio_group_fd < 0) {
+ if (errno != ENOENT) {
+ RTE_LOG(ERR, EAL, "Cannot open %s: %s\n", filename,
+ strerror(errno));
+ return -1;
+ }
+ return 0;
+ }
+ /* noiommu group found */
+ }
+
+ vfio_cfg.vfio_groups[group_idx].group_no = iommu_group_no;
+ vfio_cfg.vfio_groups[group_idx].fd = vfio_group_fd;
+ vfio_cfg.vfio_active_groups++;
+ return vfio_group_fd;
+ }
+ /* if we're in a secondary process, request group fd from the primary
+ * process via our socket
+ */
+ else {
+ int socket_fd, ret;
+
+ socket_fd = vfio_mp_sync_connect_to_primary();
+
+ if (socket_fd < 0) {
+ RTE_LOG(ERR, EAL, " cannot connect to primary process!\n");
+ return -1;
+ }
+ if (vfio_mp_sync_send_request(socket_fd, SOCKET_REQ_GROUP) < 0) {
+ RTE_LOG(ERR, EAL, " cannot request container fd!\n");
+ close(socket_fd);
+ return -1;
+ }
+ if (vfio_mp_sync_send_request(socket_fd, iommu_group_no) < 0) {
+ RTE_LOG(ERR, EAL, " cannot send group number!\n");
+ close(socket_fd);
+ return -1;
+ }
+ ret = vfio_mp_sync_receive_request(socket_fd);
+ switch (ret) {
+ case SOCKET_NO_FD:
+ close(socket_fd);
+ return 0;
+ case SOCKET_OK:
+ vfio_group_fd = vfio_mp_sync_receive_fd(socket_fd);
+ /* if we got the fd, return it */
+ if (vfio_group_fd > 0) {
+ close(socket_fd);
+ return vfio_group_fd;
+ }
+ /* fall-through on error */
+ default:
+ RTE_LOG(ERR, EAL, " cannot get container fd!\n");
+ close(socket_fd);
+ return -1;
+ }
+ }
+ return -1;
+}
+
+
+static int
+get_vfio_group_idx(int vfio_group_fd)
+{
+ int i;
+ for (i = 0; i < VFIO_MAX_GROUPS; i++)
+ if (vfio_cfg.vfio_groups[i].fd == vfio_group_fd)
+ return i;
+ return -1;
+}
+
+static void
+vfio_group_device_get(int vfio_group_fd)
+{
+ int i;
+
+ i = get_vfio_group_idx(vfio_group_fd);
+ if (i < 0 || i > (VFIO_MAX_GROUPS - 1))
+ RTE_LOG(ERR, EAL, " wrong vfio_group index (%d)\n", i);
+ else
+ vfio_cfg.vfio_groups[i].devices++;
+}
+
+static void
+vfio_group_device_put(int vfio_group_fd)
+{
+ int i;
+
+ i = get_vfio_group_idx(vfio_group_fd);
+ if (i < 0 || i > (VFIO_MAX_GROUPS - 1))
+ RTE_LOG(ERR, EAL, " wrong vfio_group index (%d)\n", i);
+ else
+ vfio_cfg.vfio_groups[i].devices--;
+}
+
+static int
+vfio_group_device_count(int vfio_group_fd)
+{
+ int i;
+
+ i = get_vfio_group_idx(vfio_group_fd);
+ if (i < 0 || i > (VFIO_MAX_GROUPS - 1)) {
+ RTE_LOG(ERR, EAL, " wrong vfio_group index (%d)\n", i);
+ return -1;
+ }
+
+ return vfio_cfg.vfio_groups[i].devices;
+}
+
+int
+clear_group(int vfio_group_fd)
+{
+ int i;
+ int socket_fd, ret;
+
+ if (internal_config.process_type == RTE_PROC_PRIMARY) {
+
+ i = get_vfio_group_idx(vfio_group_fd);
+ if (i < 0)
+ return -1;
+ vfio_cfg.vfio_groups[i].group_no = -1;
+ vfio_cfg.vfio_groups[i].fd = -1;
+ vfio_cfg.vfio_groups[i].devices = 0;
+ vfio_cfg.vfio_active_groups--;
+ return 0;
+ }
+
+ /* This is just for SECONDARY processes */
+ socket_fd = vfio_mp_sync_connect_to_primary();
+
+ if (socket_fd < 0) {
+ RTE_LOG(ERR, EAL, " cannot connect to primary process!\n");
+ return -1;
+ }
+
+ if (vfio_mp_sync_send_request(socket_fd, SOCKET_CLR_GROUP) < 0) {
+ RTE_LOG(ERR, EAL, " cannot request container fd!\n");
+ close(socket_fd);
+ return -1;
+ }
+
+ if (vfio_mp_sync_send_request(socket_fd, vfio_group_fd) < 0) {
+ RTE_LOG(ERR, EAL, " cannot send group fd!\n");
+ close(socket_fd);
+ return -1;
+ }
+
+ ret = vfio_mp_sync_receive_request(socket_fd);
+ switch (ret) {
+ case SOCKET_NO_FD:
+ RTE_LOG(ERR, EAL, " BAD VFIO group fd!\n");
+ close(socket_fd);
+ break;
+ case SOCKET_OK:
+ close(socket_fd);
+ return 0;
+ case SOCKET_ERR:
+ RTE_LOG(ERR, EAL, " Socket error\n");
+ close(socket_fd);
+ break;
+ default:
+ RTE_LOG(ERR, EAL, " UNKNOWN reply, %d\n", ret);
+ close(socket_fd);
+ }
+ return -1;
+}
+
+int
+vfio_setup_device(const char *sysfs_base, const char *dev_addr,
+ int *vfio_dev_fd, struct vfio_device_info *device_info)
+{
+ struct vfio_group_status group_status = {
+ .argsz = sizeof(group_status)
+ };
+ int vfio_group_fd;
+ int iommu_group_no;
+ int ret;
+
+ /* get group number */
+ ret = vfio_get_group_no(sysfs_base, dev_addr, &iommu_group_no);
+ if (ret == 0) {
+ RTE_LOG(WARNING, EAL, " %s not managed by VFIO driver, skipping\n",
+ dev_addr);
+ return 1;
+ }
+
+ /* if negative, something failed */
+ if (ret < 0)
+ return -1;
+
+ /* get the actual group fd */
+ vfio_group_fd = vfio_get_group_fd(iommu_group_no);
+ if (vfio_group_fd < 0)
+ return -1;
+
+ /* if group_fd == 0, that means the device isn't managed by VFIO */
+ if (vfio_group_fd == 0) {
+ RTE_LOG(WARNING, EAL, " %s not managed by VFIO driver, skipping\n",
+ dev_addr);
+ return 1;
+ }
+
+ /*
+ * at this point, we know that this group is viable (meaning, all devices
+ * are either bound to VFIO or not bound to anything)
+ */
+
+ /* check if the group is viable */
+ ret = ioctl(vfio_group_fd, VFIO_GROUP_GET_STATUS, &group_status);
+ if (ret) {
+ RTE_LOG(ERR, EAL, " %s cannot get group status, "
+ "error %i (%s)\n", dev_addr, errno, strerror(errno));
+ close(vfio_group_fd);
+ clear_group(vfio_group_fd);
+ return -1;
+ } else if (!(group_status.flags & VFIO_GROUP_FLAGS_VIABLE)) {
+ RTE_LOG(ERR, EAL, " %s VFIO group is not viable!\n", dev_addr);
+ close(vfio_group_fd);
+ clear_group(vfio_group_fd);
+ return -1;
+ }
+
+ /* check if group does not have a container yet */
+ if (!(group_status.flags & VFIO_GROUP_FLAGS_CONTAINER_SET)) {
+
+ /* add group to a container */
+ ret = ioctl(vfio_group_fd, VFIO_GROUP_SET_CONTAINER,
+ &vfio_cfg.vfio_container_fd);
+ if (ret) {
+ RTE_LOG(ERR, EAL, " %s cannot add VFIO group to container, "
+ "error %i (%s)\n", dev_addr, errno, strerror(errno));
+ close(vfio_group_fd);
+ clear_group(vfio_group_fd);
+ return -1;
+ }
+
+ /*
+ * pick an IOMMU type and set up DMA mappings for container
+ *
+ * needs to be done only once, only when first group is
+ * assigned to a container and only in primary process.
+ * Note this can happen several times with the hotplug
+ * functionality.
+ */
+ if (internal_config.process_type == RTE_PROC_PRIMARY &&
+ vfio_cfg.vfio_active_groups == 1) {
+ /* select an IOMMU type which we will be using */
+ const struct vfio_iommu_type *t =
+ vfio_set_iommu_type(vfio_cfg.vfio_container_fd);
+ if (!t) {
+ RTE_LOG(ERR, EAL,
+ " %s failed to select IOMMU type\n",
+ dev_addr);
+ close(vfio_group_fd);
+ clear_group(vfio_group_fd);
+ return -1;
+ }
+ ret = t->dma_map_func(vfio_cfg.vfio_container_fd);
+ if (ret) {
+ RTE_LOG(ERR, EAL,
+ " %s DMA remapping failed, error %i (%s)\n",
+ dev_addr, errno, strerror(errno));
+ close(vfio_group_fd);
+ clear_group(vfio_group_fd);
+ return -1;
+ }
+ }
+ }
+
+ /* get a file descriptor for the device */
+ *vfio_dev_fd = ioctl(vfio_group_fd, VFIO_GROUP_GET_DEVICE_FD, dev_addr);
+ if (*vfio_dev_fd < 0) {
+ /* if we cannot get a device fd, this implies a problem with
+ * the VFIO group or the container not having IOMMU configured.
+ */
+
+ RTE_LOG(WARNING, EAL, "Getting a vfio_dev_fd for %s failed\n",
+ dev_addr);
+ close(vfio_group_fd);
+ clear_group(vfio_group_fd);
+ return -1;
+ }
+
+ /* test and setup the device */
+ ret = ioctl(*vfio_dev_fd, VFIO_DEVICE_GET_INFO, device_info);
+ if (ret) {
+ RTE_LOG(ERR, EAL, " %s cannot get device info, "
+ "error %i (%s)\n", dev_addr, errno,
+ strerror(errno));
+ close(*vfio_dev_fd);
+ close(vfio_group_fd);
+ clear_group(vfio_group_fd);
+ return -1;
+ }
+ vfio_group_device_get(vfio_group_fd);
+
+ return 0;
+}
+
+int
+vfio_release_device(const char *sysfs_base, const char *dev_addr,
+ int vfio_dev_fd)
+{
+ struct vfio_group_status group_status = {
+ .argsz = sizeof(group_status)
+ };
+ int vfio_group_fd;
+ int iommu_group_no;
+ int ret;
+
+ /* get group number */
+ ret = vfio_get_group_no(sysfs_base, dev_addr, &iommu_group_no);
+ if (ret <= 0) {
+ RTE_LOG(WARNING, EAL, " %s not managed by VFIO driver\n",
+ dev_addr);
+ /* This is an error at this point. */
+ return -1;
+ }
+
+ /* get the actual group fd */
+ vfio_group_fd = vfio_get_group_fd(iommu_group_no);
+ if (vfio_group_fd <= 0) {
+ RTE_LOG(INFO, EAL, "vfio_get_group_fd failed for %s\n",
+ dev_addr);
+ return -1;
+ }
+
+ /* At this point we got an active group. Closing it will make the
+ * container detachment. If this is the last active group, VFIO kernel
+ * code will unset the container and the IOMMU mappings.
+ */
+
+ /* Closing a device */
+ if (close(vfio_dev_fd) < 0) {
+ RTE_LOG(INFO, EAL, "Error when closing vfio_dev_fd for %s\n",
+ dev_addr);
+ return -1;
+ }
+
+ /* An VFIO group can have several devices attached. Just when there is
+ * no devices remaining should the group be closed.
+ */
+ vfio_group_device_put(vfio_group_fd);
+ if (!vfio_group_device_count(vfio_group_fd)) {
+
+ if (close(vfio_group_fd) < 0) {
+ RTE_LOG(INFO, EAL, "Error when closing vfio_group_fd for %s\n",
+ dev_addr);
+ return -1;
+ }
+
+ if (clear_group(vfio_group_fd) < 0) {
+ RTE_LOG(INFO, EAL, "Error when clearing group for %s\n",
+ dev_addr);
+ return -1;
+ }
+ }
+
+ return 0;
+}
+
+int
+vfio_enable(const char *modname)
+{
+ /* initialize group list */
+ int i;
+ int vfio_available;
+
+ for (i = 0; i < VFIO_MAX_GROUPS; i++) {
+ vfio_cfg.vfio_groups[i].fd = -1;
+ vfio_cfg.vfio_groups[i].group_no = -1;
+ vfio_cfg.vfio_groups[i].devices = 0;
+ }
+
+ /* inform the user that we are probing for VFIO */
+ RTE_LOG(INFO, EAL, "Probing VFIO support...\n");
+
+ /* check if vfio module is loaded */
+ vfio_available = rte_eal_check_module(modname);
+
+ /* return error directly */
+ if (vfio_available == -1) {
+ RTE_LOG(INFO, EAL, "Could not get loaded module details!\n");
+ return -1;
+ }
+
+ /* return 0 if VFIO modules not loaded */
+ if (vfio_available == 0) {
+ RTE_LOG(DEBUG, EAL, "VFIO modules not loaded, "
+ "skipping VFIO support...\n");
+ return 0;
+ }
+
+ vfio_cfg.vfio_container_fd = vfio_get_container_fd();
+
+ /* check if we have VFIO driver enabled */
+ if (vfio_cfg.vfio_container_fd != -1) {
+ RTE_LOG(NOTICE, EAL, "VFIO support initialized\n");
+ vfio_cfg.vfio_enabled = 1;
+ } else {
+ RTE_LOG(NOTICE, EAL, "VFIO support could not be initialized\n");
+ }
+
+ return 0;
+}
+
+int
+vfio_is_enabled(const char *modname)
+{
+ const int mod_available = rte_eal_check_module(modname);
+ return vfio_cfg.vfio_enabled && mod_available;
+}
+
+const struct vfio_iommu_type *
+vfio_set_iommu_type(int vfio_container_fd)
+{
+ unsigned idx;
+ for (idx = 0; idx < RTE_DIM(iommu_types); idx++) {
+ const struct vfio_iommu_type *t = &iommu_types[idx];
+
+ int ret = ioctl(vfio_container_fd, VFIO_SET_IOMMU,
+ t->type_id);
+ if (!ret) {
+ RTE_LOG(NOTICE, EAL, " using IOMMU type %d (%s)\n",
+ t->type_id, t->name);
+ return t;
+ }
+ /* not an error, there may be more supported IOMMU types */
+ RTE_LOG(DEBUG, EAL, " set IOMMU type %d (%s) failed, "
+ "error %i (%s)\n", t->type_id, t->name, errno,
+ strerror(errno));
+ }
+ /* if we didn't find a suitable IOMMU type, fail */
+ return NULL;
+}
+
+int
+vfio_has_supported_extensions(int vfio_container_fd)
+{
+ int ret;
+ unsigned idx, n_extensions = 0;
+ for (idx = 0; idx < RTE_DIM(iommu_types); idx++) {
+ const struct vfio_iommu_type *t = &iommu_types[idx];
+
+ ret = ioctl(vfio_container_fd, VFIO_CHECK_EXTENSION,
+ t->type_id);
+ if (ret < 0) {
+ RTE_LOG(ERR, EAL, " could not get IOMMU type, "
+ "error %i (%s)\n", errno,
+ strerror(errno));
+ close(vfio_container_fd);
+ return -1;
+ } else if (ret == 1) {
+ /* we found a supported extension */
+ n_extensions++;
+ }
+ RTE_LOG(DEBUG, EAL, " IOMMU type %d (%s) is %s\n",
+ t->type_id, t->name,
+ ret ? "supported" : "not supported");
+ }
+
+ /* if we didn't find any supported IOMMU types, fail */
+ if (!n_extensions) {
+ close(vfio_container_fd);
+ return -1;
+ }
+
+ return 0;
+}