#include <sys/socket.h>
#include <sys/ioctl.h>
#include <sys/mman.h>
+#include <stdbool.h>
#include <rte_log.h>
#include <rte_pci.h>
};
EAL_REGISTER_TAILQ(rte_vfio_tailq)
-/* per-process VFIO config */
-static struct vfio_config vfio_cfg;
-
int
pci_vfio_read_config(const struct rte_intr_handle *intr_handle,
void *buf, size_t len, off_t offs)
/* set PCI bus mastering */
static int
-pci_vfio_set_bus_master(int dev_fd)
+pci_vfio_set_bus_master(int dev_fd, bool op)
{
uint16_t reg;
int ret;
return -1;
}
- /* set the master bit */
- reg |= PCI_COMMAND_MASTER;
+ if (op)
+ /* set the master bit */
+ reg |= PCI_COMMAND_MASTER;
+ else
+ reg &= ~(PCI_COMMAND_MASTER);
ret = pwrite64(dev_fd, ®, sizeof(reg),
VFIO_GET_REGION_ADDR(VFIO_PCI_CONFIG_REGION_INDEX) +
intr_idx = VFIO_PCI_NUM_IRQS;
/* get interrupt type from internal config (MSI-X by default, can be
- * overriden from the command line
+ * overridden from the command line
*/
switch (internal_config.vfio_intr_mode) {
case RTE_INTR_MODE_MSIX:
return -1;
}
-/* open group fd or get an existing one */
-int
-pci_vfio_get_group_fd(int iommu_group_no)
-{
- int i;
- int vfio_group_fd;
- char filename[PATH_MAX];
-
- /* check if we already have the group descriptor open */
- for (i = 0; i < vfio_cfg.vfio_group_idx; i++)
- if (vfio_cfg.vfio_groups[i].group_no == iommu_group_no)
- return vfio_cfg.vfio_groups[i].fd;
-
- /* 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 */
- }
-
- /* if the fd is valid, create a new group for it */
- if (vfio_cfg.vfio_group_idx == VFIO_MAX_GROUPS) {
- RTE_LOG(ERR, EAL, "Maximum number of VFIO groups reached!\n");
- close(vfio_group_fd);
- return -1;
- }
- vfio_cfg.vfio_groups[vfio_cfg.vfio_group_idx].group_no = iommu_group_no;
- vfio_cfg.vfio_groups[vfio_cfg.vfio_group_idx].fd = vfio_group_fd;
- 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;
-}
-
-/* parse IOMMU group number for a PCI device
- * returns 1 on success, -1 for errors, 0 for non-existent group
- */
-static int
-pci_vfio_get_group_no(const char *pci_addr, int *iommu_group_no)
-{
- return vfio_get_group_no(pci_get_sysfs_path(), pci_addr, iommu_group_no);
-}
-
-static void
-clear_current_group(void)
-{
- vfio_cfg.vfio_groups[vfio_cfg.vfio_group_idx].group_no = 0;
- vfio_cfg.vfio_groups[vfio_cfg.vfio_group_idx].fd = -1;
-}
-
-/**
- * Setup vfio_cfg for the device indentified by its address. It discovers
- * the configured I/O MMU groups or sets a new one for the device. If a new
- * groups is assigned, the DMA mapping is performed.
- * Returns 0 on success, a negative value on failure and a positive value in
- * case the given device cannot be managed this way.
- */
-static int pci_vfio_setup_device(const char *pci_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 = pci_vfio_get_group_no(pci_addr, &iommu_group_no);
- if (ret == 0) {
- RTE_LOG(WARNING, EAL, " %s not managed by VFIO driver, skipping\n",
- pci_addr);
- return 1;
- }
-
- /* if negative, something failed */
- if (ret < 0)
- return -1;
-
- /* get the actual group fd */
- vfio_group_fd = pci_vfio_get_group_fd(iommu_group_no);
- if (vfio_group_fd < 0)
- return -1;
-
- /* store group fd */
- vfio_cfg.vfio_groups[vfio_cfg.vfio_group_idx].group_no = iommu_group_no;
- vfio_cfg.vfio_groups[vfio_cfg.vfio_group_idx].fd = vfio_group_fd;
-
- /* 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",
- pci_addr);
- /* we store 0 as group fd to distinguish between existing but
- * unbound VFIO groups, and groups that don't exist at all.
- */
- vfio_cfg.vfio_group_idx++;
- 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", pci_addr, errno, strerror(errno));
- close(vfio_group_fd);
- clear_current_group();
- return -1;
- } else if (!(group_status.flags & VFIO_GROUP_FLAGS_VIABLE)) {
- RTE_LOG(ERR, EAL, " %s VFIO group is not viable!\n", pci_addr);
- close(vfio_group_fd);
- clear_current_group();
- 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", pci_addr, errno, strerror(errno));
- close(vfio_group_fd);
- clear_current_group();
- return -1;
- }
- /*
- * at this point we know that this group has been successfully
- * initialized, so we increment vfio_group_idx to indicate that we can
- * add new groups.
- */
- vfio_cfg.vfio_group_idx++;
- }
-
- /*
- * pick an IOMMU type and set up DMA mappings for container
- *
- * needs to be done only once, only when at least one group is assigned to
- * a container and only in primary process
- */
- if (internal_config.process_type == RTE_PROC_PRIMARY &&
- vfio_cfg.vfio_container_has_dma == 0) {
- /* 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", pci_addr);
- 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", pci_addr, errno, strerror(errno));
- return -1;
- }
- vfio_cfg.vfio_container_has_dma = 1;
- }
-
- /* get a file descriptor for the device */
- *vfio_dev_fd = ioctl(vfio_group_fd, VFIO_GROUP_GET_DEVICE_FD, pci_addr);
- if (*vfio_dev_fd < 0) {
- /* if we cannot get a device fd, this simply means that this
- * particular port is not bound to VFIO
- */
- RTE_LOG(WARNING, EAL, " %s not managed by VFIO driver, skipping\n",
- pci_addr);
- 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", pci_addr, errno, strerror(errno));
- close(*vfio_dev_fd);
- return -1;
- }
-
- return 0;
-}
-
/*
* map the PCI resources of a PCI device in virtual memory (VFIO version).
* primary and secondary processes follow almost exactly the same path
snprintf(pci_addr, sizeof(pci_addr), PCI_PRI_FMT,
loc->domain, loc->bus, loc->devid, loc->function);
- if ((ret = pci_vfio_setup_device(pci_addr, &vfio_dev_fd, &device_info)))
+ if ((ret = vfio_setup_device(pci_get_sysfs_path(), pci_addr,
+ &vfio_dev_fd, &device_info)))
return ret;
/* get MSI-X BAR, if any (we have to know where it is because we can't
} else {
/* if we're in a secondary process, just find our tailq entry */
TAILQ_FOREACH(vfio_res, vfio_res_list, next) {
- if (memcmp(&vfio_res->pci_addr, &dev->addr, sizeof(dev->addr)))
+ if (rte_eal_compare_pci_addr(&vfio_res->pci_addr,
+ &dev->addr))
continue;
break;
}
} else {
memreg[0].offset = reg.offset;
memreg[0].size = table_start;
- memreg[1].offset = table_end;
+ memreg[1].offset = reg.offset + table_end;
memreg[1].size = reg.size - table_end;
RTE_LOG(DEBUG, EAL,
/* if there's a second part, try to map it */
if (map_addr != MAP_FAILED
&& memreg[1].offset && memreg[1].size) {
- void *second_addr = RTE_PTR_ADD(bar_addr, memreg[1].offset);
+ void *second_addr = RTE_PTR_ADD(bar_addr,
+ memreg[1].offset -
+ (uintptr_t)reg.offset);
map_addr = pci_map_resource(second_addr,
vfio_dev_fd, memreg[1].offset,
memreg[1].size,
}
/* set bus mastering for the device */
- if (pci_vfio_set_bus_master(vfio_dev_fd)) {
+ if (pci_vfio_set_bus_master(vfio_dev_fd, true)) {
RTE_LOG(ERR, EAL, " %s cannot set up bus mastering!\n", pci_addr);
close(vfio_dev_fd);
rte_free(vfio_res);
return 0;
}
+int
+pci_vfio_unmap_resource(struct rte_pci_device *dev)
+{
+ char pci_addr[PATH_MAX] = {0};
+ struct rte_pci_addr *loc = &dev->addr;
+ int i, ret;
+ struct mapped_pci_resource *vfio_res = NULL;
+ struct mapped_pci_res_list *vfio_res_list;
+
+ struct pci_map *maps;
+
+ /* store PCI address string */
+ snprintf(pci_addr, sizeof(pci_addr), PCI_PRI_FMT,
+ loc->domain, loc->bus, loc->devid, loc->function);
+
+
+ if (close(dev->intr_handle.fd) < 0) {
+ RTE_LOG(INFO, EAL, "Error when closing eventfd file descriptor for %s\n",
+ pci_addr);
+ return -1;
+ }
+
+ if (pci_vfio_set_bus_master(dev->intr_handle.vfio_dev_fd, false)) {
+ RTE_LOG(ERR, EAL, " %s cannot unset bus mastering for PCI device!\n",
+ pci_addr);
+ return -1;
+ }
+
+ ret = vfio_release_device(pci_get_sysfs_path(), pci_addr,
+ dev->intr_handle.vfio_dev_fd);
+ if (ret < 0) {
+ RTE_LOG(ERR, EAL,
+ "%s(): cannot release device\n", __func__);
+ return ret;
+ }
+
+ vfio_res_list = RTE_TAILQ_CAST(rte_vfio_tailq.head, mapped_pci_res_list);
+ /* Get vfio_res */
+ TAILQ_FOREACH(vfio_res, vfio_res_list, next) {
+ if (memcmp(&vfio_res->pci_addr, &dev->addr, sizeof(dev->addr)))
+ continue;
+ break;
+ }
+ /* if we haven't found our tailq entry, something's wrong */
+ if (vfio_res == NULL) {
+ RTE_LOG(ERR, EAL, " %s cannot find TAILQ entry for PCI device!\n",
+ pci_addr);
+ return -1;
+ }
+
+ /* unmap BARs */
+ maps = vfio_res->maps;
+
+ RTE_LOG(INFO, EAL, "Releasing pci mapped resource for %s\n",
+ pci_addr);
+ for (i = 0; i < (int) vfio_res->nb_maps; i++) {
+
+ /*
+ * We do not need to be aware of MSI-X table BAR mappings as
+ * when mapping. Just using current maps array is enough
+ */
+ if (maps[i].addr) {
+ RTE_LOG(INFO, EAL, "Calling pci_unmap_resource for %s at %p\n",
+ pci_addr, maps[i].addr);
+ pci_unmap_resource(maps[i].addr, maps[i].size);
+ }
+ }
+
+ TAILQ_REMOVE(vfio_res_list, vfio_res, next);
+
+ return 0;
+}
+
int
pci_vfio_ioport_map(struct rte_pci_device *dev, int bar,
struct rte_pci_ioport *p)
int
pci_vfio_enable(void)
{
- /* 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;
- }
-
- /* inform the user that we are probing for VFIO */
- RTE_LOG(INFO, EAL, "Probing VFIO support...\n");
-
- /* check if vfio-pci module is loaded */
- vfio_available = rte_eal_check_module("vfio_pci");
-
- /* 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;
+ return vfio_enable("vfio_pci");
}
int
pci_vfio_is_enabled(void)
{
- return vfio_cfg.vfio_enabled;
+ return vfio_is_enabled("vfio_pci");
}
#endif
git.droids-corp.org / dpdk.git / blobdiff