X-Git-Url: http://git.droids-corp.org/?a=blobdiff_plain;f=lib%2Flibrte_eal%2Flinuxapp%2Feal%2Feal_pci.c;h=1d5a13b3312f33444bcf15e45dd7e1d0d2d6f6a9;hb=e1d09b00d7d486a6cbe2aaed999382bfc60ce1bf;hp=a422e5f6c073c58a8d8d18d6539cf9d03e71a697;hpb=67c536bdad9380ab9616bcedb6e387da3ceae816;p=dpdk.git diff --git a/lib/librte_eal/linuxapp/eal/eal_pci.c b/lib/librte_eal/linuxapp/eal/eal_pci.c index a422e5f6c0..1d5a13b331 100644 --- a/lib/librte_eal/linuxapp/eal/eal_pci.c +++ b/lib/librte_eal/linuxapp/eal/eal_pci.c @@ -33,14 +33,13 @@ #include #include -#include #include #include -#include #include #include #include +#include #include "rte_pci_dev_ids.h" #include "eal_filesystem.h" @@ -56,10 +55,8 @@ * IGB_UIO driver (or doesn't initialize, if the device wasn't bound to it). */ -struct mapped_pci_res_list *pci_res_list = NULL; - /* unbind kernel driver for this device */ -static int +int pci_unbind_kernel_driver(struct rte_pci_device *dev) { int n; @@ -69,7 +66,7 @@ pci_unbind_kernel_driver(struct rte_pci_device *dev) struct rte_pci_addr *loc = &dev->addr; /* open /sys/bus/pci/devices/AAAA:BB:CC.D/driver */ - rte_snprintf(filename, sizeof(filename), + snprintf(filename, sizeof(filename), SYSFS_PCI_DEVICES "/" PCI_PRI_FMT "/driver/unbind", loc->domain, loc->bus, loc->devid, loc->function); @@ -77,10 +74,10 @@ pci_unbind_kernel_driver(struct rte_pci_device *dev) if (f == NULL) /* device was not bound */ return 0; - n = rte_snprintf(buf, sizeof(buf), PCI_PRI_FMT "\n", + n = snprintf(buf, sizeof(buf), PCI_PRI_FMT "\n", loc->domain, loc->bus, loc->devid, loc->function); if ((n < 0) || (n >= (int)sizeof(buf))) { - RTE_LOG(ERR, EAL, "%s(): rte_snprintf failed\n", __func__); + RTE_LOG(ERR, EAL, "%s(): snprintf failed\n", __func__); goto error; } if (fwrite(buf, n, 1, f) == 0) { @@ -97,35 +94,54 @@ error: return -1; } -/* map a particular resource from a file */ -void * -pci_map_resource(void * requested_addr, int fd, off_t offset, size_t size) +static int +pci_get_kernel_driver_by_path(const char *filename, char *dri_name) { - void *mapaddr; - - /* Map the PCI memory resource of device */ - mapaddr = mmap(requested_addr, size, PROT_READ | PROT_WRITE, - MAP_SHARED, fd, offset); - if (mapaddr == MAP_FAILED || - (requested_addr != NULL && mapaddr != requested_addr)) { - RTE_LOG(ERR, EAL, "%s(): cannot mmap(%d, %p, 0x%lx, 0x%lx): %s (%p)\n", - __func__, fd, requested_addr, - (unsigned long)size, (unsigned long)offset, - strerror(errno), mapaddr); - goto fail; + int count; + char path[PATH_MAX]; + char *name; + + if (!filename || !dri_name) + return -1; + + count = readlink(filename, path, PATH_MAX); + if (count >= PATH_MAX) + return -1; + + /* For device does not have a driver */ + if (count < 0) + return 1; + + path[count] = '\0'; + + name = strrchr(path, '/'); + if (name) { + strncpy(dri_name, name + 1, strlen(name + 1) + 1); + return 0; } - RTE_LOG(DEBUG, EAL, " PCI memory mapped at %p\n", mapaddr); + return -1; +} + +void * +pci_find_max_end_va(void) +{ + const struct rte_memseg *seg = rte_eal_get_physmem_layout(); + const struct rte_memseg *last = seg; + unsigned i = 0; + + for (i = 0; i < RTE_MAX_MEMSEG; i++, seg++) { + if (seg->addr == NULL) + break; - return mapaddr; + if (seg->addr > last->addr) + last = seg; -fail: - return NULL; + } + return RTE_PTR_ADD(last->addr, last->len); } /* parse the "resource" sysfs file */ -#define IORESOURCE_MEM 0x00000200 - static int pci_parse_sysfs_resource(const char *filename, struct rte_pci_device *dev) { @@ -186,20 +202,6 @@ error: return -1; } -/* Compare two PCI device addresses. */ -static int -pci_addr_comparison(struct rte_pci_addr *addr, struct rte_pci_addr *addr2) -{ - uint64_t dev_addr = (addr->domain << 24) + (addr->bus << 16) + (addr->devid << 8) + addr->function; - uint64_t dev_addr2 = (addr2->domain << 24) + (addr2->bus << 16) + (addr2->devid << 8) + addr2->function; - - if (dev_addr > dev_addr2) - return 1; - else - return 0; -} - - /* Scan one pci sysfs entry, and fill the devices list from it. */ static int pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, @@ -208,11 +210,12 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, char filename[PATH_MAX]; unsigned long tmp; struct rte_pci_device *dev; + char driver[PATH_MAX]; + int ret; dev = malloc(sizeof(*dev)); - if (dev == NULL) { + if (dev == NULL) return -1; - } memset(dev, 0, sizeof(*dev)); dev->addr.domain = domain; @@ -221,7 +224,7 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, dev->addr.function = function; /* get vendor id */ - rte_snprintf(filename, sizeof(filename), "%s/vendor", dirname); + snprintf(filename, sizeof(filename), "%s/vendor", dirname); if (eal_parse_sysfs_value(filename, &tmp) < 0) { free(dev); return -1; @@ -229,7 +232,7 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, dev->id.vendor_id = (uint16_t)tmp; /* get device id */ - rte_snprintf(filename, sizeof(filename), "%s/device", dirname); + snprintf(filename, sizeof(filename), "%s/device", dirname); if (eal_parse_sysfs_value(filename, &tmp) < 0) { free(dev); return -1; @@ -237,7 +240,7 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, dev->id.device_id = (uint16_t)tmp; /* get subsystem_vendor id */ - rte_snprintf(filename, sizeof(filename), "%s/subsystem_vendor", + snprintf(filename, sizeof(filename), "%s/subsystem_vendor", dirname); if (eal_parse_sysfs_value(filename, &tmp) < 0) { free(dev); @@ -246,7 +249,7 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, dev->id.subsystem_vendor_id = (uint16_t)tmp; /* get subsystem_device id */ - rte_snprintf(filename, sizeof(filename), "%s/subsystem_device", + snprintf(filename, sizeof(filename), "%s/subsystem_device", dirname); if (eal_parse_sysfs_value(filename, &tmp) < 0) { free(dev); @@ -256,17 +259,24 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, /* get max_vfs */ dev->max_vfs = 0; - rte_snprintf(filename, sizeof(filename), "%s/max_vfs", dirname); + snprintf(filename, sizeof(filename), "%s/max_vfs", dirname); if (!access(filename, F_OK) && - eal_parse_sysfs_value(filename, &tmp) == 0) { + eal_parse_sysfs_value(filename, &tmp) == 0) dev->max_vfs = (uint16_t)tmp; + else { + /* for non igb_uio driver, need kernel version >= 3.8 */ + snprintf(filename, sizeof(filename), + "%s/sriov_numvfs", dirname); + if (!access(filename, F_OK) && + eal_parse_sysfs_value(filename, &tmp) == 0) + dev->max_vfs = (uint16_t)tmp; } /* get numa node */ - rte_snprintf(filename, sizeof(filename), "%s/numa_node", + snprintf(filename, sizeof(filename), "%s/numa_node", dirname); if (access(filename, R_OK) != 0) { - /* if no NUMA support just set node to 0 */ + /* if no NUMA support just set node to -1 */ dev->numa_node = -1; } else { if (eal_parse_sysfs_value(filename, &tmp) < 0) { @@ -277,26 +287,55 @@ pci_scan_one(const char *dirname, uint16_t domain, uint8_t bus, } /* parse resources */ - rte_snprintf(filename, sizeof(filename), "%s/resource", dirname); + snprintf(filename, sizeof(filename), "%s/resource", dirname); if (pci_parse_sysfs_resource(filename, dev) < 0) { RTE_LOG(ERR, EAL, "%s(): cannot parse resource\n", __func__); free(dev); return -1; } + /* parse driver */ + snprintf(filename, sizeof(filename), "%s/driver", dirname); + ret = pci_get_kernel_driver_by_path(filename, driver); + if (!ret) { + if (!strcmp(driver, "vfio-pci")) + dev->kdrv = RTE_KDRV_VFIO; + else if (!strcmp(driver, "igb_uio")) + dev->kdrv = RTE_KDRV_IGB_UIO; + else if (!strcmp(driver, "uio_pci_generic")) + dev->kdrv = RTE_KDRV_UIO_GENERIC; + else + dev->kdrv = RTE_KDRV_UNKNOWN; + } else if (ret < 0) { + RTE_LOG(ERR, EAL, "Fail to get kernel driver\n"); + free(dev); + return -1; + } else + dev->kdrv = RTE_KDRV_UNKNOWN; + /* device is valid, add in list (sorted) */ if (TAILQ_EMPTY(&pci_device_list)) { TAILQ_INSERT_TAIL(&pci_device_list, dev, next); } else { struct rte_pci_device *dev2 = NULL; + int ret; TAILQ_FOREACH(dev2, &pci_device_list, next) { - if (pci_addr_comparison(&dev->addr, &dev2->addr)) + ret = rte_eal_compare_pci_addr(&dev->addr, &dev2->addr); + if (ret > 0) continue; - else { + else if (ret < 0) { TAILQ_INSERT_BEFORE(dev2, dev, next); return 0; + } else { /* already registered */ + dev2->kdrv = dev->kdrv; + dev2->max_vfs = dev->max_vfs; + memmove(dev2->mem_resource, + dev->mem_resource, + sizeof(dev->mem_resource)); + free(dev); + return 0; } } TAILQ_INSERT_TAIL(&pci_device_list, dev, next); @@ -356,8 +395,8 @@ error: * Scan the content of the PCI bus, and the devices in the devices * list */ -static int -pci_scan(void) +int +rte_eal_pci_scan(void) { struct dirent *e; DIR *dir; @@ -380,7 +419,7 @@ pci_scan(void) &bus, &devid, &function) != 0) continue; - rte_snprintf(dirname, sizeof(dirname), "%s/%s", SYSFS_PCI_DEVICES, + snprintf(dirname, sizeof(dirname), "%s/%s", SYSFS_PCI_DEVICES, e->d_name); if (pci_scan_one(dirname, domain, bus, devid, function) < 0) goto error; @@ -393,84 +432,126 @@ error: return -1; } -/* - * If vendor/device ID match, call the devinit() function of the - * driver. - */ -int -rte_eal_pci_probe_one_driver(struct rte_pci_driver *dr, struct rte_pci_device *dev) +#ifdef RTE_PCI_CONFIG +static int +pci_config_extended_tag(struct rte_pci_device *dev) { - struct rte_pci_id *id_table; + struct rte_pci_addr *loc = &dev->addr; + char filename[PATH_MAX]; + char buf[BUFSIZ]; + FILE *f; - for (id_table = dr->id_table ; id_table->vendor_id != 0; id_table++) { + /* not configured, let it as is */ + if (strncmp(RTE_PCI_EXTENDED_TAG, "on", 2) != 0 && + strncmp(RTE_PCI_EXTENDED_TAG, "off", 3) != 0) + return 0; - /* check if device's identifiers match the driver's ones */ - if (id_table->vendor_id != dev->id.vendor_id && - id_table->vendor_id != PCI_ANY_ID) - continue; - if (id_table->device_id != dev->id.device_id && - id_table->device_id != PCI_ANY_ID) - continue; - if (id_table->subsystem_vendor_id != dev->id.subsystem_vendor_id && - id_table->subsystem_vendor_id != PCI_ANY_ID) - continue; - if (id_table->subsystem_device_id != dev->id.subsystem_device_id && - id_table->subsystem_device_id != PCI_ANY_ID) - continue; + snprintf(filename, sizeof(filename), + SYSFS_PCI_DEVICES "/" PCI_PRI_FMT "/" "extended_tag", + loc->domain, loc->bus, loc->devid, loc->function); + f = fopen(filename, "rw+"); + if (!f) + return -1; - struct rte_pci_addr *loc = &dev->addr; + fgets(buf, sizeof(buf), f); + if (strncmp(RTE_PCI_EXTENDED_TAG, "on", 2) == 0) { + /* enable Extended Tag*/ + if (strncmp(buf, "on", 2) != 0) { + fseek(f, 0, SEEK_SET); + fputs("on", f); + } + } else { + /* disable Extended Tag */ + if (strncmp(buf, "off", 3) != 0) { + fseek(f, 0, SEEK_SET); + fputs("off", f); + } + } + fclose(f); - RTE_LOG(DEBUG, EAL, "PCI device "PCI_PRI_FMT" on NUMA socket %i\n", - loc->domain, loc->bus, loc->devid, loc->function, - dev->numa_node); + return 0; +} - RTE_LOG(DEBUG, EAL, " probe driver: %x:%x %s\n", dev->id.vendor_id, - dev->id.device_id, dr->name); +static int +pci_config_max_read_request_size(struct rte_pci_device *dev) +{ + struct rte_pci_addr *loc = &dev->addr; + char filename[PATH_MAX]; + char buf[BUFSIZ], param[BUFSIZ]; + FILE *f; + /* size can be 128, 256, 512, 1024, 2048, 4096 */ + uint32_t max_size = RTE_PCI_MAX_READ_REQUEST_SIZE; - /* no initialization when blacklisted, return without error */ - if (dev->devargs != NULL && - dev->devargs->type == RTE_DEVTYPE_BLACKLISTED_PCI) { - RTE_LOG(DEBUG, EAL, " Device is blacklisted, not initializing\n"); - return 0; - } + /* not configured, let it as is */ + if (!max_size) + return 0; - if (dr->drv_flags & RTE_PCI_DRV_NEED_IGB_UIO) { - /* map resources for devices that use igb_uio */ - if (pci_uio_map_resource(dev) < 0) - return -1; - } else if (dr->drv_flags & RTE_PCI_DRV_FORCE_UNBIND && - rte_eal_process_type() == RTE_PROC_PRIMARY) { - /* unbind current driver */ - if (pci_unbind_kernel_driver(dev) < 0) - return -1; - } + snprintf(filename, sizeof(filename), + SYSFS_PCI_DEVICES "/" PCI_PRI_FMT "/" "max_read_request_size", + loc->domain, loc->bus, loc->devid, loc->function); + f = fopen(filename, "rw+"); + if (!f) + return -1; - /* reference driver structure */ - dev->driver = dr; + fgets(buf, sizeof(buf), f); + snprintf(param, sizeof(param), "%d", max_size); - /* call the driver devinit() function */ - return dr->devinit(dr, dev); + /* check if the size to be set is the same as current */ + if (strcmp(buf, param) == 0) { + fclose(f); + return 0; } - /* return positive value if driver is not found */ - return 1; + fseek(f, 0, SEEK_SET); + fputs(param, f); + fclose(f); + + return 0; } +static void +pci_config_space_set(struct rte_pci_device *dev) +{ + if (rte_eal_process_type() != RTE_PROC_PRIMARY) + return; + + /* configure extended tag */ + pci_config_extended_tag(dev); + + /* configure max read request size */ + pci_config_max_read_request_size(dev); +} +#endif + /* Init the PCI EAL subsystem */ int rte_eal_pci_init(void) { TAILQ_INIT(&pci_driver_list); TAILQ_INIT(&pci_device_list); - pci_res_list = RTE_TAILQ_RESERVE_BY_IDX(RTE_TAILQ_PCI, - mapped_pci_res_list); /* for debug purposes, PCI can be disabled */ if (internal_config.no_pci) return 0; - if (pci_scan() < 0) { + if (rte_eal_pci_scan() < 0) { RTE_LOG(ERR, EAL, "%s(): Cannot scan PCI bus\n", __func__); return -1; } +#ifdef VFIO_PRESENT + pci_vfio_enable(); + + if (pci_vfio_is_enabled()) { + + /* if we are primary process, create a thread to communicate with + * secondary processes. the thread will use a socket to wait for + * requests from secondary process to send open file descriptors, + * because VFIO does not allow multiple open descriptors on a group or + * VFIO container. + */ + if (internal_config.process_type == RTE_PROC_PRIMARY && + pci_vfio_mp_sync_setup() < 0) + return -1; + } +#endif return 0; }