From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by dpdk.org (Postfix) with ESMTP id A495D95D8 for ; Wed, 27 Jan 2016 17:50:41 +0100 (CET) Received: from fmsmga004.fm.intel.com ([10.253.24.48]) by orsmga102.jf.intel.com with ESMTP; 27 Jan 2016 08:50:40 -0800 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.22,355,1449561600"; d="scan'208";a="37156829" Received: from irvmail001.ir.intel.com ([163.33.26.43]) by fmsmga004.fm.intel.com with ESMTP; 27 Jan 2016 08:50:39 -0800 Received: from sivswdev01.ir.intel.com (sivswdev01.ir.intel.com [10.237.217.45]) by irvmail001.ir.intel.com (8.14.3/8.13.6/MailSET/Hub) with ESMTP id u0RGodcP016098; Wed, 27 Jan 2016 16:50:39 GMT Received: from sivswdev01.ir.intel.com (localhost [127.0.0.1]) by sivswdev01.ir.intel.com with ESMTP id u0RGocPw023034; Wed, 27 Jan 2016 16:50:38 GMT Received: (from aburakov@localhost) by sivswdev01.ir.intel.com with id u0RGocPp023030; Wed, 27 Jan 2016 16:50:38 GMT From: Anatoly Burakov To: dev@dpdk.org Date: Wed, 27 Jan 2016 16:50:38 +0000 Message-Id: <1453913438-22991-1-git-send-email-anatoly.burakov@intel.com> X-Mailer: git-send-email 1.7.4.1 In-Reply-To: <1453905124-28544-1-git-send-email-anatoly.burakov@intel.com> References: <1453905124-28544-1-git-send-email-anatoly.burakov@intel.com> Subject: [dpdk-dev] [PATCH v5] vfio: Support for no-IOMMU mode X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: patches and discussions about DPDK List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 27 Jan 2016 16:50:42 -0000 This commit is adding a generic mechanism to support multiple IOMMU types. For now, it's only type 1 (x86 IOMMU) and no-IOMMU (a special VFIO mode that doesn't use IOMMU at all), but it's easily extended by adding necessary DMA mapping functions to eal_pci_vfio.c. Since type 1 IOMMU module is no longer necessary to have VFIO, we fix the module check to check for vfio-pci instead. It's not ideal and triggers VFIO checks more often (and thus produces more error output, which was the reason behind the module check in the first place), so we compensate for that by providing more verbose logging, indicating whether VFIO initialization has succeeded or failed. Signed-off-by: Anatoly Burakov Signed-off-by: Santosh Shukla Tested-by: Santosh Shukla --- v5 changes: Renamed functions v4 changes: Fixed the commit message and added a missing sign-off v3 changes: Merging DMA mapping functions back into eal_pci_vfio.c Fixing and adding comments v2 changes: Compile fix (hat-tip to Santosh Shukla) Tested-by is provisional, since only superficial testing was done lib/librte_eal/linuxapp/eal/eal_pci_vfio.c | 205 +++++++++++++++++++++-------- lib/librte_eal/linuxapp/eal/eal_vfio.h | 5 + 2 files changed, 157 insertions(+), 53 deletions(-) diff --git a/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c b/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c index 74f91ba..b87819d 100644 --- a/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c +++ b/lib/librte_eal/linuxapp/eal/eal_pci_vfio.c @@ -72,11 +72,74 @@ EAL_REGISTER_TAILQ(rte_vfio_tailq) #define VFIO_DIR "/dev/vfio" #define VFIO_CONTAINER_PATH "/dev/vfio/vfio" #define VFIO_GROUP_FMT "/dev/vfio/%u" +#define VFIO_NOIOMMU_GROUP_FMT "/dev/vfio/noiommu-%u" #define VFIO_GET_REGION_ADDR(x) ((uint64_t) x << 40ULL) /* per-process VFIO config */ static struct vfio_config vfio_cfg; +/* DMA mapping function prototype. + * Takes VFIO container fd as a parameter. + * Returns 0 on success, -1 on error. + * */ +typedef int (*vfio_dma_func_t)(int); + +struct vfio_iommu_type { + int type_id; + const char *name; + vfio_dma_func_t dma_map_func; +}; + +int vfio_type1_dma_map(int); +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 */ + { VFIO_TYPE1_IOMMU, "Type 1", &vfio_type1_dma_map}, + /* IOMMU-less mode */ + { VFIO_NOIOMMU_IOMMU, "No-IOMMU", &vfio_noiommu_dma_map}, +}; + +int +vfio_type1_dma_map(int vfio_container_fd) +{ + const struct rte_memseg *ms = rte_eal_get_physmem_layout(); + int i, ret; + + /* map all DPDK segments for DMA. use 1:1 PA to IOVA mapping */ + for (i = 0; i < RTE_MAX_MEMSEG; i++) { + struct vfio_iommu_type1_dma_map dma_map; + + if (ms[i].addr == NULL) + break; + + memset(&dma_map, 0, sizeof(dma_map)); + dma_map.argsz = sizeof(struct vfio_iommu_type1_dma_map); + dma_map.vaddr = ms[i].addr_64; + dma_map.size = ms[i].len; + dma_map.iova = ms[i].phys_addr; + dma_map.flags = VFIO_DMA_MAP_FLAG_READ | VFIO_DMA_MAP_FLAG_WRITE; + + ret = ioctl(vfio_container_fd, VFIO_IOMMU_MAP_DMA, &dma_map); + + if (ret) { + RTE_LOG(ERR, EAL, " cannot set up DMA remapping, " + "error %i (%s)\n", errno, strerror(errno)); + return -1; + } + } + + return 0; +} + +int +vfio_noiommu_dma_map(int __rte_unused vfio_container_fd) +{ + /* No-IOMMU mode does not need DMA mapping */ + return 0; +} + int pci_vfio_read_config(const struct rte_intr_handle *intr_handle, void *buf, size_t len, off_t offs) @@ -208,42 +271,58 @@ pci_vfio_set_bus_master(int dev_fd) return 0; } -/* set up DMA mappings */ -static int -pci_vfio_setup_dma_maps(int vfio_container_fd) -{ - const struct rte_memseg *ms = rte_eal_get_physmem_layout(); - int i, ret; - - ret = ioctl(vfio_container_fd, VFIO_SET_IOMMU, - VFIO_TYPE1_IOMMU); - if (ret) { - RTE_LOG(ERR, EAL, " cannot set IOMMU type, " - "error %i (%s)\n", errno, strerror(errno)); - return -1; +/* pick IOMMU type. returns a pointer to vfio_iommu_type or NULL for error */ +static const struct vfio_iommu_type * +pci_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; +} - /* map all DPDK segments for DMA. use 1:1 PA to IOVA mapping */ - for (i = 0; i < RTE_MAX_MEMSEG; i++) { - struct vfio_iommu_type1_dma_map dma_map; - - if (ms[i].addr == NULL) - break; - - memset(&dma_map, 0, sizeof(dma_map)); - dma_map.argsz = sizeof(struct vfio_iommu_type1_dma_map); - dma_map.vaddr = ms[i].addr_64; - dma_map.size = ms[i].len; - dma_map.iova = ms[i].phys_addr; - dma_map.flags = VFIO_DMA_MAP_FLAG_READ | VFIO_DMA_MAP_FLAG_WRITE; - - ret = ioctl(vfio_container_fd, VFIO_IOMMU_MAP_DMA, &dma_map); +/* check if we have any supported extensions */ +static int +pci_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]; - if (ret) { - RTE_LOG(ERR, EAL, " cannot set up DMA remapping, " - "error %i (%s)\n", errno, strerror(errno)); + 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; @@ -372,17 +451,10 @@ pci_vfio_get_container_fd(void) return -1; } - /* check if we support IOMMU type 1 */ - ret = ioctl(vfio_container_fd, VFIO_CHECK_EXTENSION, VFIO_TYPE1_IOMMU); - if (ret != 1) { - if (ret < 0) - RTE_LOG(ERR, EAL, " could not get IOMMU type, " - "error %i (%s)\n", errno, - strerror(errno)); - else - RTE_LOG(ERR, EAL, " unsupported IOMMU type " - "detected in VFIO\n"); - close(vfio_container_fd); + ret = pci_vfio_has_supported_extensions(vfio_container_fd); + if (ret) { + RTE_LOG(ERR, EAL, " no supported IOMMU " + "extensions found!\n"); return -1; } @@ -432,6 +504,7 @@ pci_vfio_get_group_fd(int iommu_group_no) /* 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); @@ -442,7 +515,20 @@ pci_vfio_get_group_fd(int iommu_group_no) strerror(errno)); return -1; } - return 0; + + /* 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 */ @@ -660,14 +746,21 @@ pci_vfio_map_resource(struct rte_pci_device *dev) } /* - * set up DMA mappings for container + * 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) { - ret = pci_vfio_setup_dma_maps(vfio_cfg.vfio_container_fd); + /* select an IOMMU type which we will be using */ + const struct vfio_iommu_type *t = + pci_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)); @@ -887,35 +980,41 @@ pci_vfio_enable(void) { /* initialize group list */ int i; - int module_vfio_type1; + 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; } - module_vfio_type1 = rte_eal_check_module("vfio_iommu_type1"); + /* 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 (module_vfio_type1 == -1) { + 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 (module_vfio_type1 == 0) { - RTE_LOG(INFO, EAL, "VFIO modules not all loaded, " - "skip VFIO support...\n"); + if (vfio_available == 0) { + RTE_LOG(INFO, EAL, "VFIO modules not loaded, " + "skipping VFIO support...\n"); return 0; } vfio_cfg.vfio_container_fd = pci_vfio_get_container_fd(); /* check if we have VFIO driver enabled */ - if (vfio_cfg.vfio_container_fd != -1) + if (vfio_cfg.vfio_container_fd != -1) { + RTE_LOG(NOTICE, EAL, "VFIO support initialized\n"); vfio_cfg.vfio_enabled = 1; - else + } else { RTE_LOG(NOTICE, EAL, "VFIO support could not be initialized\n"); + } return 0; } diff --git a/lib/librte_eal/linuxapp/eal/eal_vfio.h b/lib/librte_eal/linuxapp/eal/eal_vfio.h index 72ec3f6..638ee31 100644 --- a/lib/librte_eal/linuxapp/eal/eal_vfio.h +++ b/lib/librte_eal/linuxapp/eal/eal_vfio.h @@ -52,6 +52,11 @@ #define RTE_PCI_MSIX_FLAGS_QSIZE PCI_MSIX_FLAGS_QSIZE #endif +/* older kernels may not have no-IOMMU mode */ +#ifndef VFIO_NOIOMMU_IOMMU +#define VFIO_NOIOMMU_IOMMU 8 +#endif + #define VFIO_PRESENT #endif /* kernel version */ #endif /* RTE_EAL_VFIO */ -- 2.5.0