diff options
Diffstat (limited to 'include/rdma/ib_verbs.h')
-rw-r--r-- | include/rdma/ib_verbs.h | 318 |
1 files changed, 169 insertions, 149 deletions
diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index 5ad43a487745..0f1813c13687 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -59,7 +59,8 @@ #include <linux/if_link.h> #include <linux/atomic.h> #include <linux/mmu_notifier.h> -#include <asm/uaccess.h> +#include <linux/uaccess.h> +#include <linux/cgroup_rdma.h> extern struct workqueue_struct *ib_wq; extern struct workqueue_struct *ib_comp_wq; @@ -207,6 +208,7 @@ enum ib_device_cap_flags { IB_DEVICE_MEM_WINDOW_TYPE_2A = (1 << 23), IB_DEVICE_MEM_WINDOW_TYPE_2B = (1 << 24), IB_DEVICE_RC_IP_CSUM = (1 << 25), + /* Deprecated. Please use IB_RAW_PACKET_CAP_IP_CSUM. */ IB_DEVICE_RAW_IP_CSUM = (1 << 26), /* * Devices should set IB_DEVICE_CROSS_CHANNEL if they @@ -220,6 +222,7 @@ enum ib_device_cap_flags { IB_DEVICE_ON_DEMAND_PAGING = (1ULL << 31), IB_DEVICE_SG_GAPS_REG = (1ULL << 32), IB_DEVICE_VIRTUAL_FUNCTION = (1ULL << 33), + /* Deprecated. Please use IB_RAW_PACKET_CAP_SCATTER_FCS. */ IB_DEVICE_RAW_SCATTER_FCS = (1ULL << 34), }; @@ -241,7 +244,8 @@ enum ib_atomic_cap { }; enum ib_odp_general_cap_bits { - IB_ODP_SUPPORT = 1 << 0, + IB_ODP_SUPPORT = 1 << 0, + IB_ODP_SUPPORT_IMPLICIT = 1 << 1, }; enum ib_odp_transport_cap_bits { @@ -330,6 +334,7 @@ struct ib_device_attr { uint64_t hca_core_clock; /* in KHZ */ struct ib_rss_caps rss_caps; u32 max_wq_type_rq; + u32 raw_packet_caps; /* Use ib_raw_packet_caps enum */ }; enum ib_mtu { @@ -352,6 +357,20 @@ static inline int ib_mtu_enum_to_int(enum ib_mtu mtu) } } +static inline enum ib_mtu ib_mtu_int_to_enum(int mtu) +{ + if (mtu >= 4096) + return IB_MTU_4096; + else if (mtu >= 2048) + return IB_MTU_2048; + else if (mtu >= 1024) + return IB_MTU_1024; + else if (mtu >= 512) + return IB_MTU_512; + else + return IB_MTU_256; +} + enum ib_port_state { IB_PORT_NOP = 0, IB_PORT_DOWN = 1, @@ -485,6 +504,8 @@ static inline struct rdma_hw_stats *rdma_alloc_hw_stats_struct( #define RDMA_CORE_CAP_PROT_ROCE 0x00200000 #define RDMA_CORE_CAP_PROT_IWARP 0x00400000 #define RDMA_CORE_CAP_PROT_ROCE_UDP_ENCAP 0x00800000 +#define RDMA_CORE_CAP_PROT_RAW_PACKET 0x01000000 +#define RDMA_CORE_CAP_PROT_USNIC 0x02000000 #define RDMA_CORE_PORT_IBA_IB (RDMA_CORE_CAP_PROT_IB \ | RDMA_CORE_CAP_IB_MAD \ @@ -508,6 +529,10 @@ static inline struct rdma_hw_stats *rdma_alloc_hw_stats_struct( #define RDMA_CORE_PORT_INTEL_OPA (RDMA_CORE_PORT_IBA_IB \ | RDMA_CORE_CAP_OPA_MAD) +#define RDMA_CORE_PORT_RAW_PACKET (RDMA_CORE_CAP_PROT_RAW_PACKET) + +#define RDMA_CORE_PORT_USNIC (RDMA_CORE_CAP_PROT_USNIC) + struct ib_port_attr { u64 subnet_prefix; enum ib_port_state state; @@ -1005,6 +1030,7 @@ enum ib_qp_create_flags { IB_QP_CREATE_SIGNATURE_EN = 1 << 6, IB_QP_CREATE_USE_GFP_NOIO = 1 << 7, IB_QP_CREATE_SCATTER_FCS = 1 << 8, + IB_QP_CREATE_CVLAN_STRIPPING = 1 << 9, /* reserve bits 26-31 for low level drivers' internal use */ IB_QP_CREATE_RESERVED_START = 1 << 26, IB_QP_CREATE_RESERVED_END = 1 << 31, @@ -1102,6 +1128,7 @@ enum ib_qp_attr_mask { IB_QP_RESERVED2 = (1<<22), IB_QP_RESERVED3 = (1<<23), IB_QP_RESERVED4 = (1<<24), + IB_QP_RATE_LIMIT = (1<<25), }; enum ib_qp_state { @@ -1151,6 +1178,7 @@ struct ib_qp_attr { u8 rnr_retry; u8 alt_port_num; u8 alt_timeout; + u32 rate_limit; }; enum ib_wr_opcode { @@ -1329,6 +1357,12 @@ struct ib_fmr_attr { struct ib_umem; +struct ib_rdmacg_object { +#ifdef CONFIG_CGROUP_RDMA + struct rdma_cgroup *cg; /* owner rdma cgroup */ +#endif +}; + struct ib_ucontext { struct ib_device *device; struct list_head pd_list; @@ -1361,6 +1395,8 @@ struct ib_ucontext { struct list_head no_private_counters; int odp_mrs_count; #endif + + struct ib_rdmacg_object cg_obj; }; struct ib_uobject { @@ -1368,6 +1404,7 @@ struct ib_uobject { struct ib_ucontext *context; /* associated user context */ void *object; /* containing object */ struct list_head list; /* link to context's list */ + struct ib_rdmacg_object cg_obj; /* rdmacg object */ int id; /* index into kernel idr */ struct kref ref; struct rw_semaphore mutex; /* protects .live */ @@ -1454,6 +1491,18 @@ struct ib_srq { } ext; }; +enum ib_raw_packet_caps { + /* Strip cvlan from incoming packet and report it in the matching work + * completion is supported. + */ + IB_RAW_PACKET_CAP_CVLAN_STRIPPING = (1 << 0), + /* Scatter FCS field of an incoming packet to host memory is supported. + */ + IB_RAW_PACKET_CAP_SCATTER_FCS = (1 << 1), + /* Checksum offloads are supported (for both send and receive). */ + IB_RAW_PACKET_CAP_IP_CSUM = (1 << 2), +}; + enum ib_wq_type { IB_WQT_RQ }; @@ -1477,6 +1526,11 @@ struct ib_wq { atomic_t usecnt; }; +enum ib_wq_flags { + IB_WQ_FLAGS_CVLAN_STRIPPING = 1 << 0, + IB_WQ_FLAGS_SCATTER_FCS = 1 << 1, +}; + struct ib_wq_init_attr { void *wq_context; enum ib_wq_type wq_type; @@ -1484,16 +1538,20 @@ struct ib_wq_init_attr { u32 max_sge; struct ib_cq *cq; void (*event_handler)(struct ib_event *, void *); + u32 create_flags; /* Use enum ib_wq_flags */ }; enum ib_wq_attr_mask { - IB_WQ_STATE = 1 << 0, - IB_WQ_CUR_STATE = 1 << 1, + IB_WQ_STATE = 1 << 0, + IB_WQ_CUR_STATE = 1 << 1, + IB_WQ_FLAGS = 1 << 2, }; struct ib_wq_attr { enum ib_wq_state wq_state; enum ib_wq_state curr_wq_state; + u32 flags; /* Use enum ib_wq_flags */ + u32 flags_mask; /* Use enum ib_wq_flags */ }; struct ib_rwq_ind_table { @@ -1592,17 +1650,21 @@ enum ib_flow_attr_type { /* Supported steering header types */ enum ib_flow_spec_type { /* L2 headers*/ - IB_FLOW_SPEC_ETH = 0x20, - IB_FLOW_SPEC_IB = 0x22, + IB_FLOW_SPEC_ETH = 0x20, + IB_FLOW_SPEC_IB = 0x22, /* L3 header*/ - IB_FLOW_SPEC_IPV4 = 0x30, - IB_FLOW_SPEC_IPV6 = 0x31, + IB_FLOW_SPEC_IPV4 = 0x30, + IB_FLOW_SPEC_IPV6 = 0x31, /* L4 headers*/ - IB_FLOW_SPEC_TCP = 0x40, - IB_FLOW_SPEC_UDP = 0x41 + IB_FLOW_SPEC_TCP = 0x40, + IB_FLOW_SPEC_UDP = 0x41, + IB_FLOW_SPEC_VXLAN_TUNNEL = 0x50, + IB_FLOW_SPEC_INNER = 0x100, + /* Actions */ + IB_FLOW_SPEC_ACTION_TAG = 0x1000, }; #define IB_FLOW_SPEC_LAYER_MASK 0xF0 -#define IB_FLOW_SPEC_SUPPORT_LAYERS 4 +#define IB_FLOW_SPEC_SUPPORT_LAYERS 8 /* Flow steering rule priority is set according to it's domain. * Lower domain value means higher priority. @@ -1630,7 +1692,7 @@ struct ib_flow_eth_filter { }; struct ib_flow_spec_eth { - enum ib_flow_spec_type type; + u32 type; u16 size; struct ib_flow_eth_filter val; struct ib_flow_eth_filter mask; @@ -1644,7 +1706,7 @@ struct ib_flow_ib_filter { }; struct ib_flow_spec_ib { - enum ib_flow_spec_type type; + u32 type; u16 size; struct ib_flow_ib_filter val; struct ib_flow_ib_filter mask; @@ -1669,7 +1731,7 @@ struct ib_flow_ipv4_filter { }; struct ib_flow_spec_ipv4 { - enum ib_flow_spec_type type; + u32 type; u16 size; struct ib_flow_ipv4_filter val; struct ib_flow_ipv4_filter mask; @@ -1687,7 +1749,7 @@ struct ib_flow_ipv6_filter { }; struct ib_flow_spec_ipv6 { - enum ib_flow_spec_type type; + u32 type; u16 size; struct ib_flow_ipv6_filter val; struct ib_flow_ipv6_filter mask; @@ -1701,15 +1763,36 @@ struct ib_flow_tcp_udp_filter { }; struct ib_flow_spec_tcp_udp { - enum ib_flow_spec_type type; + u32 type; u16 size; struct ib_flow_tcp_udp_filter val; struct ib_flow_tcp_udp_filter mask; }; +struct ib_flow_tunnel_filter { + __be32 tunnel_id; + u8 real_sz[0]; +}; + +/* ib_flow_spec_tunnel describes the Vxlan tunnel + * the tunnel_id from val has the vni value + */ +struct ib_flow_spec_tunnel { + u32 type; + u16 size; + struct ib_flow_tunnel_filter val; + struct ib_flow_tunnel_filter mask; +}; + +struct ib_flow_spec_action_tag { + enum ib_flow_spec_type type; + u16 size; + u32 tag_id; +}; + union ib_flow_spec { struct { - enum ib_flow_spec_type type; + u32 type; u16 size; }; struct ib_flow_spec_eth eth; @@ -1717,6 +1800,8 @@ union ib_flow_spec { struct ib_flow_spec_ipv4 ipv4; struct ib_flow_spec_tcp_udp tcp_udp; struct ib_flow_spec_ipv6 ipv6; + struct ib_flow_spec_tunnel tunnel; + struct ib_flow_spec_action_tag flow_tag; }; struct ib_flow_attr { @@ -1755,59 +1840,17 @@ enum ib_mad_result { #define IB_DEVICE_NAME_MAX 64 +struct ib_port_cache { + struct ib_pkey_cache *pkey; + struct ib_gid_table *gid; + u8 lmc; + enum ib_port_state port_state; +}; + struct ib_cache { rwlock_t lock; struct ib_event_handler event_handler; - struct ib_pkey_cache **pkey_cache; - struct ib_gid_table **gid_cache; - u8 *lmc_cache; -}; - -struct ib_dma_mapping_ops { - int (*mapping_error)(struct ib_device *dev, - u64 dma_addr); - u64 (*map_single)(struct ib_device *dev, - void *ptr, size_t size, - enum dma_data_direction direction); - void (*unmap_single)(struct ib_device *dev, - u64 addr, size_t size, - enum dma_data_direction direction); - u64 (*map_page)(struct ib_device *dev, - struct page *page, unsigned long offset, - size_t size, - enum dma_data_direction direction); - void (*unmap_page)(struct ib_device *dev, - u64 addr, size_t size, - enum dma_data_direction direction); - int (*map_sg)(struct ib_device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction direction); - void (*unmap_sg)(struct ib_device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction direction); - int (*map_sg_attrs)(struct ib_device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction direction, - unsigned long attrs); - void (*unmap_sg_attrs)(struct ib_device *dev, - struct scatterlist *sg, int nents, - enum dma_data_direction direction, - unsigned long attrs); - void (*sync_single_for_cpu)(struct ib_device *dev, - u64 dma_handle, - size_t size, - enum dma_data_direction dir); - void (*sync_single_for_device)(struct ib_device *dev, - u64 dma_handle, - size_t size, - enum dma_data_direction dir); - void *(*alloc_coherent)(struct ib_device *dev, - size_t size, - u64 *dma_handle, - gfp_t flag); - void (*free_coherent)(struct ib_device *dev, - size_t size, void *cpu_addr, - u64 dma_handle); + struct ib_port_cache *ports; }; struct iw_cm_verbs; @@ -1820,8 +1863,6 @@ struct ib_port_immutable { }; struct ib_device { - struct device *dma_device; - char name[IB_DEVICE_NAME_MAX]; struct list_head event_handler_list; @@ -1933,7 +1974,8 @@ struct ib_device { struct ib_udata *udata); int (*dealloc_pd)(struct ib_pd *pd); struct ib_ah * (*create_ah)(struct ib_pd *pd, - struct ib_ah_attr *ah_attr); + struct ib_ah_attr *ah_attr, + struct ib_udata *udata); int (*modify_ah)(struct ib_ah *ah, struct ib_ah_attr *ah_attr); int (*query_ah)(struct ib_ah *ah, @@ -2070,7 +2112,6 @@ struct ib_device { struct ib_rwq_ind_table_init_attr *init_attr, struct ib_udata *udata); int (*destroy_rwq_ind_table)(struct ib_rwq_ind_table *wq_ind_table); - struct ib_dma_mapping_ops *dma_ops; struct module *owner; struct device dev; @@ -2097,6 +2138,10 @@ struct ib_device { struct attribute_group *hw_stats_ag; struct rdma_hw_stats *hw_stats; +#ifdef CONFIG_CGROUP_RDMA + struct rdmacg_device cg_device; +#endif + /** * The following mandatory functions are used only at device * registration. Keep functions such as these at the end of this @@ -2254,6 +2299,13 @@ static inline u8 rdma_end_port(const struct ib_device *device) return rdma_cap_ib_switch(device) ? 0 : device->phys_port_cnt; } +static inline int rdma_is_port_valid(const struct ib_device *device, + unsigned int port) +{ + return (port >= rdma_start_port(device) && + port <= rdma_end_port(device)); +} + static inline bool rdma_protocol_ib(const struct ib_device *device, u8 port_num) { return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_IB; @@ -2286,6 +2338,16 @@ static inline bool rdma_ib_or_roce(const struct ib_device *device, u8 port_num) rdma_protocol_roce(device, port_num); } +static inline bool rdma_protocol_raw_packet(const struct ib_device *device, u8 port_num) +{ + return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_RAW_PACKET; +} + +static inline bool rdma_protocol_usnic(const struct ib_device *device, u8 port_num) +{ + return device->port_immutable[port_num].core_cap_flags & RDMA_CORE_CAP_PROT_USNIC; +} + /** * rdma_cap_ib_mad - Check if the port of a device supports Infiniband * Management Datagrams. @@ -2581,6 +2643,24 @@ void ib_dealloc_pd(struct ib_pd *pd); struct ib_ah *ib_create_ah(struct ib_pd *pd, struct ib_ah_attr *ah_attr); /** + * ib_get_gids_from_rdma_hdr - Get sgid and dgid from GRH or IPv4 header + * work completion. + * @hdr: the L3 header to parse + * @net_type: type of header to parse + * @sgid: place to store source gid + * @dgid: place to store destination gid + */ +int ib_get_gids_from_rdma_hdr(const union rdma_network_hdr *hdr, + enum rdma_network_type net_type, + union ib_gid *sgid, union ib_gid *dgid); + +/** + * ib_get_rdma_header_version - Get the header version + * @hdr: the L3 header to parse + */ +int ib_get_rdma_header_version(const union rdma_network_hdr *hdr); + +/** * ib_init_ah_from_wc - Initializes address handle attributes from a * work completion. * @device: Device on which the received message arrived. @@ -2927,9 +3007,7 @@ static inline int ib_req_ncomp_notif(struct ib_cq *cq, int wc_cnt) */ static inline int ib_dma_mapping_error(struct ib_device *dev, u64 dma_addr) { - if (dev->dma_ops) - return dev->dma_ops->mapping_error(dev, dma_addr); - return dma_mapping_error(dev->dma_device, dma_addr); + return dma_mapping_error(&dev->dev, dma_addr); } /** @@ -2943,9 +3021,7 @@ static inline u64 ib_dma_map_single(struct ib_device *dev, void *cpu_addr, size_t size, enum dma_data_direction direction) { - if (dev->dma_ops) - return dev->dma_ops->map_single(dev, cpu_addr, size, direction); - return dma_map_single(dev->dma_device, cpu_addr, size, direction); + return dma_map_single(&dev->dev, cpu_addr, size, direction); } /** @@ -2959,28 +3035,7 @@ static inline void ib_dma_unmap_single(struct ib_device *dev, u64 addr, size_t size, enum dma_data_direction direction) { - if (dev->dma_ops) - dev->dma_ops->unmap_single(dev, addr, size, direction); - else - dma_unmap_single(dev->dma_device, addr, size, direction); -} - -static inline u64 ib_dma_map_single_attrs(struct ib_device *dev, - void *cpu_addr, size_t size, - enum dma_data_direction direction, - unsigned long dma_attrs) -{ - return dma_map_single_attrs(dev->dma_device, cpu_addr, size, - direction, dma_attrs); -} - -static inline void ib_dma_unmap_single_attrs(struct ib_device *dev, - u64 addr, size_t size, - enum dma_data_direction direction, - unsigned long dma_attrs) -{ - return dma_unmap_single_attrs(dev->dma_device, addr, size, - direction, dma_attrs); + dma_unmap_single(&dev->dev, addr, size, direction); } /** @@ -2997,9 +3052,7 @@ static inline u64 ib_dma_map_page(struct ib_device *dev, size_t size, enum dma_data_direction direction) { - if (dev->dma_ops) - return dev->dma_ops->map_page(dev, page, offset, size, direction); - return dma_map_page(dev->dma_device, page, offset, size, direction); + return dma_map_page(&dev->dev, page, offset, size, direction); } /** @@ -3013,10 +3066,7 @@ static inline void ib_dma_unmap_page(struct ib_device *dev, u64 addr, size_t size, enum dma_data_direction direction) { - if (dev->dma_ops) - dev->dma_ops->unmap_page(dev, addr, size, direction); - else - dma_unmap_page(dev->dma_device, addr, size, direction); + dma_unmap_page(&dev->dev, addr, size, direction); } /** @@ -3030,9 +3080,7 @@ static inline int ib_dma_map_sg(struct ib_device *dev, struct scatterlist *sg, int nents, enum dma_data_direction direction) { - if (dev->dma_ops) - return dev->dma_ops->map_sg(dev, sg, nents, direction); - return dma_map_sg(dev->dma_device, sg, nents, direction); + return dma_map_sg(&dev->dev, sg, nents, direction); } /** @@ -3046,10 +3094,7 @@ static inline void ib_dma_unmap_sg(struct ib_device *dev, struct scatterlist *sg, int nents, enum dma_data_direction direction) { - if (dev->dma_ops) - dev->dma_ops->unmap_sg(dev, sg, nents, direction); - else - dma_unmap_sg(dev->dma_device, sg, nents, direction); + dma_unmap_sg(&dev->dev, sg, nents, direction); } static inline int ib_dma_map_sg_attrs(struct ib_device *dev, @@ -3057,12 +3102,7 @@ static inline int ib_dma_map_sg_attrs(struct ib_device *dev, enum dma_data_direction direction, unsigned long dma_attrs) { - if (dev->dma_ops) - return dev->dma_ops->map_sg_attrs(dev, sg, nents, direction, - dma_attrs); - else - return dma_map_sg_attrs(dev->dma_device, sg, nents, direction, - dma_attrs); + return dma_map_sg_attrs(&dev->dev, sg, nents, direction, dma_attrs); } static inline void ib_dma_unmap_sg_attrs(struct ib_device *dev, @@ -3070,12 +3110,7 @@ static inline void ib_dma_unmap_sg_attrs(struct ib_device *dev, enum dma_data_direction direction, unsigned long dma_attrs) { - if (dev->dma_ops) - return dev->dma_ops->unmap_sg_attrs(dev, sg, nents, direction, - dma_attrs); - else - dma_unmap_sg_attrs(dev->dma_device, sg, nents, direction, - dma_attrs); + dma_unmap_sg_attrs(&dev->dev, sg, nents, direction, dma_attrs); } /** * ib_sg_dma_address - Return the DMA address from a scatter/gather entry @@ -3117,10 +3152,7 @@ static inline void ib_dma_sync_single_for_cpu(struct ib_device *dev, size_t size, enum dma_data_direction dir) { - if (dev->dma_ops) - dev->dma_ops->sync_single_for_cpu(dev, addr, size, dir); - else - dma_sync_single_for_cpu(dev->dma_device, addr, size, dir); + dma_sync_single_for_cpu(&dev->dev, addr, size, dir); } /** @@ -3135,10 +3167,7 @@ static inline void ib_dma_sync_single_for_device(struct ib_device *dev, size_t size, enum dma_data_direction dir) { - if (dev->dma_ops) - dev->dma_ops->sync_single_for_device(dev, addr, size, dir); - else - dma_sync_single_for_device(dev->dma_device, addr, size, dir); + dma_sync_single_for_device(&dev->dev, addr, size, dir); } /** @@ -3150,19 +3179,10 @@ static inline void ib_dma_sync_single_for_device(struct ib_device *dev, */ static inline void *ib_dma_alloc_coherent(struct ib_device *dev, size_t size, - u64 *dma_handle, + dma_addr_t *dma_handle, gfp_t flag) { - if (dev->dma_ops) - return dev->dma_ops->alloc_coherent(dev, size, dma_handle, flag); - else { - dma_addr_t handle; - void *ret; - - ret = dma_alloc_coherent(dev->dma_device, size, &handle, flag); - *dma_handle = handle; - return ret; - } + return dma_alloc_coherent(&dev->dev, size, dma_handle, flag); } /** @@ -3174,12 +3194,9 @@ static inline void *ib_dma_alloc_coherent(struct ib_device *dev, */ static inline void ib_dma_free_coherent(struct ib_device *dev, size_t size, void *cpu_addr, - u64 dma_handle) + dma_addr_t dma_handle) { - if (dev->dma_ops) - dev->dma_ops->free_coherent(dev, size, cpu_addr, dma_handle); - else - dma_free_coherent(dev->dma_device, size, cpu_addr, dma_handle); + dma_free_coherent(&dev->dev, size, cpu_addr, dma_handle); } /** @@ -3357,4 +3374,7 @@ int ib_sg_to_pages(struct ib_mr *mr, struct scatterlist *sgl, int sg_nents, void ib_drain_rq(struct ib_qp *qp); void ib_drain_sq(struct ib_qp *qp); void ib_drain_qp(struct ib_qp *qp); + +int ib_resolve_eth_dmac(struct ib_device *device, + struct ib_ah_attr *ah_attr); #endif /* IB_VERBS_H */ |