diff options
Diffstat (limited to 'include/rdma')
| -rw-r--r-- | include/rdma/ib.h | 2 | ||||
| -rw-r--r-- | include/rdma/ib_umem.h | 2 | ||||
| -rw-r--r-- | include/rdma/ib_umem_odp.h | 58 | ||||
| -rw-r--r-- | include/rdma/ib_verbs.h | 85 | ||||
| -rw-r--r-- | include/rdma/iw_portmap.h | 3 | ||||
| -rw-r--r-- | include/rdma/opa_port_info.h | 2 | ||||
| -rw-r--r-- | include/rdma/rdma_netlink.h | 10 | ||||
| -rw-r--r-- | include/rdma/rdma_vt.h | 1 | ||||
| -rw-r--r-- | include/rdma/rdmavt_cq.h | 1 | ||||
| -rw-r--r-- | include/rdma/rdmavt_qp.h | 44 | ||||
| -rw-r--r-- | include/rdma/restrack.h | 3 | ||||
| -rw-r--r-- | include/rdma/signature.h | 2 |
12 files changed, 168 insertions, 45 deletions
diff --git a/include/rdma/ib.h b/include/rdma/ib.h index 4f385ec54f80..fe2fc9e91588 100644 --- a/include/rdma/ib.h +++ b/include/rdma/ib.h @@ -36,6 +36,8 @@ #include <linux/types.h> #include <linux/sched.h> #include <linux/cred.h> +#include <linux/uaccess.h> +#include <linux/fs.h> struct ib_addr { union { diff --git a/include/rdma/ib_umem.h b/include/rdma/ib_umem.h index 1052d0d62be7..a91b2af64ec4 100644 --- a/include/rdma/ib_umem.h +++ b/include/rdma/ib_umem.h @@ -42,7 +42,7 @@ struct ib_ucontext; struct ib_umem_odp; struct ib_umem { - struct ib_ucontext *context; + struct ib_device *ibdev; struct mm_struct *owning_mm; size_t length; unsigned long address; diff --git a/include/rdma/ib_umem_odp.h b/include/rdma/ib_umem_odp.h index 479db5c98ff6..253df1a1fa54 100644 --- a/include/rdma/ib_umem_odp.h +++ b/include/rdma/ib_umem_odp.h @@ -37,11 +37,6 @@ #include <rdma/ib_verbs.h> #include <linux/interval_tree.h> -struct umem_odp_node { - u64 __subtree_last; - struct rb_node rb; -}; - struct ib_umem_odp { struct ib_umem umem; struct ib_ucontext_per_mm *per_mm; @@ -72,7 +67,15 @@ struct ib_umem_odp { int npages; /* Tree tracking */ - struct umem_odp_node interval_tree; + struct interval_tree_node interval_tree; + + /* + * An implicit odp umem cannot be DMA mapped, has 0 length, and serves + * only as an anchor for the driver to hold onto the per_mm. FIXME: + * This should be removed and drivers should work with the per_mm + * directly. + */ + bool is_implicit_odp; struct completion notifier_completion; int dying; @@ -88,14 +91,13 @@ static inline struct ib_umem_odp *to_ib_umem_odp(struct ib_umem *umem) /* Returns the first page of an ODP umem. */ static inline unsigned long ib_umem_start(struct ib_umem_odp *umem_odp) { - return ALIGN_DOWN(umem_odp->umem.address, 1UL << umem_odp->page_shift); + return umem_odp->interval_tree.start; } /* Returns the address of the page after the last one of an ODP umem. */ static inline unsigned long ib_umem_end(struct ib_umem_odp *umem_odp) { - return ALIGN(umem_odp->umem.address + umem_odp->umem.length, - 1UL << umem_odp->page_shift); + return umem_odp->interval_tree.last + 1; } static inline size_t ib_umem_odp_num_pages(struct ib_umem_odp *umem_odp) @@ -120,25 +122,20 @@ static inline size_t ib_umem_odp_num_pages(struct ib_umem_odp *umem_odp) #ifdef CONFIG_INFINIBAND_ON_DEMAND_PAGING struct ib_ucontext_per_mm { - struct ib_ucontext *context; - struct mm_struct *mm; + struct mmu_notifier mn; struct pid *tgid; - bool active; struct rb_root_cached umem_tree; /* Protects umem_tree */ struct rw_semaphore umem_rwsem; - - struct mmu_notifier mn; - unsigned int odp_mrs_count; - - struct list_head ucontext_list; - struct rcu_head rcu; }; -int ib_umem_odp_get(struct ib_umem_odp *umem_odp, int access); -struct ib_umem_odp *ib_alloc_odp_umem(struct ib_umem_odp *root_umem, - unsigned long addr, size_t size); +struct ib_umem_odp *ib_umem_odp_get(struct ib_udata *udata, unsigned long addr, + size_t size, int access); +struct ib_umem_odp *ib_umem_odp_alloc_implicit(struct ib_udata *udata, + int access); +struct ib_umem_odp *ib_umem_odp_alloc_child(struct ib_umem_odp *root_umem, + unsigned long addr, size_t size); void ib_umem_odp_release(struct ib_umem_odp *umem_odp); int ib_umem_odp_map_dma_pages(struct ib_umem_odp *umem_odp, u64 start_offset, @@ -163,8 +160,17 @@ int rbt_ib_umem_for_each_in_range(struct rb_root_cached *root, * Find first region intersecting with address range. * Return NULL if not found */ -struct ib_umem_odp *rbt_ib_umem_lookup(struct rb_root_cached *root, - u64 addr, u64 length); +static inline struct ib_umem_odp * +rbt_ib_umem_lookup(struct rb_root_cached *root, u64 addr, u64 length) +{ + struct interval_tree_node *node; + + node = interval_tree_iter_first(root, addr, addr + length - 1); + if (!node) + return NULL; + return container_of(node, struct ib_umem_odp, interval_tree); + +} static inline int ib_umem_mmu_notifier_retry(struct ib_umem_odp *umem_odp, unsigned long mmu_seq) @@ -185,9 +191,11 @@ static inline int ib_umem_mmu_notifier_retry(struct ib_umem_odp *umem_odp, #else /* CONFIG_INFINIBAND_ON_DEMAND_PAGING */ -static inline int ib_umem_odp_get(struct ib_umem_odp *umem_odp, int access) +static inline struct ib_umem_odp *ib_umem_odp_get(struct ib_udata *udata, + unsigned long addr, + size_t size, int access) { - return -EINVAL; + return ERR_PTR(-EINVAL); } static inline void ib_umem_odp_release(struct ib_umem_odp *umem_odp) {} diff --git a/include/rdma/ib_verbs.h b/include/rdma/ib_verbs.h index c5f8a9f17063..6a47ba85c54c 100644 --- a/include/rdma/ib_verbs.h +++ b/include/rdma/ib_verbs.h @@ -98,15 +98,54 @@ void ibdev_info(const struct ib_device *ibdev, const char *format, ...); #if defined(CONFIG_DYNAMIC_DEBUG) #define ibdev_dbg(__dev, format, args...) \ dynamic_ibdev_dbg(__dev, format, ##args) -#elif defined(DEBUG) -#define ibdev_dbg(__dev, format, args...) \ - ibdev_printk(KERN_DEBUG, __dev, format, ##args) #else __printf(2, 3) __cold static inline void ibdev_dbg(const struct ib_device *ibdev, const char *format, ...) {} #endif +#define ibdev_level_ratelimited(ibdev_level, ibdev, fmt, ...) \ +do { \ + static DEFINE_RATELIMIT_STATE(_rs, \ + DEFAULT_RATELIMIT_INTERVAL, \ + DEFAULT_RATELIMIT_BURST); \ + if (__ratelimit(&_rs)) \ + ibdev_level(ibdev, fmt, ##__VA_ARGS__); \ +} while (0) + +#define ibdev_emerg_ratelimited(ibdev, fmt, ...) \ + ibdev_level_ratelimited(ibdev_emerg, ibdev, fmt, ##__VA_ARGS__) +#define ibdev_alert_ratelimited(ibdev, fmt, ...) \ + ibdev_level_ratelimited(ibdev_alert, ibdev, fmt, ##__VA_ARGS__) +#define ibdev_crit_ratelimited(ibdev, fmt, ...) \ + ibdev_level_ratelimited(ibdev_crit, ibdev, fmt, ##__VA_ARGS__) +#define ibdev_err_ratelimited(ibdev, fmt, ...) \ + ibdev_level_ratelimited(ibdev_err, ibdev, fmt, ##__VA_ARGS__) +#define ibdev_warn_ratelimited(ibdev, fmt, ...) \ + ibdev_level_ratelimited(ibdev_warn, ibdev, fmt, ##__VA_ARGS__) +#define ibdev_notice_ratelimited(ibdev, fmt, ...) \ + ibdev_level_ratelimited(ibdev_notice, ibdev, fmt, ##__VA_ARGS__) +#define ibdev_info_ratelimited(ibdev, fmt, ...) \ + ibdev_level_ratelimited(ibdev_info, ibdev, fmt, ##__VA_ARGS__) + +#if defined(CONFIG_DYNAMIC_DEBUG) +/* descriptor check is first to prevent flooding with "callbacks suppressed" */ +#define ibdev_dbg_ratelimited(ibdev, fmt, ...) \ +do { \ + static DEFINE_RATELIMIT_STATE(_rs, \ + DEFAULT_RATELIMIT_INTERVAL, \ + DEFAULT_RATELIMIT_BURST); \ + DEFINE_DYNAMIC_DEBUG_METADATA(descriptor, fmt); \ + if (DYNAMIC_DEBUG_BRANCH(descriptor) && __ratelimit(&_rs)) \ + __dynamic_ibdev_dbg(&descriptor, ibdev, fmt, \ + ##__VA_ARGS__); \ +} while (0) +#else +__printf(2, 3) __cold +static inline +void ibdev_dbg_ratelimited(const struct ib_device *ibdev, const char *format, ...) {} +#endif + union ib_gid { u8 raw[16]; struct { @@ -451,6 +490,16 @@ enum ib_port_state { IB_PORT_ACTIVE_DEFER = 5 }; +enum ib_port_phys_state { + IB_PORT_PHYS_STATE_SLEEP = 1, + IB_PORT_PHYS_STATE_POLLING = 2, + IB_PORT_PHYS_STATE_DISABLED = 3, + IB_PORT_PHYS_STATE_PORT_CONFIGURATION_TRAINING = 4, + IB_PORT_PHYS_STATE_LINK_UP = 5, + IB_PORT_PHYS_STATE_LINK_ERROR_RECOVERY = 6, + IB_PORT_PHYS_STATE_PHY_TEST = 7, +}; + enum ib_port_width { IB_WIDTH_1X = 1, IB_WIDTH_2X = 16, @@ -1417,11 +1466,6 @@ struct ib_ucontext { bool cleanup_retryable; - void (*invalidate_range)(struct ib_umem_odp *umem_odp, - unsigned long start, unsigned long end); - struct mutex per_mm_list_lock; - struct list_head per_mm_list; - struct ib_rdmacg_object cg_obj; /* * Implementation details of the RDMA core, don't use in drivers: @@ -2378,6 +2422,8 @@ struct ib_device_ops { u64 iova); int (*unmap_fmr)(struct list_head *fmr_list); int (*dealloc_fmr)(struct ib_fmr *fmr); + void (*invalidate_range)(struct ib_umem_odp *umem_odp, + unsigned long start, unsigned long end); int (*attach_mcast)(struct ib_qp *qp, union ib_gid *gid, u16 lid); int (*detach_mcast)(struct ib_qp *qp, union ib_gid *gid, u16 lid); struct ib_xrcd *(*alloc_xrcd)(struct ib_device *device, @@ -2647,7 +2693,9 @@ struct ib_client { const union ib_gid *gid, const struct sockaddr *addr, void *client_data); - struct list_head list; + + refcount_t uses; + struct completion uses_zero; u32 client_id; /* kverbs are not required by the client */ @@ -3711,6 +3759,25 @@ static inline struct ib_cq *ib_alloc_cq(struct ib_device *dev, void *private, NULL); } +struct ib_cq *__ib_alloc_cq_any(struct ib_device *dev, void *private, + int nr_cqe, enum ib_poll_context poll_ctx, + const char *caller); + +/** + * ib_alloc_cq_any: Allocate kernel CQ + * @dev: The IB device + * @private: Private data attached to the CQE + * @nr_cqe: Number of CQEs in the CQ + * @poll_ctx: Context used for polling the CQ + */ +static inline struct ib_cq *ib_alloc_cq_any(struct ib_device *dev, + void *private, int nr_cqe, + enum ib_poll_context poll_ctx) +{ + return __ib_alloc_cq_any(dev, private, nr_cqe, poll_ctx, + KBUILD_MODNAME); +} + /** * ib_free_cq_user - Free kernel/user CQ * @cq: The CQ to free diff --git a/include/rdma/iw_portmap.h b/include/rdma/iw_portmap.h index b9fee7feeeb5..c89535047c42 100644 --- a/include/rdma/iw_portmap.h +++ b/include/rdma/iw_portmap.h @@ -33,6 +33,9 @@ #ifndef _IW_PORTMAP_H #define _IW_PORTMAP_H +#include <linux/socket.h> +#include <linux/netlink.h> + #define IWPM_ULIBNAME_SIZE 32 #define IWPM_DEVNAME_SIZE 32 #define IWPM_IFNAME_SIZE 16 diff --git a/include/rdma/opa_port_info.h b/include/rdma/opa_port_info.h index 7147a9263011..bdbfe25d3854 100644 --- a/include/rdma/opa_port_info.h +++ b/include/rdma/opa_port_info.h @@ -33,6 +33,8 @@ #if !defined(OPA_PORT_INFO_H) #define OPA_PORT_INFO_H +#include <rdma/opa_smi.h> + #define OPA_PORT_LINK_MODE_NOP 0 /* No change */ #define OPA_PORT_LINK_MODE_OPA 4 /* Port mode is OPA */ diff --git a/include/rdma/rdma_netlink.h b/include/rdma/rdma_netlink.h index 6631624e4d7c..ab22759de7ea 100644 --- a/include/rdma/rdma_netlink.h +++ b/include/rdma/rdma_netlink.h @@ -76,28 +76,32 @@ int ibnl_put_attr(struct sk_buff *skb, struct nlmsghdr *nlh, /** * Send the supplied skb to a specific userspace PID. + * @net: Net namespace in which to send the skb * @skb: The netlink skb * @pid: Userspace netlink process ID * Returns 0 on success or a negative error code. */ -int rdma_nl_unicast(struct sk_buff *skb, u32 pid); +int rdma_nl_unicast(struct net *net, struct sk_buff *skb, u32 pid); /** * Send, with wait/1 retry, the supplied skb to a specific userspace PID. + * @net: Net namespace in which to send the skb * @skb: The netlink skb * @pid: Userspace netlink process ID * Returns 0 on success or a negative error code. */ -int rdma_nl_unicast_wait(struct sk_buff *skb, __u32 pid); +int rdma_nl_unicast_wait(struct net *net, struct sk_buff *skb, __u32 pid); /** * Send the supplied skb to a netlink group. + * @net: Net namespace in which to send the skb * @skb: The netlink skb * @group: Netlink group ID * @flags: allocation flags * Returns 0 on success or a negative error code. */ -int rdma_nl_multicast(struct sk_buff *skb, unsigned int group, gfp_t flags); +int rdma_nl_multicast(struct net *net, struct sk_buff *skb, + unsigned int group, gfp_t flags); /** * Check if there are any listeners to the netlink group diff --git a/include/rdma/rdma_vt.h b/include/rdma/rdma_vt.h index 525848e227dc..ac5a9430abb6 100644 --- a/include/rdma/rdma_vt.h +++ b/include/rdma/rdma_vt.h @@ -116,6 +116,7 @@ struct rvt_ibport { u64 n_unaligned; u64 n_rc_dupreq; u64 n_rc_seqnak; + u64 n_rc_crwaits; u16 pkey_violations; u16 qkey_violations; u16 mkey_violations; diff --git a/include/rdma/rdmavt_cq.h b/include/rdma/rdmavt_cq.h index 04c519ef6d71..574eb7278f46 100644 --- a/include/rdma/rdmavt_cq.h +++ b/include/rdma/rdmavt_cq.h @@ -53,6 +53,7 @@ #include <linux/kthread.h> #include <rdma/ib_user_verbs.h> +#include <rdma/ib_verbs.h> /* * Define an ib_cq_notify value that is not valid so we know when CQ diff --git a/include/rdma/rdmavt_qp.h b/include/rdma/rdmavt_qp.h index 0eeea520a853..b550ae89bf85 100644 --- a/include/rdma/rdmavt_qp.h +++ b/include/rdma/rdmavt_qp.h @@ -608,7 +608,7 @@ static inline void rvt_qp_wqe_reserve( /** * rvt_qp_wqe_unreserve - clean reserved operation * @qp - the rvt qp - * @wqe - the send wqe + * @flags - send wqe flags * * This decrements the reserve use count. * @@ -620,11 +620,9 @@ static inline void rvt_qp_wqe_reserve( * the compiler does not juggle the order of the s_last * ring index and the decrementing of s_reserved_used. */ -static inline void rvt_qp_wqe_unreserve( - struct rvt_qp *qp, - struct rvt_swqe *wqe) +static inline void rvt_qp_wqe_unreserve(struct rvt_qp *qp, int flags) { - if (unlikely(wqe->wr.send_flags & RVT_SEND_RESERVE_USED)) { + if (unlikely(flags & RVT_SEND_RESERVE_USED)) { atomic_dec(&qp->s_reserved_used); /* insure no compiler re-order up to s_last change */ smp_mb__after_atomic(); @@ -853,6 +851,7 @@ rvt_qp_complete_swqe(struct rvt_qp *qp, u32 byte_len, last; int flags = wqe->wr.send_flags; + rvt_qp_wqe_unreserve(qp, flags); rvt_put_qp_swqe(qp, wqe); need_completion = @@ -974,6 +973,41 @@ static inline void rvt_free_rq(struct rvt_rq *rq) rq->wq = NULL; } +/** + * rvt_to_iport - Get the ibport pointer + * @qp: the qp pointer + * + * This function returns the ibport pointer from the qp pointer. + */ +static inline struct rvt_ibport *rvt_to_iport(struct rvt_qp *qp) +{ + struct rvt_dev_info *rdi = ib_to_rvt(qp->ibqp.device); + + return rdi->ports[qp->port_num - 1]; +} + +/** + * rvt_rc_credit_avail - Check if there are enough RC credits for the request + * @qp: the qp + * @wqe: the request + * + * This function returns false when there are not enough credits for the given + * request and true otherwise. + */ +static inline bool rvt_rc_credit_avail(struct rvt_qp *qp, struct rvt_swqe *wqe) +{ + lockdep_assert_held(&qp->s_lock); + if (!(qp->s_flags & RVT_S_UNLIMITED_CREDIT) && + rvt_cmp_msn(wqe->ssn, qp->s_lsn + 1) > 0) { + struct rvt_ibport *rvp = rvt_to_iport(qp); + + qp->s_flags |= RVT_S_WAIT_SSN_CREDIT; + rvp->n_rc_crwaits++; + return false; + } + return true; +} + struct rvt_qp_iter *rvt_qp_iter_init(struct rvt_dev_info *rdi, u64 v, void (*cb)(struct rvt_qp *qp, u64 v)); diff --git a/include/rdma/restrack.h b/include/rdma/restrack.h index b0fc6b26bdf5..83df1ec6664e 100644 --- a/include/rdma/restrack.h +++ b/include/rdma/restrack.h @@ -105,8 +105,7 @@ struct rdma_restrack_entry { }; int rdma_restrack_count(struct ib_device *dev, - enum rdma_restrack_type type, - struct pid_namespace *ns); + enum rdma_restrack_type type); void rdma_restrack_kadd(struct rdma_restrack_entry *res); void rdma_restrack_uadd(struct rdma_restrack_entry *res); diff --git a/include/rdma/signature.h b/include/rdma/signature.h index f24cc2a1d3c5..d16b0fcc8344 100644 --- a/include/rdma/signature.h +++ b/include/rdma/signature.h @@ -6,6 +6,8 @@ #ifndef _RDMA_SIGNATURE_H_ #define _RDMA_SIGNATURE_H_ +#include <linux/types.h> + enum ib_signature_prot_cap { IB_PROT_T10DIF_TYPE_1 = 1, IB_PROT_T10DIF_TYPE_2 = 1 << 1, |