}
}
+/**
+ * rvt_alloc_rq - allocate memory for user or kernel buffer
+ * @rq: receive queue data structure
+ * @size: number of request queue entries
+ * @node: The NUMA node
+ * @udata: True if user data is available or not false
+ *
+ * Return: If memory allocation failed, return -ENONEM
+ * This function is used by both shared receive
+ * queues and non-shared receive queues to allocate
+ * memory.
+ */
+int rvt_alloc_rq(struct rvt_rq *rq, u32 size, int node,
+ struct ib_udata *udata)
+{
+ if (udata) {
+ rq->wq = vmalloc_user(sizeof(struct rvt_rwq) + size);
+ if (!rq->wq)
+ goto bail;
+ /* need kwq with no buffers */
+ rq->kwq = kzalloc_node(sizeof(*rq->kwq), GFP_KERNEL, node);
+ if (!rq->kwq)
+ goto bail;
+ rq->kwq->curr_wq = rq->wq->wq;
+ } else {
+ /* need kwq with buffers */
+ rq->kwq =
+ vzalloc_node(sizeof(struct rvt_krwq) + size, node);
+ if (!rq->kwq)
+ goto bail;
+ rq->kwq->curr_wq = rq->kwq->wq;
+ }
+
+ spin_lock_init(&rq->lock);
+ return 0;
+bail:
+ rvt_free_rq(rq);
+ return -ENOMEM;
+}
+
/**
* rvt_init_qp - initialize the QP state to the reset state
* @qp: the QP to init or reinit
qp->s_tail_ack_queue = 0;
qp->s_acked_ack_queue = 0;
qp->s_num_rd_atomic = 0;
- if (qp->r_rq.wq) {
- qp->r_rq.wq->head = 0;
- qp->r_rq.wq->tail = 0;
- }
qp->r_sge.num_sge = 0;
atomic_set(&qp->s_reserved_used, 0);
}
qp->r_rq.max_sge = init_attr->cap.max_recv_sge;
sz = (sizeof(struct ib_sge) * qp->r_rq.max_sge) +
sizeof(struct rvt_rwqe);
- if (udata)
- qp->r_rq.wq = vmalloc_user(
- sizeof(struct rvt_rwq) +
- qp->r_rq.size * sz);
- else
- qp->r_rq.wq = vzalloc_node(
- sizeof(struct rvt_rwq) +
- qp->r_rq.size * sz,
- rdi->dparms.node);
- if (!qp->r_rq.wq)
+ err = rvt_alloc_rq(&qp->r_rq, qp->r_rq.size * sz,
+ rdi->dparms.node, udata);
+ if (err) {
+ ret = ERR_PTR(err);
goto bail_driver_priv;
+ }
}
/*
rvt_free_qpn(&rdi->qp_dev->qpn_table, qp->ibqp.qp_num);
bail_rq_wq:
- if (!qp->ip)
- vfree(qp->r_rq.wq);
+ rvt_free_rq(&qp->r_rq);
bail_driver_priv:
rdi->driver_f.qp_priv_free(rdi, qp);
}
wc.status = IB_WC_WR_FLUSH_ERR;
- if (qp->r_rq.wq) {
- struct rvt_rwq *wq;
+ if (qp->r_rq.kwq) {
u32 head;
u32 tail;
+ struct rvt_rwq *wq = NULL;
+ struct rvt_krwq *kwq = NULL;
spin_lock(&qp->r_rq.lock);
-
+ /* qp->ip used to validate if there is a user buffer mmaped */
+ if (qp->ip) {
+ wq = qp->r_rq.wq;
+ head = RDMA_READ_UAPI_ATOMIC(wq->head);
+ tail = RDMA_READ_UAPI_ATOMIC(wq->tail);
+ } else {
+ kwq = qp->r_rq.kwq;
+ head = kwq->head;
+ tail = kwq->tail;
+ }
/* sanity check pointers before trusting them */
- wq = qp->r_rq.wq;
- head = wq->head;
if (head >= qp->r_rq.size)
head = 0;
- tail = wq->tail;
if (tail >= qp->r_rq.size)
tail = 0;
while (tail != head) {
tail = 0;
rvt_cq_enter(ibcq_to_rvtcq(qp->ibqp.recv_cq), &wc, 1);
}
- wq->tail = tail;
-
+ if (qp->ip)
+ RDMA_WRITE_UAPI_ATOMIC(wq->tail, tail);
+ else
+ kwq->tail = tail;
spin_unlock(&qp->r_rq.lock);
} else if (qp->ibqp.event_handler) {
ret = 1;
if (qp->ip)
kref_put(&qp->ip->ref, rvt_release_mmap_info);
- else
- vfree(qp->r_rq.wq);
+ kvfree(qp->r_rq.kwq);
rdi->driver_f.qp_priv_free(rdi, qp);
kfree(qp->s_ack_queue);
rdma_destroy_ah_attr(&qp->remote_ah_attr);
const struct ib_recv_wr **bad_wr)
{
struct rvt_qp *qp = ibqp_to_rvtqp(ibqp);
- struct rvt_rwq *wq = qp->r_rq.wq;
+ struct rvt_krwq *wq = qp->r_rq.kwq;
unsigned long flags;
int qp_err_flush = (ib_rvt_state_ops[qp->state] & RVT_FLUSH_RECV) &&
!qp->ibqp.srq;
next = wq->head + 1;
if (next >= qp->r_rq.size)
next = 0;
- if (next == wq->tail) {
+ if (next == READ_ONCE(wq->tail)) {
spin_unlock_irqrestore(&qp->r_rq.lock, flags);
*bad_wr = wr;
return -ENOMEM;
* Make sure queue entry is written
* before the head index.
*/
- smp_wmb();
- wq->head = next;
+ smp_store_release(&wq->head, next);
}
spin_unlock_irqrestore(&qp->r_rq.lock, flags);
}
const struct ib_recv_wr **bad_wr)
{
struct rvt_srq *srq = ibsrq_to_rvtsrq(ibsrq);
- struct rvt_rwq *wq;
+ struct rvt_krwq *wq;
unsigned long flags;
for (; wr; wr = wr->next) {
}
spin_lock_irqsave(&srq->rq.lock, flags);
- wq = srq->rq.wq;
+ wq = srq->rq.kwq;
next = wq->head + 1;
if (next >= srq->rq.size)
next = 0;
- if (next == wq->tail) {
+ if (next == READ_ONCE(wq->tail)) {
spin_unlock_irqrestore(&srq->rq.lock, flags);
*bad_wr = wr;
return -ENOMEM;
for (i = 0; i < wr->num_sge; i++)
wqe->sg_list[i] = wr->sg_list[i];
/* Make sure queue entry is written before the head index. */
- smp_wmb();
- wq->head = next;
+ smp_store_release(&wq->head, next);
spin_unlock_irqrestore(&srq->rq.lock, flags);
}
return 0;
return 0;
}
+/**
+ * get_rvt_head - get head indices of the circular buffer
+ * @rq: data structure for request queue entry
+ * @ip: the QP
+ *
+ * Return - head index value
+ */
+static inline u32 get_rvt_head(struct rvt_rq *rq, void *ip)
+{
+ u32 head;
+
+ if (ip)
+ head = RDMA_READ_UAPI_ATOMIC(rq->wq->head);
+ else
+ head = rq->kwq->head;
+
+ return head;
+}
+
/**
* rvt_get_rwqe - copy the next RWQE into the QP's RWQE
* @qp: the QP
{
unsigned long flags;
struct rvt_rq *rq;
+ struct rvt_krwq *kwq;
struct rvt_rwq *wq;
struct rvt_srq *srq;
struct rvt_rwqe *wqe;
void (*handler)(struct ib_event *, void *);
u32 tail;
+ u32 head;
int ret;
+ void *ip = NULL;
if (qp->ibqp.srq) {
srq = ibsrq_to_rvtsrq(qp->ibqp.srq);
handler = srq->ibsrq.event_handler;
rq = &srq->rq;
+ ip = srq->ip;
} else {
srq = NULL;
handler = NULL;
rq = &qp->r_rq;
+ ip = qp->ip;
}
spin_lock_irqsave(&rq->lock, flags);
ret = 0;
goto unlock;
}
+ if (ip) {
+ wq = rq->wq;
+ tail = RDMA_READ_UAPI_ATOMIC(wq->tail);
+ } else {
+ kwq = rq->kwq;
+ tail = kwq->tail;
+ }
- wq = rq->wq;
- tail = wq->tail;
/* Validate tail before using it since it is user writable. */
if (tail >= rq->size)
tail = 0;
- if (unlikely(tail == wq->head)) {
+
+ head = get_rvt_head(rq, ip);
+ if (unlikely(tail == head)) {
ret = 0;
goto unlock;
}
- /* Make sure entry is read after head index is read. */
+ /* Make sure entry is read after the count is read. */
smp_rmb();
wqe = rvt_get_rwqe_ptr(rq, tail);
/*
*/
if (++tail >= rq->size)
tail = 0;
- wq->tail = tail;
+ if (ip)
+ RDMA_WRITE_UAPI_ATOMIC(wq->tail, tail);
+ else
+ kwq->tail = tail;
if (!wr_id_only && !init_sge(qp, wqe)) {
ret = -1;
goto unlock;
* Validate head pointer value and compute
* the number of remaining WQEs.
*/
- n = wq->head;
+ n = get_rvt_head(rq, ip);
if (n >= rq->size)
n = 0;
if (n < tail)
const struct ib_recv_wr **bad_wr);
int rvt_wss_init(struct rvt_dev_info *rdi);
void rvt_wss_exit(struct rvt_dev_info *rdi);
+int rvt_alloc_rq(struct rvt_rq *rq, u32 size, int node,
+ struct ib_udata *udata);
#endif /* DEF_RVTQP_H */
} else {
u32 min, max, x;
u32 credits;
- struct rvt_rwq *wq = qp->r_rq.wq;
u32 head;
u32 tail;
/* sanity check pointers before trusting them */
- head = wq->head;
+ if (qp->ip) {
+ head = RDMA_READ_UAPI_ATOMIC(qp->r_rq.wq->head);
+ tail = RDMA_READ_UAPI_ATOMIC(qp->r_rq.wq->tail);
+ } else {
+ head = READ_ONCE(qp->r_rq.kwq->head);
+ tail = READ_ONCE(qp->r_rq.kwq->tail);
+ }
if (head >= qp->r_rq.size)
head = 0;
- tail = wq->tail;
if (tail >= qp->r_rq.size)
tail = 0;
/*
#include "srq.h"
#include "vt.h"
-
+#include "qp.h"
/**
* rvt_driver_srq_init - init srq resources on a per driver basis
* @rdi: rvt dev structure
srq->rq.max_sge = srq_init_attr->attr.max_sge;
sz = sizeof(struct ib_sge) * srq->rq.max_sge +
sizeof(struct rvt_rwqe);
- srq->rq.wq = udata ?
- vmalloc_user(sizeof(struct rvt_rwq) + srq->rq.size * sz) :
- vzalloc_node(sizeof(struct rvt_rwq) + srq->rq.size * sz,
- dev->dparms.node);
- if (!srq->rq.wq) {
+ if (rvt_alloc_rq(&srq->rq, srq->rq.size * sz,
+ dev->dparms.node, udata)) {
ret = -ENOMEM;
goto bail_srq;
}
bail_ip:
kfree(srq->ip);
bail_wq:
- vfree(srq->rq.wq);
+ rvt_free_rq(&srq->rq);
bail_srq:
return ret;
}
{
struct rvt_srq *srq = ibsrq_to_rvtsrq(ibsrq);
struct rvt_dev_info *dev = ib_to_rvt(ibsrq->device);
- struct rvt_rwq *wq;
+ struct rvt_rq tmp_rq = {};
int ret = 0;
if (attr_mask & IB_SRQ_MAX_WR) {
- struct rvt_rwq *owq;
+ struct rvt_krwq *okwq = NULL;
+ struct rvt_rwq *owq = NULL;
struct rvt_rwqe *p;
u32 sz, size, n, head, tail;
((attr_mask & IB_SRQ_LIMIT) ?
attr->srq_limit : srq->limit) > attr->max_wr)
return -EINVAL;
-
sz = sizeof(struct rvt_rwqe) +
srq->rq.max_sge * sizeof(struct ib_sge);
size = attr->max_wr + 1;
- wq = udata ?
- vmalloc_user(sizeof(struct rvt_rwq) + size * sz) :
- vzalloc_node(sizeof(struct rvt_rwq) + size * sz,
- dev->dparms.node);
- if (!wq)
+ if (rvt_alloc_rq(&tmp_rq, size * sz, dev->dparms.node,
+ udata))
return -ENOMEM;
-
/* Check that we can write the offset to mmap. */
if (udata && udata->inlen >= sizeof(__u64)) {
__u64 offset_addr;
* validate head and tail pointer values and compute
* the number of remaining WQEs.
*/
- owq = srq->rq.wq;
- head = owq->head;
- tail = owq->tail;
+ if (udata) {
+ owq = srq->rq.wq;
+ head = RDMA_READ_UAPI_ATOMIC(owq->head);
+ tail = RDMA_READ_UAPI_ATOMIC(owq->tail);
+ } else {
+ okwq = srq->rq.kwq;
+ head = okwq->head;
+ tail = okwq->tail;
+ }
if (head >= srq->rq.size || tail >= srq->rq.size) {
ret = -EINVAL;
goto bail_unlock;
goto bail_unlock;
}
n = 0;
- p = wq->wq;
+ p = tmp_rq.kwq->curr_wq;
while (tail != head) {
struct rvt_rwqe *wqe;
int i;
if (++tail >= srq->rq.size)
tail = 0;
}
- srq->rq.wq = wq;
+ srq->rq.kwq = tmp_rq.kwq;
+ if (udata) {
+ srq->rq.wq = tmp_rq.wq;
+ RDMA_WRITE_UAPI_ATOMIC(tmp_rq.wq->head, n);
+ RDMA_WRITE_UAPI_ATOMIC(tmp_rq.wq->tail, 0);
+ } else {
+ tmp_rq.kwq->head = n;
+ tmp_rq.kwq->tail = 0;
+ }
srq->rq.size = size;
- wq->head = n;
- wq->tail = 0;
if (attr_mask & IB_SRQ_LIMIT)
srq->limit = attr->srq_limit;
spin_unlock_irq(&srq->rq.lock);
vfree(owq);
+ kvfree(okwq);
if (srq->ip) {
struct rvt_mmap_info *ip = srq->ip;
struct rvt_dev_info *dev = ib_to_rvt(srq->ibsrq.device);
u32 s = sizeof(struct rvt_rwq) + size * sz;
- rvt_update_mmap_info(dev, ip, s, wq);
+ rvt_update_mmap_info(dev, ip, s, tmp_rq.wq);
/*
* Return the offset to mmap.
bail_unlock:
spin_unlock_irq(&srq->rq.lock);
bail_free:
- vfree(wq);
+ rvt_free_rq(&tmp_rq);
return ret;
}
spin_unlock(&dev->n_srqs_lock);
if (srq->ip)
kref_put(&srq->ip->ref, rvt_release_mmap_info);
- else
- vfree(srq->rq.wq);
+ kvfree(srq->rq.kwq);
}
#include <rdma/ib_pack.h>
#include <rdma/ib_verbs.h>
#include <rdma/rdmavt_cq.h>
+#include <rdma/rvt-abi.h>
/*
* Atomic bit definitions for r_aflags.
*/
struct rvt_sge sg_list[0];
};
-/*
- * Receive work request queue entry.
- * The size of the sg_list is determined when the QP (or SRQ) is created
- * and stored in qp->r_rq.max_sge (or srq->rq.max_sge).
+/**
+ * struct rvt_krwq - kernel struct receive work request
+ * @head: index of next entry to fill
+ * @tail: index of next entry to pull
+ * @count: count is aproximate of total receive enteries posted
+ * @rvt_rwqe: struct of receive work request queue entry
+ *
+ * This structure is used to contain the head pointer,
+ * tail pointer and receive work queue entries for kernel
+ * mode user.
*/
-struct rvt_rwqe {
- u64 wr_id;
- u8 num_sge;
- struct ib_sge sg_list[0];
-};
-
-/*
- * This structure is used to contain the head pointer, tail pointer,
- * and receive work queue entries as a single memory allocation so
- * it can be mmap'ed into user space.
- * Note that the wq array elements are variable size so you can't
- * just index into the array to get the N'th element;
- * use get_rwqe_ptr() instead.
- */
-struct rvt_rwq {
+struct rvt_krwq {
u32 head; /* new work requests posted to the head */
u32 tail; /* receives pull requests from here. */
- struct rvt_rwqe wq[0];
+ struct rvt_rwqe *curr_wq;
+ struct rvt_rwqe wq[];
};
struct rvt_rq {
struct rvt_rwq *wq;
+ struct rvt_krwq *kwq;
u32 size; /* size of RWQE array */
u8 max_sge;
/* protect changes in this struct */
static inline struct rvt_rwqe *rvt_get_rwqe_ptr(struct rvt_rq *rq, unsigned n)
{
return (struct rvt_rwqe *)
- ((char *)rq->wq->wq +
+ ((char *)rq->kwq->curr_wq +
(sizeof(struct rvt_rwqe) +
rq->max_sge * sizeof(struct ib_sge)) * n);
}
ibcq_to_rvtcq(send_cq)->kqueue->head;
}
+/**
+ * rvt_free_rq - free memory allocated for rvt_rq struct
+ * @rvt_rq: request queue data structure
+ *
+ * This function should only be called if the rvt_mmap_info()
+ * has not succeeded.
+ */
+static inline void rvt_free_rq(struct rvt_rq *rq)
+{
+ kvfree(rq->kwq);
+ rq->kwq = NULL;
+ vfree(rq->wq);
+ rq->wq = NULL;
+}
+
struct rvt_qp_iter *rvt_qp_iter_init(struct rvt_dev_info *rdi,
u64 v,
void (*cb)(struct rvt_qp *qp, u64 v));
#include <linux/types.h>
#include <rdma/ib_user_verbs.h>
+#include <rdma/ib_verbs.h>
#ifndef RDMA_ATOMIC_UAPI
#define RDMA_ATOMIC_UAPI(_type, _name) struct{ _type val; } _name
#endif
struct ib_uverbs_wc uqueue[];
};
+/*
+ * Receive work request queue entry.
+ * The size of the sg_list is determined when the QP (or SRQ) is created
+ * and stored in qp->r_rq.max_sge (or srq->rq.max_sge).
+ */
+struct rvt_rwqe {
+ __u64 wr_id;
+ __u8 num_sge;
+ __u8 padding[7];
+ struct ib_sge sg_list[];
+};
+
+/*
+ * This structure is used to contain the head pointer, tail pointer,
+ * and receive work queue entries as a single memory allocation so
+ * it can be mmap'ed into user space.
+ * Note that the wq array elements are variable size so you can't
+ * just index into the array to get the N'th element;
+ * use get_rwqe_ptr() for user space and rvt_get_rwqe_ptr()
+ * for kernel space.
+ */
+struct rvt_rwq {
+ /* new work requests posted to the head */
+ RDMA_ATOMIC_UAPI(__u32, head);
+ /* receives pull requests from here. */
+ RDMA_ATOMIC_UAPI(__u32, tail);
+ struct rvt_rwqe wq[];
+};
#endif /* RVT_ABI_USER_H */
projects / linux.git / commitdiff