#include "rds.h"
#include "rdma_transport.h"
-#define RDS_FMR_SIZE 256
-#define RDS_FMR_POOL_SIZE 8192
+#define RDS_FMR_1M_POOL_SIZE (8192 / 2)
+#define RDS_FMR_1M_MSG_SIZE 256
+#define RDS_FMR_8K_MSG_SIZE 2
+#define RDS_MR_8K_SCALE (256 / (RDS_FMR_8K_MSG_SIZE + 1))
+#define RDS_FMR_8K_POOL_SIZE (RDS_MR_8K_SCALE * (8192 / 2))
#define RDS_IB_MAX_SGE 8
#define RDS_IB_RECV_SGE 2
#define RDS_IB_RECYCLE_BATCH_COUNT 32
+#define RDS_IB_WC_MAX 32
+#define RDS_IB_SEND_OP BIT_ULL(63)
+
extern struct rw_semaphore rds_ib_devices_lock;
extern struct list_head rds_ib_devices;
struct rds_ib_send_work {
void *s_op;
- struct ib_send_wr s_wr;
+ union {
+ struct ib_send_wr s_wr;
+ struct ib_rdma_wr s_rdma_wr;
+ struct ib_atomic_wr s_atomic_wr;
+ };
struct ib_sge s_sge[RDS_IB_MAX_SGE];
unsigned long s_queued;
};
atomic_t w_free_ctr;
};
+/* Rings are posted with all the allocations they'll need to queue the
+ * incoming message to the receiving socket so this can't fail.
+ * All fragments start with a header, so we can make sure we're not receiving
+ * garbage, and we can tell a small 8 byte fragment from an ACK frame.
+ */
+struct rds_ib_ack_state {
+ u64 ack_next;
+ u64 ack_recv;
+ unsigned int ack_required:1;
+ unsigned int ack_next_valid:1;
+ unsigned int ack_recv_valid:1;
+};
+
+
struct rds_ib_device;
struct rds_ib_connection {
/* alphabet soup, IBTA style */
struct rdma_cm_id *i_cm_id;
struct ib_pd *i_pd;
- struct ib_mr *i_mr;
struct ib_cq *i_send_cq;
struct ib_cq *i_recv_cq;
+ struct ib_wc i_send_wc[RDS_IB_WC_MAX];
+ struct ib_wc i_recv_wc[RDS_IB_WC_MAX];
+
+ /* interrupt handling */
+ struct tasklet_struct i_send_tasklet;
+ struct tasklet_struct i_recv_tasklet;
/* tx */
struct rds_ib_work_ring i_send_ring;
atomic_t i_signaled_sends;
/* rx */
- struct tasklet_struct i_recv_tasklet;
struct mutex i_recv_mutex;
struct rds_ib_work_ring i_recv_ring;
struct rds_ib_incoming *i_ibinc;
struct rds_ib_ipaddr {
struct list_head list;
__be32 ipaddr;
+ struct rcu_head rcu;
+};
+
+enum {
+ RDS_IB_MR_8K_POOL,
+ RDS_IB_MR_1M_POOL,
};
struct rds_ib_device {
struct list_head conn_list;
struct ib_device *dev;
struct ib_pd *pd;
- struct ib_mr *mr;
- struct rds_ib_mr_pool *mr_pool;
- unsigned int fmr_max_remaps;
unsigned int max_fmrs;
+ struct rds_ib_mr_pool *mr_1m_pool;
+ struct rds_ib_mr_pool *mr_8k_pool;
+ unsigned int fmr_max_remaps;
+ unsigned int max_8k_fmrs;
+ unsigned int max_1m_fmrs;
int max_sge;
unsigned int max_wrs;
unsigned int max_initiator_depth;
struct rds_ib_statistics {
uint64_t s_ib_connect_raced;
uint64_t s_ib_listen_closed_stale;
- uint64_t s_ib_tx_cq_call;
+ uint64_t s_ib_evt_handler_call;
+ uint64_t s_ib_tasklet_call;
uint64_t s_ib_tx_cq_event;
uint64_t s_ib_tx_ring_full;
uint64_t s_ib_tx_throttle;
uint64_t s_ib_tx_sg_mapping_failure;
uint64_t s_ib_tx_stalled;
uint64_t s_ib_tx_credit_updates;
- uint64_t s_ib_rx_cq_call;
uint64_t s_ib_rx_cq_event;
uint64_t s_ib_rx_ring_empty;
uint64_t s_ib_rx_refill_from_cq;
uint64_t s_ib_ack_send_delayed;
uint64_t s_ib_ack_send_piggybacked;
uint64_t s_ib_ack_received;
- uint64_t s_ib_rdma_mr_alloc;
- uint64_t s_ib_rdma_mr_free;
- uint64_t s_ib_rdma_mr_used;
- uint64_t s_ib_rdma_mr_pool_flush;
- uint64_t s_ib_rdma_mr_pool_wait;
- uint64_t s_ib_rdma_mr_pool_depleted;
+ uint64_t s_ib_rdma_mr_8k_alloc;
+ uint64_t s_ib_rdma_mr_8k_free;
+ uint64_t s_ib_rdma_mr_8k_used;
+ uint64_t s_ib_rdma_mr_8k_pool_flush;
+ uint64_t s_ib_rdma_mr_8k_pool_wait;
+ uint64_t s_ib_rdma_mr_8k_pool_depleted;
+ uint64_t s_ib_rdma_mr_1m_alloc;
+ uint64_t s_ib_rdma_mr_1m_free;
+ uint64_t s_ib_rdma_mr_1m_used;
+ uint64_t s_ib_rdma_mr_1m_pool_flush;
+ uint64_t s_ib_rdma_mr_1m_pool_wait;
+ uint64_t s_ib_rdma_mr_1m_pool_depleted;
uint64_t s_ib_atomic_cswp;
uint64_t s_ib_atomic_fadd;
};
* doesn't define it.
*/
static inline void rds_ib_dma_sync_sg_for_cpu(struct ib_device *dev,
- struct scatterlist *sg, unsigned int sg_dma_len, int direction)
+ struct scatterlist *sglist,
+ unsigned int sg_dma_len,
+ int direction)
{
+ struct scatterlist *sg;
unsigned int i;
- for (i = 0; i < sg_dma_len; ++i) {
+ for_each_sg(sglist, sg, sg_dma_len, i) {
ib_dma_sync_single_for_cpu(dev,
- ib_sg_dma_address(dev, &sg[i]),
- ib_sg_dma_len(dev, &sg[i]),
+ ib_sg_dma_address(dev, sg),
+ ib_sg_dma_len(dev, sg),
direction);
}
}
#define ib_dma_sync_sg_for_cpu rds_ib_dma_sync_sg_for_cpu
static inline void rds_ib_dma_sync_sg_for_device(struct ib_device *dev,
- struct scatterlist *sg, unsigned int sg_dma_len, int direction)
+ struct scatterlist *sglist,
+ unsigned int sg_dma_len,
+ int direction)
{
+ struct scatterlist *sg;
unsigned int i;
- for (i = 0; i < sg_dma_len; ++i) {
+ for_each_sg(sglist, sg, sg_dma_len, i) {
ib_dma_sync_single_for_device(dev,
- ib_sg_dma_address(dev, &sg[i]),
- ib_sg_dma_len(dev, &sg[i]),
+ ib_sg_dma_address(dev, sg),
+ ib_sg_dma_len(dev, sg),
direction);
}
}
void rds_ib_dev_put(struct rds_ib_device *rds_ibdev);
extern struct ib_client rds_ib_client;
-extern unsigned int fmr_message_size;
+extern unsigned int rds_ib_fmr_1m_pool_size;
+extern unsigned int rds_ib_fmr_8k_pool_size;
extern unsigned int rds_ib_retry_count;
extern spinlock_t ib_nodev_conns_lock;
void rds_ib_add_conn(struct rds_ib_device *rds_ibdev, struct rds_connection *conn);
void rds_ib_remove_conn(struct rds_ib_device *rds_ibdev, struct rds_connection *conn);
void rds_ib_destroy_nodev_conns(void);
-struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *);
+struct rds_ib_mr_pool *rds_ib_create_mr_pool(struct rds_ib_device *rds_dev,
+ int npages);
void rds_ib_get_mr_info(struct rds_ib_device *rds_ibdev, struct rds_info_rdma_connection *iinfo);
void rds_ib_destroy_mr_pool(struct rds_ib_mr_pool *);
void *rds_ib_get_mr(struct scatterlist *sg, unsigned long nents,
void rds_ib_sync_mr(void *trans_private, int dir);
void rds_ib_free_mr(void *trans_private, int invalidate);
void rds_ib_flush_mrs(void);
+int rds_ib_fmr_init(void);
+void rds_ib_fmr_exit(void);
/* ib_recv.c */
int rds_ib_recv_init(void);
int rds_ib_recv(struct rds_connection *conn);
int rds_ib_recv_alloc_caches(struct rds_ib_connection *ic);
void rds_ib_recv_free_caches(struct rds_ib_connection *ic);
-void rds_ib_recv_refill(struct rds_connection *conn, int prefill);
+void rds_ib_recv_refill(struct rds_connection *conn, int prefill, gfp_t gfp);
void rds_ib_inc_free(struct rds_incoming *inc);
int rds_ib_inc_copy_to_user(struct rds_incoming *inc, struct iov_iter *to);
-void rds_ib_recv_cq_comp_handler(struct ib_cq *cq, void *context);
+void rds_ib_recv_cqe_handler(struct rds_ib_connection *ic, struct ib_wc *wc,
+ struct rds_ib_ack_state *state);
void rds_ib_recv_tasklet_fn(unsigned long data);
void rds_ib_recv_init_ring(struct rds_ib_connection *ic);
void rds_ib_recv_clear_ring(struct rds_ib_connection *ic);
void rds_ib_attempt_ack(struct rds_ib_connection *ic);
void rds_ib_ack_send_complete(struct rds_ib_connection *ic);
u64 rds_ib_piggyb_ack(struct rds_ib_connection *ic);
+void rds_ib_set_ack(struct rds_ib_connection *ic, u64 seq, int ack_required);
/* ib_ring.c */
void rds_ib_ring_init(struct rds_ib_work_ring *ring, u32 nr);
extern wait_queue_head_t rds_ib_ring_empty_wait;
/* ib_send.c */
-char *rds_ib_wc_status_str(enum ib_wc_status status);
void rds_ib_xmit_complete(struct rds_connection *conn);
int rds_ib_xmit(struct rds_connection *conn, struct rds_message *rm,
unsigned int hdr_off, unsigned int sg, unsigned int off);
-void rds_ib_send_cq_comp_handler(struct ib_cq *cq, void *context);
+void rds_ib_send_cqe_handler(struct rds_ib_connection *ic, struct ib_wc *wc);
void rds_ib_send_init_ring(struct rds_ib_connection *ic);
void rds_ib_send_clear_ring(struct rds_ib_connection *ic);
int rds_ib_xmit_rdma(struct rds_connection *conn, struct rm_rdma_op *op);
Code Review / kvmfornfv.git / blobdiff