X-Git-Url: https://git.kernel.dk/?p=fio.git;a=blobdiff_plain;f=engines%2Frdma.c;h=2569a8e32e6a0d34f61fb3c9a46e13be56dfcf27;hp=87ba4658ceca94f6c6aec683cc4e56fc5d0c4d63;hb=ec9e13345f3f433f3693891864cf794206cdd057;hpb=d220c761f78bc04bf34355560a0b6b7b85fba0e8

diff --git a/engines/rdma.c b/engines/rdma.c
index 87ba4658..2569a8e3 100644
--- a/engines/rdma.c
+++ b/engines/rdma.c
@@ -30,7 +30,7 @@
 #include <netinet/in.h>
 #include <arpa/inet.h>
 #include <netdb.h>
-#include <sys/poll.h>
+#include <poll.h>
 #include <sys/types.h>
 #include <sys/socket.h>
 #include <sys/time.h>
@@ -44,7 +44,6 @@
 #include "../optgroup.h"
 
 #include <rdma/rdma_cma.h>
-#include <infiniband/arch.h>
 
 #define FIO_RDMA_MAX_IO_DEPTH    512
 
@@ -60,6 +59,7 @@ struct rdmaio_options {
 	struct thread_data *td;
 	unsigned int port;
 	enum rdma_io_mode verb;
+	char *bindname;
 };
 
 static int str_hostname_cb(void *data, const char *input)
@@ -82,6 +82,16 @@ static struct fio_option options[] = {
 		.category = FIO_OPT_C_ENGINE,
 		.group	= FIO_OPT_G_RDMA,
 	},
+	{
+		.name	= "bindname",
+		.lname	= "rdma engine bindname",
+		.type	= FIO_OPT_STR_STORE,
+		.off1	= offsetof(struct rdmaio_options, bindname),
+		.help	= "Bind for RDMA IO engine",
+		.def    = "",
+		.category = FIO_OPT_C_ENGINE,
+		.group	= FIO_OPT_G_RDMA,
+	},
 	{
 		.name	= "port",
 		.lname	= "rdma engine port",
@@ -191,7 +201,7 @@ struct rdmaio_data {
 
 static int client_recv(struct thread_data *td, struct ibv_wc *wc)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	unsigned int max_bs;
 
 	if (wc->byte_len != sizeof(rd->recv_buf)) {
@@ -216,7 +226,7 @@ static int client_recv(struct thread_data *td, struct ibv_wc *wc)
 		rd->rmt_nr = ntohl(rd->recv_buf.nr);
 
 		for (i = 0; i < rd->rmt_nr; i++) {
-			rd->rmt_us[i].buf = ntohll(rd->recv_buf.rmt_us[i].buf);
+			rd->rmt_us[i].buf = be64_to_cpu(rd->recv_buf.rmt_us[i].buf);
 			rd->rmt_us[i].rkey = ntohl(rd->recv_buf.rmt_us[i].rkey);
 			rd->rmt_us[i].size = ntohl(rd->recv_buf.rmt_us[i].size);
 
@@ -232,7 +242,7 @@ static int client_recv(struct thread_data *td, struct ibv_wc *wc)
 
 static int server_recv(struct thread_data *td, struct ibv_wc *wc)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	unsigned int max_bs;
 
 	if (wc->wr_id == FIO_RDMA_MAX_IO_DEPTH) {
@@ -257,7 +267,7 @@ static int server_recv(struct thread_data *td, struct ibv_wc *wc)
 
 static int cq_event_handler(struct thread_data *td, enum ibv_wc_opcode opcode)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct ibv_wc wc;
 	struct rdma_io_u_data *r_io_u_d;
 	int ret;
@@ -368,7 +378,7 @@ static int cq_event_handler(struct thread_data *td, enum ibv_wc_opcode opcode)
  */
 static int rdma_poll_wait(struct thread_data *td, enum ibv_wc_opcode opcode)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct ibv_cq *ev_cq;
 	void *ev_ctx;
 	int ret;
@@ -405,7 +415,7 @@ again:
 
 static int fio_rdmaio_setup_qp(struct thread_data *td)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct ibv_qp_init_attr init_attr;
 	int qp_depth = td->o.iodepth * 2;	/* 2 times of io depth */
 
@@ -415,7 +425,7 @@ static int fio_rdmaio_setup_qp(struct thread_data *td)
 		rd->pd = ibv_alloc_pd(rd->cm_id->verbs);
 
 	if (rd->pd == NULL) {
-		log_err("fio: ibv_alloc_pd fail\n");
+		log_err("fio: ibv_alloc_pd fail: %m\n");
 		return 1;
 	}
 
@@ -424,7 +434,7 @@ static int fio_rdmaio_setup_qp(struct thread_data *td)
 	else
 		rd->channel = ibv_create_comp_channel(rd->cm_id->verbs);
 	if (rd->channel == NULL) {
-		log_err("fio: ibv_create_comp_channel fail\n");
+		log_err("fio: ibv_create_comp_channel fail: %m\n");
 		goto err1;
 	}
 
@@ -438,12 +448,12 @@ static int fio_rdmaio_setup_qp(struct thread_data *td)
 		rd->cq = ibv_create_cq(rd->cm_id->verbs,
 				       qp_depth, rd, rd->channel, 0);
 	if (rd->cq == NULL) {
-		log_err("fio: ibv_create_cq failed\n");
+		log_err("fio: ibv_create_cq failed: %m\n");
 		goto err2;
 	}
 
 	if (ibv_req_notify_cq(rd->cq, 0) != 0) {
-		log_err("fio: ibv_create_cq failed\n");
+		log_err("fio: ibv_req_notify_cq failed: %m\n");
 		goto err3;
 	}
 
@@ -459,13 +469,13 @@ static int fio_rdmaio_setup_qp(struct thread_data *td)
 
 	if (rd->is_client == 0) {
 		if (rdma_create_qp(rd->child_cm_id, rd->pd, &init_attr) != 0) {
-			log_err("fio: rdma_create_qp failed\n");
+			log_err("fio: rdma_create_qp failed: %m\n");
 			goto err3;
 		}
 		rd->qp = rd->child_cm_id->qp;
 	} else {
 		if (rdma_create_qp(rd->cm_id, rd->pd, &init_attr) != 0) {
-			log_err("fio: rdma_create_qp failed\n");
+			log_err("fio: rdma_create_qp failed: %m\n");
 			goto err3;
 		}
 		rd->qp = rd->cm_id->qp;
@@ -485,19 +495,19 @@ err1:
 
 static int fio_rdmaio_setup_control_msg_buffers(struct thread_data *td)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 
 	rd->recv_mr = ibv_reg_mr(rd->pd, &rd->recv_buf, sizeof(rd->recv_buf),
 				 IBV_ACCESS_LOCAL_WRITE);
 	if (rd->recv_mr == NULL) {
-		log_err("fio: recv_buf reg_mr failed\n");
+		log_err("fio: recv_buf reg_mr failed: %m\n");
 		return 1;
 	}
 
 	rd->send_mr = ibv_reg_mr(rd->pd, &rd->send_buf, sizeof(rd->send_buf),
 				 0);
 	if (rd->send_mr == NULL) {
-		log_err("fio: send_buf reg_mr failed\n");
+		log_err("fio: send_buf reg_mr failed: %m\n");
 		ibv_dereg_mr(rd->recv_mr);
 		return 1;
 	}
@@ -529,7 +539,7 @@ static int get_next_channel_event(struct thread_data *td,
 				  struct rdma_event_channel *channel,
 				  enum rdma_cm_event_type wait_event)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct rdma_cm_event *event;
 	int ret;
 
@@ -561,7 +571,7 @@ static int get_next_channel_event(struct thread_data *td,
 
 static int fio_rdmaio_prep(struct thread_data *td, struct io_u *io_u)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct rdma_io_u_data *r_io_u_d;
 
 	r_io_u_d = io_u->engine_data;
@@ -604,7 +614,7 @@ static int fio_rdmaio_prep(struct thread_data *td, struct io_u *io_u)
 
 static struct io_u *fio_rdmaio_event(struct thread_data *td, int event)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct io_u *io_u;
 	int i;
 
@@ -622,7 +632,7 @@ static struct io_u *fio_rdmaio_event(struct thread_data *td, int event)
 static int fio_rdmaio_getevents(struct thread_data *td, unsigned int min,
 				unsigned int max, const struct timespec *t)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	enum ibv_wc_opcode comp_opcode;
 	struct ibv_cq *ev_cq;
 	void *ev_ctx;
@@ -684,7 +694,7 @@ again:
 static int fio_rdmaio_send(struct thread_data *td, struct io_u **io_us,
 			   unsigned int nr)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct ibv_send_wr *bad_wr;
 #if 0
 	enum ibv_wc_opcode comp_opcode;
@@ -731,7 +741,7 @@ static int fio_rdmaio_send(struct thread_data *td, struct io_u **io_us,
 		}
 
 		if (ibv_post_send(rd->qp, &r_io_u_d->sq_wr, &bad_wr) != 0) {
-			log_err("fio: ibv_post_send fail\n");
+			log_err("fio: ibv_post_send fail: %m\n");
 			return -1;
 		}
 
@@ -747,7 +757,7 @@ static int fio_rdmaio_send(struct thread_data *td, struct io_u **io_us,
 static int fio_rdmaio_recv(struct thread_data *td, struct io_u **io_us,
 			   unsigned int nr)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct ibv_recv_wr *bad_wr;
 	struct rdma_io_u_data *r_io_u_d;
 	int i;
@@ -759,7 +769,7 @@ static int fio_rdmaio_recv(struct thread_data *td, struct io_u **io_us,
 			r_io_u_d = io_us[i]->engine_data;
 			if (ibv_post_recv(rd->qp, &r_io_u_d->rq_wr, &bad_wr) !=
 			    0) {
-				log_err("fio: ibv_post_recv fail\n");
+				log_err("fio: ibv_post_recv fail: %m\n");
 				return 1;
 			}
 		}
@@ -767,7 +777,7 @@ static int fio_rdmaio_recv(struct thread_data *td, struct io_u **io_us,
 		   || (rd->rdma_protocol == FIO_RDMA_MEM_WRITE)) {
 		/* re-post the rq_wr */
 		if (ibv_post_recv(rd->qp, &rd->rq_wr, &bad_wr) != 0) {
-			log_err("fio: ibv_post_recv fail\n");
+			log_err("fio: ibv_post_recv fail: %m\n");
 			return 1;
 		}
 
@@ -781,9 +791,10 @@ static int fio_rdmaio_recv(struct thread_data *td, struct io_u **io_us,
 	return i;
 }
 
-static int fio_rdmaio_queue(struct thread_data *td, struct io_u *io_u)
+static enum fio_q_status fio_rdmaio_queue(struct thread_data *td,
+					  struct io_u *io_u)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 
 	fio_ro_check(td, io_u);
 
@@ -801,8 +812,8 @@ static int fio_rdmaio_queue(struct thread_data *td, struct io_u *io_u)
 static void fio_rdmaio_queued(struct thread_data *td, struct io_u **io_us,
 			      unsigned int nr)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
-	struct timeval now;
+	struct rdmaio_data *rd = td->io_ops_data;
+	struct timespec now;
 	unsigned int i;
 
 	if (!fio_fill_issue_time(td))
@@ -824,7 +835,7 @@ static void fio_rdmaio_queued(struct thread_data *td, struct io_u **io_us,
 
 static int fio_rdmaio_commit(struct thread_data *td)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct io_u **io_us;
 	int ret;
 
@@ -856,7 +867,7 @@ static int fio_rdmaio_commit(struct thread_data *td)
 
 static int fio_rdmaio_connect(struct thread_data *td, struct fio_file *f)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct rdma_conn_param conn_param;
 	struct ibv_send_wr *bad_wr;
 
@@ -866,7 +877,7 @@ static int fio_rdmaio_connect(struct thread_data *td, struct fio_file *f)
 	conn_param.retry_count = 10;
 
 	if (rdma_connect(rd->cm_id, &conn_param) != 0) {
-		log_err("fio: rdma_connect fail\n");
+		log_err("fio: rdma_connect fail: %m\n");
 		return 1;
 	}
 
@@ -881,7 +892,7 @@ static int fio_rdmaio_connect(struct thread_data *td, struct fio_file *f)
 	rd->send_buf.nr = htonl(td->o.iodepth);
 
 	if (ibv_post_send(rd->qp, &rd->sq_wr, &bad_wr) != 0) {
-		log_err("fio: ibv_post_send fail");
+		log_err("fio: ibv_post_send fail: %m\n");
 		return 1;
 	}
 
@@ -907,7 +918,7 @@ static int fio_rdmaio_connect(struct thread_data *td, struct fio_file *f)
 
 static int fio_rdmaio_accept(struct thread_data *td, struct fio_file *f)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct rdma_conn_param conn_param;
 	struct ibv_send_wr *bad_wr;
 	int ret = 0;
@@ -918,7 +929,7 @@ static int fio_rdmaio_accept(struct thread_data *td, struct fio_file *f)
 	conn_param.initiator_depth = 1;
 
 	if (rdma_accept(rd->child_cm_id, &conn_param) != 0) {
-		log_err("fio: rdma_accept\n");
+		log_err("fio: rdma_accept: %m\n");
 		return 1;
 	}
 
@@ -932,7 +943,7 @@ static int fio_rdmaio_accept(struct thread_data *td, struct fio_file *f)
 	ret = rdma_poll_wait(td, IBV_WC_RECV) < 0;
 
 	if (ibv_post_send(rd->qp, &rd->sq_wr, &bad_wr) != 0) {
-		log_err("fio: ibv_post_send fail");
+		log_err("fio: ibv_post_send fail: %m\n");
 		return 1;
 	}
 
@@ -952,7 +963,7 @@ static int fio_rdmaio_open_file(struct thread_data *td, struct fio_file *f)
 
 static int fio_rdmaio_close_file(struct thread_data *td, struct fio_file *f)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct ibv_send_wr *bad_wr;
 
 	/* unregister rdma buffer */
@@ -965,7 +976,7 @@ static int fio_rdmaio_close_file(struct thread_data *td, struct fio_file *f)
 				     || (rd->rdma_protocol ==
 					 FIO_RDMA_MEM_READ))) {
 		if (ibv_post_send(rd->qp, &rd->sq_wr, &bad_wr) != 0) {
-			log_err("fio: ibv_post_send fail");
+			log_err("fio: ibv_post_send fail: %m\n");
 			return 1;
 		}
 
@@ -1005,30 +1016,53 @@ static int fio_rdmaio_close_file(struct thread_data *td, struct fio_file *f)
 	return 0;
 }
 
+static int aton(struct thread_data *td, const char *host,
+		     struct sockaddr_in *addr)
+{
+	if (inet_aton(host, &addr->sin_addr) != 1) {
+		struct hostent *hent;
+
+		hent = gethostbyname(host);
+		if (!hent) {
+			td_verror(td, errno, "gethostbyname");
+			return 1;
+		}
+
+		memcpy(&addr->sin_addr, hent->h_addr, 4);
+	}
+	return 0;
+}
+
 static int fio_rdmaio_setup_connect(struct thread_data *td, const char *host,
 				    unsigned short port)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
+	struct rdmaio_options *o = td->eo;
+	struct sockaddr_storage addrb;
 	struct ibv_recv_wr *bad_wr;
 	int err;
 
 	rd->addr.sin_family = AF_INET;
 	rd->addr.sin_port = htons(port);
 
-	if (inet_aton(host, &rd->addr.sin_addr) != 1) {
-		struct hostent *hent;
+	err = aton(td, host, &rd->addr);
+	if (err)
+		return err;
 
-		hent = gethostbyname(host);
-		if (!hent) {
-			td_verror(td, errno, "gethostbyname");
-			return 1;
-		}
+	/* resolve route */
+	if (strcmp(o->bindname, "") != 0) {
+		addrb.ss_family = AF_INET;
+		err = aton(td, o->bindname, (struct sockaddr_in *)&addrb);
+		if (err)
+			return err;
+		err = rdma_resolve_addr(rd->cm_id, (struct sockaddr *)&addrb,
+					(struct sockaddr *)&rd->addr, 2000);
 
-		memcpy(&rd->addr.sin_addr, hent->h_addr, 4);
+	} else {
+		err = rdma_resolve_addr(rd->cm_id, NULL,
+					(struct sockaddr *)&rd->addr, 2000);
 	}
 
-	/* resolve route */
-	err = rdma_resolve_addr(rd->cm_id, NULL, (struct sockaddr *)&rd->addr, 2000);
 	if (err != 0) {
 		log_err("fio: rdma_resolve_addr: %d\n", err);
 		return 1;
@@ -1072,24 +1106,29 @@ static int fio_rdmaio_setup_connect(struct thread_data *td, const char *host,
 
 static int fio_rdmaio_setup_listen(struct thread_data *td, short port)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
+	struct rdmaio_options *o = td->eo;
 	struct ibv_recv_wr *bad_wr;
 	int state = td->runstate;
 
 	td_set_runstate(td, TD_SETTING_UP);
 
 	rd->addr.sin_family = AF_INET;
-	rd->addr.sin_addr.s_addr = htonl(INADDR_ANY);
 	rd->addr.sin_port = htons(port);
 
+	if (strcmp(o->bindname, "") == 0)
+		rd->addr.sin_addr.s_addr = htonl(INADDR_ANY);
+	else
+		rd->addr.sin_addr.s_addr = htonl(*o->bindname);
+
 	/* rdma_listen */
 	if (rdma_bind_addr(rd->cm_id, (struct sockaddr *)&rd->addr) != 0) {
-		log_err("fio: rdma_bind_addr fail\n");
+		log_err("fio: rdma_bind_addr fail: %m\n");
 		return 1;
 	}
 
 	if (rdma_listen(rd->cm_id, 3) != 0) {
-		log_err("fio: rdma_listen fail\n");
+		log_err("fio: rdma_listen fail: %m\n");
 		return 1;
 	}
 
@@ -1110,7 +1149,7 @@ static int fio_rdmaio_setup_listen(struct thread_data *td, short port)
 
 	/* post recv buf */
 	if (ibv_post_recv(rd->qp, &rd->rq_wr, &bad_wr) != 0) {
-		log_err("fio: ibv_post_recv fail\n");
+		log_err("fio: ibv_post_recv fail: %m\n");
 		return 1;
 	}
 
@@ -1156,7 +1195,8 @@ static int compat_options(struct thread_data *td)
 {
 	// The original RDMA engine had an ugly / seperator
 	// on the filename for it's options. This function
-	// retains backwards compatibility with it.100
+	// retains backwards compatibility with it. Note we do not
+	// support setting the bindname option is this legacy mode.
 
 	struct rdmaio_options *o = td->eo;
 	char *modep, *portp;
@@ -1207,7 +1247,7 @@ bad_host:
 
 static int fio_rdmaio_init(struct thread_data *td)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 	struct rdmaio_options *o = td->eo;
 	unsigned int max_bs;
 	int ret, i;
@@ -1238,13 +1278,13 @@ static int fio_rdmaio_init(struct thread_data *td)
 
 	rd->cm_channel = rdma_create_event_channel();
 	if (!rd->cm_channel) {
-		log_err("fio: rdma_create_event_channel fail\n");
+		log_err("fio: rdma_create_event_channel fail: %m\n");
 		return 1;
 	}
 
 	ret = rdma_create_id(rd->cm_channel, &rd->cm_id, rd, RDMA_PS_TCP);
 	if (ret) {
-		log_err("fio: rdma_create_id fail\n");
+		log_err("fio: rdma_create_id fail: %m\n");
 		return 1;
 	}
 
@@ -1295,12 +1335,12 @@ static int fio_rdmaio_init(struct thread_data *td)
 				      IBV_ACCESS_REMOTE_READ |
 				      IBV_ACCESS_REMOTE_WRITE);
 		if (io_u->mr == NULL) {
-			log_err("fio: ibv_reg_mr io_u failed\n");
+			log_err("fio: ibv_reg_mr io_u failed: %m\n");
 			return 1;
 		}
 
 		rd->send_buf.rmt_us[i].buf =
-		    htonll((uint64_t) (unsigned long)io_u->buf);
+		    cpu_to_be64((uint64_t) (unsigned long)io_u->buf);
 		rd->send_buf.rmt_us[i].rkey = htonl(io_u->mr->rkey);
 		rd->send_buf.rmt_us[i].size = htonl(max_bs);
 
@@ -1316,7 +1356,7 @@ static int fio_rdmaio_init(struct thread_data *td)
 
 static void fio_rdmaio_cleanup(struct thread_data *td)
 {
-	struct rdmaio_data *rd = td->io_ops->data;
+	struct rdmaio_data *rd = td->io_ops_data;
 
 	if (rd)
 		free(rd);
@@ -1332,12 +1372,12 @@ static int fio_rdmaio_setup(struct thread_data *td)
 		td->o.open_files++;
 	}
 
-	if (!td->io_ops->data) {
+	if (!td->io_ops_data) {
 		rd = malloc(sizeof(*rd));
 
 		memset(rd, 0, sizeof(*rd));
 		init_rand_seed(&rd->rand_state, (unsigned int) GOLDEN_RATIO_PRIME, 0);
-		td->io_ops->data = rd;
+		td->io_ops_data = rd;
 	}
 
 	return 0;