diff --git a/drivers/infiniband/hw/cxgb4/cq.c b/drivers/infiniband/hw/cxgb4/cq.c
index f04a838b65c7..de9bcf2e6d30 100644
--- a/drivers/infiniband/hw/cxgb4/cq.c
+++ b/drivers/infiniband/hw/cxgb4/cq.c
@@ -633,11 +633,15 @@ proc_cqe:
wq->sq.cidx = (uint16_t)idx;
PDBG("%s completing sq idx %u\n", __func__, wq->sq.cidx);
*cookie = wq->sq.sw_sq[wq->sq.cidx].wr_id;
+ if (c4iw_wr_log)
+ c4iw_log_wr_stats(wq, hw_cqe);
t4_sq_consume(wq);
} else {
PDBG("%s completing rq idx %u\n", __func__, wq->rq.cidx);
*cookie = wq->rq.sw_rq[wq->rq.cidx].wr_id;
BUG_ON(t4_rq_empty(wq));
+ if (c4iw_wr_log)
+ c4iw_log_wr_stats(wq, hw_cqe);
t4_rq_consume(wq);
goto skip_cqe;
}
diff --git a/drivers/infiniband/hw/cxgb4/device.c b/drivers/infiniband/hw/cxgb4/device.c
index 8386678f1159..df1f1b52c7ec 100644
--- a/drivers/infiniband/hw/cxgb4/device.c
+++ b/drivers/infiniband/hw/cxgb4/device.c
@@ -55,6 +55,15 @@ module_param(allow_db_coalescing_on_t5, int, 0644);
MODULE_PARM_DESC(allow_db_coalescing_on_t5,
"Allow DB Coalescing on T5 (default = 0)");
+int c4iw_wr_log = 0;
+module_param(c4iw_wr_log, int, 0444);
+MODULE_PARM_DESC(c4iw_wr_log, "Enables logging of work request timing data.");
+
+int c4iw_wr_log_size_order = 12;
+module_param(c4iw_wr_log_size_order, int, 0444);
+MODULE_PARM_DESC(c4iw_wr_log_size_order,
+ "Number of entries (log2) in the work request timing log.");
+
struct uld_ctx {
struct list_head entry;
struct cxgb4_lld_info lldi;
@@ -103,6 +112,117 @@ static ssize_t debugfs_read(struct file *file, char __user *buf, size_t count,
return simple_read_from_buffer(buf, count, ppos, d->buf, d->pos);
}
+void c4iw_log_wr_stats(struct t4_wq *wq, struct t4_cqe *cqe)
+{
+ struct wr_log_entry le;
+ int idx;
+
+ if (!wq->rdev->wr_log)
+ return;
+
+ idx = (atomic_inc_return(&wq->rdev->wr_log_idx) - 1) &
+ (wq->rdev->wr_log_size - 1);
+ le.poll_sge_ts = cxgb4_read_sge_timestamp(wq->rdev->lldi.ports[0]);
+ getnstimeofday(&le.poll_host_ts);
+ le.valid = 1;
+ le.cqe_sge_ts = CQE_TS(cqe);
+ if (SQ_TYPE(cqe)) {
+ le.qid = wq->sq.qid;
+ le.opcode = CQE_OPCODE(cqe);
+ le.post_host_ts = wq->sq.sw_sq[wq->sq.cidx].host_ts;
+ le.post_sge_ts = wq->sq.sw_sq[wq->sq.cidx].sge_ts;
+ le.wr_id = CQE_WRID_SQ_IDX(cqe);
+ } else {
+ le.qid = wq->rq.qid;
+ le.opcode = FW_RI_RECEIVE;
+ le.post_host_ts = wq->rq.sw_rq[wq->rq.cidx].host_ts;
+ le.post_sge_ts = wq->rq.sw_rq[wq->rq.cidx].sge_ts;
+ le.wr_id = CQE_WRID_MSN(cqe);
+ }
+ wq->rdev->wr_log[idx] = le;
+}
+
+static int wr_log_show(struct seq_file *seq, void *v)
+{
+ struct c4iw_dev *dev = seq->private;
+ struct timespec prev_ts = {0, 0};
+ struct wr_log_entry *lep;
+ int prev_ts_set = 0;
+ int idx, end;
+
+#define ts2ns(ts) ((ts) * dev->rdev.lldi.cclk_ps / 1000)
+
+ idx = atomic_read(&dev->rdev.wr_log_idx) &
+ (dev->rdev.wr_log_size - 1);
+ end = idx - 1;
+ if (end < 0)
+ end = dev->rdev.wr_log_size - 1;
+ lep = &dev->rdev.wr_log[idx];
+ while (idx != end) {
+ if (lep->valid) {
+ if (!prev_ts_set) {
+ prev_ts_set = 1;
+ prev_ts = lep->poll_host_ts;
+ }
+ seq_printf(seq, "%04u: sec %lu nsec %lu qid %u opcode "
+ "%u %s 0x%x host_wr_delta sec %lu nsec %lu "
+ "post_sge_ts 0x%llx cqe_sge_ts 0x%llx "
+ "poll_sge_ts 0x%llx post_poll_delta_ns %llu "
+ "cqe_poll_delta_ns %llu\n",
+ idx,
+ timespec_sub(lep->poll_host_ts,
+ prev_ts).tv_sec,
+ timespec_sub(lep->poll_host_ts,
+ prev_ts).tv_nsec,
+ lep->qid, lep->opcode,
+ lep->opcode == FW_RI_RECEIVE ?
+ "msn" : "wrid",
+ lep->wr_id,
+ timespec_sub(lep->poll_host_ts,
+ lep->post_host_ts).tv_sec,
+ timespec_sub(lep->poll_host_ts,
+ lep->post_host_ts).tv_nsec,
+ lep->post_sge_ts, lep->cqe_sge_ts,
+ lep->poll_sge_ts,
+ ts2ns(lep->poll_sge_ts - lep->post_sge_ts),
+ ts2ns(lep->poll_sge_ts - lep->cqe_sge_ts));
+ prev_ts = lep->poll_host_ts;
+ }
+ idx++;
+ if (idx > (dev->rdev.wr_log_size - 1))
+ idx = 0;
+ lep = &dev->rdev.wr_log[idx];
+ }
+#undef ts2ns
+ return 0;
+}
+
+static int wr_log_open(struct inode *inode, struct file *file)
+{
+ return single_open(file, wr_log_show, inode->i_private);
+}
+
+static ssize_t wr_log_clear(struct file *file, const char __user *buf,
+ size_t count, loff_t *pos)
+{
+ struct c4iw_dev *dev = ((struct seq_file *)file->private_data)->private;
+ int i;
+
+ if (dev->rdev.wr_log)
+ for (i = 0; i < dev->rdev.wr_log_size; i++)
+ dev->rdev.wr_log[i].valid = 0;
+ return count;
+}
+
+static const struct file_operations wr_log_debugfs_fops = {
+ .owner = THIS_MODULE,
+ .open = wr_log_open,
+ .release = single_release,
+ .read = seq_read,
+ .llseek = seq_lseek,
+ .write = wr_log_clear,
+};
+
static int dump_qp(int id, void *p, void *data)
{
struct c4iw_qp *qp = p;
@@ -604,6 +724,12 @@ static int setup_debugfs(struct c4iw_dev *devp)
if (de && de->d_inode)
de->d_inode->i_size = 4096;
+ if (c4iw_wr_log) {
+ de = debugfs_create_file("wr_log", S_IWUSR, devp->debugfs_root,
+ (void *)devp, &wr_log_debugfs_fops);
+ if (de && de->d_inode)
+ de->d_inode->i_size = 4096;
+ }
return 0;
}
@@ -717,6 +843,16 @@ static int c4iw_rdev_open(struct c4iw_rdev *rdev)
pr_err(MOD "error allocating status page\n");
goto err4;
}
+ if (c4iw_wr_log) {
+ rdev->wr_log = kzalloc((1 << c4iw_wr_log_size_order) *
+ sizeof(*rdev->wr_log), GFP_KERNEL);
+ if (rdev->wr_log) {
+ rdev->wr_log_size = 1 << c4iw_wr_log_size_order;
+ atomic_set(&rdev->wr_log_idx, 0);
+ } else {
+ pr_err(MOD "error allocating wr_log. Logging disabled\n");
+ }
+ }
return 0;
err4:
c4iw_rqtpool_destroy(rdev);
@@ -730,6 +866,7 @@ err1:
static void c4iw_rdev_close(struct c4iw_rdev *rdev)
{
+ kfree(rdev->wr_log);
free_page((unsigned long)rdev->status_page);
c4iw_pblpool_destroy(rdev);
c4iw_rqtpool_destroy(rdev);
diff --git a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
index 75541cb833c6..69f047cdba6a 100644
--- a/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
+++ b/drivers/infiniband/hw/cxgb4/iw_cxgb4.h
@@ -150,6 +150,18 @@ struct c4iw_hw_queue {
int t4_stat_len;
};
+struct wr_log_entry {
+ struct timespec post_host_ts;
+ struct timespec poll_host_ts;
+ u64 post_sge_ts;
+ u64 cqe_sge_ts;
+ u64 poll_sge_ts;
+ u16 qid;
+ u16 wr_id;
+ u8 opcode;
+ u8 valid;
+};
+
struct c4iw_rdev {
struct c4iw_resource resource;
unsigned long qpshift;
@@ -169,6 +181,9 @@ struct c4iw_rdev {
struct c4iw_stats stats;
struct c4iw_hw_queue hw_queue;
struct t4_dev_status_page *status_page;
+ atomic_t wr_log_idx;
+ struct wr_log_entry *wr_log;
+ int wr_log_size;
};
static inline int c4iw_fatal_error(struct c4iw_rdev *rdev)
@@ -1011,6 +1026,8 @@ void c4iw_ev_dispatch(struct c4iw_dev *dev, struct t4_cqe *err_cqe);
extern struct cxgb4_client t4c_client;
extern c4iw_handler_func c4iw_handlers[NUM_CPL_CMDS];
+extern void c4iw_log_wr_stats(struct t4_wq *wq, struct t4_cqe *cqe);
+extern int c4iw_wr_log;
extern int db_fc_threshold;
extern int db_coalescing_threshold;
extern int use_dsgl;
diff --git a/drivers/infiniband/hw/cxgb4/qp.c b/drivers/infiniband/hw/cxgb4/qp.c
index 0de3cf64eb5e..fd66bd9a9db0 100644
--- a/drivers/infiniband/hw/cxgb4/qp.c
+++ b/drivers/infiniband/hw/cxgb4/qp.c
@@ -823,6 +823,11 @@ int c4iw_post_send(struct ib_qp *ibqp, struct ib_send_wr *wr,
qhp->sq_sig_all;
swsqe->flushed = 0;
swsqe->wr_id = wr->wr_id;
+ if (c4iw_wr_log) {
+ swsqe->sge_ts = cxgb4_read_sge_timestamp(
+ qhp->rhp->rdev.lldi.ports[0]);
+ getnstimeofday(&swsqe->host_ts);
+ }
init_wr_hdr(wqe, qhp->wq.sq.pidx, fw_opcode, fw_flags, len16);
@@ -886,6 +891,13 @@ int c4iw_post_receive(struct ib_qp *ibqp, struct ib_recv_wr *wr,
}
qhp->wq.rq.sw_rq[qhp->wq.rq.pidx].wr_id = wr->wr_id;
+ if (c4iw_wr_log) {
+ qhp->wq.rq.sw_rq[qhp->wq.rq.pidx].sge_ts =
+ cxgb4_read_sge_timestamp(
+ qhp->rhp->rdev.lldi.ports[0]);
+ getnstimeofday(
+ &qhp->wq.rq.sw_rq[qhp->wq.rq.pidx].host_ts);
+ }
wqe->recv.opcode = FW_RI_RECV_WR;
wqe->recv.r1 = 0;
diff --git a/drivers/infiniband/hw/cxgb4/t4.h b/drivers/infiniband/hw/cxgb4/t4.h
index dd45186d1c55..c9f7034e6647 100644
--- a/drivers/infiniband/hw/cxgb4/t4.h
+++ b/drivers/infiniband/hw/cxgb4/t4.h
@@ -269,6 +269,8 @@ struct t4_swsqe {
int signaled;
u16 idx;
int flushed;
+ struct timespec host_ts;
+ u64 sge_ts;
};
static inline pgprot_t t4_pgprot_wc(pgprot_t prot)
@@ -306,6 +308,8 @@ struct t4_sq {
struct t4_swrqe {
u64 wr_id;
+ struct timespec host_ts;
+ u64 sge_ts;
};
struct t4_rq {
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
index ba7d13de5e83..9c7e4f0a7683 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_main.c
@@ -3898,6 +3898,19 @@ err:
}
EXPORT_SYMBOL(cxgb4_read_tpte);
+u64 cxgb4_read_sge_timestamp(struct net_device *dev)
+{
+ u32 hi, lo;
+ struct adapter *adap;
+
+ adap = netdev2adap(dev);
+ lo = t4_read_reg(adap, SGE_TIMESTAMP_LO);
+ hi = GET_TSVAL(t4_read_reg(adap, SGE_TIMESTAMP_HI));
+
+ return ((u64)hi << 32) | (u64)lo;
+}
+EXPORT_SYMBOL(cxgb4_read_sge_timestamp);
+
static struct pci_driver cxgb4_driver;
static void check_neigh_update(struct neighbour *neigh)
@@ -4161,6 +4174,7 @@ static void uld_attach(struct adapter *adap, unsigned int uld)
lli.wr_cred = adap->params.ofldq_wr_cred;
lli.adapter_type = adap->params.chip;
lli.iscsi_iolen = MAXRXDATA_GET(t4_read_reg(adap, TP_PARA_REG2));
+ lli.cclk_ps = 1000000000 / adap->params.vpd.cclk;
lli.udb_density = 1 << QUEUESPERPAGEPF0_GET(
t4_read_reg(adap, SGE_EGRESS_QUEUES_PER_PAGE_PF) >>
(adap->fn * 4));
diff --git a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
index c7170d68bf63..79a84de1d204 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/cxgb4_uld.h
@@ -243,6 +243,7 @@ struct cxgb4_lld_info {
unsigned char fw_api_ver; /* FW API version */
unsigned int fw_vers; /* FW version */
unsigned int iscsi_iolen; /* iSCSI max I/O length */
+ unsigned int cclk_ps; /* Core clock period in psec */
unsigned short udb_density; /* # of user DB/page */
unsigned short ucq_density; /* # of user CQs/page */
unsigned short filt_mode; /* filter optional components */
@@ -297,5 +298,6 @@ int cxgb4_flush_eq_cache(struct net_device *dev);
void cxgb4_disable_db_coalescing(struct net_device *dev);
void cxgb4_enable_db_coalescing(struct net_device *dev);
int cxgb4_read_tpte(struct net_device *dev, u32 stag, __be32 *tpte);
+u64 cxgb4_read_sge_timestamp(struct net_device *dev);
#endif /* !__CXGB4_OFLD_H */
diff --git a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
index ae7776471ceb..3b244abbf907 100644
--- a/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
+++ b/drivers/net/ethernet/chelsio/cxgb4/t4_regs.h
@@ -251,6 +251,12 @@
#define V_NOCOALESCE(x) ((x) << S_NOCOALESCE)
#define F_NOCOALESCE V_NOCOALESCE(1U)
+#define SGE_TIMESTAMP_LO 0x1098
+#define SGE_TIMESTAMP_HI 0x109c
+#define S_TSVAL 0
+#define M_TSVAL 0xfffffffU
+#define GET_TSVAL(x) (((x) >> S_TSVAL) & M_TSVAL)
+
#define SGE_TIMER_VALUE_0_AND_1 0x10b8
#define TIMERVALUE0_MASK 0xffff0000U
#define TIMERVALUE0_SHIFT 16