Manual Pages  — IBV_SRQ_PINGPONG

ibv_srq_pingpong - simple InfiniBand shared receive queue test

SYNOPSIS DESCRIPTION OPTIONS SEE ALSO AUTHORS BUGS

ibv_srq_pingpong [-p port] [-d device] [-i ib port] [-s size] [-m size] [-q num QPs] [-r rx depth] [-n iters] [-l sl] [-e] [-g gid index] HOSTNAME

ibv_srq_pingpong [-p port] [-d device] [-i ib port] [-s size] [-m size] [-q num QPs] [-r rx depth] [-n iters] [-l sl] [-e] [-g gid index]

Run a simple ping-pong test over InfiniBand via the reliable connected (RC) transport, using multiple queue pairs (QPs) and a single shared receive queue (SRQ).

-p, --port=PORT
	use TCP port PORT for initial synchronization (default 18515)
-d, --ib-dev=DEVICE
	use IB device DEVICE (default first device found)
-i, --ib-port=PORT
	use IB port PORT (default port 1)
-s, --size=SIZE
	ping-pong messages of size SIZE (default 4096)
-m, --mtu=SIZE
	path MTU SIZE (default 1024)
-q, --num-qp=NUM
	use NUM queue pairs for test (default 16)
-r, --rx-depth=DEPTH
	post DEPTH receives at a time (default 1000)
-n, --iters=ITERS
	perform ITERS message exchanges (default 1000)
-l, --sl=SL
	use SL as the service level value of the QPs (default 0)
-e, --events
	sleep while waiting for work completion events (default is to poll for completions)
-g, --gid-idx=GIDINDEX
	local port GIDINDEX

Roland Dreier
	<rolandd@cisco.com>

The network synchronization between client and server instances is weak, and does not prevent incompatible options from being used on the two instances. The method used for retrieving work completions is not strictly correct, and race conditions may cause failures on some systems.