Sean Hefty [Tue, 5 Jun 2012 22:28:18 +0000 (15:28 -0700)]
rsocket: Use configuration files to specify default settings
Give an administrator control over the maximum and default settings
used by rsockets. Use files under rdma/rsocket as shown:
mem_default - default size of receive buffer
mem_max - max size of the receive buffer (restricts setsockopt)
wmem_default - default size of send buffer
wmem_max - max size of the send buffer (restricts setsockopt)
sqsize_default - default size of send queue
sqsize_max - max size of send queue (restricts setsockopt)
rqsize_default - default size of receive queue
rqsize_max - max size of receive queue (restricts setsockopt)
inline_default - default size of inline data
inline_max - max size of inline data (restricts setsockopt)
If configuration files are not available, rsockets will continue to
use internal defaults, with maximums limited by hardware.
Sean Hefty [Mon, 4 Jun 2012 21:51:41 +0000 (14:51 -0700)]
rsocket: Spin before blocking on an rsocket
The latency cost of blocking is significant compared to round
trip ping-pong time. Spin briefly on rsockets before calling
into the kernel and blocking.
The time to spin before blocking is read from an rsocket
configuration file rdma/rsocket/polling_time. This is user
adjustable or may be set automatically by ibacm.
As a completely unintentional side effect, this just happens to
improve application performance in benchmarks, like netpipe. ;)
Sean Hefty [Mon, 4 Jun 2012 21:51:41 +0000 (14:51 -0700)]
rsocket: Spin before blocking on an rsocket
The latency cost of blocking is significant compared to round
trip ping-pong time. Spin briefly on rsockets before calling
into the kernel and blocking.
The time to spin before blocking is read from an rsocket
configuration file rdma/rsocket/polling_time. This is user
adjustable or may be set automatically by ibacm.
As a completely unintentional side effect, this just happens to
improve application performance in benchmarks, like netpipe.
Sean Hefty [Mon, 4 Jun 2012 21:51:41 +0000 (14:51 -0700)]
rsocket: Spin before blocking on an rsocket
The latency cost of blocking is significant compared to round
trip ping-pong time. Spin briefly on rsockets before calling
into the kernel and blocking.
The time to spin before blocking is read from an rsocket
configuration file rdma/rsocket/polling_time. This is user
adjustable or may be set automatically by ibacm.
As a completely unintentional side effect, this just happens to
improve application performance in benchmarks, like netpipe.
Sean Hefty [Mon, 4 Jun 2012 21:51:41 +0000 (14:51 -0700)]
rsocket: Spin before blocking on an rsocket
The latency cost of blocking is significant compared to round
trip ping-pong time. Spin briefly on rsockets before calling
into the kernel and blocking.
The time to spin before blocking is read from an rsocket
configuration file rdma/rsocket/polling_time. This is user
adjustable or may be set automatically by ibacm.
As a completely unintentional side effect, this just happens to
improve application performance in benchmarks, like netpipe.
Sean Hefty [Mon, 4 Jun 2012 21:51:41 +0000 (14:51 -0700)]
rsocket: Spin before blocking on an rsocket
The latency cost of blocking is significant compared to round
trip ping-pong time. Spin briefly on rsockets before calling
into the kernel and blocking.
The time to spin before blocking is read from an rsocket
configuration file rdma/rsocket/polling_time. This is user
adjustable or may be set automatically by ibacm.
As a completely unintentional side effect, this just happens to
improve application performance in benchmarks, like netpipe.
Sean Hefty [Mon, 4 Jun 2012 21:51:41 +0000 (14:51 -0700)]
rsocket: Spin before blocking on an rsocket
The latency cost of blocking is significant compared to round
trip ping-pong time. Spin briefly on rsockets before calling
into the kernel and blocking.
The time to spin before blocking is read from an rsocket
configuration file rdma/rsocket/polling_time. This is user
adjustable or may be set automatically by ibacm.
As a completely unintentional side effect, this just happens to
improve application performance in benchmarks, like netpipe.
Sean Hefty [Mon, 4 Jun 2012 21:51:41 +0000 (14:51 -0700)]
rsocket: Spin before blocking on an rsocket
The latency cost of blocking is significant compared to round
trip ping-pong time. Spin briefly on rsockets before calling
into the kernel and blocking.
The time to spin before blocking is read from an rsocket
configuration file rdma/rsocket/polling_time. This is user
adjustable or may be set automatically by ibacm.
As a completely unintentional side effect, this just happens to
improve application performance in benchmarks, like netpipe.
Sean Hefty [Mon, 4 Jun 2012 21:51:41 +0000 (14:51 -0700)]
rsocket: Spin before blocking on an rsocket
The latency cost of blocking is significant compared to round
trip ping-pong time. Spin briefly on rsockets before calling
into the kernel and blocking.
The time to spin before blocking is read from an rsocket
configuration file rdma/rsocket/polling_time. This is user
adjustable or may be set automatically by ibacm.
As a completely unintentional side effect, this just happens to
improve application performance in benchmarks, like netpipe.
Sean Hefty [Mon, 4 Jun 2012 21:51:41 +0000 (14:51 -0700)]
rsocket: Spin before blocking on an rsocket
The latency cost of blocking is significant compared to round
trip ping-pong time. Spin briefly on rsockets before calling
into the kernel and blocking.
We calculate the time to delay blocking by timing the round
trip ping-pong time on a QP configured for loopback traffic. The
delay time is a multiple of this time to include wire transfer
time, plus minimal response time on the remote side.
As a completely unintentional side effect, this just happens to
improve application performance in benchmarks, like netpipe.
Sean Hefty [Mon, 4 Jun 2012 21:51:41 +0000 (14:51 -0700)]
rsocket: Spin before blocking on an rsocket
The latency cost of blocking is significant compared to round
trip ping-pong time. Spin briefly on rsockets before calling
into the kernel and blocking.
We calculate the time to delay blocking by timing the round
trip ping-pong time on a QP configured for loopback traffic. The
delay time is a multiple of this time to include wire transfer
time, plus minimal response time on the remote side.
projects / ~shefty / librdmacm.git / log