cslater@ibmpa.UUCP (Charlie Slater) (06/09/89)
Status: IBM/4.3 4-port async problem
The driver for the 4-port buffered asynchronous communications card has
been modified. With the new driver, uucp is stable at 9.6 Kbps and has
held the connection for as long as one hour at 19.2 Kbps. Throughput
measurements for the transfer of a one-megabyte file between two RTs
connected via a null modem are as follows:
uucp of a 1007616-byte file
Speed Setting Time Line Utilization
9.6Kbps (typical) 1311 sec 80%
9.6Kbps (best) 1183 89
19.2Kbps (typical) 759 69
19.2Kbps (best) 666 79
The RT's used had 4 Megabytes of memory and no APC. Note that these
performance measurements were done with the buffered 4-port card
(which contains 4 national 16550 UARTs each with a 16-byte FIFO).
The older, unbufferred 4-port card will not support uucp at speeds in
excess of 2400 bps.
Two characters per millisecond seems to be as much as the tested
configuration could handle. For now at least, it would be a good idea
to plan on an aggregate limit of 19.2Kbps for simultaneous uucp
connections (2x9.6Kbps, 4x4.8Kbps, 8x2.4Kbps, or any other mixture
which totals to less than or eqaul to 19.2Kbps).
This driver has been used on "ibmsupt" to download netnews from uunet
using a Telebit Trailblazer modem with the uucp "g" protocol enabled.
Throughput measurements for batched net news files (~50Kbytes each) are
shown below:
Speed Setting Line Utilization
9.6Kbps (typical) 80%
9.6Kbps (best) 89
19.2Kbps (typical) 55
19.2Kbps (best) 67
All of the kernel changes needed to implement the new driver have been
isolated to asynchronous communications support, and do not depend on,
or interfere with, other bug fixes.
The code is now being beta tested at an IBM Research location. We
regret that it has taken so long to overcome the problem, and hope to
distribute the solution shortly.
charlie slater
uunet!ibmsupt!cslater
(415) 855-4413tropp@ce.chalmers.se. (Ulf Tropp) (06/13/89)
In article <1259@ibmpa.UUCP> cslater@ibmpa.UUCP (Charlie Slater) writes: > >Status: IBM/4.3 4-port async problem > >The driver for the 4-port buffered asynchronous communications card has >been modified. With the new driver, uucp is stable at 9.6 Kbps and has >held the connection for as long as one hour at 19.2 Kbps. That was posted at the same day I had my modifications working after over a week of work... I haven't tested uucp, since I don't want to set that up, but cat-ing a 50k file to a port and reading it back via a loopback plug seems to work sometimes at 19200 and almost always at 9600 so I guess uucp would work. Unbuffered 4-port and Serial/Parallel cards kind of work too. >All of the kernel changes needed to implement the new driver have been >isolated to asynchronous communications support, and do not depend on, >or interfere with, other bug fixes. My fixes affect caio/asy.c, ca/softint.h and ca/trap.c. caio/psp.c would be affected too, I just haven't done it yet (we don't use psp's). >The code is now being beta tested at an IBM Research location. We >regret that it has taken so long to overcome the problem, and hope to >distribute the solution shortly. > >charlie slater >uunet!ibmsupt!cslater >(415) 855-4413 I guess we should give IBM time to test this out once and for all, so if any of you out there are in a real hurry, drop me a mail and I will send you my fixes. We have had two systems running for a couple of days without problems. Also, beware of a hardware bug/feature on all cards and possibly the psp ports too. See another article from me, coming soon. Ulf Tropp, tropp@ce.chalmers.se