|
|
1.1 root 1: .\" Copyright (c) 1983 Regents of the University of California.
2: .\" All rights reserved. The Berkeley software License Agreement
3: .\" specifies the terms and conditions for redistribution.
4: .\"
5: .\" @(#)hy.4 6.1 (Berkeley) 5/15/85
6: .\"
7: .TH HY 4 "May 15, 1985"
8: .UC 5
9: .SH NAME
10: hy \- Network Systems Hyperchannel interface
11: .SH SYNOPSIS
12: .B "device hy0 at uba0 csr 0172410 vector hyint"
13: .SH DESCRIPTION
14: The
15: .I hy
16: interface provides access to a Network
17: Systems Corporation Hyperchannel Adapter.
18: .PP
19: The network to which the interface is attached
20: is specified at boot time with an SIOCSIFADDR ioctl.
21: The host's address is discovered by reading the adapter status
22: register. The interface will not transmit or receive
23: packets until the network number is known.
24: .SH DIAGNOSTICS
25: \fBhy%d: unit number 0x%x port %d type %x microcode level 0x%x\fP.
26: Identifies the device during autoconfiguration.
27: .PP
28: \fBhy%d: can't handle af%d\fP. The interface was handed
29: a message with addresses formatted in an unsuitable address
30: family; the packet was dropped.
31: .PP
32: \fBhy%d: can't initialize\fP.
33: The interface was unable to allocate UNIBUS resources. This
34: is usually due to having too many network devices on an 11/750
35: where there are only 3 buffered data paths.
36: .PP
37: \fBhy%d: NEX - Non Existent Memory\fP.
38: Non existent memory error returned from hardware.
39: .PP
40: \fBhy%d: BAR overflow\fP. Bus address register
41: overflow error returned from hardware.
42: .PP
43: \fBhy%d: Power Off bit set, trying to reset\fP.
44: Adapter has lost power, driver will reset the bit
45: and see if power is still out in the adapter.
46: .PP
47: \fBhy%d: Power Off Error, network shutdown\fP.
48: Power was really off in the adapter, network
49: connections are dropped.
50: Software does not shut down the network unless
51: power has been off for a while.
52: .PP
53: \fBhy%d: RECVD MP > MPSIZE (%d)\fP.
54: A message proper was received that is too big.
55: Probable a driver bug.
56: Shouldn't happen.
57: .PP
58: \fBhy%d: xmit error \- len > hy_olen [%d > %d]\fP.
59: Probable driver error.
60: Shouldn't happen.
61: .PP
62: \fBhy%d: DRIVER BUG \- INVALID STATE %d\fP.
63: The driver state machine reached a non-existent state.
64: Definite driver bug.
65: .PP
66: \fBhy%d: watchdog timer expired\fP.
67: A command in the adapter has taken too long to complete.
68: Driver will abort and retry the command.
69: .PP
70: \fBhy%d: adapter power restored\fP.
71: Software was able to reset the power off bit,
72: indicating that the power has been restored.
73: .SH SEE ALSO
74: intro(4N), inet(4F)
75: .SH BUGS
76: If the adapter does not respond to the status command
77: issued during autoconfigure, the adapter is assumed down.
78: A reboot is required to recognize it.
79: .PP
80: The adapter power fail interrupt seems to occur
81: sporadically when power has, in fact, not failed.
82: The driver will believe that power has failed
83: only if it can not reset the power fail latch after
84: a ``reasonable'' time interval.
85: These seem to appear about 2-4 times a day on some machines.
86: There seems to be no correlation with adapter
87: rev level, number of ports used etc. and whether a
88: machine will get these ``bogus powerfails''.
89: They don't seem to cause any real problems so they have
90: been ignored.
This archive runs on limited infrastructure. Preserving old code on modern bandwidth. Automated agents are requested to crawl responsibly.