Jim
4b75d003bd
devpnc: handle case -2 (shutdown) to avoid "bad func in devpnc" error
2011-11-09 14:49:21 -05:00
Jim
6f8555e3f5
Move dongle subdir to src
2011-11-07 15:52:46 -05:00
Jim
c6d4fd1724
emdev.h: mt backspace at EOT kept returning EOT
2011-11-03 17:03:25 -04:00
Jim
6198e38d70
makefile: add targets for broken dongle tests
2011-11-03 14:56:23 -04:00
Jim
33a4b7a459
devpnc: bs problems fixed, corrected rcv pkt hdr update
2011-10-25 20:48:01 -04:00
Jim
77471613b2
makefile: make Intel the default, p suffix for PowerPC
2011-10-25 17:10:11 -04:00
Jim
7b2ea3828b
Don't print memory contents on a fatal error. Diag CPU.TIMER causes a
...
fatal error in get16trap, and printing the memory contents causes a
nested fatal error.
2011-10-24 22:59:04 -04:00
Jim
5ae86cca58
bs: fix major bug in get64r: plp jimmy>date_af would fail, l command
...
(ld -packed -sort_name) would fail after running jimmy>lines>lines.run
2011-10-24 16:47:10 -04:00
Jim
d829b0efd9
Hey, it runs Primos again! This time on Intel. :)
2011-10-21 18:55:43 -04:00
Jim
53893d85ae
bs: more changes, functions for symbolic register access, incl DMX
2011-10-20 17:23:15 -04:00
Jim
e43a2167f0
bs: flt.pt. fixes, fix ecb copy in PCL
2011-10-20 10:44:04 -04:00
Jim
7075729d8c
lrs: don't try to shift 32 bits: works on PPC, fails on Intel
2011-10-20 09:42:33 -04:00
Jim
ff791c1b9c
Remove bogus var from ea64v.h, add quit handler to flush trace buffers
2011-10-19 11:02:23 -04:00
Jim
ec9496ec75
bs: change add32 to a function, use get/put in callers
2011-10-18 10:43:00 -04:00
Jim
3ed4d26624
bs: change tch and tcr to functions, change callers to use get/put
2011-10-17 22:45:33 -04:00
Jim
f77c58a69b
bs: remove remaining crs[] references: BI/DX/Y, I/DRX, ...
2011-10-17 22:39:30 -04:00
Jim
8b5d47c733
bs: change add16 to a function, change all callers to use get/put
2011-10-17 22:30:28 -04:00
Jim
fa097d21fc
BS rvec, memory accesses; order of RPH/RPL is byte-order dependent,
...
fix bug in get32 when -DFAST isn't used
2011-10-17 22:14:55 -04:00
Jim
7b150eed02
bs: replace pmep C pointer with Prime physical address
2011-10-17 15:15:35 -04:00
Jim
5790718ed7
Add swap calls to get/put functions & macros, add tracei target for Intel
2011-10-17 15:06:55 -04:00
Jim
2a89fb6ab8
First set of byte-swap changes; should compile equal to version 194
2011-10-17 10:53:58 -04:00
Jim
69a37aef55
Remove crs[] and crsl[] references in trace output, add vfy target to
...
makefile to allow comparing emulator binaries
2011-10-17 10:53:12 -04:00
Jim
aad6c9f94c
Don't delay on first BDX * to avoid Prime bug in SCHED.PMA causing
...
backstop process to delay on the first cycle
2011-10-11 17:39:32 -04:00
Jim
b9980bfe30
devamlc: add outbound dedicated lines, programmable clock line,
...
move local vars to the block that uses them
2011-10-10 00:07:51 -04:00
Jim
77ee732144
Changes to get rev 18 to boot
2011-10-01 22:42:59 -04:00
Jim
d1596b5e82
devamlc: fix ttymsg bug
2011-09-17 15:21:13 -04:00
Jim
ff48e0c07a
fatal: display message on nested call
2011-09-17 12:00:56 -04:00
Jim
3a64f463f0
devamlc: minor edits
2011-09-16 09:46:51 -04:00
Jim
f912f5dcc5
devamlc: major revision to interrupt only on the clock line board,
...
and in general, optimize operation of the AMLC subsystem.
2011-09-15 18:27:33 -04:00
Jim
efada91773
Fixed several AMLC issues:
...
- Kermit/telnet negotiations work now: fd var was not being set, so
write() was not actually working
- with many controllers, Primos would halt with tumble table overflow:
you can't do read processing until lines are enabled
- only look for new connections when handling the clock line
controller; there is no sense doing this more than 10x per second, and
it probably should still have a timer in case polls are being sped up
- use buf[] for tty message, not another static buf
- when a new connection occurs, set devpoll for the controller. The
first connection to a controller was taking 5 seconds when more than 1
controller was present
- when status is read and an eor occurred, do read processing again
rather than waiting for the next poll
- divide tumble table space up by # of connected lines. The old way
read MAXREAD (64) chars from each line until the tumble table was
full. But with 8 AMLC boards configured, there is only 53 words of TT
space in each double buffer. A select() on connected fd's would be
better.
- allow # as first character of amlc.cfg for comment lines
- turn on TCP/IP NODELAY (disable Nagle) to improve character echo
2011-09-13 14:02:28 -04:00
Jim
88d7a3fba6
Add a way for device to be polled when CPU is idle. Maybe this can be used
...
to interleave user processes with device & DIM processes, for example, during
a file upload.
2011-09-05 16:20:26 -04:00
Jim
023edffb04
Remove "room left in input buffer" concept, remove DMT support
2011-09-05 16:17:48 -04:00
Jim
20e43665bd
Add mtread/mtwrite: read/write .tap files from/to physical tape
2011-08-31 13:40:22 -04:00
Jim
5d43e41c12
Revert #139 (eafa); emacs, esc75* inserted garbage, not 75 *'s
2011-08-31 12:46:53 -04:00
Jim
871568778f
Re-add broken Kermit negotiation, but now telnet works again!
2011-08-27 20:51:19 -04:00
Jim
4e59c7728b
Re-add Kermit telnet negotiation, but leave it disabled
2011-08-27 20:35:12 -04:00
Jim
cc4d7561ed
devamlc comments and cleanups
2011-08-27 20:23:32 -04:00
Jim
7230a331f3
Back out telnet changes for Kermit; they screwed up regular telnet
2011-08-27 20:11:37 -04:00
Jim
927d8c9c13
Tried to fix Kermit telnet negotiation so /nowait isn't required; worked
...
for most things, but not for DO KERMIT / WILL KERMIT - still hangs. :(
2011-08-27 17:04:59 -04:00
Jim
d7dc1847a8
amlc.cfg: can dedicate AMLC lines to an IP address. This is useful
...
for serial device servers connected to a serial printer for example,
where the SDS initiates the telnet connect to the Prime when the
printer is turned on.
2011-08-27 08:29:44 -04:00
Jim
51aaaeb341
Reverse order of devpnc.h and devamlc.h: both use MAXHOSTLEN
2011-08-27 07:34:52 -04:00
Jim
5a5be64d17
Split devamlc into .h file, correct devpnc.h brace bug (hobby)
2011-08-24 17:42:35 -04:00
Jim
b643ed80ed
Add Unix versions of smad, smag for Prime pdevs
2011-08-24 17:15:16 -04:00
Jim
d398a93143
Removed PNCXSNAK - don't need it
2011-08-24 17:12:36 -04:00
Jim
50a95cc1ba
Use close(): shutdown() causes a descriptor leak. Add a trace when a
...
connection is accepted.
2011-08-19 17:04:31 -04:00
Jim
49eaf4b21b
Don't zero uid of disabled node, to avoid "can't find uid" error.
2011-08-19 16:15:16 -04:00
Jim
a3a1f06d6f
When a packet is too big to receive, disable the node it came from.
...
This happens for example if a node is configured with a 1024-word
packet size and sends to a node with a 256-word packet size. In the
emulator PNC design, the packet has already been ACK'd at the sending
side, so there is no way to indicate a transmission error other than
taking down the connection. Since the error will just occur again,
the best course is to disable the node until the next reboot.
Primenet will disconnect the sending and receive nodes from each other
in about 30 seconds.
Also removed pnchexuid: useful in the beginning, not so much now.
2011-08-19 16:04:39 -04:00
Jim
2a698b038d
Don't NAK xmit packets. prmnt1 seems to indicate that a packet no one
...
accepts is NAK'd, but if we do that, it keeps getting sent over and
over. If we're not connected, retries don't make sense.
2011-08-19 13:37:02 -04:00
Jim
6b8abd3a8b
Use current hg rev when building
2011-08-18 20:58:18 -04:00
Jim
aba2688490
Don't enable tsrc$$ trace hack whenever tracing is enabled.
2011-08-18 14:07:23 -04:00
