DTR has gone high / DTR has gone low - Loop

[hakkanpersson]

Running WinVice 3.6.1 with the settings of;

ACIA RS232, Base: $DE00, IRQ: NMI, Emulation mode: Swiftlink
RS232 Device: IP:25232, Baud: 38400 (Swiftlink/Turbo232)
IP232: Checked

TCPSER v 1.1.4 with command line;
tcpser -v 25232 -s 38400 -p 64128 -l 4

ASUS Router with port forwarding to designated ports.

TCPSER responding perfectly yo CCGMS, HTERM etc. Can call out.

CBase Versions CBase 3.1 Holy Moses and C*Base TAO 3.3.8 all loops between DTR gone high / DTR gone low
Control Lines: DSR:1 DCD:0 CTS:0
Stuck on ATZ in both BBS Softwares. It just loops.

What am I doing wrrong? Have tested 1.1.4 on Linux and also other versions on Window 10, same result.

Please advice.

[go4retro]

Can you share the output of "-l 7 -tiIsS" added to your command line?

[go4retro]

Also, can you check running tcpser with the user port rs232 in VICE? I know the VICE folks worked on the user port rs232 code, but not sure if they also worked on the Swiftlink code.

Jim

[hakkanpersson]

I was able to run via the Userport RS232 and it now reponds to ATZ.

WinVice 3.6.1 with the settings of;
UP232, 2400

TCPSER v 1.1.5alpha with command line;
tcpser -v 25232 -s 2400 -p 64128 -l 7 -t iIsS

BBS
C*Base 3.1 HM with UP2400 modem and corresponding NMI file

Tried to call and getting "Already have ip232 connection, rejecting new". In my log there is a local login after the rejects because I wanted to test if it answered to ATZ again after logging off.

/Hakan

UP2400.txt

[hakkanpersson]

I was able to replicate the "loop".

WinVice 3.6.1 with the settings of;
ACIA RS232, Base: $DE00, IRQ: NMI, Emulation mode: Swiftlink
RS232 Device: IP:25232, Baud: 38400 (Swiftlink/Turbo232)
IP232: Checked

TCPSER v 1.1.5alpha with command line;
tcpser -v 25232 -s 2400 -p 64128 -l 7 -t iIsS

BBS
CBase 3.1 HM with Swiftlink 38400 and corresponding NMI file to match $DE00
CBase TAO 3.3.8 Swiftlink 19200 and corresponding NMI file to match $DE00

Looks like "ATZ" in read as "ATH" or am I missunderstanding?

/Hakan

SWIFT38400-Loop.txt

[go4retro]

This replication still uses the Swiftlink BBS "driver" and the virtual Swiftlink in VICE. Your test just changed the bps rate of the virtual swiftlink, but I would expect to see the same behavior. I'm asking that you reconfigure the BBS to use the user port, switch to using the virtual user port rs232 config in VICE, and see if the problem continues. If it does not, then it's either in the swiftlink "driver" in the BBS app, or the virtual Swiftlink device in VICE.

[go4retro]

It is weird, as VICE appears to pushing DTR up and down continuously. I also think it is odd that VICE sends 3 0x01 bytes after the ATZ and before the ATH. It seems to continue to do DTR down, DTR up, 0x01, 0x01, 0x01, ATH, repeat, as if it cannot understand the ATH.

[hakkanpersson]

Sorry for long time since feedback. Could not get it to work, but with a tool (null-modem emulator) called com0com the communication between TCPSer and WinVice/C*Base works like a charm! Hope this can help anyone who might have similar problems. Link: https://sourceforge.net/projects/com0com/

[mrdudz]

Did you report this on the VICE bugtracker? :) Can you link to a .d64 with a minimal c*base setup on it that we can use to reproduce the problem?