Skip to contentSkip to conversation

1.3.1 and data carrier detect

DISCUSSION Stas Sergeev 2 participants 4 messages Nov 28, 2004 – Dec 2, 2004

lore
Share & tools
lore
#0 lore raw

Hello.

Quoted 2 lines · Maurilio Longo

If I switch from 1.2.2 to 1.3.1 my dos app (no other changes to dosemu.conf or any other files) stops detecting DCD when a connection is established via modem.

Thats odd, I can't recall anything that could cause this.

Quoted 1 line

What can I do?

Open the bug-report and attach the -D9+s logs from both the 1.2.2 and 1.3.1 sessions.

Quoted 1 line

PS. Adding the RTC patch from sourceforge helps a lot serial sending

How exactly it helps? Making it faster? Have you tried setting $_hogthreshold=(0) and see if that helps either?

#1 lore raw

Stas,

Stas Sergeev ha scritto:

Quoted 6 lines

Hello.

Quoted 3 lines · Maurilio Longo

If I switch from 1.2.2 to 1.3.1 my dos app (no other changes to dosemu.conf or any other files) stops detecting DCD when a connection is established via modem.

Quoted 7 lines

Thats odd, I can't recall anything
that could cause this.

Quoted 1 line

What can I do?

Open the bug-report and attach the
-D9+s logs from both the 1.2.2 and
1.3.1 sessions.

I've opened bug id 1075068 adding three logs, if you want I can send them to you, they're 20kb in size.

Quoted 2 lines
Quoted 1 line

PS. Adding the RTC patch from sourceforge helps a lot serial sending

How exactly it helps? Making it faster?

Making it smoother (I have a serial printer and a modem to handle) and making it able to sustain a serial transmission.

Quoted 3 lines

Have you tried setting $_hogthreshold=(0)
and see if that helps either?

Yes, hogthreshold to 0 does not make it faster nor smoother. I've also redefined INT2F_ and INT15_ defines to INT28_ which is a lot smaller, this help serial receiving.

I think that there should be a thread which handles serial I/O and which never sleep but blocks on receiving from real serial port ( /dev/ttySx) not to consume cpu (receiving is the most demanding task), the current implementation which usleep()s here and there isn't able to cope with a 19200 baud transmission on a 3Ghz P4 if I don't change those two defines from timers.h.

Note that telepathy comm library of my clipper app writes directly to the serial hardware and does not use bios.

regards.

--
__________

|  |  | |__| [email protected]
|_|_|_|____| Team OS/2 Italia
#2 lore raw
In reply to Stas Sergeev

Stas,

After you latest fix CDC is again handled correctly.

Serial transmission, though, is still not as good as it should (IMHO), in particular serial sending has problem even at 9600 baud while serial receiving shows crc errors which, I fear, comes from a too slow incoming chars handling.

Since my app writes directly to the harware I think there is something wrong there as well, how can I show you this fact?

Which kind of log do I have to send?

thanks a lot for your help so far.

maurilio.

Stas Sergeev ha scritto:

Quoted 15 lines

Hello.

Quoted 3 lines · Maurilio Longo

If I switch from 1.2.2 to 1.3.1 my dos app (no other changes to dosemu.conf or any other files) stops detecting DCD when a connection is established via modem.

Thats odd, I can't recall anything
that could cause this.

Quoted 1 line

What can I do?

Open the bug-report and attach the
-D9+s logs from both the 1.2.2 and
1.3.1 sessions.

Quoted 1 line

PS. Adding the RTC patch from sourceforge helps a lot serial sending

Quoted 8 lines

How exactly it helps? Making it faster?
Have you tried setting $_hogthreshold=(0)
and see if that helps either?

-
To unsubscribe from this list: send the line "unsubscribe linux-msdos" in the body of a message to [email protected]
More majordomo info at http://vger.kernel.org/majordomo-info.html

--
__________

|  |  | |__| [email protected]
|_|_|_|____| Team OS/2 Italia
#3 lore raw

Hello.

Quoted 3 lines · Maurilio Longo

Since my app writes directly to the harware I think there is something wrong there as well, how can I show you this fact?
Which kind of log do I have to send?

Try starting with the bug-report and -D9+sr
log to it, and we'll see. Another similar
log but with the RTC path may also help.