[pups] BSD 2.11 on an 11/73 + RD54

[pups] BSD 2.11 on an 11/73 + RD54

[Steven M. Schultz]

Hi -

	A few days ago the "getty plays parity games" topic came up
	again.

	I've been thinking about it and looked at the kernel a bit more.
	7e1 is rather firmly in place and I'm not sure turning the 'console'
	(kl/dl) driver into a "LITOUT only" one is a 100% correct way to
	go (the other drivers such as dz, dh, etc remain 7e1 output by
	default).

	Here's a patch for getty/main.c which I think will do what's 
	wanted without having to modify the kernel:

------------snip-------------
*** main.c.dist	Mon Sep  9 16:52:24 2002
--- main.c	Mon Sep  9 16:56:55 2002
***************
*** 384,391 ****
  
  	c = cc;
  	c |= partab[c&0177] & 0200;
! 	if (OP)
! 		c ^= 0200;
  	if (!UB) {
  		outbuf[obufcnt++] = c;
  		if (obufcnt >= OBUFSIZ)
--- 384,401 ----
  
  	c = cc;
  	c |= partab[c&0177] & 0200;
! /*
!  * If "any" parity do nothing otherwise set even parity unless OP is
!  * set.  Since 'ap' is set in the "default" entry of /etc/gettytab this
!  * has the effect of disabling parity on output without having to change
!  * the kernel.
! */
! 	if (!AP) {
! 	   c |= partab[c & 0177] & 0200;
! 	   if (OP)
! 	      c ^= 0200;
! 	}
! 
  	if (!UB) {
  		outbuf[obufcnt++] = c;
  		if (obufcnt >= OBUFSIZ)
--------------snip---------------

	What this does is check the "AP" (AnyParity) flag from /etc/gettytab
	and if NOT set then do the 'even' (or 'odd' if OP is set) parity.

	Since 'ap' is present in the default line of /etc/gettytab the above
	block effectively becomes a no-op unless /etc/gettytab is explicitly
	set for 'ep' or 'op'.

	If someone could test this and report back I'd appreciate it.

	Thanks.

	Steven Schultz
	sms at 2bsd.com

[pups] BSD 2.11 on an 11/73 + RD54

[Tobias Russell]

Hi,

Hmm, still no joy with stty, I get unknown mode errors back is response to
the stty commands you suggested. 

I've just been having a look through the stty man page
(http://www.freebsd.org/cgi/man.cgi?query=stty&apropos=0&sektion=0&manpath=2
.11+BSD&format=html). Does the pass8 switch solve the byte size problem?

Cheers,
Toby

-----Original Message-----
From: Fred N. van Kempen [mailto:pups-admin@minnie.tuhs.org]On Behalf Of
Fred N. van Kempen
Sent: 08 September 2002 16:07
To: Tobias Russell; pups at minnie.tuhs.org
Subject: RE: [pups] BSD 2.11 on an 11/73 + RD54


Hi,

> Currently when I boot, I start off with my terminal set to 19,200 baud
> 8-N-1 but once BSD has booted I have to switch to 19,200 7-E-1
The hardware runs on 8-bit clean channels.  Most UNIX kernels
kinda prefer to use 7e1 or 7o1.


> Can anyone tell me why this is happening and how I fix it so 
> the setting remains on 8-N-1? (stty?)
Once you're logged in to the system, type "stty -parenb bits8" or
"stty -parenb cs8" or "stty -parenb 8" to go back to 8-bit mode.
Then reset your terminal program again :)

> Also, if I can't fix this, will vtserver/vtc run on 7-E-1 comms?
Nope.  VTc requires an 8-bit clean channel (for now).

--fred

-------------- next part --------------
A non-text attachment was scrubbed...
Name: winmail.dat
Type: application/ms-tnef
Size: 1188 bytes
Desc: not available
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20020909/43687856/attachment.bin>

[pups] BSD 2.11 on an 11/73 + RD54

[Tom Ivar Helbekkmo]

"Tobias Russell" <toby at taer.com> writes:

> Currently when I boot, I start off with my terminal set to 19,200 baud
> 8-N-1 but once BSD has booted I have to switch to 19,200 7-E-1

The PDP-11s tend to like 8n1, but UNIX used to prefer 7e1.  I got
tired of 2.11BSD behaving like this, so I changed it thus (note that
the relative directory names in the patch headers are probably not
relative to the same place; you'll have to figure it out):

*** getty/main.c.ORIG	Thu Dec 29 17:22:13 1994
--- getty/main.c	Thu Dec 29 17:21:28 1994
***************
*** 383,391 ****
--- 383,393 ----
  	char c;
  
  	c = cc;
+ #ifdef notdef /* hack to get rid of parity in getty */
  	c |= partab[c&0177] & 0200;
  	if (OP)
  		c ^= 0200;
+ #endif /* parity hack */
  	if (!UB) {
  		outbuf[obufcnt++] = c;
  		if (obufcnt >= OBUFSIZ)

*** pdp/cons.c.ORIG	Sun May 11 11:21:01 1997
--- pdp/cons.c	Sun May 11 11:26:05 1997
***************
*** 62,68 ****
  	if ((tp->t_state&TS_ISOPEN) == 0) {
  		ttychars(tp);
  		tp->t_state = TS_ISOPEN|TS_CARR_ON;
! 		tp->t_flags = EVENP|ECHO|XTABS|CRMOD;
  	}
  	if (tp->t_state&TS_XCLUDE && u.u_uid != 0)
  		return (EBUSY);
--- 62,68 ----
  	if ((tp->t_state&TS_ISOPEN) == 0) {
  		ttychars(tp);
  		tp->t_state = TS_ISOPEN|TS_CARR_ON;
! 		tp->t_flags = ANYP|ECHO|XTABS|CRMOD;
  	}
  	if (tp->t_state&TS_XCLUDE && u.u_uid != 0)
  		return (EBUSY);

*** sys/tty.c.ORIG	Sun May 11 11:21:40 1997
--- sys/tty.c	Sun May 11 11:27:40 1997
***************
*** 48,53 ****
--- 48,54 ----
   */
  
  char partab[] = {
+ #ifdef notdef /* even parity setup */
  	0001,0201,0201,0001,0201,0001,0001,0201,
  	0202,0004,0003,0201,0005,0206,0201,0001,
  	0201,0001,0001,0201,0001,0201,0201,0001,
***************
*** 64,69 ****
--- 65,88 ----
  	0200,0000,0000,0200,0000,0200,0200,0000,
  	0200,0000,0000,0200,0000,0200,0200,0000,
  	0000,0200,0200,0000,0200,0000,0000,0201,
+ #else /* no parity setup follows */
+ 	0001,0001,0001,0001,0001,0001,0001,0001,
+ 	0002,0004,0003,0001,0005,0006,0001,0001,
+ 	0001,0001,0001,0001,0001,0001,0001,0001,
+ 	0001,0001,0001,0001,0001,0001,0001,0001,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0000,
+ 	0000,0000,0000,0000,0000,0000,0000,0001,
+ #endif /* end of parity selection stuff */
  
  	/*
  	 * 7 bit ascii ends with the last character above,

-tih
-- 
Popularity is the hallmark of mediocrity.  --Niles Crane, "Frasier"

[pups] BSD 2.11 on an 11/73 + RD54

[Fred N. van Kempen]

Hi,

> Currently when I boot, I start off with my terminal set to 19,200 baud
> 8-N-1 but once BSD has booted I have to switch to 19,200 7-E-1
The hardware runs on 8-bit clean channels.  Most UNIX kernels
kinda prefer to use 7e1 or 7o1.


> Can anyone tell me why this is happening and how I fix it so 
> the setting remains on 8-N-1? (stty?)
Once you're logged in to the system, type "stty -parenb bits8" or
"stty -parenb cs8" or "stty -parenb 8" to go back to 8-bit mode.
Then reset your terminal program again :)

> Also, if I can't fix this, will vtserver/vtc run on 7-E-1 comms?
Nope.  VTc requires an 8-bit clean channel (for now).

--fred
-------------- next part --------------
A non-text attachment was scrubbed...
Name: winmail.dat
Type: application/ms-tnef
Size: 2419 bytes
Desc: not available
URL: <http://minnie.tuhs.org/pipermail/tuhs/attachments/20020908/3ee85e12/attachment.bin>

[pups] BSD 2.11 on an 11/73 + RD54

[Tobias Russell]

Hi,

I've managed to get a BSD 2.11 root filesystem onto my 11/73 via VTserver
and I am preparing to use vtc to transfer the /usr components.

Currently when I boot, I start off with my terminal set to 19,200 baud
8-N-1 but once BSD has booted I have to switch to 19,200 7-E-1

Can anyone tell me why this is happening and how I fix it so the setting
remains on 8-N-1? (stty?)

Also, if I can't fix this, will vtserver/vtc run on 7-E-1 comms?


Toby