Return to BSD News archive
Newsgroups: comp.unix.bsd
Path: sserve!manuel!munnari.oz.au!uunet!gatech!bloom-beacon!bloom-picayune.mit.edu!athena.mit.edu!eichin
From: eichin@athena.mit.edu (Mark Eichin)
Subject: reading 0.0->0.1 source diffs
Message-ID: <EICHIN.92Jul21092841@tweedledumber.mit.edu>
Sender: news@athena.mit.edu (News system)
Nntp-Posting-Host: cambridge.cygnus.com
Organization: Massachusetts Institute of Technology
Date: Tue, 21 Jul 1992 13:28:52 GMT
Lines: 87
I've been using 0.0 and have had some particular problems that I was
waiting for 0.1 to fix... now that I've read the changes to the
source, I'm pretty sure they're not fixed (I'll have to build a kernel
to be sure) so I'll get to work on them. (In particular, UDP/NFS over
SLIP gives up and times out when the packets are heading into the pc,
but works fine the other way. TCP/NFS (and other TCP protocols)
doesn't have this problem.)
Reading the diffs is educational. There are lots of neat
things. A few oddities: in tty_ring.c, the arguments to nextc are
still switched as they were in the original 0.0 release; I thought I
reported this directly (I know it was discussed on the net.) Also, it
seems that the some of the copyright messages are slightly mangled.
(diff below shows both problems; paycheck.backup2 is a backup of my
machine running 0.0)
Also, I noticed that the com driver always jumps back to spl0
rather than bracketing with {s=spltty();splx(s);} I'd be curious to
know why this was done (I believe that some of the splx code was added
by Dan Muntz, but I don't know if what I'm running is the latest
version of his driver.) I thought someone mentioned that the com
driver in 0.1 isn't the latest; if that's so, I'll keep my old one.
This is by no means a thorough reading (or thorough commentary
on what reading I've done) but it was interesting, nonetheless.
_Mark_ <eichin@athena.mit.edu>
MIT Student Information Processing Board
Cygnus Support <eichin@cygnus.com>
ps. I emailed the above comments with the bugnfix.form but thought it
would be good to discuss a bit on the net (and it's hard to tell if
internet mail is getting to the the authors.)
diff -rcbtw paycheck.backup2/src/sys.386bsd/kern/tty_ring.c 0.1/usr/src/sys.386b
sd/kern/tty_ring.c
*** paycheck.backup2/src/sys.386bsd/kern/tty_ring.c Sun Mar 29 22:09:20 1992
--- 0.1/usr/src/sys.386bsd/kern/tty_ring.c Mon Feb 24 16:38:10 1992
***************
*** 13,19 ****
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This software is a component of "386BSD" developed by
! * William F. Jolitz, TeleMuse.
* 4. Neither the name of the developer nor the name "386BSD"
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
--- 13,19 ----
* 3. All advertising materials mentioning features or use of this software
* must display the following acknowledgement:
* This software is a component of "386BSD" developed by
! William F. Jolitz, TeleMuse.
* 4. Neither the name of the developer nor the name "386BSD"
* may be used to endorse or promote products derived from this software
* without specific prior written permission.
***************
*** 22,28 ****
* AND IS INTENDED FOR RESEARCH AND EDUCATIONAL PURPOSES ONLY. THIS
* SOFTWARE SHOULD NOT BE CONSIDERED TO BE A COMMERCIAL PRODUCT.
* THE DEVELOPER URGES THAT USERS WHO REQUIRE A COMMERCIAL PRODUCT
! * NOT MAKE USE OF THIS WORK.
*
* FOR USERS WHO WISH TO UNDERSTAND THE 386BSD SYSTEM DEVELOPED
* BY WILLIAM F. JOLITZ, WE RECOMMEND THE USER STUDY WRITTEN
--- 22,28 ----
* AND IS INTENDED FOR RESEARCH AND EDUCATIONAL PURPOSES ONLY. THIS
* SOFTWARE SHOULD NOT BE CONSIDERED TO BE A COMMERCIAL PRODUCT.
* THE DEVELOPER URGES THAT USERS WHO REQUIRE A COMMERCIAL PRODUCT
! * NOT MAKE USE THIS WORK.
*
* FOR USERS WHO WISH TO UNDERSTAND THE 386BSD SYSTEM DEVELOPED
* BY WILLIAM F. JOLITZ, WE RECOMMEND THE USER STUDY WRITTEN
***************
*** 80,86 ****
return (c);
}
! nextc(rbp, cpp) struct ringb *rbp; char **cpp; {
if (*cpp == rbp->rb_tl) return (0);
else { char *cp;
--- 80,86 ----
return (c);
}
! nextc(cpp, rbp) struct ringb *rbp; char **cpp; {
if (*cpp == rbp->rb_tl) return (0);
else { char *cp;