Jordan K. Hubbard [Thu, 8 Jul 1993 21:57:48 +0000 (21:57 +0000)]
Changed Makefile to use Nate's selective obj/noobj build scheme.
Nate Willams [Thu, 8 Jul 1993 17:58:20 +0000 (17:58 +0000)]
Removed Fixed Links
Nate Willams [Thu, 8 Jul 1993 17:50:03 +0000 (17:50 +0000)]
Fixed up the make w/out obj directories in libc.
(cgd@sun-lamp.cs.berkeley.edu)
remember curdir as something other than ".", so that makes
of things which include files in directories won't get hosed...
...
made .CURDIR be absolute, not relative, thereby eliminating
the need for some of the makefile trickery when making objects,
and fixing a few .PATH bugs
Paul Richards [Thu, 8 Jul 1993 17:02:21 +0000 (17:02 +0000)]
Removed link to cksum.tahoe.c, just copied it to cksum.i386.c
The link was screwing up build using obj in some circumstances.
J.T. Conklin [Thu, 8 Jul 1993 16:18:59 +0000 (16:18 +0000)]
Gnu grep 2.0 does fgrep too!
Paul Richards [Thu, 8 Jul 1993 14:18:02 +0000 (14:18 +0000)]
Made top level of /usr/include safe (only mpool..h and stab.h needed
doing)
Paul Richards [Thu, 8 Jul 1993 12:17:08 +0000 (12:17 +0000)]
Removed the core.* before disaster strikes.
I removed core as well since it's pretty redundant.
Jordan K. Hubbard [Thu, 8 Jul 1993 03:50:56 +0000 (03:50 +0000)]
Added less to list of things to make.
Jordan K. Hubbard [Thu, 8 Jul 1993 01:15:33 +0000 (01:15 +0000)]
Removed man from list of targets.
Jordan K. Hubbard [Wed, 7 Jul 1993 23:07:02 +0000 (23:07 +0000)]
This commit was generated by cvs2svn to track changes on a CVS vendor
branch.
Jordan K. Hubbard [Wed, 7 Jul 1993 23:07:02 +0000 (23:07 +0000)]
Newer new man page reader.
Nate Willams [Wed, 7 Jul 1993 22:27:07 +0000 (22:27 +0000)]
New man system changes
Nate Willams [Wed, 7 Jul 1993 22:26:28 +0000 (22:26 +0000)]
New man page system Makefile changes
Nate Willams [Wed, 7 Jul 1993 22:24:43 +0000 (22:24 +0000)]
More obj/no-obj fixes to Makefile
Nate Willams [Wed, 7 Jul 1993 22:20:53 +0000 (22:20 +0000)]
Added missing symlinked file into Makefile
Nate Willams [Wed, 7 Jul 1993 21:42:46 +0000 (21:42 +0000)]
Cleaned up header files and added core.* to clean directives
Nate Willams [Wed, 7 Jul 1993 20:38:17 +0000 (20:38 +0000)]
A different (and temporarily more portable) way of obj/noobj
Nate Willams [Wed, 7 Jul 1993 20:30:34 +0000 (20:30 +0000)]
What a mess. Got tn3270 to work with/without obj directories, and you
aren't allowed to look at these Makefiles (too embarassing for both
Charles and I) :-(
David Greenman [Wed, 7 Jul 1993 14:02:39 +0000 (14:02 +0000)]
Fixed panics and process hangs resulting from several bugs in the
socket code and a bug in the pipe code. Suggested by Yuval Yurom on 2/93.
See test programs crashtest3.c and crashtest4.c.
Paul Richards [Wed, 7 Jul 1993 07:13:40 +0000 (07:13 +0000)]
Added gcc1 to gnu makefile
Paul Richards [Wed, 7 Jul 1993 06:05:36 +0000 (06:05 +0000)]
Stopped gcc1 linking itself to gcc
Paul Richards [Wed, 7 Jul 1993 06:01:05 +0000 (06:01 +0000)]
Changed dir names back since cvs wouldn't rename the actual dirs.
Paul Richards [Wed, 7 Jul 1993 05:53:09 +0000 (05:53 +0000)]
Changed name of manual page
Paul Richards [Wed, 7 Jul 1993 05:50:44 +0000 (05:50 +0000)]
Fixed gcc2 to co-exist with gcc1
Jordan K. Hubbard [Wed, 7 Jul 1993 00:43:39 +0000 (00:43 +0000)]
Changed makedb rule to use catpages. NOTE: This will still not produce
the althogether desired effect since it will only build the database for
formatted pages, and we now format on demand. We need some sort of
"catman" command we can run first, then this will work as expected.
Jordan K. Hubbard [Wed, 7 Jul 1993 00:38:56 +0000 (00:38 +0000)]
Fixed a NetBSD anacronism.
Jordan K. Hubbard [Wed, 7 Jul 1993 00:37:45 +0000 (00:37 +0000)]
Added Makefile.
Jordan K. Hubbard [Wed, 7 Jul 1993 00:35:19 +0000 (00:35 +0000)]
Descend into librpc properly
J.T. Conklin [Tue, 6 Jul 1993 23:24:53 +0000 (23:24 +0000)]
silence annoying compiler warning messages.
J.T. Conklin [Tue, 6 Jul 1993 22:00:37 +0000 (22:00 +0000)]
Posix D11.2.2.3 requires that the system's true and false be accessed
instead of searching the $PATH. The best way to satisfy this requirement
is to make them a builtin.
true was allready builtin, this patch adds false, and fixes a minor
manpage bug.
Nate Willams [Tue, 6 Jul 1993 21:46:25 +0000 (21:46 +0000)]
Fixed mkbuiltins to work with/without obj directories
Nate Willams [Tue, 6 Jul 1993 19:46:03 +0000 (19:46 +0000)]
Really cleaned up obj/noobj this time (same as I did with CVS)
Nate Willams [Tue, 6 Jul 1993 19:38:11 +0000 (19:38 +0000)]
We don't need librcs.a in /usr/lib, so make install not do anything
Nate Willams [Tue, 6 Jul 1993 19:23:15 +0000 (19:23 +0000)]
Added a link entry for cksum.i386.c, since it is a symlink in the original
distribution and CVS loses symlinks.
Nate Willams [Tue, 6 Jul 1993 19:12:57 +0000 (19:12 +0000)]
Really fixed the obj/noobj problems in CVS, and also made it so a
make install did not install libcvs in /usr/lib, since it has no
reason to do that.
Nate Willams [Tue, 6 Jul 1993 18:53:50 +0000 (18:53 +0000)]
Removed egrep since the new grep is just a link to it w/out any changes
anymore
Nate Willams [Tue, 6 Jul 1993 18:45:59 +0000 (18:45 +0000)]
Left an ugly in the Makefile
Nate Willams [Tue, 6 Jul 1993 18:45:26 +0000 (18:45 +0000)]
Update GNU grep to version 2.0
Nate Willams [Tue, 6 Jul 1993 18:36:50 +0000 (18:36 +0000)]
Don't echo anything when doing null args, since not doing anything is
more like the standard behavior.
Nate Willams [Tue, 6 Jul 1993 18:06:49 +0000 (18:06 +0000)]
#ifdef out an unecessary function prototype that conflicted with the
system prototype. (Needed for gcc1)
Nate Willams [Tue, 6 Jul 1993 18:00:18 +0000 (18:00 +0000)]
#ifdef out an unecessary function prototype that conflicted with the
system prototype. (Needed for gcc1)
Jordan K. Hubbard [Tue, 6 Jul 1993 16:46:02 +0000 (16:46 +0000)]
Updated slattach to new header-compression enabled version (from NetBSD)
Jordan K. Hubbard [Tue, 6 Jul 1993 00:46:12 +0000 (00:46 +0000)]
Do a proper ANSI declaration.
Nate Willams [Mon, 5 Jul 1993 22:54:53 +0000 (22:54 +0000)]
Bad hack to get _abort messages gone.
I suspect that bdiv() was never called until we added the new math
libraries, and now that it is called (and calls abort()) we need ld
to also contain another link to libc.a as well to pick up abort()
which is contained in libc.a
HACK/HACK/HACK
Jordan K. Hubbard [Mon, 5 Jul 1993 22:54:43 +0000 (22:54 +0000)]
User proper whatis database name.
Nate Willams [Mon, 5 Jul 1993 21:43:35 +0000 (21:43 +0000)]
Fixed netstart to have a hard-coded localhost address.
This fixes situations where you don't have a correct localhost in
your nameserver or no name-server.
David Greenman [Mon, 5 Jul 1993 13:50:18 +0000 (13:50 +0000)]
undid a NetBSD fix because without other support code it does nothing, and
also because it's untested.
David Greenman [Mon, 5 Jul 1993 13:47:49 +0000 (13:47 +0000)]
fixed incorrect signed var type in select that can lead to a panic
Nate Willams [Sun, 4 Jul 1993 04:07:39 +0000 (04:07 +0000)]
Added scandir.c back into the Makefile, since the new directory file
still requires it.
Rick Macklem [Sat, 3 Jul 1993 21:21:35 +0000 (21:21 +0000)]
Increased default data size (DFLDSIZ) to 16MB. Need to rebuild libutil,
kernel, ps and w for this to work!
Rick Macklem [Sat, 3 Jul 1993 12:39:22 +0000 (12:39 +0000)]
Arghh... someone forgot a \ on a line...
Rick Macklem [Sat, 3 Jul 1993 03:09:19 +0000 (03:09 +0000)]
Removed DPADD of libcompat, added LINKS and MLINKS for page -> more
Rick Macklem [Sat, 3 Jul 1993 03:03:40 +0000 (03:03 +0000)]
Enable traversal into the gnu tree at the top level now.
Rick Macklem [Sat, 3 Jul 1993 02:47:17 +0000 (02:47 +0000)]
Fix again for obj dir case..
Rick Macklem [Sat, 3 Jul 1993 02:01:30 +0000 (02:01 +0000)]
Updated since the links for the gnu stuff are no longer needed.
Rick Macklem [Sat, 3 Jul 1993 01:24:54 +0000 (01:24 +0000)]
Another missing ${COPY} knob
Rick Macklem [Sat, 3 Jul 1993 01:12:35 +0000 (01:12 +0000)]
Removed rm -rf ./strip from install:, this should be done by a make clean
Rick Macklem [Sat, 3 Jul 1993 01:06:26 +0000 (01:06 +0000)]
Added ${COPY} knob to install, added CLEANFILES+= termcap so things get
installed and cleaned up properly.
Rick Macklem [Sat, 3 Jul 1993 00:53:52 +0000 (00:53 +0000)]
New manual page system, missed this one on the first round
Rick Macklem [Sat, 3 Jul 1993 00:40:59 +0000 (00:40 +0000)]
Added ${COPY} knob to install
Rick Macklem [Sat, 3 Jul 1993 00:25:14 +0000 (00:25 +0000)]
Added the ${COPY} knob to install
Rick Macklem [Sat, 3 Jul 1993 00:21:21 +0000 (00:21 +0000)]
Added ${COPY} knob to the install commands
Rick Macklem [Sat, 3 Jul 1993 00:12:24 +0000 (00:12 +0000)]
Rename manual page for new manual system, I missed this when doing
the commits.
Nate Willams [Sat, 3 Jul 1993 00:01:21 +0000 (00:01 +0000)]
The 4.4 version needs alot more stuff than exists in the current interim
sources. Back to the old version again.
Nate Willams [Fri, 2 Jul 1993 22:05:24 +0000 (22:05 +0000)]
Chris Provenzan (proven@Athena.MIT.EDU)
Removed bugs and memory leaks from the directory functions.
Nate Willams [Fri, 2 Jul 1993 21:24:51 +0000 (21:24 +0000)]
guido@gvr.win.tue.nl (Guido van Rooij)
Bug:
telnet SEND command broken in client (e.g., SEND AYT)
Diagnosis::
In commands.c:sendcmd(), options that don't have a special
handler function are sent by NET2ADD(IAC, what).
The local variable "what" is never assigned, resulting in junk
being sent to the telnet server.
Nate Willams [Fri, 2 Jul 1993 21:16:12 +0000 (21:16 +0000)]
davidb@otto.bf.rmit.oz.au (David Burren [Athos])
The following patch to /usr/src/usr.bin/netstat/route.c fixes what
I saw as two bugs in `netstat -r`:
1. The output of "Route Tree for Protocol Family" specified the
PF as an int rather than a name.
My fix for this was a quick hack.
2. The gateways in the routing table were always reported as
numeric addresses, even when "-n" was ommitted.
The fix for this is simple and clean.
Rick Macklem [Fri, 2 Jul 1993 12:02:53 +0000 (12:02 +0000)]
Added err.3 to manual page list, got dropped when I did a cvs udpate/merge
of Jordans work, and Nates work.
Rick Macklem [Fri, 2 Jul 1993 11:42:30 +0000 (11:42 +0000)]
Added ed, it has been missed so many times now!
Andrew Moore [Fri, 2 Jul 1993 10:55:17 +0000 (10:55 +0000)]
added missing parens in buffer allocation (caused seg violation)
Nate Willams [Fri, 2 Jul 1993 10:25:21 +0000 (10:25 +0000)]
We really need atof somewhere in the tree, so tell the Makefile to
to find it.
Nate Willams [Fri, 2 Jul 1993 09:16:49 +0000 (09:16 +0000)]
1) Bug description =============================================
Programs compiled on 386bsd using the stock gcc 1.39 do not cast
doubles to ints correctly. K&R says: "When the value of a
floating type is converted to integral type, the fractional part
is discarded...". However, under 386bsd, doubles get ROUNDED to
the nearest int during a cast (instead of getting truncated).
This did not seem to occur under gcc 2.3.3 (on ref.tfs.com).
2) Fix by steve@robobar.co.uk (Steve Bleazard) ==================
"...the bug appears to be in the __fix{uns}dfsi routines that
convert from double to long. The 386 versions of these routines
do not appear to set the rounding mode. [...] gcc 2.3.3 appears to
have it's own C version of these routines...."
Nate Willams [Fri, 2 Jul 1993 09:01:31 +0000 (09:01 +0000)]
1) Bug description =============================================
Programs compiled on 386bsd using the stock gcc 1.39 do not cast
doubles to ints correctly. K&R says: "When the value of a
floating type is converted to integral type, the fractional part
is discarded...". However, under 386bsd, doubles get ROUNDED to
the nearest int during a cast (instead of getting truncated).
This did not seem to occur under gcc 2.3.3 (on ref.tfs.com).
2) Fix by steve@robobar.co.uk (Steve Bleazard) ==================
"...the bug appears to be in the __fix{uns}dfsi routines that
convert from double to long. The 386 versions of these routines
do not appear to set the rounding mode. [...] gcc 2.3.3 appears to
have it's own C version of these routines...."
David Greenman [Fri, 2 Jul 1993 08:56:07 +0000 (08:56 +0000)]
Modified Bruce Evan's patch to useracc to disallow access to UPAGES.
Originally suggested by Yuval Yurom.
Modified calls to round_page() that pass a param that's off by one.
Suggested by Ralf Friedl and Guido Van Rooij.
Rick Macklem [Fri, 2 Jul 1993 08:51:38 +0000 (08:51 +0000)]
New manual page system
Rick Macklem [Fri, 2 Jul 1993 07:27:37 +0000 (07:27 +0000)]
New manual page system
Nate Willams [Fri, 2 Jul 1993 06:53:14 +0000 (06:53 +0000)]
Added diffs from NetBSD version to allow find to compile
Rick Macklem [Fri, 2 Jul 1993 06:44:42 +0000 (06:44 +0000)]
New manual page system
Andrew Moore [Fri, 2 Jul 1993 06:16:28 +0000 (06:16 +0000)]
Consolidate mark code - no functional changes or fixes.
Rick Macklem [Fri, 2 Jul 1993 06:14:54 +0000 (06:14 +0000)]
New manual page system
Nate Willams [Fri, 2 Jul 1993 06:13:38 +0000 (06:13 +0000)]
Updated to newest version on uunet
Nate Willams [Fri, 2 Jul 1993 06:04:27 +0000 (06:04 +0000)]
Date: Sun, 4 Apr 1993 15:43:24 +1000
From: Bruce Evans <bde@runx.oz.au>
Subject: strtod and hugeval patches
... a decent implementation of HUGE_VAL by me.
Nate Willams [Fri, 2 Jul 1993 05:53:33 +0000 (05:53 +0000)]
Bruce Evans minor (cleanup) fixes
Rick Macklem [Fri, 2 Jul 1993 05:48:40 +0000 (05:48 +0000)]
New manual page system
Nate Willams [Fri, 2 Jul 1993 05:45:18 +0000 (05:45 +0000)]
Bruce Evans:
I still don't understand the overflow possibilities mentioned next to
the XXX. errtmp was added by my patch to 0.0. Bill deleted the
volatile but it is necessary to stop gcc -O from optimizing away the
assignments to errtmp. Of course volatile is not portable to old
compilers. No doubt that's why it got deleted. My minix version of the
library has volatile errtmp's in several other files but my 386BSD
version only has them for pow.c (they came with this fix). The point of
the errtmp expressions was to set various IEEE flags but we are a long
way from supporting this properly. First we need to support ANSI
behaviour.
Rick Macklem [Fri, 2 Jul 1993 05:41:22 +0000 (05:41 +0000)]
New manual page system
Rick Macklem [Fri, 2 Jul 1993 04:54:10 +0000 (04:54 +0000)]
New manual page system
Rick Macklem [Fri, 2 Jul 1993 04:34:59 +0000 (04:34 +0000)]
New manual page system
Rick Macklem [Fri, 2 Jul 1993 04:22:44 +0000 (04:22 +0000)]
New manual page system
Rick Macklem [Fri, 2 Jul 1993 01:54:14 +0000 (01:54 +0000)]
New manual page system
Nate Willams [Thu, 1 Jul 1993 23:27:01 +0000 (23:27 +0000)]
Fixed up gcc1 to compile where it's at.
Nate Willams [Thu, 1 Jul 1993 20:49:02 +0000 (20:49 +0000)]
Fixed missing entries for assert and for the strtod fixes
Nate Willams [Thu, 1 Jul 1993 18:19:23 +0000 (18:19 +0000)]
L Jonas Olsson <ljo@vger.EEAP.CWRU.Edu>
My stdio.h patch has showed up! I've discussed it with Bruce and the
conclusion is that a nicer patch to use is ...
Nate Willams [Thu, 1 Jul 1993 00:17:55 +0000 (00:17 +0000)]
Bruce Evans:
The enclosed diffs implement printing of the floating point state for
the version of gdb-3.5 in 386BSD-0.0. I don't have gdb for 386BSD-0.1
but I've been told that it is also missing this feature.
The changes are small. Code to read the FP state from the kernel was
#ifdef'ed out, but it essentially works. Code to change the FP regs
is still #ifdef'ed out. It is close to working too. Printing of the
FP regs was broken because hard reg numbers were confused with stack
offsets.
4. The emulator does not handle FP errors right, and it does not
communicate the emulated FP state to the rest of the kernel, so
"info float" shows garbage.
Nate Willams [Thu, 1 Jul 1993 00:15:11 +0000 (00:15 +0000)]
todd@flex.eng.mcmaster.ca (Todd Pfaff)
There are a few errors in the new pc3 termcap entry supplied with pk-0.2.4.
The entries for the k1-9 functions should have an '=' rather than a ':'.
I discovered this when trying to convert the termcap entry to a terminfo
entry on a Sun.
Nate Willams [Thu, 1 Jul 1993 00:00:26 +0000 (00:00 +0000)]
Whoops, need to add the code to the Makefile I suppose. :-)
Nate Willams [Wed, 30 Jun 1993 23:45:11 +0000 (23:45 +0000)]
Thomas Russo (russo@nile.chem.columbia.edu)
Upon successful completion fgetpos() ...return 0.... Otherwise, -1 is
returned and the global variable errno is set to indicate the error.
It is fixed now.
Nate Willams [Wed, 30 Jun 1993 23:33:41 +0000 (23:33 +0000)]
Andrew A. Chernov (ache@astral.msk.su)
The t_flags field NEVER contain CRTSCTS, but t_cflag filed does.
Fix help me to minimize "silo overflow".
Nate Willams [Wed, 30 Jun 1993 23:25:41 +0000 (23:25 +0000)]
Added new 4.4 assert that Chris Torek posted to the net.
Nate Willams [Wed, 30 Jun 1993 22:59:18 +0000 (22:59 +0000)]
Added (protection) around negative constants, in case a program wants
to use the negative of that constant.
#define NEG_NUM -3
#define SAFE_NEG_NUM (-3)
i = -NEG_NUM; /* Error --3 */
j = -SAFE_NEG_NUM /* Okay -(-3) */