ea4defd6da
do the Net/2 thing with startrtclock() for non-i386 architectures.
...
i386's startrtclock should be moved down, as well, but i believe it
does some magic...
1993-08-07 06:25:43 +00:00
f5ec218873
the swtch() calls here should not need to be splclock() protected,
...
and certainly don't need to be bracketed by splclock()/splnone(),
the latter of which is a jolitz invention.
1993-08-07 05:16:38 +00:00
bbc8c11fd5
Nuke an extra `||' Chris inserted.
1993-08-03 02:45:20 +00:00
d6b4910ac2
fix from Garrett Wollman <wollman@emba.uvm.edu> to return EPROTONOTSUPP
...
if user tries to get a socket for a protocol with no usrreq function
1993-08-03 01:36:10 +00:00
5cc9898c12
Cosmetic change to VOP_ADVLOCK() fix.
1993-08-03 00:11:29 +00:00
f9ce597695
Collapse a bunch of `if (a & x) b |= x; else b &= ~x;' statements.
...
Whoever wrote this fugly code must've been on drugs.
1993-08-02 23:37:56 +00:00
4862b84c92
Add RCS identifiers (this time on the correct side of the branch), and
...
incorporate recent changes in netbsd-0-9 branch.
1993-08-01 19:22:24 +00:00
2680a043c7
general cleanup (get close to KNF), and remove a few fuzzies
...
in preparation for The Grand Re-rewriting.
1993-07-28 04:08:27 +00:00
7b2afa7fe6
incorporate changes from 0-9-base to 0-9-ALPHA
1993-07-28 02:20:34 +00:00
96ec00e106
Use all of freebufspace
1993-07-24 22:09:50 +00:00
6a12d4b9b2
Clean up deleted files.
1993-07-19 16:39:09 +00:00
9bd082f70e
replace jolitz's vfs__bio with a better one from CMU via mw.
...
so, replace vfs__bio, and deal with attendant changes.
1993-07-19 16:38:35 +00:00
57ee917973
allocate out of kmem_map, not buffer_map. buffer_map is for
...
*buffers*... hopefully this won't fragment kmem_map too much.
it should probably be put into its own map, but doesn't need to be
right now.
1993-07-19 09:57:22 +00:00
7aecc812a5
Move flushq() macro into tty.h.
1993-07-19 05:52:27 +00:00
69fdebd9fa
Make sure there are character in the ring before flushing.
1993-07-19 05:49:13 +00:00
ec99160f49
Use flushq() macro instead.
1993-07-19 05:48:01 +00:00
b8756ae04c
Use ndflush(), not while(getc()).
1993-07-19 05:10:30 +00:00
5dbe2d9d5b
* don't used copyout() to relocate icode - use bcopy() instead
1993-07-18 08:17:36 +00:00
23b68e4c79
Clean up deleted files.
1993-07-18 07:21:13 +00:00
a2587ffb83
Nuke a kluge from Net/2. The argument list ocreat() creates for open() can
...
now be a struct open_args; no need to redefine the structure.
1993-07-18 06:28:01 +00:00
41b40c6792
Only declare these variables extern in buf.h; define them only once in
...
vfs__bio.c.
1993-07-18 05:15:25 +00:00
331148c31a
Make these files GCC 2 happy.
1993-07-17 16:20:24 +00:00
97cae397ee
Finish moving struct definitions outside of function declarations.
1993-07-17 15:24:33 +00:00
9c0456d7d7
gcc2 cleanup, and break args out of procedure def'ns
1993-07-15 22:56:23 +00:00
082beb9cc0
make args & locals of select unsigned where appropriate.
1993-07-15 22:26:11 +00:00
74f0cb89c9
Add 'ps' command. Add -more- pager to output from Mach ddb.
1993-07-15 18:36:32 +00:00
5287574e0a
get rid of bogus retrymalloc code (and it was bogus)...
1993-07-15 13:33:23 +00:00
c0334ef92c
break args structs out, into syscallname_args structs, so gcc2 doesn't
...
whine so much.
1993-07-13 22:13:15 +00:00
7ef782c7d7
Clean up deleted files.
1993-07-12 11:36:21 +00:00
41b03a4a6e
Change tty code to use clist interface, but with ring buffer implementation.
...
Also, fix a couple of bugs in tty.c and pccons.c, and some gross kluginess
in the hp300 stuff.
1993-07-12 11:33:54 +00:00
bfc1b68bb5
Change tty code to use clist interface, but with ring buffer implementation.
...
Also, fix a couple of bugs in tty.c and pccons.c, and some gross kluginess
in the hp300 stuff.
1993-07-12 11:16:07 +00:00
daacc668f9
re-add two changes which had been deleted by commit of r1.7
1993-07-11 07:22:52 +00:00
5a84b475d6
handle the initflags problem in a simple (if twisted) way.
...
also, remind the pagedaemon that it's a daemon, not an r... 8-)
1993-07-10 03:25:03 +00:00
842b58edb5
Change the names of processes 0 and 2.
1993-07-10 02:48:57 +00:00
b5236e7d0f
Roundup stack len to naturally aligned address. This fixes emacs aborting due to an odd address in alloca'ed memory.
1993-07-10 02:07:48 +00:00
1c33fddc0b
Oops.
1993-07-08 22:53:05 +00:00
31d24dc82e
Ignore the flags field in midmag.
1993-07-08 16:48:18 +00:00
98bb5ccd73
if mincnt returns 0 (i.e. off end of disk, or something), punt.
1993-07-08 10:53:40 +00:00
0ffdd71a7e
mark the buffer busy immediately when we acquire it,
...
also, sleep at a slightly higher priority (so says mw...)
1993-07-08 10:29:26 +00:00
8f29c4b498
specification of mountroot now handled by config
1993-07-07 10:55:00 +00:00
06d372686f
revert code base back to net/2. in particular,
...
fix jolitz's trapsignal() change (it's only supposed to work on curproc)
and add a panic if called w/something other than curproc, and also
fix a relic of his crappy exec implementation
1993-07-07 08:12:33 +00:00
38e4cf5f34
Clean up deleted files.
1993-07-07 06:03:57 +00:00
419b16b0f6
add rcsid, and get rid of extraneous debugging printf
1993-07-07 06:00:15 +00:00
ac8fe993ef
add rmap routines written by Wolfgang Solfrank to the tree,
...
so that they can be used to (eventually) replace rlists.
1993-07-06 00:25:26 +00:00
b274bb4126
add support for checking entry against max user address
1993-07-05 01:53:56 +00:00
955ae6a915
change exit() to kexit(), so prototypes are "safe" with stdio.h, etc.
1993-07-04 23:30:33 +00:00
b21025e4f3
call setregs with a retval, too, so the hp300 & mac, and more works.
1993-07-04 09:48:29 +00:00
2f1182cf07
biodone returns void...
1993-07-04 04:01:38 +00:00
1fd0e7e71f
shminit has void return type.
1993-07-04 03:59:55 +00:00
c5fdaf369a
Blasted ftpd!
1993-07-02 03:11:11 +00:00
50974047ed
Fix bugs in rb_write and rb_cwrite, and make tab handling use rb_cwrite.
1993-07-02 03:04:51 +00:00
c414ffef06
Fix a situation where we might forget to splx().
1993-07-01 00:22:10 +00:00
789d3e468d
Fix broken prototype.
1993-06-30 13:43:03 +00:00
09ff7fc9f4
#endif was somehow missing from the end of a DDB conditional!
1993-06-27 06:22:32 +00:00
3e701a1438
* ansifications
...
* Yuval Yarom's socket recv(2) fixes - access rights problems (see also
uipc_socket.c).
1993-06-27 06:08:40 +00:00
d46fb2c3fb
* ansifications
...
* Yuval Yarom's socket recv(2) fixes, to prevent incorrect blocking and
lack thereof with recv(2) and MSG_WAITALL. Fixes a sbdrop() panic during
some MSG_WAITALL recv(2) sleeps. Access rights fix (also in
uipc_syscalls.c) too. A test program which shows these problems is
available.
1993-06-27 06:08:15 +00:00
6535028529
* ansifications
...
* fixed the arguments given to the line discipline l_open function in
ptsopen()
* should ptsstart() return 1 if (tp->t_state & TS_TTSTOP)?
1993-06-27 06:06:47 +00:00
4d08b29158
* ansifications
...
* question about whether the "tp->t_state |= TS_ISOPEN" in ttyopen() should
be shifted to the end of the block in which it appears.
1993-06-27 06:06:14 +00:00
f74843ee98
* ansifications
...
* cleaned up hardclock() to avoid checking "p" multiple times, and avoid a
gcc2 possible-use-before-initialisation warning.
* changed softclock() timeout callback functions to be of type timeout_t -
a pointer to a void fn(int). No-one was using the second, tick, argument
that was being passed to these callbacks - it is much cleaner to drop the
thing entirely, rather than add a whole heap of casts of dubious
correctness to calls to timeout(), etc. The old style is kept in an
#ifdef, for future reference.
1993-06-27 06:05:32 +00:00
0e6cb953ea
ANSIfications - removed all implicit function return types and argument
...
definitions. Ensured that all files include "systm.h" to gain access to
general prototypes. Casts where necessary.
1993-06-27 06:01:27 +00:00
c34ec14c78
another NDDB -> DDB change. why did DDB invade kern/*?
1993-06-27 01:05:56 +00:00
0ee0a87c29
Fix uninitialized variable.
1993-06-24 21:34:54 +00:00
b2285d45d0
special case raw disk handling return values
1993-06-22 05:19:14 +00:00
b7545aa929
fixing more ansi-style function definitions...
1993-06-22 02:33:21 +00:00
ac6d9dc22e
more damn ansi function declarations. damn you chris....
1993-06-22 01:59:39 +00:00
ee8d969be4
ansi-style function definitions
1993-06-22 01:31:53 +00:00
1ca1339919
> NetBSD 0.8a (TDR) #2 : Mon Jun 21 11:06:03 MDT 1993
...
produces "uname -v" output "TDR#2"
"uname -a" then is..
> NetBSD gecko 0.8a TDR#2 i386
1993-06-21 02:06:10 +00:00
4c0169f317
Fixed ECHONL.
1993-06-20 04:52:53 +00:00
c8bdad0560
convert magic numbers to network byte order, and attendent changes
1993-06-18 06:49:30 +00:00
ce4562fd4b
Find version number for uname.
1993-06-18 02:11:20 +00:00
9ef827c1e0
Clean up deleted files.
1993-06-16 22:06:45 +00:00
0dde03704d
use/add new physio...
1993-06-16 22:06:33 +00:00
9751bf77a7
include the (stupid and ugly) patch for the i386 interrupt stuff.
...
it seems to help stability, when included.
1993-06-15 01:28:29 +00:00
53eb1ac38c
a few changes to select. most notable: don't sleep forever unless
...
user actually wants to. (it could happen in strange circumstances...)
1993-06-14 05:11:58 +00:00
85387a3dea
Believe it or not, selrecord() was broken...
1993-06-13 07:39:32 +00:00
4d785d18cc
handle jolitz physio bug which i mentioned in mail;
...
if you malloc, and get nothing back, panic.
this is totally done wrong, anyway.
1993-06-12 19:54:35 +00:00
d790308a20
Yuval Yarom's 'panic:closef: count < 0' fix to unp_discard().
1993-06-12 14:01:23 +00:00
224ca11730
Leave a hole for AFS
1993-06-10 16:25:01 +00:00
b907fbcbb9
add preliminary support for devfs (e.g. MOUNT_DEVFS #define, etc.)
1993-06-10 01:02:59 +00:00
77e24b59b2
allow modules to be loaded only if device is open for writing
1993-06-07 23:10:09 +00:00
57d3a6d525
add support for terry lambert's loadable kernel modules.
...
needs a bit of cleanup, but overall: SLICK!
1993-06-07 19:51:59 +00:00
24fe3e4563
fixed bug where extra spaces at the end of a shell name would
...
cause shell scripts to be executed incorrectly. shells would
receive an *empty* arg for name of script, open it, the kernel
would say "oh, that's ." or something similar, and the results
would be interesting... found, diagnosed, and killed by me.
1993-06-07 03:09:29 +00:00
f1b8c74730
make getc() and ungetc() be rb{un,}getc(), so getc() and ungetc()
...
don't conflict w/ansi prototypes...
1993-06-06 23:04:42 +00:00
f17f8f678b
make strlen return a size_t
1993-06-06 03:35:39 +00:00
7489eff03c
fix stoopid location to label problem
1993-06-06 02:36:22 +00:00
f67174d697
patch from David Greenman and Rod Grimes (davidg@implode.rain.com and
...
rgrimes@agora.rain.com ) to support 57600 and 115200 baud
1993-06-05 22:40:40 +00:00
165b917efe
add fixes supplied by markus wild
1993-06-05 19:38:13 +00:00
23f01c5982
call a cpu-dependent function to check executable's a_mid type.
1993-06-03 01:31:43 +00:00
dccfb7e206
make strcpy "normal" and get rid of extra (random) spacing...
1993-06-02 23:47:23 +00:00
462e59fb61
two fixes from ws:
...
if resource cur/max limits hosed, fix
copy the correct amount from the rusage struct
1993-06-02 23:46:28 +00:00
d6e1f01ad4
nextpid & maxproc fixes from ws
1993-06-02 23:10:52 +00:00
c26bdfdb86
kill processes outright if they've exceeded their hard CPU limit.
1993-06-02 23:03:45 +00:00
654e213804
when copying out signal trampoline code, make sure that
...
the copyout actually succeeds.
1993-06-02 04:43:54 +00:00
d2bf51450c
fixed minor typo in sigcode copying code
1993-06-02 03:15:36 +00:00
af4017c23f
general cleanup including, but not limited to:
...
better-integreate vmcmd building and exec hdr interpretation,
get stack sigcode copying right
etc.
1993-06-02 02:49:40 +00:00
956c046db5
do address mapping in a new way...
...
it works, but the code needs to be cleaned some...
1993-06-02 00:20:07 +00:00
4206d7f100
break before letting child run, if tracing, and do the right
...
thing with stack limits
1993-06-01 04:52:40 +00:00
0656a4bbbe
final patchups, also add support for finding argv/env via PS_STRINGS
1993-06-01 01:31:39 +00:00
2be76c2f59
Clean up deleted files.
1993-05-31 01:44:57 +00:00
eb4c4de385
replace exec.
...
new one based on one by Wolfgang Solfrank, but majorly hacked
(as it: it didn't have a chance in hell of working, before; didn't
even bother copying out argc, for one...)
1993-05-31 01:44:55 +00:00
2b422f4029
have setregs copy the stack pointer, too
1993-05-30 18:52:39 +00:00
7cef199976
make the sleeps on socket open interruptable.
1993-05-27 15:38:18 +00:00
ada4945e6b
patches to make kernel malloc() retryable. from <MDICKSON@CSI.compuserve.com>
1993-05-27 14:34:29 +00:00
b671aede02
slight functionality change to make the rbpack() and
...
rbunpack() code that I wrote before work cleaner. Many old drivers
call nnn = q_to_b(queue, buf, len), where nnn gets set to len.
for those who are porting:
q_to_b(queue, buf, len) --> rbunpack(queue, buf, len)
b_to_q(buf, queue, len) --> rbpack(buf, queue, len)
ie. they are essentialy drop in replacements now.
1993-05-27 09:29:40 +00:00
2f0c0eb674
tty dynamic allocation
1993-05-26 10:06:41 +00:00
a8a67cbfaf
fix dependencies on M_LAST's old off-by-one error...
...
(arguably, the old error wasn't an error, but, given some
of the old usage of M_LAST and its name, i say it was)
1993-05-25 18:04:17 +00:00
44dcf397e8
move sipcreg into i386/include/reg.h where it belongs, right next to ipcreg.
1993-05-24 11:37:12 +00:00
fe1802950b
add include of select.h if necessary for protos, or delete if extraneous
1993-05-22 11:40:42 +00:00
6fb28b3c74
more disklabel changes
1993-05-20 23:12:53 +00:00
4be87870bf
the id strings are now made immutable by rcs.
1993-05-20 16:20:12 +00:00
9b611f165b
do something better with version strings for created files,
...
also inclusion protect syscall.h
1993-05-20 16:17:38 +00:00
cdcb81b16e
hack on the uname "machine name" stuff for hopefully the last time.
...
now it uses MACHINE, as defined in param.h
1993-05-20 15:36:20 +00:00
b95f963973
add rcs ids, and clean up headers where necessary
1993-05-20 03:53:21 +00:00
48e0fa217b
fix for new syscalls.master versioning
1993-05-20 02:59:55 +00:00
342129da5d
kill the "chmod 444" of the syscall-dependent files. makes running
...
makesyscalls.sh again somewhat more difficult than necessary, sometimes.
1993-05-20 02:57:57 +00:00
230dcf0d05
add $Id$ strings, and clean up file headers where necessary
1993-05-20 02:54:09 +00:00
99a13fab8d
make uname stuff in init_main machine independent
1993-05-20 02:05:39 +00:00
8d6c77881c
make kernel select interface be one-stop shopping & clean it all up.
1993-05-18 18:18:40 +00:00
23af7ea092
clean up comment/rcsid...
1993-05-18 15:38:35 +00:00
f9350c73f0
getpgrp actually takes void. ask the IEEE...
1993-05-16 01:51:51 +00:00
5760558f80
various 8-bit patches from Andrew Chernov <ache@astral.msk.su>
...
tty_compat.c is cleaned up, as is STOP+TIOCSTI in tty.c
1993-05-13 20:58:26 +00:00
37b890c63c
from Luke Mewburn <zak@rmit.edu.au>:
...
add TIOCSTAT ioctl to give load average stats if requested (for tcsh)
1993-05-13 06:42:56 +00:00
62781e28ba
fix kernel printf %b format.
...
from Gordon Burditt <gordon@sneaky.lonestar.org>
1993-05-13 06:17:51 +00:00
64f341ec2a
ring buffer now uses rbchar's (shorts) instead of chars.
1993-05-10 23:15:37 +00:00
e9be58fa40
add dcfclock support from patchkit patch 113
1993-05-07 07:15:19 +00:00
e27c0ba7bc
update for bde's useracc changes. from patchkit patch 137.
1993-05-07 07:06:42 +00:00
7fb852403d
fix uname initialization
1993-05-07 04:09:18 +00:00
c5d886aff7
diffs for uname (posix!) system call, provided by John Brezak <brezak@osf.org>
1993-05-06 10:46:12 +00:00
e811352a74
oops, forgot a c!
1993-05-04 05:48:17 +00:00
501ba00c04
fix accidental 2dd, and fix comm handling for debug
1993-05-03 02:49:27 +00:00
a70911f1aa
add starting, elapsed, user, and system times, and clean up a bit
1993-05-02 20:40:31 +00:00
bd05104f48
add barebones accounting, enabled w/"options ACCOUNTING", via /dev/acct
1993-05-02 09:22:31 +00:00
58f76c7974
Clean up deleted files.
1993-05-02 09:14:24 +00:00
a823bc627d
fix other half of the text file busy problem:
...
don't allow user to exec a file if it's open for writing
1993-04-29 10:12:42 +00:00
c94345dd9a
JOLITZ MANGLED MY PROFILING PATCH!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!!
...
get the i386 profiling assembly in the right ifdef...
1993-04-29 05:38:52 +00:00
c6be5fba43
Give processes 0 and 2 more appropriate names (`scheduler' and `swapper', respectively).
1993-04-28 04:19:25 +00:00
bd4084637f
Add rb_read, to read part of a ring buffer into an ordinary buffer.
1993-04-22 07:49:18 +00:00
e76bede1e3
Don't try to copy in from NULL pointer.
1993-04-19 12:42:39 +00:00
24b644ac3d
Map QMAGIC execs at correct address.
1993-04-18 07:34:34 +00:00
f7450657ec
Allow argument to `#!' (i.e. `#!/bin/foo -argument').
1993-04-17 08:07:03 +00:00
aabb2e137e
Add proper cast.
1993-04-17 01:38:51 +00:00
ed4a9f1a3c
i hate NDDB......
1993-04-15 07:15:25 +00:00
9c736c1dcf
Added QMAGIC, which is a "compact demand load" format: first page is
...
unmapped, and the a.out header is inside the text section.
1993-04-15 05:48:22 +00:00
a434508e6d
version's not supposed to be printed here; it's supposed to be printed
...
in machdep.c
1993-04-10 20:46:16 +00:00
550b9e0bef
forgot to update syscalls.master. the makesyscalls.sh generated stuff
...
is now updated
1993-04-10 20:27:10 +00:00
ff2c06d490
fixed '#ifdef NFS' appropriately.
1993-04-10 20:18:57 +00:00
034707f9e5
fixed to be compliant, subservient, and to take advantage of the newly
...
hacked config(8)
1993-04-10 12:04:35 +00:00
5cc382fc66
add support for PCFS, the MSDOS filesystem.
...
written by Paul Popelka (paulp@uts.amdahl.com ) (patch 129)
invoke w/"options PCFS" in kernel config file.
1993-04-09 19:37:49 +00:00
a128c34f9c
From: Guido van Rooij <guido@gvr.win.tue.nl>
...
when mmapping a file, permissions are checked as it should be. When
mprotect()-ing the address range afterwards, no protection was checked
regarding the protection of the file originally opened. So
when you open /usr/bin/su RDONLY and SHARED you could afterwards change
the mmapped region to READ|WRITE. This gave the possibility to obtain
root privs obviously.
1993-04-09 17:16:30 +00:00
5b67f9cea5
changed order of copyright/version notice (to match 4.4 boot string)...
1993-04-06 14:23:00 +00:00
8d58080ba3
used by makesyscalls.sh to produce init_sysent.c, syscalls.c ../sys/syscall.h
...
edit this to add/delete/ifdef system calls
1993-04-06 10:39:51 +00:00
2a0fbbf7d6
added 'makesyscalls.sh' from BNR/2. Edit only 'syscalls.master'
...
recreates init_sysent.c syscalls.c ../sys/syscall.h which should
then be checked back in.
1993-04-06 10:37:09 +00:00
e9e7ae8a6e
now uses `maxfdescs' to bound `openfiles' resource limit.
1993-04-04 04:32:14 +00:00
95d13869d1
got rid of accidental extra newline
1993-04-03 11:19:42 +00:00
5d07e08579
added changes from Steven Reiz <sreiz@aie.nl> (based on
...
those by Poul-Henning Kamp <phk@data.fls.dk>) to get the kernel
to compile properly when gcc2.* is cc. (should still work
when gcc1.39 is in use.)
1993-04-03 02:15:18 +00:00
fafe485efe
change from make tinguely (tinguely@hookie.cs.ndsu.NoDak.edu) to make
...
opening running programs for writing return ETXTBUSY.
1993-04-03 01:56:45 +00:00
a381c1fd17
now just prints out version. also, got rid of kernel_version,
...
and fixed wfj's trampling on UCB copyright notices.
1993-04-03 01:41:32 +00:00
6efd5159e0
getdomainname fix. had to declare get/setdomainname() implicitly extern.
1993-04-02 20:38:22 +00:00
b02974b037
Jay Fenlason <hack@datacube.com>:
...
1: the fi_readers and fi_writers fields of the fifoinfo structure were not
being initialized to 0. This caused the driver to not sleep the first
process to open the fifo--it thought there was already another process to
talk to (most of the time.)
2: fifo_open() was calling tsleep() without unlocking the inode of the fifo
file. This caused *any* subsequent access to the file (even an ls (!)) to
hang forever. Note that this bug was usually masked by bug #1 above.
1993-04-02 20:19:48 +00:00
355265a866
added: the entire domainanme set. domainmame(1), domainname(2), and a spiffy
...
man page.
1993-04-02 12:24:21 +00:00
942f89c21f
if PANICWAIT defined, will always wait for a key press before reboot
...
after panic. (note that this only makes a difference if DDB is not
installed.)
1993-03-28 08:40:33 +00:00
edca1943a4
fixed typo (in a comment)
1993-03-26 23:40:42 +00:00
e0cd0652b6
removed unnecessary #ifdefing, fixed conf/files appropriately
1993-03-26 22:16:30 +00:00
8cfd6d23f2
changes made by David Greenman's (davidg@implode.rain.com) BPF package.
1993-03-24 23:55:29 +00:00
0be333bc99
Oops. Inserted at the wrong place.
1993-03-24 20:22:51 +00:00
e4eea462cf
Handle one-word cases in word-erase.
1993-03-24 19:41:48 +00:00
93156ce06e
modified files to support kernfs and fdesc fs
1993-03-23 23:56:07 +00:00
690826b089
got rid of hightlighted test, and changed copyright/kernel version
...
string delcarations
1993-03-23 22:16:43 +00:00
e541169ce2
after 0.2.2 "stable" patches applied
1993-03-21 18:04:42 +00:00
61f282557f
initial import of 386bsd-0.1 sources
1993-03-21 09:45:37 +00:00
cgd