4d3fb006eb
kill damned off_t weirdness, also upgrade get- and setrlimit.
1994-05-01 05:02:24 +00:00
e0db92993a
change timeout/untimeout/wakeup/sleep/tsleep args to void *
1994-04-29 23:15:51 +00:00
531f171602
kill an FDSCRIPTS bug which nobody had found yet.
1994-04-29 08:35:29 +00:00
f50f031448
kill syscall name aliases. no user-visible changes
1994-04-29 04:41:02 +00:00
6029e550f1
kill 'aliases'
1994-04-29 04:29:14 +00:00
009e6dd4e9
rearrange some splfoo
1994-04-28 01:17:00 +00:00
6ce04ec890
Routine to display callout table from DDB.
1994-04-27 10:47:51 +00:00
ed7e1b86da
clean up a little bit, and minor optimization...
1994-04-26 05:40:35 +00:00
7f3f130206
minor cleanup
1994-04-25 09:51:56 +00:00
f43af3a756
Remove another bit of that.
1994-04-25 08:47:50 +00:00
ccb0412b7c
Remove a piece of the previous patch.
1994-04-25 08:41:03 +00:00
e4af8f69a5
Minor cleanup.
1994-04-25 08:22:07 +00:00
efe5996f50
Set the file type correctly for stat().
1994-04-25 08:09:59 +00:00
cd6e6bc193
Remove sbselqueue().
1994-04-25 08:08:56 +00:00
22aa92e0a9
kill some code that became obsolete when berknet died...
1994-04-25 05:48:41 +00:00
d071d1cf05
some prototype cleanup, eliminate/replace bogus types (e.g. quad and
...
u_quad) -> use better types (e.g. quad_t & u_quad_t in inodes),
some cleanup.
1994-04-25 03:49:27 +00:00
b1f4730729
some more queue code (that's #ifdef DEBUG)
1994-04-23 08:41:05 +00:00
4917d8beec
make fs types consistent over new kernels. also, some proto foo.
1994-04-23 07:54:38 +00:00
dd617665cc
Clear trampoline bits in execsigs().
1994-04-22 22:33:59 +00:00
3dda0064a5
Convert mount, vnode, and buf structs to use <sys/queue.h>. Also,
...
some knf and structure frobbing to do along with it.
1994-04-21 07:47:31 +00:00
21c5c73db5
I suppose a SIGKILL or SIGTERM should make a process go about its
...
business fast. But the previous situation would allow
"sigblock(sigmask(SIGTERM)); kill(SIGTERM)" to thwart priority policies.
1994-04-19 20:50:30 +00:00
3fe93ccc24
don't let cons dev vnode get subsumed by a 'real' vnode. the current
...
scheme of vnode aliasing just *has* to go.
1994-04-18 21:03:14 +00:00
551e92c16a
start to phase out temp. off_t syscalls
1994-04-16 23:19:06 +00:00
3b4faf4521
slightly loosen lseek restriction
1994-04-16 06:51:58 +00:00
0f8b2efbad
the packet header is at the start of the mbuf chain, not the end.
1994-04-14 21:34:17 +00:00
4be7b669e2
fs types are names now; accompanying changes.
1994-04-14 04:05:28 +00:00
c71822f4b8
FIONREAD returns int, not off_t
1994-04-12 19:41:48 +00:00
b6ad9ec8d5
FIONREAD returns int not off_t. (ssize_t prefered, but standards may
...
dictate otherwise)
1994-04-12 19:18:33 +00:00
2566987063
Fix numerous bugs; flags in the wrong places, modes getting clobbered, etc.
1994-04-12 17:18:59 +00:00
11dad0a608
kill MOUNT_AFS; it's not going into the source tree in the near (long?) term.
1994-04-12 05:03:03 +00:00
6076d8a10d
Fix typo.
1994-04-12 02:23:14 +00:00
0600b23926
Remove a bogus optimization I did.
1994-04-11 23:43:04 +00:00
913fdbc06d
slight optimization, kill unnecessary label.
1994-04-11 22:03:17 +00:00
3b2b4bc307
Add some comments.
1994-04-11 21:54:29 +00:00
bfefadcc74
vcount(): Don't try to vgone() our argument.
1994-04-11 21:37:10 +00:00
47894a50ac
light clean
1994-04-10 00:16:58 +00:00
55dacae6ea
FIONREAD plays with an int, not an off_t.
1994-04-09 22:35:00 +00:00
b9bf37a91f
if MNT_USER is set, let fs authenticate unmount
1994-04-07 07:20:31 +00:00
675a199909
SUGID semantics, similar to 4.4BSD
1994-04-07 00:40:14 +00:00
2396958784
kill two LIBCOMPATs
1994-04-07 00:15:29 +00:00
ed740c7de8
frob arguments a little bit
1994-04-02 08:39:20 +00:00
8abbe136e3
some temporary syscall shuffling
1994-04-02 08:03:17 +00:00
236c2fe9f8
gah. shoot me.
1994-04-02 05:17:04 +00:00
47610af098
some type-changing, simplification, and re-instate casts for phil.
1994-04-01 09:52:30 +00:00
e9dc62de63
Support BSS in LKMs
...
Don't panic when modunloading an empty slot
1994-03-31 20:31:19 +00:00
304e374292
expand uid_t/gid_t/off_t
1994-03-27 09:08:02 +00:00
4a28c23c2f
do the syscall shuffle
1994-03-27 08:34:59 +00:00
ebf601d2d0
bad comment. fix from vdlinden@fwi.uva.nl (Frank van der Linden)
1994-03-19 18:38:33 +00:00
044478d0f7
Ooops. Forgot to move something.
1994-03-18 20:15:39 +00:00
8f86452554
Standardize #includes.
1994-03-18 19:08:28 +00:00
1dbf6bf32e
Clean up uname(2) code some more.
1994-03-18 18:57:13 +00:00
905558ae87
Sparc support.
1994-03-18 08:35:56 +00:00
02dd2aa8ab
fix somebody's typo
1994-03-18 04:01:15 +00:00
fa2051edb1
add hw input flow control support
1994-03-18 03:48:30 +00:00
7dcd783f0c
as pointed out by Phil: all instances of bcopy used in this file should
...
be 'ovbcopy', because they can (and sometimes do) overlap. arguably should
be rewritten to work differently.
1994-03-14 21:15:07 +00:00
b2d32f38eb
A zero scale factor turns profiling off.
1994-03-11 09:12:22 +00:00
ce516ff13d
Make FFS optional
1994-03-09 21:21:37 +00:00
535c793257
syscalls.master updated
1994-03-08 23:57:47 +00:00
ba8c3b1ce4
move LKM syscalls to 210 -> 219, and fix comment
1994-03-08 23:57:00 +00:00
676007bcd2
Don't do TIOCHPCL if !COMPAT_43. In the new world, you use termios.
1994-03-05 22:42:54 +00:00
1827fc3820
small improvement in output formatting of scdebug routines
1994-03-01 07:58:28 +00:00
dc58832610
kill XXX's i marked long ago
1994-02-27 11:53:31 +00:00
a247ecab97
whoops, typo
1994-02-24 01:37:18 +00:00
2ec6ab0aca
hp300||amiga||da30 -> m68k
1994-02-24 01:10:11 +00:00
570c081077
Add da30 to the conditionals for m68k code.
...
(Maybe this should become #if defined(m68k).)
1994-02-24 01:04:50 +00:00
8599bd99bc
Parity setting wasn't preserved across TIOCGETP / TIOCSETP if you
...
had CS7, PARENB and IGNPAR set: TIOCGETP gave you ANYP, which
TIOCSETP interpreted as no parity.
1994-02-24 01:03:15 +00:00
0622d87221
fix incorect check of nprocs vs. maxproc.
1994-02-21 23:10:46 +00:00
f663fac02d
Return a legitimate value from ttylclose().
1994-02-20 17:24:19 +00:00
6adb7d5015
Don't send SIGINFO if ISIG off.
1994-02-20 16:56:56 +00:00
678e80720a
Some formatting changes.
1994-02-20 16:55:10 +00:00
75029eb7b4
#include proc.h.
1994-02-17 14:50:38 +00:00
96fc0a6188
Check `uio_resid' separately (and panic if invalid).
1994-02-16 17:34:01 +00:00
c90756358c
read(): return EINVAL if `count' argument is negative.
1994-02-16 17:23:03 +00:00
a790e23e87
simplify error returns, and fix bugs
1994-02-16 01:21:00 +00:00
d78c3ff463
Some cleanup.
1994-02-15 14:17:07 +00:00
e69b4e2e08
We don't use Aztec C.
1994-02-15 14:02:59 +00:00
83be6ac4f5
Some more cleanup.
1994-02-15 13:35:53 +00:00
022a8ce9c5
add a proc ptr to the syscall debug info
1994-02-15 06:58:07 +00:00
95e17fef9f
add scdebug variable, to enable printing, defaulting to on, and change args
1994-02-15 06:52:25 +00:00
9740a53c8b
add SYSCALL_DEBUG stuff for glass
1994-02-15 06:38:33 +00:00
fbf99acae8
(Hopefully) do the right thing with VTIME > 0 and select
1994-02-14 18:22:39 +00:00
f2fe1704b1
Fix typo.
1994-02-13 11:32:46 +00:00
3406139426
Format police, the end.
1994-02-13 11:31:16 +00:00
22a8203482
Format police, second pass.
1994-02-13 10:20:02 +00:00
e3c4154717
Format police, first pass.
1994-02-13 09:57:12 +00:00
aa8b8cf7c4
Format police.
1994-02-13 09:36:09 +00:00
78bca19b28
KNFify uname code.
1994-02-13 04:18:06 +00:00
a0bdba3231
always good to think about patches before they're included:
...
NEVER pass a "NULL" proc to an ioctl; you never know when a 'real'
proc will be needed. pass curproc, if you don't have one.
1994-02-13 00:10:47 +00:00
2afef02671
undefined variable fix from Mark_Weaver@brown.edu
1994-02-12 15:52:49 +00:00
974e9ba226
Fix typo in last change.
1994-02-12 07:19:11 +00:00
3a46cb73c6
ditto on the last; weird failure modes are the cause of the day.
1994-02-12 07:02:34 +00:00
64517d6f7f
don't forget to deallocate vmcmds if exec fails bounds checks.
...
thanks to charles for pointing it out.
1994-02-12 04:13:43 +00:00
32c2dba8da
Mike Karels' solution to Andrew Herbert <andrew@werple.apana.org.au>'s
...
kern_exit race condition problem. this is a hack:
don't vrele() or vgoneall() if we don't have the vnode
any more. no vgoneall() is necessary, because if the refcnt
is zero, there's nobody to take it away from!
a possible "real" long term solution would be to change the extra ref
into a bit in the vnode's flags, and DTRT in the places where the last
ref is killed. That has some problems, too.
1994-02-11 03:09:17 +00:00
f834e39cf5
Don't need back pointers for disksort().
1994-02-10 15:43:20 +00:00
bbb3e7618c
All ioctl routines take a struct proc * now.
1994-02-09 21:06:46 +00:00
a8aa70d4a5
Remove l_rend and l_meta.
1994-02-09 20:18:17 +00:00
9ea6d6571f
kill support for 'dcfclk' ; shouldn't pollute tree for things not in NetBSD.
1994-02-07 07:33:49 +00:00
57892d4faf
Remove another use of b_actl.
1994-02-06 17:39:48 +00:00
adeebc448e
Use b_actf, not av_forw.
1994-02-06 10:00:30 +00:00
114cf35108
dtrt with exec header size
1994-02-05 02:23:30 +00:00
a2e2253e6c
Remove out-dated comments
1994-02-03 16:17:25 +00:00
425f5d3170
more untimouts needed, from someone at freebsd
1994-02-01 15:09:52 +00:00
c1d3660b28
pass around a `clockframe *' instead of a `clockframe'
1994-02-01 01:50:38 +00:00
1b37a40b3d
Fix that last bug correctly.
1994-02-01 01:04:07 +00:00
92f1bd1c60
Replace a bogus pointer-dereference with something that at least *looks*
...
more sensible.
1994-02-01 00:29:14 +00:00
fa2aa77517
make panic a varargs function
1994-01-30 13:22:05 +00:00
5235180083
Fix spelling error in Copyright notice
1994-01-28 23:43:26 +00:00
55d2821ec0
make sure PT_STEP's are in #ifdefs
1994-01-28 19:06:17 +00:00
6aac4eb9f1
undo totally misguided changes from Andrew Chernov in rev. 1.7:
...
RTS has nothing to do with ttyblock()
also, close a race.
1994-01-28 11:32:43 +00:00
1b2497a000
if you define PT_[GS]ETFPREGS, then you'd better define struct fpreg, etc.
1994-01-28 06:51:13 +00:00
4dc27cd628
add new type
1994-01-28 04:55:41 +00:00
1758eaf2cb
kill hack for vfs_bio jolitz-ish hack
1994-01-27 07:45:21 +00:00
07c0d9025d
get rid of jolitz hack, and add panic() where appropriate
1994-01-27 07:39:32 +00:00
ad869e40c0
kill two patchkit bogons
1994-01-27 03:26:48 +00:00
1a3cc08a27
All hp300 ifdefs apply for amiga too.
1994-01-26 21:58:43 +00:00
9c81b541aa
amiga wants RTC started early, too (like i386 and mac)
1994-01-26 21:55:44 +00:00
ff3c7667ae
remove warning
1994-01-23 19:11:25 +00:00
abf6a6bfdd
more COMPAT_SUNOS changes.
1994-01-23 06:06:21 +00:00
316b930486
pull in COMPAT_SUNOS stuff from magnum
1994-01-23 05:56:27 +00:00
aca2b87e68
don't forget to free old pathname buffer
1994-01-22 06:15:54 +00:00
2cee3e073f
Make procfs really work for debugging.
...
Implement not & notepg files in procfs.
1994-01-20 21:22:47 +00:00
8c996398fc
clean up, break script handling out of check_exec(), and comment a bit.
1994-01-16 03:09:59 +00:00
e11a5f7688
casting for exec_package field changes
1994-01-16 03:08:18 +00:00
92cb33abed
add function for FD handling
1994-01-16 03:07:33 +00:00
135ef10a6d
prevent an explicit splimp->splnet lowering. (which blows up on
...
machines with a levelled-spl system that check for lowering)
de-ansify two function definitions.
1994-01-14 22:20:25 +00:00
af01174969
`extern int cpu' isn't used at all.
1994-01-14 03:41:49 +00:00
bc53d82fa7
fix utimes() to deal with NULL timeval ptr
1994-01-13 23:51:43 +00:00
8ca18c3fa8
use exec_map
1994-01-13 06:23:57 +00:00
7422e2a40c
trim debugging code, for now, and kill dead code (unused options)
1994-01-13 02:33:57 +00:00
f023b7235d
trim extra debugging code, for now
1994-01-13 02:33:28 +00:00
6c7058a8ea
EXEC_DEBUG -> DEBUG, and trim extra debugging code, for now
1994-01-13 02:29:32 +00:00
497cad8b12
Remove the evil #ifndef; we no longer need it.
1994-01-11 23:19:00 +00:00
4448a68e2c
Get rid of disklabel indirection functions.
1994-01-11 16:35:44 +00:00
a537084311
Cleanup.
1994-01-10 06:10:30 +00:00
27a2653bc6
*Really* fix the wakeup code this time.
1994-01-10 05:58:37 +00:00
b58ba6927f
Wake up the blasted process when we detach!
1994-01-10 04:20:29 +00:00
3fbd67b7d3
implement PT_ATTACH and PT_DETACH.
1994-01-09 23:57:15 +00:00
30a6426c82
Fix typo in last
1994-01-09 18:46:17 +00:00
4c8ff3970f
try until error != ENOEXEC, so that formats can report errros reasonably
1994-01-09 17:11:42 +00:00
b0bc24471f
PT_STEP is actually machine dependent. not all machines have it!
1994-01-09 16:52:11 +00:00
528eaca557
Really observe lmc_loadbuf.cnt and
...
correct argument order on execsw restoration.
1994-01-09 16:47:40 +00:00
fd601bd3d6
Ugh. Missed the other. mac=>mac68k...
1994-01-09 15:56:22 +00:00
e53963b58b
mac => mac68k
1994-01-09 15:54:20 +00:00
34ea9285e9
de-macroify kill_vmcmds()
1994-01-08 18:05:29 +00:00
ec0fa1f388
Use NULLVP, just for Correctness (tm).
1994-01-08 16:06:04 +00:00
c6afda5ec5
Trivial stylistic change for consistency.
1994-01-08 15:24:02 +00:00
2079970043
SHM_LOCK and SHM_UNLOCK aren't just HPUXCOMPAT
1994-01-08 15:09:01 +00:00
47529628bb
core reorg
1994-01-08 11:18:56 +00:00
957ce06f24
far reaching but relatively minor cleanup and slight reorg of exec code
1994-01-08 07:14:24 +00:00
d3106cb44a
Fix some casting.
1994-01-08 05:51:17 +00:00
155811b49c
#include vm_kern.h.
1994-01-08 05:29:15 +00:00
e879c04f57
#include vm_user.h.
1994-01-08 04:22:36 +00:00
8955aeedd5
#include vm_user.h.
1994-01-08 04:15:41 +00:00
b418b2c18f
oops; fix that last
1994-01-07 19:26:44 +00:00
163a1d473e
do the *right* thing with resident set size
1994-01-07 19:13:17 +00:00
f70a340fd9
cleanup a printf
1994-01-07 18:03:56 +00:00
d4e74a7cda
really fix the tab code
1994-01-07 17:51:24 +00:00
582eb13c4a
from Mike Karels <karels@bsdi.com>: clear the PENDIN flag if ICANON
...
is cleared, and retain the PENDIN flag if set when ICANON is set
1994-01-05 16:25:37 +00:00
a83f6577b0
minor cleanup; extra spaces, patchkit info, etc.
1994-01-05 13:49:52 +00:00
2c5c3a2774
add support for union and loopback mounts, from jsp
1994-01-04 14:10:48 +00:00
ba67215d35
generalize dupfdopen() to allow dups and moves. from jsp
1994-01-04 12:26:21 +00:00
914aac9116
field name change
1994-01-04 11:29:55 +00:00
1424954c16
print out what we think is the resident set size. very nasty; the field
...
in the proc's vmspace struct should be updated, but isn't...
Also, if the process is a zombie or infantile, don't print, because
that could cause a null pointer deref.
1993-12-30 12:05:16 +00:00
6e81a663bd
OXTABS expansion was putting fewer than the required spaces if the clists
...
became full. we now retry the tab expansion later.
1993-12-24 01:07:40 +00:00
42d0ba8c0a
move getnewfsid() to where jsp says it belongs. update for header
...
files; clear new field and delete old vtype name.
1993-12-22 13:51:14 +00:00
87dfd3227e
cast to match header (changed back...)
1993-12-22 13:46:28 +00:00
146889c8fd
add portal and lofs entries
1993-12-22 13:43:18 +00:00
c3ee44a1b9
various bug fixes from BSDI, via jsp. Kill some iffy fixes
...
by wfj (and replace them with better ones from BSDI). add support
for p_vnode, and add proc_reparent() function, from jsp.
1993-12-22 13:42:27 +00:00
a67d24d94b
add support for p_vnode, from jsp
1993-12-22 13:39:27 +00:00
afce447ada
p_spare is in the 'zero range' now
1993-12-21 10:39:17 +00:00
adbe4b0962
kill two wrong 'case's
1993-12-21 07:19:14 +00:00
621f7b658f
more of the same; gah!
1993-12-21 06:39:12 +00:00
d797e926cf
kill a billism
1993-12-21 06:30:28 +00:00
a6eecc8e2a
load average changes from magnum
1993-12-20 12:42:34 +00:00
bdabe240d0
replace VREF as appropriate to avoid logical recusion
1993-12-20 10:42:39 +00:00
842286f9c4
make searching a bit easier
1993-12-20 05:42:37 +00:00
7f50bd1829
Canonicalize all #includes.
1993-12-18 04:21:37 +00:00
21edb9924e
Canonicalize all #includes.
1993-12-18 03:59:02 +00:00
b4d3382694
Canonicalize all #includes.
1993-12-17 07:56:32 +00:00
493603cecd
fix from Daniel Harris <daniel@werple.apana.org.au>
...
VTIME code must untimeout
1993-12-16 18:09:08 +00:00
e7356e6993
Use proc pointer passed in uio rather then curproc for credentials (should
...
perhaps be done for b_proc also).
1993-12-16 16:28:43 +00:00
252dd4bb0e
VMIN/VTIME support from Marc Teitelbaum <marc@vangogh.cs.berkeley.edu>
1993-12-13 14:40:31 +00:00
e5fc8b4460
update to new syscalls.master
1993-12-12 19:42:13 +00:00
c38503af4d
sparc must flush register windows before vm_deallocate or else fireworks
...
must deallocate shm's
set/reset emulation environment at the right times
keep arg buffer around for later use
canonicalize all #includes.
support STACKGAP for COMPAT_SUNOS code
support OMAGIC/NMAGIC-style memory loading
don't assume VM_MIN_ADDRESS is 0.
(changes come from magnum branch)
1993-12-12 19:38:37 +00:00
9dd14b7930
make syscall#0 "syscall", like in 4.4 (causes SYS_syscall to be defined)
1993-12-12 19:28:21 +00:00
9e62501f7b
support OMAGIC/NMAGIC a.out's
...
cannonicalize #includes
change VTEXT check
(this stuff comes from the magnum branch)
1993-12-12 19:26:18 +00:00
94dd8a0dfb
dtrt with 'error' in setpriority()
1993-12-10 07:19:16 +00:00
3cf1d8505e
echo ^V^? correctly as ^?. From Edward Wang <edward@homer.CS.Berkeley.EDU>
1993-12-09 22:58:17 +00:00
9610135cb8
Do the right thing when arguments are NULL.
...
Don't leave filesystems marked busy after copyout() errors.
1993-12-09 18:00:18 +00:00
644f409712
Properly initialize the credentials fields in the buffer used in physio().
...
Actually use the credentials of `curproc' when doing IO (among other things,
this allows us to read /dev/drum when swapping on an NFS mounted file).
1993-12-07 23:46:13 +00:00
620a222f9f
If EXEC_DEBUG, display name of file we're trying to exec.
1993-12-06 14:18:58 +00:00
86b5a367cf
use progname.core for core dump names
1993-12-04 07:11:11 +00:00
5419d0bd49
initialize uninitialized variable, from Andy Heffernan <ahh@netcom.com>.
1993-12-01 22:05:04 +00:00
392cf19708
return from rmfree() if we loose space with no options left.
1993-11-29 13:21:00 +00:00
2a9aacbf1c
don't permit TIOCSCTTY on /dev/tty, or kernel recursion happens
...
from Lars Fredriksen <ludvig!lars@wixer.bga.com>
1993-11-24 03:40:10 +00:00
bd783b4d80
initialize pseudo devices with pdevinit[], not with a bunch of
...
#include/#ifdef pairs..
1993-11-23 07:49:04 +00:00
facd69a0ee
do something better with lookup return values; suggested by BSDI's msdosfs mod
1993-11-20 09:40:32 +00:00
5838270f1e
patch from Ukai Fumitoshi <ukai@kmc.kyoto-u.ac.jp>
...
to do the right thing with NFS fsid's and getnewfsid()
1993-11-19 02:31:23 +00:00
dde22728d4
drag changes down from magnum
1993-11-15 10:21:08 +00:00
33d816fdbc
add a ptyattach() function, as in magnum.
1993-11-15 09:17:05 +00:00
cgd