12347b2657
Make uvm_vslock() return the error code from uvm_fault_wire(). All places
...
which use uvm_vslock() should now test the return value. If it's not
KERN_SUCCESS, wiring the pages failed, so the operation which is using
uvm_vslock() should error out.
XXX We currently just EFAULT a failed uvm_vslock(). We may want to do
more about translating error codes in the future.
1999-06-17 15:47:22 +00:00
f4628e2fda
FILE_UNUSE needs two args, not one.
1999-06-13 21:34:32 +00:00
6e2b65ec1d
regen
1999-06-09 12:38:50 +00:00
b4e4c530db
add "sys_" prefix to LFS syscalls
1999-06-09 12:33:18 +00:00
354ba222e6
Regen
1999-06-09 04:50:32 +00:00
b7a0cd53a2
Propagate kern changes
1999-06-09 04:50:10 +00:00
4ebea99d8e
Regen
1999-06-09 04:28:24 +00:00
1e6c35821b
propagate changes from kern
1999-06-09 04:27:44 +00:00
3da80f5ebd
don't vrele a bogus pointer if the namei() of the loader fails
1999-05-28 04:39:14 +00:00
bd84136493
copyinst() takes a `size_t *'.
1999-05-27 14:03:05 +00:00
c0567a6bc7
Add support for SETALL and GETALL to the Linux emulation of semctl()
...
and improve handling of SETVAL. Patches supplied by YAMAMOTO Jiro in
PR kern/7621.
1999-05-27 13:30:40 +00:00
497248ca55
XXX Pass VM_PROT_NONE to uvm_vslock() as access_type. Why are we even
...
vslocking here?! copyout() on its own seems to suffice just about everwhere
else, and it's not like the process is going to exit; it's in a system
call!
1999-05-26 01:07:06 +00:00
a2dc12021c
Oops, completely forgot to translate the exit signal in clone(2).
1999-05-17 19:26:33 +00:00
0f74e48ca5
Emulate the Linux {get,set}resgid(2) system calls.
1999-05-14 18:45:31 +00:00
8a61761cc5
Emulate the Linux {get,set}resuid(2) system calls.
1999-05-14 18:44:50 +00:00
dd623ce325
Regen.
1999-05-14 18:44:15 +00:00
a03622b33f
Emulate the Linux vfork(2) (which is like NetBSD's __vfork14(2)),
...
{get,set}resuid(2), and {get,set}resgid(2). Note a few more system
calls that we don't yet emulate.
1999-05-14 18:43:58 +00:00
5f1b28961e
Regen.
1999-05-14 18:43:20 +00:00
e0a6808cef
Emulate the Linux vfork(2) (which is like NetBSD's __vfork14(2)),
...
{get,set}resuid(2), and {get,set}resgid(2). Note a few more system
calls that we don't yet emulate.
1999-05-14 18:43:01 +00:00
64d90992b4
Regen.
1999-05-14 18:41:20 +00:00
ba8fcafa16
Emulate the Linux vfork(2) (which is like OSF/1 vfork(2) and NetBSD's
...
__vfork14(2)) and {get,set}resuid(2). Linux does not have {get,set}resgid(2)
on their Alpha port!
1999-05-14 18:39:38 +00:00
087adcea52
Regen.
1999-05-13 23:43:15 +00:00
fa1c5fed1a
Emulate the Linux clone(2) system call.
1999-05-13 23:42:34 +00:00
053ec76d86
Fully emulate the Linux wait4(2) system call, including the __WCLONE
...
flag (which is equivalent to our new WALTSIG).
1999-05-13 01:00:50 +00:00
5512f35a12
Define the flags used by the Linux wait4() system call.
1999-05-13 00:31:57 +00:00
e2c674e2c6
Add scheduling-related definitions. Currently we have the flags passed
...
to the __clone(2) system call.
1999-05-12 19:49:09 +00:00
c283289b46
For lack of a bit, the letter was lost. For lack of the letter, the
...
macro was lost. For lack of the macro, the function was lost. For lack
of the function, the syscall was lost. For lack of the syscall, the
application was lost. For lack of the application, my time was lost.
1999-05-10 21:41:07 +00:00
9ee892559b
regen
1999-05-10 06:00:10 +00:00
798cd0a332
start at (new) sendmsg and recvmsg. right now, the former doesn't
...
handle control messages and the latter is just a stub.
1999-05-10 05:58:44 +00:00
55c7aec3cc
rename xopen-ish iovec and msghdr structs
1999-05-10 04:18:22 +00:00
21c7b563a8
msghdr and x/open iovec structs
1999-05-10 03:58:50 +00:00
99cc65e9ba
regen
1999-05-10 03:34:20 +00:00
4bf16b94f4
old accept -> compat_43_sys_accept
1999-05-10 03:33:04 +00:00
bf29d6c987
regen
1999-05-10 03:00:27 +00:00
1c05179cd4
usleep_thread: round ticks up, don't sleep forever, use better wchan
1999-05-10 02:55:57 +00:00
1e2f18cb6a
listen is the same as native
1999-05-10 02:54:57 +00:00
f7c70eaf0e
regen
1999-05-10 01:59:41 +00:00
726fed408a
plink, plink.... socketpair!
1999-05-10 01:58:37 +00:00
fe0ce0cd42
Need to explicitly include <machine/cpu.h> for a cpu_wait() macro definition.
1999-05-08 01:23:01 +00:00
192b3c733c
Build libraries for kernel/standalone code from within their compilation
...
directories and use .PATH to lookup the source files.
(Formerly, the libs were built from the source dirs, with MAKEOBJDIR
set to the compilation directory.)
This solves 2 problems:
-"mkdep" and "make" are now consistent about the file lookup, this fixes
bad interactions with amd reported in PR bin/7374 (Arne Juul) and
lossage reported by Andrew Gillham ("obj" dirs and relative paths still
don't work well together)
-kernel compile trees can be moved around without forcing a new
"make depend" - fixing PR kern/4021 by Martin Husemann
1999-05-07 14:28:50 +00:00
e3669c3393
Add "use counting" to file entries. When closing a file, and it's reference
...
count is 0, wait for use count to drain before finishing the close.
This is necessary in order for multiple processes to safely share file
descriptor tables.
1999-05-05 20:01:01 +00:00
3b5a8a7db7
shuffle (sort)
1999-05-05 01:51:31 +00:00
5e1a5ebf21
clean up sigset_t manipulation macros
1999-05-05 00:57:43 +00:00
ceff265279
shuffle (alphabetize)
1999-05-05 00:48:30 +00:00
ff38fb6396
regen
1999-05-04 05:23:05 +00:00
e7f2a6f938
stub classcntl(). for now, it returns ENOSYS. (a stub for classcntl()
...
was suggested by Anders Magnusson <ragge@ludd.luth.se>.)
1999-05-04 05:17:22 +00:00
e24a4116bc
add a few XXXs
1999-05-04 05:11:03 +00:00
debeb21304
regen
1999-05-04 05:00:21 +00:00
ab7cdebda6
emulate shm*(), but only so far as to return ENOSYS for now (until
...
they're really needed). put them in their own file because when we
need to emulate them, they'll need a bunch of code.
1999-05-04 04:59:21 +00:00
2165378111
add a 'clean' target to nuke files built from syscalls.master, so that
...
it's easier to commit non-generated files all at once.
1999-05-04 04:56:44 +00:00
d0937ee8b6
don't pull in osf1_syscall.h
1999-05-04 02:45:35 +00:00
15e8e0d207
IPC key_t
1999-05-04 02:44:43 +00:00
759a1e3cd7
regen
1999-05-04 02:12:33 +00:00
bd7a54454d
pathconf/fpathconf
1999-05-04 02:12:15 +00:00
a0a7b0de30
protection-related bits -> osf1_prot.c
1999-05-01 05:49:01 +00:00
ca7738a519
clean up ordering, add a comment or two
1999-05-01 05:43:52 +00:00
12df8bf0da
{get,set}rlimit, getrusage -> osf1_resource.c
1999-05-01 05:41:56 +00:00
8d9e0684d2
descriptor-frobbing syscalls (fstat, fcntl(), lseek(), etc.) ->
...
osf1_descrip.c.
1999-05-01 05:34:59 +00:00
39ba96ee96
utimes() belongs here
1999-05-01 05:33:36 +00:00
789e810c20
get/settimeofday, interval timers -> osf1_time.c
1999-05-01 05:25:37 +00:00
bf9d06e035
typical filename using syscalls (open(), access(), execve(), etc.) -> osf1_file.c
1999-05-01 05:18:01 +00:00
deeb289def
sys_generic-ish bits move to osf1_generic.c
1999-05-01 05:06:46 +00:00
f493e92106
clean up vm-related headers
1999-05-01 04:59:42 +00:00
d3b6da7da3
and the socket calls move
1999-05-01 04:57:15 +00:00
ffc61cf092
shuffle some conversion bits around that were missed earlier
1999-05-01 04:47:00 +00:00
9cc30ed036
needs mount.h
1999-05-01 04:38:39 +00:00
2d25ceb2d2
pull in compat_util.h
1999-05-01 04:38:20 +00:00
e42f7d331f
move mmap() and related functions into their own file
1999-05-01 04:34:20 +00:00
170b003d0b
shuffle bits around. separate conversion conversion functions and tables
...
from the code that uses them, just as I previously separated out the
OSF/1 definitions and structures from the code that translates to and
from them.
1999-05-01 02:57:08 +00:00
38da682259
Moved to osf1_cvt_errno.c,v
1999-05-01 02:16:01 +00:00
333154d794
Moved from osf1_errno.c,v
1999-05-01 02:16:00 +00:00
af8a9059d3
compute the value to be used for e_arglen via howmany(value, sizeof (char *)),
...
rather than just "value / sizeof (char *)". While in these cases the latter
works as well, the former's a better example for other uses to copy (since
it handles the non-integral multiple case correctly).
1999-04-30 23:07:01 +00:00
1bce8c40fc
ep_arglen is in units of 'sizeof (char *)', not in units of bytes. use
...
howmany(value, sizeof (char *)) to get the right value.
1999-04-30 23:02:06 +00:00
5a93f3c9b1
emulate wait4(), put some guts into sysinfo(), and redo fcntl()
...
F_{GET,SET}FL emulation.
1999-04-30 05:25:34 +00:00
28410181b2
return error if unsupported sigaltstack flags used
1999-04-30 05:24:04 +00:00
a928d381b2
regen
1999-04-30 04:36:04 +00:00
d2d8e961fe
emulate wait4, clean up sigreturn args, emulate {get,set}domainname
1999-04-30 04:33:51 +00:00
e836b093b4
nuke most fcntl() F* flags, and note that O_* should be used instead.
1999-04-30 04:31:45 +00:00
c21c5d0c2d
systeminfo.h SI_* constants, wait.h W{NOHANG,UNTRACED}, plus spacing cleanup
1999-04-30 02:12:03 +00:00
afe914207b
we can actually setitimer() now
1999-04-30 01:57:27 +00:00
94a4e36ed2
use osf1-specific sigcode
1999-04-30 01:46:09 +00:00
6abb3f7b37
regen
1999-04-29 23:39:55 +00:00
59c243a0b7
emulate setitimer(), but for now just return EINVAL. stub sysinfo() (XXX).
1999-04-29 23:37:23 +00:00
e9bc2c7ae3
need to emulate setitimer and sysinfo
1999-04-29 22:08:49 +00:00
2b9f4ca12d
in NetBSD, top 32 bits of ioctl com are always 0
1999-04-29 17:34:49 +00:00
0792acce0d
Regen.
1999-04-29 17:20:37 +00:00
fb1f070eba
Add entry for set_program_attributes().
1999-04-29 17:19:47 +00:00
8ef6e2a635
Implement set_program_attributes().
1999-04-29 17:18:59 +00:00
4c1bb7794a
Need sunos_sigcode and sunos_esigcode now that our sigreturn is
...
different.
1999-04-29 16:14:04 +00:00
a38ee86714
Make ss_size something reasonable (not 0) so sigaltstack does not fail.
...
Pull me up plz!
1999-04-29 16:09:12 +00:00
a509481404
itimerval and related constants
1999-04-29 05:54:13 +00:00
ee325de1bb
regen for syscalls.master
1999-04-29 02:11:51 +00:00
1041a1c770
emulate access() and mprotect(), hook into COMPAT_43 get{peer,sock}name
1999-04-29 02:10:07 +00:00
f4507cddca
mprotect() and access()
1999-04-29 02:08:58 +00:00
cc4f9f6afb
access() flags
1999-04-29 02:06:50 +00:00
52b80ff167
fix some bugs (good to get to the point where bug-fixing is an issue!):
...
* when copying out the timezone in gettimeofday, don't accidentally copy
out the timeval again (into the timezone buffer).
* mmap() PROT_WRITE and PROT_EXECUTE imply PROT_READ.
1999-04-28 06:01:51 +00:00
278f92716f
regen for updated syscalls.master
1999-04-28 05:10:38 +00:00
d3708a1699
more timeval-related lossage. fixed some fns, added settimeofday & utimes
1999-04-28 05:09:47 +00:00
fc1142ac44
be more careful about getting the exec_name emul arg field
1999-04-28 02:49:38 +00:00
a61beae31e
regen for changed syscalls.master
1999-04-28 02:34:47 +00:00
a9deaa593a
emulate select(), to get the timeval right.
...
And CGD said, "Let there be eyes!" And there were eyes, and it was good.
1999-04-28 02:34:25 +00:00
4c9a03c3b3
do easy 'f'-class ioctls
1999-04-28 02:16:07 +00:00
3cca776d42
regen for changed syscalls.master
1999-04-28 02:03:45 +00:00
28561f9679
can't use system gettimeofday, and need to implement uname
1999-04-28 02:03:13 +00:00
93e2cb1a1e
uname, gettimeofday
1999-04-28 02:02:50 +00:00
ccd0018aca
uname and timezone structures, and some minor cleanup
1999-04-28 02:00:36 +00:00
6e5469dc42
implement osf1_sys_madvise()
1999-04-27 18:45:22 +00:00
3cb15b5b9b
regen for changed syscalls.master
1999-04-27 18:41:37 +00:00
df6ae494ef
fix madvise args
1999-04-27 18:40:53 +00:00
99df19a561
madvise operation definitions
1999-04-27 18:38:07 +00:00
3d0c9ec2a8
kill a couple of debugging printfs i accidentally left in
1999-04-27 17:56:52 +00:00
8f40430549
regen for changed syscalls.master
1999-04-27 17:53:11 +00:00
6b9bb2a3cb
fix getrusage args
1999-04-27 17:52:44 +00:00
009c460d25
emulate getrusage, add some XXXs
1999-04-27 17:50:59 +00:00
9a71b695bd
osf1_timeval and osf1_rusage and related constants
1999-04-27 17:50:23 +00:00
44cbebec30
note that his should be shot from a cannon
1999-04-27 16:09:28 +00:00
1f9cfb5d53
update for current reality
1999-04-27 16:08:40 +00:00
b69c7692af
From cgd again: pbuf == path on failure.
1999-04-27 15:42:37 +00:00
c4f1bab08e
Make the usage consistant and document it:
...
if the path was given in a user buffer, return the new path in
a user buffer.
if the path was given in a kernel buffer, allocate a new kernel buffer
to return the new path; the caller is responsible for freeing the buffer
if we fail always set the new path to NULL.
This usage broke when I adde the /../ shortcut. Noted by cgd.
1999-04-27 11:36:47 +00:00
f355b96b7a
note that this document is now out of date
1999-04-27 06:39:49 +00:00
7e3ad9e963
evil hack to work around non-fixed mmap requests
1999-04-27 06:37:12 +00:00
6ef8f59ae6
dynamic executable/loader handling, done right
1999-04-27 05:38:08 +00:00
743038a6cb
clean up for ecoff_machdep.h changes, look for the loader
1999-04-27 03:19:44 +00:00
fe6198f6e2
do special OSF/1-specific ECOFF recognition (not yet complete), and
...
implement a copyargs function to do special argument copying for
emulated processes.
1999-04-27 01:45:03 +00:00
bd33916784
exec internals have been moved (and improved) to osf1_exec.c. Not much
...
there yet, but more to come.
1999-04-26 18:34:05 +00:00
bd947212ac
major(), minor(), and makedev()
1999-04-26 06:10:36 +00:00
9524149336
use fcntl() flags from osf1.h, map all of them
1999-04-26 05:57:53 +00:00
1904f78fb5
fcntl() flags
1999-04-26 05:56:34 +00:00
6d24aedc2c
table-based translation of sigaction and sigaltstack flags
1999-04-26 05:35:08 +00:00
635a7843a1
copyright notice
1999-04-26 05:32:18 +00:00
6dbaac99f9
slight cleanup of spacing
1999-04-26 04:29:43 +00:00
35a22b65fc
rlimit stuff
1999-04-26 03:30:48 +00:00
646c8f519a
resource limit definitions
1999-04-26 03:29:05 +00:00
5d3c1d0c49
use osf1.h definitions and some table-based flag translation
1999-04-26 03:10:58 +00:00
32c9cd9866
exec and mount structs/defns, and some cleanup
1999-04-26 03:07:40 +00:00
445556f304
add a section of OSF/1 types, and add many more OSF/1 structures and
...
definitions.
1999-04-26 01:24:26 +00:00
1d0207c926
const-ify flags translation tables.
...
kill some cruft, and move various flags and structure definitions into
osf1.h (not done yet).
Do a better job of translating fcntl(), and translate flags in sento()
emulation.
1999-04-26 01:23:01 +00:00
196564f8e3
remove some structs and def'ns; they're moving to osf1.h
1999-04-26 01:22:18 +00:00
637226f0ad
pull in osf1.h
1999-04-26 01:21:43 +00:00
ad10b95967
translate EAGAIN/EWOULDBLOCK properly
1999-04-26 01:16:04 +00:00
8a8e8d144f
Shut 'gcc -Wuninitialized' up.
1999-04-25 02:42:01 +00:00
96b63fc788
Don't defopt COMPAT_NETBSD32 here; it's done in sys/conf/files.
1999-04-24 19:17:33 +00:00
45a020863e
substantial cleanup, including:
...
* do more flags conversions, using the table driven converter.
* check for alternate (emulation) path in lots more places.
1999-04-24 07:23:54 +00:00
ed106195ea
cleanup: kill syscall args comments
1999-04-24 07:13:23 +00:00
d081124ca2
cleanup: nuke register
1999-04-24 07:09:49 +00:00
d7c211e583
cleanup: kill syscall args comments
1999-04-24 07:06:35 +00:00
dd4d52e331
slight hdr inclusion cleanup, get ioctl-related def'ns from osf1.h
1999-04-24 06:59:11 +00:00
8b1bcd5e9c
pull in osf1.h, not osf1_errno.h
1999-04-24 06:58:26 +00:00
2fc0019c7a
Collected OSF/1 definitions and structures, sorted by OSF/1 header.
...
Error numbers (errno.h) aren't here, since they're likely to change
(additions) more often.
1999-04-24 06:56:25 +00:00
5b217baeb1
implement a simple function, emul_flags_translate() to do table-based
...
flags translation.
1999-04-24 02:56:06 +00:00
thorpej