92390243e6
always update length; from vixie.
1996-10-08 04:09:58 +00:00
53ded8ebcb
add a hack so that old- and new-toolchain alphas do the right thing:
...
include bsd.own.mk early on, and if !alpha or if !ELF_TOOLCHAIN,
build c++rt0.o. ELF alphas don't use the 'normal' C++ runtime startup
routines. Eventually, there should be better ways to distinguish what
toolchain and object format is in use, and the Alpha C++ startup code
changes should be merged with that used by the other ports.
1996-10-08 00:36:28 +00:00
c7a4944620
update so this compiles with the latest ELF ld.so. Include bsd.own.mk
...
at the top, for ELF_TOOLCHAIN definition, so this can be better
conditionalized for new/old toolchains.
1996-10-08 00:29:48 +00:00
6f72cdccd4
Remove outdated BUGS section (from Mike Long, PR 2804)
1996-10-04 16:59:46 +00:00
8cf4d8ab67
some spaces -> tabs
1996-10-04 03:34:09 +00:00
61dc0e200c
fix pasto: Elf_{Off,Word} rather than Elf32_{Off,Word}
1996-10-03 04:55:31 +00:00
1d7d5ec2f8
add support for real KV addresses, and clean up some.
1996-10-01 21:12:05 +00:00
4540a28ede
Real alpha support, minus actual KV -> physical translation. (Does K0SEG ->
...
physical translation, though, which is enough for e.g. dmesg and vmstat -i.)
1996-10-01 19:04:02 +00:00
2af2958a7d
add code so that if kvm_nlist() failed because __fdnlist() failed,
...
kvm_geterr() actually provides useful information rather than just an
empty string.
1996-10-01 18:56:11 +00:00
ec641b91a7
Fix spelling error, add alpha to list of ports that use new libkvm
1996-10-01 14:38:28 +00:00
51e3315ea1
stub _kvm_pa2off so that new libkvm links
1996-10-01 14:37:00 +00:00
e5629b37cc
clone from nlist_elf32.c (with ECOFF mods). Adds comments,
...
optimization, cleanup.
1996-10-01 13:37:05 +00:00
802cbd1ba7
clean up comments a bit.
1996-10-01 13:35:34 +00:00
561e189aaa
"do over." Rewrite from scratch, or, more precisely, based on a high-speed
...
merge of libc's ECOFF nlist support (written by me, a while ago), and the
(relatively new, but again written by me) kvm_mkdb ELF support.
1996-10-01 00:32:52 +00:00
571692b849
fix a thinko in the check-removal changes
1996-10-01 00:27:38 +00:00
e5699c2fe3
kill recently added 'check' functions, since they only add code to be more
...
expensive, i.e. they save nothing and add code. also, beat it with the
style stick.
1996-09-30 23:51:05 +00:00
80fde3aab4
kill recently added 'check' functions, since they only add code to be more
...
expensive, i.e. they save nothing and add code.
1996-09-30 23:49:27 +00:00
e324d0c9f6
Reflect reality.
1996-09-28 19:08:22 +00:00
9a5eb71903
slightly tune what gets compiled in: On the Alpha, we need ECOFF and
...
ELF64. On other architectures only include a.out support, since we don't
know whether or not they'll have the headers necessary for ECOFF and/or
ELF, and since they only want a.out for now (execpt the mips archs, but
they do Special things). There should be a better way to select which
bits of nlist code get compiled in, but currently there is not.
1996-09-27 22:41:59 +00:00
d1bae5c5d5
rename rcsid to e64rcsid, so it doesn't conflict with the one in nlist_elf32.c
1996-09-27 22:38:06 +00:00
c8365a76b4
Seriously revamp libc's nlist()/__fdnlist() support:
...
(1) split nlist() into multiple files, for clarity and to make
ELF 32/64 support easier,
(2) support multiple executable types at the same time, and
(3) add support for 32- and 64-bit ELF (32-bit ELF support
originally from OpenBSD, but with several bug fixes so
that it actually handles symbols types more correctly
(and therefore _works_ for some of the more tricky uses
of the nlist routines) and changes for 64-bit ELF support).
1996-09-27 22:23:03 +00:00
e9b92001e2
RCS id policing.
1996-09-26 23:45:22 +00:00
b92278aef8
The pc532 port now has string functions in assembler.
...
strcmp.S, strcpy.S and strlen.S are contributed by Randy Hyde.
1996-09-24 09:29:52 +00:00
9ea65c1c28
Changed KMSRCS to reflect new reality (Stringfunctions in assembler).
1996-09-24 09:18:29 +00:00
6a01b1234b
Make it possible to have a md version of memccpy.
1996-09-24 09:14:18 +00:00
95c870f69b
Update for the new internal name for __getlogin().
1996-09-23 02:43:11 +00:00
a500813c14
RCS id police.
1996-09-19 19:39:36 +00:00
f2caf8ea31
Make a quick 64-bit safeness sweep. S/Key now works on the Alpha.
1996-09-19 19:36:51 +00:00
6c68d47b78
Renamed _getlogin syscall to __getlogin, as _getlogin is the namespace
...
clean name for getlogin.
Moved _getlogin.o (now __getlogin.o) from PSEUDO to ASM, as this
syscall really does have a return value (although it will "always"
succeed since it should only be called from libc's getlogin())
1996-09-19 04:50:02 +00:00
731ac8686e
set the GP before accessing errno.
1996-09-17 21:16:12 +00:00
9f3b05068c
* Add Mips ABI changes for ELf shared library support.
...
From OpenBSD, patched into NetBSD by <bouyer@ensta.fr>.
* Include generic NetBSD/mips <mips/*.h> files, instead of old Sprite names.
1996-09-17 01:32:22 +00:00
81b108b45f
Add NetBSD RCSid headers, and change assembly-code LIBC_SCCS id tag string
...
from "Id:" to "NetBSD:".
1996-09-16 18:10:35 +00:00
a5adf9b9ba
reload GP _before_ error check+branch
1996-09-15 18:18:26 +00:00
95c287709b
set HAVE_STRERROR
1996-09-13 17:46:49 +00:00
9118c05a03
RCS ID police: add NetBSD RCS IDs to everything. Since it was imported with
...
normal RCS IDs apparently unmodified (i.e. they were clobbered), remove
the normal Id tags so they cause fewer conflicts later.
1996-09-13 00:29:49 +00:00
9867074973
remove this file, since it's not particularly useful since we've removed
...
most of the files it mentions, and it doesn't really describe the rest.
1996-09-13 00:23:43 +00:00
bd21bf3909
remove some more unused files.
1996-09-12 23:54:37 +00:00
98a5e78a03
clean up 'make test' so that it works even in the presence of
...
'obj' dirs, and added minigzip to CLEANFILES.
1996-09-12 22:33:40 +00:00
5da64b50ae
make it into a shared library, too.
1996-09-12 22:23:40 +00:00
a2157537ee
add libz
1996-09-12 22:21:21 +00:00
7b1b031b12
We do not use these.
1996-09-12 20:10:50 +00:00
e94c324095
Add a NO_DEFLATE conditional that allows this to be compiled without
...
reference to the "deflate" compression module. This can be used to
build a very small zcat for things like install media. (An example
of such a small zcat may be found in src/distrib/utils/zcat).
1996-09-12 19:50:28 +00:00
03e6c4e30f
We do not use these.
1996-09-12 19:40:30 +00:00
de831ea222
Separate interface from implementation so that zlib.h can be
...
included without the other, private header files. Only zlib.h
should be installed in /usr/include; all others are private.
The Makefile has a "test" rule to verify that minigzip.c can
be compiled using only zlib.h (and that it actually works).
Note: This file has extensive modifications.
1996-09-12 19:33:53 +00:00
4222075279
Eliminate use of things private to the library.
1996-09-12 19:26:44 +00:00
ad91ec121b
Fix a "const discarded" warning.
1996-09-12 19:09:00 +00:00
62a507468c
We do not use these.
1996-09-12 19:03:45 +00:00
cccb7a88b4
Reimplement in C, adding ELF support. if compiled with -DECOFF_COMPAT
...
(currently the default), then it compiles on both an ECOFF toolchain
and on an ELF toolchain (but w/o support for dynamic linking or C++
constructors & destructors). If compiled without -DECOFF_COMPAT, it
supports statically- or dynamically-linked ELF binaries (and shared
libraries).
1996-09-12 16:59:01 +00:00
431817b1aa
Import the "zlib" package (version 1.0.4) as libz
...
(That appears to be the usual installed name on *nix machines)
1996-09-12 15:33:09 +00:00
9b9f8829d1
sync with tzcode96l
1996-09-10 22:04:29 +00:00
b27646e2c2
import tzcode96l
1996-09-10 21:59:47 +00:00
7332aa83f6
These files contains stub functions that just called abort() are no
...
longer needed now we have working assembly versions.
1996-09-09 22:06:25 +00:00
10756ef086
Use the new assembly files for fpgetmask(), fpsetmask(), fpgetsticky()
...
and fpsetsticky().
1996-09-09 22:03:24 +00:00
7f53925054
Initial commit of working versions of the functions for manipulating
...
the floating point cumulative exception flags and exception enable masks.
These functions replace the dummy C functions that just called abort().
1996-09-09 22:00:39 +00:00
44279d38cb
user better local labels, be smarter with noat and use of the at register.
...
Suggested by Matt Thomas.
1996-09-09 01:03:38 +00:00
8df506ad69
user better names for local labels, as suggested by Matt Thomas and
...
slightly tweaked.
1996-09-09 00:57:59 +00:00
86a813e847
remove unused local label
1996-09-09 00:51:27 +00:00
dc5f9bfad3
Increment minor version.
1996-09-08 00:03:56 +00:00
c49d89171e
Mention that timeout is in milliseconds.
1996-09-07 21:53:08 +00:00
28a375c45a
Cross-reference poll(2).
1996-09-07 21:51:53 +00:00
56b990d61e
Document the use of INFTIM.
1996-09-07 21:50:07 +00:00
52528cc4f8
Add a poll(2) man page.
1996-09-07 19:05:32 +00:00
f6f3a3525f
Document EFAULT as a valid error.
1996-09-07 17:54:26 +00:00
65e2460e9e
Build a stub for poll(2).
1996-09-07 14:25:51 +00:00
4063ad5adb
#ifdef _KERNEL, #include <lib/libkern/libkern.h> instead of <string.h>.
1996-08-15 22:45:17 +00:00
1e6bbdf957
Add memcmp.c and memset.c to KSRCS (files copied to libkern).
1996-08-15 22:36:59 +00:00
3dc0d5258f
fix xdr_ypmap_parms (missing "&"'s in calls to xdr_ypdomain_wrap_string
...
and xdr_ypmap_wrap_string). prevents segv in ypserv.
1996-08-15 21:43:03 +00:00
082a70252b
Add an RCS id.
1996-08-13 20:13:35 +00:00
961775ef83
Build librmt.
1996-08-13 20:11:18 +00:00
dc06f7ce36
Change an instance of "int" to "unsigned long" (the correct type for
...
ioctl commands under NetBSD) so that this compiles on the Alpha.
1996-08-13 20:09:50 +00:00
caae5cd93d
Use <machine/asm.h>
1996-08-13 17:00:28 +00:00
5b3a9aad8d
Cute little bug with operator precedence in our stdio code. It appears
...
that flushing wasn't done right when input buffers were refilled, due
to a check like
if (foo == _A | _B) which, since _A and _B are both != 0, always
evaluates to true.
Found by proven@cygnus.com
1996-08-13 07:54:00 +00:00
a9347d2469
Fix description of ceil(3). Closes pr lib/2690
1996-08-12 22:39:34 +00:00
aac4c7bf01
add -DFDDI ; closes pr lib/2578
1996-08-10 22:31:19 +00:00
6dd52f02da
as the new reboot(2) will only use the bootstr option if the new RB_STRING
...
flag is set, the major number does not need to change. backing this out.
1996-08-10 00:25:36 +00:00
18393cb155
document RB_STRING.
1996-08-10 00:22:29 +00:00
5ef055ef5c
Phrasing/formatting tweaks, from Kenneth Stailey <kstailey@dol-esa.gov>
...
in PR #2677 .
1996-08-09 19:09:32 +00:00
9c0b5cc2eb
reboot(2)/getpgid(2) changes.
1996-08-09 14:10:45 +00:00
74693079c6
update for new reboot(2).
1996-08-09 10:32:11 +00:00
bb516fa80e
Fix some potential type-size problems.
1996-08-09 10:06:02 +00:00
7693327da9
rsh is now in /usr/bin, not /usr/ucb
1996-08-09 04:17:36 +00:00
8ee4077ea0
minor changes for better integration into NetBSD
1996-08-09 03:39:00 +00:00
67ead26177
Remote mag tape library from volume 18 of comp.sources.unix.
1996-08-09 03:35:17 +00:00
d5df3db65b
Make this with whether __STDC__ is defined or not.
1996-08-07 17:39:44 +00:00
4e058e1f10
Make sure errno is always zeroed before the syscall.
...
Don't trash r0 on exit.
1996-08-07 17:38:16 +00:00
74dedb6ee2
Correct tputs() prototype (PR#2645).
1996-08-03 18:55:40 +00:00
8bcbf9b3bb
Make it correctly return unslept seconds if interrupted for any other
...
reason than SIGALRM. (Was ALWAYS returning 0.)
1996-08-02 17:52:50 +00:00
943f3d2f5c
The argument to isspace() must be representable as an unsigned char.
...
This change casts the argument to an unsigned char so that this
function will work with strings with their 8th bit set.
1996-07-20 01:09:05 +00:00
2d903c854c
Changing type of "c" to unsigned did not fix the problem --- a signed
...
char is sign extended before it is assigned to an unsigned int. This
fix, which has been tested with a different testcase, adds casts to
signed chars which results in proper behavior.
1996-07-20 01:03:54 +00:00
83481cc8e8
Changing type of "c" to unsigned did not fix the problem --- a signed
...
char is sign extended before it is assigned to an unsigned int. This
fix, which has been tested with a different testcase, adds explicit
casts to unsigned char before the value of a character is copied.
1996-07-20 01:00:55 +00:00
24ff249a08
document AUTONICETIME and AUTONICEVAL sysctls
1996-07-17 18:06:36 +00:00
60bbd380f2
pull up bug fix from libkern:
...
>frob the mask so that only the low set bit is set, to avoid lossage.
1996-07-15 05:21:09 +00:00
2f6bb217b1
add an assembly version of ffs().
1996-07-15 03:13:15 +00:00
24de9e4a12
The argument to isfoo() must be representable as an unsigned char, so
...
type of "c" has been changed to unsigned. This change allows these
functions will work with strings with their 8th bit set.
1996-07-11 01:47:38 +00:00
6d8b15101f
Install linkaddr.3 as link_addr.3 and link_ntoa.3 too.
1996-07-10 20:15:48 +00:00
500e663387
Now that setreuid() and setregid() are system calls again, don't call
...
the C library "private" versions. Doing so results in unresolved
symbols the next time a dynamically linked program is run after the
new C library is installed.
1996-07-08 16:42:49 +00:00
688ff1ddf9
Correct think-o; this is an assembler source file (and the Makefile
...
expected as such).
1996-07-08 03:27:25 +00:00
ca7561ba59
Fix typo
1996-07-06 00:15:45 +00:00
7e4136b303
Add a continuation slash in the commented-out i387 section
1996-07-05 22:26:51 +00:00
a0d2fe2548
Thanks to the calling conventions used by some cpus, the argument lists
...
passed to the execl(), execle(), and execlp() functions are in the form
of the argument vector to be passed to execve(). On these architectures,
it is unnecessary to count the number of arguments, allocate space on
the stack, copy the arguments, etc. The vector already on the stack
can be used instead.
When some arguments are passed in registers (like most RISC CPU's), it
should be possible to allocate stack space adjacent such that the
registers can be copied to that memory. After that, the same
approach described above can be used.
This change takes advantage of this on the i386, m68k, and ns32k. It
is probably true on the vax, but I am unable to check. RISC CPU's
probably need assembly language implementations to ensure everything
is placed exactly as needed when registers are copied to the stack...
1996-07-04 07:18:56 +00:00
614420255a
Split all exec*() functions out of exec.c into their own files.
1996-07-03 21:41:46 +00:00
a174b58e2c
Added a new VA_START macro to hide the differences between varargs.h
...
and stdargs.h. This cleans up a lot of conditionals throughout this
file.
1996-07-03 19:53:46 +00:00
9fc92b384e
Add new i387 math functions to ARCH_SRCS
1996-07-03 18:02:43 +00:00
a5e2dfa95f
use one less fp stack slot
1996-07-03 17:31:28 +00:00
0ab3a0b137
It's necessary to use 64 bit precision to avoid errors in the least
...
significant bit(s) (for some operands).
1996-07-03 17:27:34 +00:00
88420f2e4d
this simple adaptation exp() does not work
1996-07-03 17:20:16 +00:00
5f9accdf53
Assembly versions of expf(), logf(),log10f(), scalbf(), expm1() and expm1f().
1996-07-03 16:50:20 +00:00
347a3b2582
Copy over from libkern:
...
> Prevent recursive invocation of mcount() while in kernel by setting profiling
> state to BUSY while in mcount(). This allows profiling to work when mcount()
> calls a profiled routine (for example, udiv on the SPARC).
1996-06-29 14:32:58 +00:00
b021f209c3
Changed to conform to POSIX.2 B.3.2.2 "popen() shall ensure that any streams
...
from previous popen() calls that remain open in the parent process are closed
in the new child process." Fixes PR #2575 .
1996-06-29 00:51:49 +00:00
16e8217f5a
new implementation that simply clears most significant bit
1996-06-26 15:05:40 +00:00
290756c01b
use fscalel instead of ftwotoxl/fmulx
1996-06-26 15:04:06 +00:00
991d9c139a
we don't need to branch
1996-06-26 14:44:05 +00:00
6f5348dbbd
Enable mc68881 version of fmod
1996-06-26 14:39:49 +00:00
3a2de776bf
add m68881 specific fmod()
1996-06-26 14:36:14 +00:00
6655b8bb1f
enable alpha-specific libm files
1996-06-26 07:38:42 +00:00
c13e010e3c
Alpha versions of copysign() and copysignf()
1996-06-26 07:29:59 +00:00
5743dc5d7a
Fix from Bruce Evans. Commited to FreeBSD earlier this month, but was
...
never forwarded to me :-(.
> Clean up the FP stack before returning. The i387 exp() leaked an FP
> register on its first call. Subsequent calls reused the register so
> the leak didn't accumulate.
1996-06-25 02:07:09 +00:00
fa72190309
Fix the empty error message problem more completely. When the magic-check
...
fails on the cpu_hdr, return 0 instead of -1. This allows the caller to
distinguise between fatal errors and 'nothing there'.
1996-06-23 20:28:05 +00:00
cca068a144
Fix the mysterious empty error message when savecore is executed and no
...
dump is present. This was caused by the fact that kvm_dump_mkheader() was
called *before* savecore checks the dump magic and kvm_dump_mkheader() returned
-1 without setting an error message. The latter is fixed now.
1996-06-23 13:56:54 +00:00
601306abca
Resurrect the setre[ug]id() stubs.
1996-06-23 11:52:49 +00:00
23558a0acd
Clean up removed files.
1996-06-23 11:50:33 +00:00
f699ec835f
Remove the old (broken) setre[ug]id() emulation.
1996-06-23 11:27:59 +00:00
3a375fb249
Merged in changes from branch (make sure all memory, file descriptors,
...
mappings, etc. are freed and return (nl_catd) -1 on error). Also fix
one minor bug and clean up NLSPATH parsing a bit more.
1996-06-21 06:21:04 +00:00
749fcbc59b
Ensure user can't make us overrun fixed sized buffers.
1996-06-20 18:47:08 +00:00
c40394d175
Fix NLSPATH processing --- catopen was not processing beyond the first
...
colon. Try all paths in NLSPATH as per XPG4.2. Expand %l, %t, and %c
to nothing. Should be language, territory, and codeset.
1996-06-20 14:54:38 +00:00
0df251764e
Changed clnt_{s,}p{errno,error,createerror} so that the functions that
...
output emit newlines and those that return a pointer to string do not
append them.
The existing behavior was inherited from RPC4.0, but this seems to have
been corrected sometime before SunOS 4.1.
The documentation has been updated to reflect the new behavior.
1996-06-19 20:38:43 +00:00
1d4349e7c8
ypset(8) stopped working; reason:
...
xdr_ypbind_setdom() would always fail because the test for
xdr_ypdomain_wrap_string() was reversed. Fixed this and in
the process made all the function tests to be of the form:
if (!xdr_foo())
return FALSE;
....
return TRUE;
instead of having some of them like above and others like:
if (xdr_foo() == FALSE)
return FALSE;
...
return xdr_bar();
this is more consistant now and hopefully in the future people
will pattern match correctly and not introduce spurious errors.
1996-06-18 20:05:59 +00:00
418b1071e0
include bsd.own.mk at the top, so checks of EXPORTABLE_SYSTEM work as
...
intended.
1996-06-14 17:22:05 +00:00
0396d19a02
use p->hashfraction when doing non-time-critical calculations, rather than
...
using HASHFRACTION directly. in time-critical calculations, if HASHFRACTION
is a power of two, check that p->hashfraction == HASHFRACTION and if so do
the calculation with the compiled-in value so that the compiler can optimize
out (potentially) expensive divisions. if p->hashfraction != HASHFRACTION,
actually do the division. This has the result that on machines with slow
division, the division can be optimized out of the common case, but that
if HASHFRACTION changes from the compiled-in value (for whatever reason),
profiling will still work. Changes suggested by Chris Torek.
1996-06-12 04:15:34 +00:00
88fce69ea9
s/KERNEL/_KERNEL/ to match libkern version.
1996-06-12 04:03:33 +00:00
29811afbb6
Set r1 to -1 on exit as well as r0 otherwise syscalls returning 64 bit
...
results will not read -1.
1996-06-05 19:02:12 +00:00
84fc678011
Use setne instead of setnel, new versions of gas balk at the bad mnemonic.
...
Wrong mask was used in s_finitef.S.
1996-06-04 18:00:34 +00:00
23beafab63
Document that the caller must do a pw_abort() if pw_mkdb() fails.
1996-06-02 19:26:18 +00:00
e257708bf2
Don't delete the lock file if pwd_mkdb fails. The caller will take care
...
of it by calling pw_abort() or pw_error() as appropriate.
1996-06-02 19:25:43 +00:00
ae94fc4d74
Note functions that probably can be removed
1996-06-01 21:39:21 +00:00
7c04139bcd
Fix computation of ISO 8601 Week Of Year
1996-06-01 21:23:25 +00:00
8e1fdaa50e
Don't special case shift counts > QUAD_BITS or == 0. The former is
...
undefined behavior (so we can do whatever we want) while the latter
doesn't happen much in actual practice.
These changes make these functions considerably smaller and faster.
1996-06-01 21:21:50 +00:00
ab5dfe6dc8
merge bugfix from 1.2 branch: use includes target for include files
1996-06-01 19:59:30 +00:00
17c5522d18
Re-arranged code a bit so that gcc can optimize this a bit better.
...
This change results in better code on the i386, m68k, & ns32k, and
equivalent code on the sparc.
1996-06-01 02:24:29 +00:00
fb600c331a
Fix typo in last change
1996-06-01 02:06:23 +00:00
5bbcde6a0d
Enable use of assembly language shift routines on the m68k
1996-06-01 01:14:00 +00:00
57ceb7a883
Added ashldi3.S, ashrdi3.S, and lshrdi3.S to KMSRCS
1996-06-01 01:08:02 +00:00
ff3bd4df33
m68k assembly versions of 64 bit shift functions
1996-05-31 02:05:33 +00:00
0351ca5efa
Add some casts so that we're not bitten by type promotions on 64-bit
...
systems. Not a great concern since the only 64-bit system currently
supported by NetBSD (Alpha) uses assembly versions of these routines.
1996-05-29 23:23:27 +00:00
7a26f6a8ac
Pull down yp/xdr fixes from release branch.
1996-05-29 20:00:52 +00:00
c0e6f9b4f3
use 'ALIGN' from <sys/param.h> rather than generating a custom alignment expr
1996-05-29 04:05:52 +00:00
d792665e0d
Fix typo (PR 2461)
1996-05-28 13:34:39 +00:00
343a54bb75
Change iuserok() to take u_int32_t. Some netgroup fixes (partially
...
from OpenBSD and Theo).
1996-05-28 02:07:32 +00:00
ebd16dc1e1
switch from long to int32_t, where appropriate, so that this actually
...
works on NetBSD/Alpha for non-trivial filters.
1996-05-26 23:50:37 +00:00
409a9590f3
Avoid using clnt_perror() on the first rpc failure, since it is going to be
...
retried anyway. Only report it every _yplib_nerrs times.
1996-05-23 13:48:59 +00:00
mrg