28a8ddfdd3
Need <stdlib.h> for malloc() protoype (to avoid warning on LP64).
...
XXX This patch probably isn't right, but I'm not interested in
detangling the GNU include file nightmare right now.
2001-10-05 18:47:42 +00:00
c3aaaae7a0
Provide an extern decl for strlen() all the time, not just if __alpha__
...
(so this builds on sparc64). XXX Why we don't just use <string.h> is
beyond me...
2001-10-05 07:38:22 +00:00
4e3296004c
Need <string.h> for memset()/memcpy().
2001-10-05 07:35:04 +00:00
044932d6cc
Need <string.h> when building the exception handler stuff on LP64
...
systems (like sparc64).
2001-10-05 07:04:04 +00:00
bf8eb2fc2d
Generate ldcripts as "native" for a native linker. However, disable
...
installation into /usr/share/ldscripts at the moment, as the scripts will
no longer be shareable on all targets. This will be tweaked at a later
date to generate "cross style" scripts for all targets (native ones are
compiled into the ld binary) so that they will indeed be shareable.
Should fix PR bin/14114, pkg/14122, and related issues.
2001-10-04 17:46:22 +00:00
eff7f464bc
Pick up warning fixes from rev 1.115 of bintutils cvs. Part of the following
...
change, though I'm only picking up elf32-mips.c here.
2001-09-21 Nick Clifton <nickc@cambridge.redhat.com>
* elfxx-ia64.c: Fix compile time warning messages.
* coff-mcore.c: Fix compile time warning messages.
* coff-ppc.c: Fix compile time warning messages.
* coffcode.h: Fix compile time warning messages.
* elf32-mips.c: Fix compile time warning messages.
* elf64-alpha.c: Fix compile time warning messages.
* libbfd.c: Fix compile time warning messages.
* bfd-in2.h: Regenerate.
2001-10-04 04:03:50 +00:00
34876aa7e9
Implement `info float'.
2001-09-26 03:41:52 +00:00
5554debffd
Add a missing va_end.
2001-09-24 14:33:50 +00:00
4c99916337
va_{start,end} audit:
...
Make sure that each va_start has one and only one matching va_end,
especially in error cases.
If the va_list is used multiple times, do multiple va_starts/va_ends.
If a function gets va_list as argument, don't let it use va_end (since
it's the callers responsibility).
Improved by comments from enami and christos -- thanks!
Heimdal/krb4/KAME changes already fed back, rest to follow.
Inspired by, but not not based on, OpenBSD.
2001-09-24 13:22:25 +00:00
b77ae95a9e
For ns32k-netbsd, don't add ns32k/xm-netbsd.h to $xm_file; it gets
...
included in tconfig.h anyway.
2001-09-23 13:55:13 +00:00
8e868c5d17
Don't define mem{cmp,cpy,set}; from gcc-current after the 2.95.3 release.
2001-09-23 13:42:39 +00:00
531012d19e
Skip the "includes" phase of gnu/libexec, gnu/usr.bin, and gnu/usr.sbin.
2001-09-21 21:04:59 +00:00
3ab74bd348
Set SHLIB_{MAJOR,MINOR} explicitly instead of using shlib_version files.
...
(This currently means that checkver doesn't work with these libs, but that
will be fixed.)
2001-09-21 15:38:15 +00:00
456dff6cb8
Spell 'occurred' with two 'r's.
2001-09-16 16:34:23 +00:00
1f8d9cbdd1
pull patch from openbsd Errata 033:
...
A security hole exists in uucp(1), uux(1) and uuxqt(1) that may allow
an attacker on the local machine to run arbitrary commands with
root privileges.
Fix:
Pay more attention to commandline parsing of long options.
Patch from jbj@redhat.com via millert@.
2001-09-12 07:51:03 +00:00
2db48f4c88
Add "Remember to update distrib/sets..." lines (and NetBSD RCS IDs in
...
some cases).
2001-09-10 11:18:41 +00:00
6be82855aa
Final config file for VAX ELF.
2001-09-10 06:04:02 +00:00
2067808a2a
compare IPv6 address correctly, for net_interfaces directive.
...
KAME PR 382.
2001-09-09 02:52:44 +00:00
f6225f91fb
Rebuild USE_NEW_TOOLCHAIN files for vax. Add binutils, etc. ..
2001-09-08 18:26:05 +00:00
71e81eb955
Also apply a fix if there are no symbols involved (which means that no
...
reloc will be emitted for the fix).
2001-09-08 16:38:32 +00:00
649981af00
The don't apply fix if not NO_RELOC only applies to OBJ_ELF.
2001-09-08 04:38:34 +00:00
f23306e36f
We need md_apply_fix is there is NO_RELOC.
2001-09-08 04:33:58 +00:00
239fd58a71
md_apply_fix is a NOP for ELF. So macro it away. Simplify the resulting
...
one for the non-ELF case.
2001-09-07 17:14:24 +00:00
b3b6be7160
Only use data from the r_addend, not the relocation slot itself. Make sure
...
pcrel relocations enable pcrel_offset and augment that for VAX specific
needs.
2001-09-07 17:12:15 +00:00
3a6f34a1f2
fix part of previous: do pass -mcmodel=medlow to the 64 bit cc1 if -p*.
2001-09-07 07:19:52 +00:00
bb46238ed7
Merge in 2.11.2 changes into my ELF support
2001-09-07 05:06:51 +00:00
e934346ca9
Make -mvec as an alternative for -m7400 since -mvec existed in binutil 2.9
2001-09-06 21:07:07 +00:00
b3fe7bae05
Add the AltiVec data stream instructions that seemingly got lost between
...
binutils 2.9 and 2.11.
2001-09-06 21:02:29 +00:00
798e998541
Add (commented out) some language-related FILES that could also
...
be installed.
2001-09-02 23:00:30 +00:00
b47aecd969
Install some more locale files.
2001-09-02 22:55:58 +00:00
dac7d101b7
Reduce diff to config.h created by texinfo-4.0.
2001-09-02 22:18:51 +00:00
e405b87d8f
Normalize man page handling: Man pages are now done in the same dir as the
...
program they are referring to, instead of being all done in the main
Makefile.
2001-09-02 22:16:58 +00:00
cbf30c4010
Minimize diff to texinfo-4.0.
2001-09-02 21:55:03 +00:00
d56ab1d29b
Fix merge botch.
2001-09-02 21:42:49 +00:00
fecd6a27e1
Not in texinfo-4.0, either.
2001-09-02 21:26:49 +00:00
0eea4da15e
Not in texinfo-4.0.
2001-09-02 21:24:38 +00:00
376efde5a3
These files are not part of texinfo-4.0.
2001-09-02 21:19:47 +00:00
6e2daf5a64
fix the -m32 code to use /usr/lib/32 paths. don't pass -mcmodel to the 32bit cc1, it doesn't like that.
2001-08-30 09:30:40 +00:00
115cd493b5
Do not compare to chars to >= 0 since this will cause a warning where
...
chars are unsigned. instead, change it from ((c) < foo && (c) >= 0) to
((unsigned char)(c) < foo) which does the same thing but avoids the warning
2001-08-30 02:15:42 +00:00
33ebc9435e
Regenerate so that ${DIST} is not expanded.
2001-08-30 01:31:49 +00:00
61cb93c98c
Commit NEW_TOOLCHAIN files. Do not attempt to use these unless you know
...
what you are doing!
2001-08-28 07:08:54 +00:00
bef08d29d9
Also accept (and ignore) -export-dynamic (already accepted --export-dynamic).
2001-08-26 00:02:46 +00:00
69879ba2e5
regenerate.
2001-08-25 12:23:49 +00:00
251b3464be
heirarchy -> hierarchy
2001-08-24 10:24:45 +00:00
3b7e3e08b4
upgrade to 8.11.6. fixes security issue in command line parsing.
2001-08-21 07:13:20 +00:00
1e378c4c12
precede, not preceed.
2001-08-20 12:00:46 +00:00
ee333eaabb
add missing o: to the strip(1) getopt format string.
2001-08-20 04:47:29 +00:00
87cdecebcb
Update new toolchain that now toolchain2netbsd has been
...
run natively on an ARM ELF system.
2001-08-19 17:48:49 +00:00
c757135288
When making the ldscripts, set MACHINE to "". elf.sc makes bad scripts
...
if MACHINE is set.
2001-08-19 17:19:46 +00:00
4af2f898e7
preliminary new toolchain support for ARM ELF.
2001-08-19 16:21:21 +00:00
cd345950b6
Prelim ARM ELF support for binutils.
2001-08-19 16:18:59 +00:00
f5c0264a5d
New toolchain support for ARM ELF.
2001-08-19 16:05:02 +00:00
05a8e5fe04
build & install info & man page documentation.
2001-08-19 14:23:55 +00:00
c5716adacf
copy across our gcov.1 from egcs.
2001-08-19 14:22:24 +00:00
21207fbaae
build c89 from gnu/usr.bin.
2001-08-19 14:21:02 +00:00
7f85f1f0c3
build & install info documentation.
2001-08-19 14:19:38 +00:00
a0831b3afc
Doesn't need libz: File compression handling is done by forking gzip.
2001-08-18 14:50:03 +00:00
4b5fcdbfec
Use our getopt_long.
2001-08-18 14:48:55 +00:00
f05a7e7c12
fix the NetBSD comment char (already fixed in net binutils.)
2001-08-17 11:12:34 +00:00
e6de26d11c
include ../../Makefile.inc to set BINDIR.
2001-08-16 16:59:24 +00:00
318911587e
regenerate.
2001-08-15 02:44:58 +00:00
7157e2832f
XXX - force handing down of -no-gcc to off, so that cpp will predefine
...
__GNUC__ automatically.
Note that this is probably not the Right solution; we should fix any in-tree
uses of /usr/bin/cpp to use $CC -E. However, this is being added for now
for backwards compatibility.
2001-08-14 13:46:14 +00:00
0bcc7933e9
Trim the gcc frontend's paths way down to size. We know where system-supplied
...
libraries and so forth are, so we don't need to go looking in 9+ locations
for it.
2001-08-14 13:36:09 +00:00
8d35d68bd3
backout this change. we aren't yet ready for it.
...
2000-08-16 Andrew Macleod <amacleod@cygnus.com>
* elf64-sparc.c (sparc64_elf_adjust_dynamic_symbol): Don't allocate
four extra entries at the beginning of the .rela.plt section.
(sparc64_elf_finish_dynamic_symbol): Adjust the offset in the .rela.plt
section to account for the four reserved entries in the .plt section.
2001-08-14 11:37:10 +00:00
5e6e5d8bf4
Nuke; this was never set up to be built.
2001-08-14 07:28:46 +00:00
8f264ae2fc
Regen.
2001-08-14 05:18:26 +00:00
e9ca096740
Typo -- fix sed substitution to make sure it's actually a version number.
2001-08-14 05:17:59 +00:00
75c2396dcc
Regen.
2001-08-14 05:14:59 +00:00
2634f75160
Strip NetBSD version numbers from target triplets in all generated files.
2001-08-14 05:11:50 +00:00
daaabffb64
Nuke DRIVER_DEFINES from list of gcc variables.
2001-08-14 04:59:32 +00:00
53f1af26cd
Resolve conflicts from binutils 2.11.2 import
...
(Also fix some cosmetic garbage added by autoconf, automake, or CVS merges
to get our files closer to the FSF versions.)
2001-08-14 04:43:05 +00:00
cb093d3c1d
Import binutils 2.11.2.
2001-08-14 02:57:43 +00:00
2e1e353192
Import binutils 2.11.2.
2001-08-14 02:47:42 +00:00
dabda76181
For now, just to make sure we don't conflict with an out-of-the-box
...
gcc 3.0 build (as noted by mrg), bump shlib major again, to version 4.0.
There might be a better solution to this kind of thing in the future; I'll
have to think about it.
2001-08-14 01:21:23 +00:00
6c78eab38c
Add the LINTCOMMENT hack, unconditional (no #ifdefs). Since gcc 3.0's cpp
...
does this Right (without even the overwrite-with-whitespace kludge), we
don't need to worry about breaking anything here. Both -C and -CC will
result in full lint-capable comments.
2001-08-14 01:13:34 +00:00
ff23a23f1d
Fix comment delimiting start of gdb section to say `gdb' instead of `gcc'
2001-08-13 03:25:41 +00:00
feaf6caf3c
build sparc & sparc64 binutils & gdb.
2001-08-10 15:35:37 +00:00
5b25536c3b
build libbfd for sparc & sparc64.
2001-08-10 12:42:37 +00:00
14c389e8b1
regenerate
2001-08-10 12:41:19 +00:00
82f7a2e6a2
build in gcc directory as well.
2001-08-10 12:34:45 +00:00
a9c88b6f92
don't build in here if USE_NEW_TOOLCHAIN.
2001-08-10 12:33:55 +00:00
8d54deacdd
regenerate.
2001-08-10 12:01:48 +00:00
5f38c885be
build libstdc++ for sparc & sparc64.
2001-08-10 11:59:50 +00:00
8375332080
regenerate.
2001-08-10 11:59:22 +00:00
220ef62d75
with USE_NEW_TOOLCHAIN, install all required libstdc++ headers. this needs
...
toolchain2netbsd to be re-run for each platform. (sparc, sparc64 and i386
updates coming shortly.)
2001-08-10 11:57:48 +00:00
eda65d255d
deal with no TEMPLATE_NAME being defined for a target
2001-08-10 10:39:10 +00:00
4ce1ea0d33
USE_NEW_TOOLCHAIN powerpc files for libg2c & libstdc++
2001-08-10 06:46:31 +00:00
f88f89f0d4
Update/add new files for powerpc with USE_NEW_TOOLCHAIN
2001-08-10 06:44:06 +00:00
19a0a03e84
install the C++ headers with the new toolchain.
2001-08-09 16:10:56 +00:00
4a850cf2d7
include bsd.own.mk before testing USE_NEW_TOOLCHAIN.
2001-08-09 16:09:52 +00:00
4ccab74b56
enable dwarf2 exception handling.
2001-08-09 16:08:46 +00:00
8655405d18
call this GDB 5.0-nb1
2001-08-09 16:07:20 +00:00
3c86caea97
sync wth 20010228-pl04.
2001-08-09 08:10:17 +00:00
48a8a26166
postfix 20010228-pl04
2001-08-09 07:59:02 +00:00
e89a6ab2cd
Add powerpc-*-netbsd support.
2001-08-08 21:33:45 +00:00
7a1835dd36
Fix typo.
2001-08-08 14:59:26 +00:00
b15841eee9
Add MKPIC=no explicitly for old-toolchain build structure, to avoid shlib
...
version mismatch with new toolchain.
2001-08-07 03:28:07 +00:00
16083edcc0
Add new toolchain gdb build structure, in new subdir to go along with
...
gdbserver and gdbreplay in the future.
2001-08-06 23:28:53 +00:00
1ac83e35b9
Add gdb bits (but not yet gdbserver/gdbreplay).
2001-08-06 23:26:28 +00:00
1b95f4bdaf
Cast &readline_line_completion_function to a (Function *) to make it the
...
proper fully qualified ANSI type.
XXX: Cast it again to a (void *) because libedit's readline emulation thinks
that rl_completion_entry_function should be a (CPFunction *) -- perhaps more
correct, but it breaks compatibility. libedit might be better to change to
conform to what readline wants.
2001-08-06 22:50:07 +00:00
thorpej