Aren
/
NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
131,457 Commits 606 Branches 0 Tags 3.1 GiB
Commit Graph

4790 Commits

Author SHA1 Message Date
mrg 18e7b0d6a5 initial import of GCC 3.3.3, which fixes a few more bugs from the prior 
3.3.3-pre snapshot we were using.

visit http://gcc.gnu.org/gcc-3.3/changes.html#3.3.3 for more details.
 2004-03-01 03:43:04 +00:00
mrg ec14d423fc initial import of GCC 3.3.3, which fixes a few more bugs from the prior 
3.3.3-pre snapshot we were using.

visit http://gcc.gnu.org/gcc-3.3/changes.html#3.3.3 for more details.
 2004-03-01 03:32:51 +00:00
simonb aa48fe87a3 Pull up: 
2003-06-22  Jason Thorpe  <thorpej@wasabisystems.com>

	* config/tc-ns32k.c (md_begin): Initialize inst_hash_table after
	all locals have been declared.

from the binutils mainline.

Problem (and also fix) reported by Eyal Lebedinsky on port-pc532.
 2004-03-01 00:01:16 +00:00
wiz 15a3e59b1f Add FreeBSD 5.2 definition for ath(4). 2004-02-29 18:47:12 +00:00
skrll ce461bed68 Fix PR 24593. 
XXX NOPIC and MKPIC need separating
 2004-02-29 17:05:31 +00:00
skrll 0aae540257 Backout recent changes made to gnu/lib/libgcc3 for a forced MKPIC=no build. 
Fix the behaviour of native and tools gcc when MKPIC=no is specified for
platforms that mknative has determined support shared libraries.

XXX distrib/sets/sets.subr doesn't support MKPIC=no
 2004-02-27 14:52:01 +00:00
matt a9a2fcf9dd Turns out we need libgcc_s for MKPIC=no too (C++ wants it). 2004-02-25 22:11:14 +00:00
matt e88bf5d8c7 Always build libgcc_eh. It's still needed when MKPIC=no 2004-02-25 21:53:49 +00:00
skrll 869ddf30e8 Whitespace 2004-02-23 21:41:05 +00:00
uwe 7a1ea12099 Work around toolchain/22489 (gcc optimization/11937) by compiling 
function.c with -O0.
 2004-02-22 01:24:25 +00:00
cl 661c78132a Fix flow control: 
The condition of the check if we are at a safe point to pause (i.e. we're not
holding a lock) was inverted.  Suspending the client did result in the
server using upto datalimit memory, because the server child process did
continue producing output while the server parent was holding back output
to the client.
 2004-02-21 15:06:11 +00:00
skrll f8314ad5e5 mknative sweep. mainly version string differences and STDC_HEADERs. 
no need for the signbit stuff on sparc as libc got fixed.
 2004-02-19 08:11:31 +00:00
skrll 7354ecab79 Re-run mknative for mipseb (sgimips). Should close PR 24438. 2004-02-16 16:06:38 +00:00
skrll b263dbdd73 Be more gcc like in the way we deal with libgcc so that sh3/hppa millicode 
and exception handling have a chance of working properly.

        - creates libgcc, libgcc_eh and libgcc_s

        - updates LIBGCC_SPEC to use them appropriately.

          There's a hack in here at the moment with respect to libgcc_so in that it
          is preferable to link against libgcc_so will only when -shared-libgcc is
          specified (the c++ frontend does this automatically.) Configurations where
          LINK_EH_SPEC is defined already do this. The gcc configuration for
          NetBSD/alpha and another NetBSD platform (I forget which) actually define
          LINK_EH_SPEC probably by accident rather than design.

        - updates share/mk to use the compiler's knowledge of what needs linking into
          libraries and executables. This removes an hppa hack.

        - updates the sets for the newly created libgcc* files.

	- support for linking against the _pg version of libgcc has been removed.
 2004-02-15 19:52:27 +00:00
simonb 3e7ae7aefb Re-run mknative with latest mknative-gcc. 2004-02-14 08:39:07 +00:00
mrg e38f9bc603 pass all required -I paths: 
+CPPFLAGS+=    -I${G_EXTRA_HEADERS:H:u}

this fixes ppc builds (reported by hannken & briggs)
 2004-02-12 23:17:59 +00:00
skrll 13145e5f28 Results of re-running mknative for a few MACHINE_ARCHs 2004-02-12 22:34:57 +00:00
skrll f6f3dd9b5f Pass -Bsymbolic down to the linker when -symbolic is used. 
OK'd by Jason a while ago.
 2004-02-12 18:51:41 +00:00
mrg 41ba62f48c - some libgcc3/Makefile updates for new mknative 
- re-run mknative for ~nearly everyone
 2004-02-11 16:22:51 +00:00
mrg cf10ab1d3c minor NETBSD_NATIVE adjustments needed for GCC3.3.3. 2004-02-11 16:01:48 +00:00
mrg 0e4a80d700 regenerate with GCC 3.3.3-pre20040209. 
XXX: missing vax, sh3*, hppa
 2004-02-11 04:04:02 +00:00
mrg bce4dd7994 re-add generated files our builds need. 
(not sure why these are not in the 20040209 tarball.)
 2004-02-10 13:41:16 +00:00
mrg 5722a41a7b merge cornflakes for gcc 3.3.2-pre20040209. 2004-02-10 12:56:21 +00:00
mrg ca85c986ec import GCC 3.3.3-pre20040209. this fixes dozens of bugs from the 
GCC database, several logged by netbsd users/developers.  total
count in various area:
	3 - Bootstrap failures and issues
	19 - Internal compiler errors (multi-platform)
	13 - C and optimization bugs
	42 - C++ compiler and library
	1 - Java compiler and library
	1 - Objective-C compiler and library
	3 - Fortran compiler and library
	5 - x86-specific (Intel/AMD)
	4 - PowerPC-specific
	3 - SPARC-specific
	1 - ARM-specific
	4 - ia64-specific
	3 - Alpha-specific
	2 - HPPA-specific
	1 - S390-specific
	6 - SH-specific
	5 - Other embedded target specific
	1 - GNU HURD-specific
	1 - Tru64 Unix specific
	2 - AIX-specific
	1 - IRIX-specific
	1 - Solaris-specific
	2 - Testsuite problems (compiler is not affected)
	1 - Miscellaneous

total:  125
 2004-02-10 12:15:15 +00:00
mrg 2d9d64e9bd import GCC 3.3.3-pre20040209. this fixes dozens of bugs from the 
GCC database, several logged by netbsd users/developers.  total
count in various area:
	3 - Bootstrap failures and issues
	19 - Internal compiler errors (multi-platform)
	13 - C and optimization bugs
	42 - C++ compiler and library
	1 - Java compiler and library
	1 - Objective-C compiler and library
	3 - Fortran compiler and library
	5 - x86-specific (Intel/AMD)
	4 - PowerPC-specific
	3 - SPARC-specific
	1 - ARM-specific
	4 - ia64-specific
	3 - Alpha-specific
	2 - HPPA-specific
	1 - S390-specific
	6 - SH-specific
	5 - Other embedded target specific
	1 - GNU HURD-specific
	1 - Tru64 Unix specific
	2 - AIX-specific
	1 - IRIX-specific
	1 - Solaris-specific
	2 - Testsuite problems (compiler is not affected)
	1 - Miscellaneous

total:  125
 2004-02-10 12:05:11 +00:00
skrll 95cfd78c59 Updates so that mknative gets the right things. 
- Disable symbol versioning (for now)
	- Make sure that libiberty knows its being configured with a
	  cross compiler.
	- The CXX_* variables are no longer needed/used.
	- LIB2FUNCS_EXTRA gets pulled in via LIB2ADD
	- Get LIB1ASMFUNCS and LIB2ASMSRC
	- MAYBE_USE_COLLECT2 got renamed to USE_COLLECT2 (but might not
	  be used)
	- Get EXTRA_HEADERS so that we get generate the right paths for
	  CPPFLAGS
	- Get some variables related to shared libgcc
 2004-02-10 09:37:58 +00:00
lukem 78f3b1bf18 use uname=${BINOWN} gname=${BINGRP} instead of uname=root gname=wheel 2004-02-09 06:52:34 +00:00
lukem 55e3a56f6d consistency tweaks in .stamp files 2004-02-08 22:28:27 +00:00
lukem 5ed2a261e5 cosmetic tweaks and MAKEVERBOSE support 2004-02-08 22:23:45 +00:00
lukem 8ee2b897bd Speed up make by caching the objdirs of the 4 libraries. 
On my P4 2.4 an empty "cleandir" goes from 6.63s to 1.43s.
 2004-02-07 05:10:44 +00:00
lukem 4d8a527fc5 When creating multiple sources from the one operation, use a 
dummy "timestamp" file and depend upon that.
Prevents (unnecessarily) multiple invocations with parallel make.
 2004-02-07 04:46:59 +00:00
fredb cc03b614c7 Let "cu" connect at 230400 bps, if the port supports it. 2004-02-03 14:22:26 +00:00
lukem bc42f94b01 Actually, proxymap can run chrooted even though it defeats most of the 
purpose of the service.
(Use updated text from postfix 2.0.18)
 2004-01-30 04:37:39 +00:00
lukem b3437c9c77 proxymap can't run chrooted either 2004-01-30 04:33:05 +00:00
lukem 25d2de94fe Document what BROKEN_SPOOLER_FLAGS 'add 16' is for, and add it to the 
default broken flags.  groff 1.19 defaulted to this value off, whereas
groff 1.18.1 and earlier defaulted to this being on.  Setting this value
fixes postscript printing to my HP LJ4m.
Note: BROKEN_SPOOLER_FLAGS was previously `7', so we were already enabling
workarounds for other issues...
 2004-01-27 01:00:09 +00:00
christos e9edfecbb5 Don't run commit_prep UNSETXID in server mode, because the created #cvs.lastdir 
file is owned by the committing user, so that cvs cannot later remove it,
leaving junk in /tmp
 2004-01-24 23:46:30 +00:00
mrg 574bbc73cf add -Bforcearchive as an alias for --whole-archive, for previous 
netbsd compatibility.
 2004-01-22 04:49:57 +00:00
mrg aaacfb4a0a this file is obsolete 2004-01-12 04:47:22 +00:00
lukem 01e14c2ace fix previous (MKSHARE=no support) 2004-01-12 01:43:54 +00:00
lukem f48f3fcdfc support MKDOC=no 2004-01-12 01:33:51 +00:00
lukem e7f5146b76 support MKSHARE=no 2004-01-12 01:01:55 +00:00
christos 5cd9414514 PR/24021: Greg Troxel: cvs(1) doesn't work with mode 770 repositories for 
secondary gids.
Implement proper group membership testing as suggested in the PR.
 2004-01-08 17:51:33 +00:00
lukem 5f82e1b832 deprecate NO_SENDMAIL and consistently test MKSENDMAIL 2004-01-06 02:10:43 +00:00
itojun 88aff973a4 bump version to 1.11.11. sync w/ config.h.in. noted by naoki fukaumi 2004-01-04 18:37:47 +00:00
lukem 986979b8c9 Consistently check ${TOOLCHAIN_MISSING} against "no". 2004-01-03 14:04:27 +00:00
nathanw 58e744664c Enable alpha kcore support. 2004-01-02 22:42:05 +00:00
nathanw 240c6ac9cf Re-implement fetch_kcore_registers() for alpha. 2004-01-02 22:41:17 +00:00
skrll 2879279c43 Move libgcc3/*.mk to libgcc3/arch. 
Update mknative-gcc.
 2004-01-02 11:58:32 +00:00
itojun ce7a5bd52c upgrade to 1.11.11 for security reason 2004-01-02 11:34:40 +00:00
itojun 977212b1bf import 1.11.11 for security reasons 
from GENTOO LINUX SECURITY ANNOUNCEMENT 200312-08

   "Stable CVS 1.11.11 has been released. Stable releases contain only
   bug fixes from previous versions of CVS. This release adds code to
   the CVS server to prevent it from continuing as root after a user
   login, as an extra failsafe against a compromise of the
   CVSROOT/passwd file. Previously, any user with the ability to write
   the CVSROOT/passwd file could execute arbitrary code as the root
   user on systems with CVS pserver access enabled. We recommend this
   upgrade for all CVS servers!"
 2004-01-02 11:29:56 +00:00
mrg 416b3a63d6 do not link xmalloc.c or xexit.c. 2003-12-21 07:05:25 +00:00
simonb 72964fcdbe Apply rev 1.225 from binutils mainline: 
date: 2003/07/09 01:27:30;  author: cgd;  state: Exp;  lines: +3 -2
2003-07-08  Chris Demetriou  <cgd@broadcom.com>

	* config/tc-mips.c (mips_validate_fix): Do not warn about branch
	target being a global symbol if not compiling SVR4 PIC code.

Fixes warnings compiling MIPS kernels.  Problem noticed by Izumi Tsutsui
on the port-pmax list.
 2003-12-15 00:54:41 +00:00
wiz 90d2703a98 Remove superfluous "is". Reported by Brian Chase in PR 23739. 2003-12-14 09:38:29 +00:00
rearnsha 79d79358ab Pull up the following patch from the main gcc tree. 
PR optimization/13037
	* loop.c (update_giv_derive): Ignore redundant sets of a biv
	when calculating how to derive a giv from a biv.
This fixes the underlying problem in toolchain/23002.
 2003-12-13 12:20:53 +00:00
matt a29393ca12 Pull in the following change from gcc mainline: 
2003-11-30  Alan Modra  <amodra@bigpond.net.au>

        * config/rs6000/rs6000.c (rs6000_elf_section_type_flags): Don't
        set SECTION_WRITE on TARGET_RELOCATABLE.
 2003-12-12 17:43:37 +00:00
jmc ce438b1c67 Needs USETBL or the ms man page comes out wrong. 2003-12-12 06:08:30 +00:00
matt 020ecacc80 Don't use -traditional-cpp when assembling the PowerPC .S files. 2003-12-11 22:37:27 +00:00
nathanw 243cf32568 Alpha uses software single-step. Lost in the 5.0->5.3 conversion. 
Addresses PR toolchain/23672.
 2003-12-11 21:06:27 +00:00
nathanw de13a4e603 Catch up alpha_software_single_step() prototype with reality. 2003-12-11 21:02:24 +00:00
dyoung 4758291178 Fix the checkflist for builds without Kerberos 4 (MKKERBEROS4=no) 
and without Kerberos 4 & 5 (MKKERBEROS=no). Previously checkflist
complained of missing files.

* move kerberos- and kerberos 4-only files into new flists,
  distrib/sets/lists/*/krb.*

* make the flist generators grok MKKERBEROS{,4} variables

* fix Makefiles which treat MKKERBEROS=no as MKKERBEROS5=no.
  9 out of 10 experts agree that it is ludicrous to build w/
  KERBEROS4 and w/o KERBEROS5.

* fix header files, also, which treat MKKERBEROS=no as MKKERBEROS5=no.

* omit some Kerberos-only subdirectories from the build as
  MKKERBEROS{,4} indicate

(I acknowledge the sentiment that flists are the wrong way to go,
and that the makefiles should produce the metalog directly.  That
sounds to me like the right way to go, but I am not prepared to do
revamp all the makefiles.  While my approach is expedient, it fits
painlessly within the current build architecture until we are
delivered from flist purgatory, and it does not postpone our
delivery. Fair enough?)
 2003-12-11 09:46:26 +00:00
itojun a96b475fd1 upgrade to 1.11.10. includes security fix. 2003-12-10 06:09:52 +00:00
itojun 4f87169ef4 cvs 1.11.10, includes security fix 2003-12-10 05:58:05 +00:00
mrg 12dc5d733c now that medlow is the defalut, no need to force it in -p/-pg. 
suggested and tested by martin.
 2003-12-08 13:15:01 +00:00
matt cfbf1414f1 initial GCC3 support for VAX. (barely tested). No PIC support (aka no 
shared libraries).  This is an initial step to full GCC3 support for VAX.
 2003-12-08 03:33:13 +00:00
matt f158bd2750 Make nan matches __builtin_nan(const char *). 2003-12-08 03:18:07 +00:00
scw b8191da53a Gcc3 for sh5 gets an ICE compiling libbfd's elf32.c and elf64.c at -O2. 
Drop to -O1 for now when building for sh5.
 2003-12-07 21:13:04 +00:00
scw 9867e34ef7 - Work around a gcc3 ICE when building reload1.c at -O2 on sh5: 
Compile reload1.c at -O1.

 - Work around a fatal warning when building function.c on sh5:
   Define NOGCCERROR to disable -Werror.
 2003-12-07 19:56:59 +00:00
scw a72cb1203e MD glue for binutils on sh5. 2003-12-07 19:42:03 +00:00
simonb affca5fc17 Use -O1 to compile cppmacro.c on ns32k to avoid an ICE. 2003-12-07 12:51:43 +00:00
simonb df1bc575d0 Include gthr-single.h instead of gthr-posix.h on ns32k now. 2003-12-07 12:44:54 +00:00
simonb 32820e3ec5 Use gthr-single.h ("no real pthreads support") on ns32k for now. 2003-12-07 12:25:43 +00:00
simonb 3155813857 ns32k c++ (probably a.out?) currently doesn't support the 
-ffunction-sections andd -fdata-sections flags; disabled them
for now.
 2003-12-07 12:23:35 +00:00
scw 03aed749f0 Update for gcc 3.3.2 support (with in-tree binutils). 2003-12-07 10:02:45 +00:00
scw 0e54a63381 Recognise sh5 and sh5le. 2003-12-07 09:58:56 +00:00
matt ed5964b611 Install the MMX/SSE/Altivec include files that gcc provides. 2003-12-05 18:56:11 +00:00
simonb 3952da3765 Add ns32k binutils 2.14 support. 2003-12-05 12:08:47 +00:00
simonb f0467a5c13 Add ns32k GCC3 support. 2003-12-05 12:05:32 +00:00
simonb 172612b7e5 Apply rev 1.228 from binutils mainline (via patch): 
date: 2003/09/14 15:24:49;  author: ths;  state: Exp;  lines: +37 -30
        * config/tc-mips.c (HAVE_64BIT_ADDRESS_CONSTANTS): Remove.
        (macro_build_ldst_constoffset): Sign-extend 32-bit constants. Change
        the function prototype.
        (load_register): Likewise. Simplify the checks for sign-extended
        constants.
        (macro): Likewise. Fix code generation for 64-bit address constants
        outside the 32-bit compatibility space. Adjust
        macro_build_ldst_constoffset calls.
        (s_cprestore): Adjust macro_build_ldst_constoffset call.

Fixes problems compiling r3k kernel cache routines.
 2003-12-05 06:52:57 +00:00
mrg 4dd4b4772e regenerate with a working compiler 2003-12-05 04:53:27 +00:00
mrg 492975536e XXX include <stdint.h> on NetBSD 1.6 and greater. 
XXX this file is generated, but this change is needed for GCC2 hosts to
XXX to able to build this file without warnings.
 2003-12-05 03:25:36 +00:00
mrg 51b0c7cbfa regenerate these such that they (mostly) work.. 2003-12-05 02:57:02 +00:00
keihan 3a8b7119a9 netbsd.org -> NetBSD.org 2003-12-04 23:32:37 +00:00
mrg 228f85d62e regenerate non-broken 2003-12-04 14:39:09 +00:00
mrg abda1a5442 libopcodes.so.1/libbfd.so.6 2003-12-04 12:18:57 +00:00
mrg de5c931206 update some local code for binutils 2.14: 
s/boolean/bfd_boolean/;
	s/true/TRUE/;
 2003-12-04 10:11:12 +00:00
mrg ce29aa2370 build lib{opcodes,bfd,iberty} back in gnu/lib instead of gnu/usr.bin/binutils 
as they are needed earlier than when gnu/usr.bin/binutils is built when
"make -j" is used.
 2003-12-04 09:29:31 +00:00
mrg 293d502fe4 delete binutils 2.13 sources. 2003-12-04 02:00:07 +00:00
mrg 18f0c92069 delete these long long obsolete versions 2003-12-03 10:29:32 +00:00
mrg 2aa1a44c7a switch everyone to binutils 2.14. it's confirmed to work on all the 
platforms that actually build at the moment...
 2003-12-03 10:26:45 +00:00
mrg 099de62e9e new configurations for all supported platforms tested and known working 
on these platforms so far:
	- alpha, arm, i386, mipseb, mipsel, sparc64
built but not yet tested on:
	- armeb, sparc, amd64
build errors (mostly not related) on:
	- hppa, m68000, m68k, macppc, sh3el, sh3eb, vax

XXX again this is not everything required; many other changes in
XXX already used files are left out.
 2003-12-03 06:53:48 +00:00
nathanw 1a4f33e2db Move _initialize_corelow() back to before _initialize_nbsd_thread() so that 
threaded core debugging is enabled.
 2003-12-01 17:54:41 +00:00
simonb 9c8a6768e6 Revert 2.14 mknative (with expanded local paths!) while we're forced at 
2.13.2.1; needed to get clean -current building again.
 2003-11-28 05:16:19 +00:00
simonb d95b3733ab Revert part of binutils 2.14 framework; needed to get clean -current 
building again.
 2003-11-28 05:14:18 +00:00
mrg d409bb8e80 framework to build binutils 2.14. note lib{bfd,opcodes,iberty} framework 
is moving here, keeping everything together.

tested with: vax (old config), mac68k (old config), i386 (new config)
and shark (new config).  tested i386 binaries only so far, but the
system seems to be surviving the self-hosted test.

XXX: note that this isn't *all* the bits required to run a binutils
2.14 world for arm or i386; these will come soon enough...
 2003-11-27 15:02:21 +00:00
mrg d2f6146b45 find libraries etc in a binutils 2.14 world. 2003-11-27 14:36:45 +00:00
mrg 65e91005fa don't build libopcodes/bfd/iberty here if we have binutils 2.14 2003-11-27 13:53:11 +00:00
mrg eb9e243e98 include "getopt.h" as we refer to struct option. 2003-11-27 13:52:30 +00:00
mrg e47eb0663b backout most of the previous; it is no longer necessary. 2003-11-27 10:54:17 +00:00
mrg 8533021b27 gdb 5.0 is gone. 2003-11-27 10:52:02 +00:00
mrg 3a3b2d3c50 delete GDB 5.0. it is no longer used by supported platforms (sorry pc532 :-) 2003-11-27 10:32:25 +00:00
mrg 7764cab06d merge the non-aout-shlib changes forward from gnu/dist/toolchain. 2003-11-27 09:45:33 +00:00
mrg f8f47744d8 disable rules that modify the source tree. 2003-11-27 09:40:38 +00:00
mrg ebf56cd11f re-add binutils 2.14 files. 2003-11-26 12:29:43 +00:00
mrg 1a898f987d initial import of binutils 2.14 into gnu/dist/binutils. 2003-11-26 11:33:51 +00:00
mrg 2e50f8ac41 initial import of binutils 2.14 into gnu/dist/binutils. 2003-11-26 11:31:11 +00:00
uwe 633913b2b3 Add ifdefed inits for TUI and if 0'ed _initialize_nbsd_thread (not yet). 2003-11-24 01:59:59 +00:00
uwe c9a82b254b DEFAULT_BFD_VEC should be little endian bfd_elf32_shlnbsd_vec. 2003-11-24 01:58:22 +00:00
uwe 10ff81335f Set ENTRY=__start to be consistent with other ELF ports. The produced 
binaries are the same, as we provide both _start and __start symbols
for the entry point in crt0.

PR#22073 by Izumi Tsutsui.
Ok by Matthew Green <mrg@netbsd>.
 2003-11-23 13:54:11 +00:00
nathanw d8637f542a Use $(build_exeext) in both the source and destination, like the other 
rules.
 2003-11-20 18:43:58 +00:00
nathanw 0dcab12a43 Add a missing $(build_exeext) 2003-11-19 23:45:25 +00:00
lukem b7b564a26d add fini.lo to CLEANFILES 2003-11-16 22:49:48 +00:00
lukem 263786182a support MAKEVERBOSE 2003-11-16 16:29:04 +00:00
lukem 59c9b264d0 Use .lo for host tool objects. 
Use CPPFLAGS.file correctly (CPPFLAGS.foo.c _not_ CPPFLAGS.foo.o !)
Support MAKEVERBOSE.
Be consistent about make(1) variable use.
 2003-11-16 16:24:48 +00:00
matt 4e90dd053f Add kcore and netbsd thread support. 2003-11-14 19:10:00 +00:00
matt 8056b67d49 Add kcore & netbsd thread support 2003-11-14 19:08:49 +00:00
briggs 5e0fd052da Fix from Ian Lance Taylor to allow building a cross-gdb with arm--netbsdelf 
target.  Patch as seen on tech-toolchain@.
 2003-11-14 15:06:15 +00:00
dbj be929a0a55 update to most recent versions 2003-11-13 03:06:40 +00:00
martin 4b0b317837 Remove unused variables. 2003-11-11 16:28:59 +00:00
martin 6354cf63d2 Fix register offset calculations. From Christian Limpach. 2003-11-11 16:24:23 +00:00
scw 9bf3f153c5 Apply a patch from Richard Earnshaw which fixes a half-word load/store 
codegen bug on armeb.
 2003-11-11 12:04:13 +00:00
matt 8c60671ca9 Add thread support. 2003-11-10 21:59:58 +00:00
matt 0ef98a7702 Make version 5.3nb1 2003-11-10 21:49:14 +00:00
matt e2bf0dc236 Enable nbsd-thread. 
Make version 5.3nb1.
 2003-11-10 21:48:48 +00:00
mrg 0bcb648614 set the arm breakpoint depending on the byte order. now armeb gdb works. 2003-11-09 02:52:07 +00:00
ragge 587d7055bd Set optimizing to -O0 for vax (was -O1 before), otherwise awk 
gets internal errors when used.
 2003-11-08 17:17:59 +00:00
mrg b6f7d4362f default armeb*netbsd* to ARM_FLAG_BIG_END 2003-11-08 06:57:26 +00:00
uwe af9cced758 The "pref @rN" instruction was introduced in sh3, not sh4, so correct 
its sh_opcode_info entry accordingly.
 2003-11-07 14:43:18 +00:00
mrg 57ddb40e82 be sure to default to bfd_elf32_bigarm_vec not bfd_elf32_littlearm_vec 2003-11-07 07:31:44 +00:00
mrg 68002eaa4a build ra-colorize.c with -O0 until the bug that causes it to have dozens 
of undefined references to local symbols is fixed.
 2003-11-05 00:05:48 +00:00
hannken 83bd54e877 Add "target kcore" to gdb53/sparc64. 
Taken verbatim from the old gdb in src/gnu/dist/toolchain/gdb.

Approved by: Christos Zoulas <christos@netbsd.org>
 2003-11-04 19:33:39 +00:00
scw 437384a077 Regen for TARGET_ENDIAN_DEFAULT=ARM_FLAG_BIG_END fix in config.gcc. 2003-11-04 14:46:59 +00:00
scw 2a38410f5e Need to add TARGET_ENDIAN_DEFAULT=ARM_FLAG_BIG_END to tm_defines if 
configuring for armeb.
 2003-11-04 14:45:16 +00:00
wiz 9678ca7caa Install grohtml(1). 2003-11-03 17:05:28 +00:00
skrll c1946cebd4 Index fpregsetp->fpr correctly. 2003-11-03 15:59:11 +00:00
ragge 3a99ca5371 Set USER_LABEL_PREFIX to "", otherwise native compilation of some GNU 
utilities (like texinfo) fails.
 2003-11-03 10:07:34 +00:00
hannken 3f72313b8e Add "target kcore" to gdb53/sparc. 
Taken verbatim from the old gdb in src/gnu/dist/toolchain/gdb.

Approved by: Christos Zoulas <zoulas@netbsd.org>
 2003-11-02 20:59:28 +00:00
uwe 1d729d7bc4 First cut at implementing software single stepping for sh3. We need 
it because sh3 ports don't provide PT_STEP.  It is finally possible to
run programs under gdb now, though there are still issues to be
resolved.  It was deemed useful enough to commit early, while the work
is still in progress.  Tested on hpcsh.

Blessed by christos@.
 2003-10-31 00:11:27 +00:00
mycroft b9c785bd44 As seen on tech-toolchain: 
So the problem is thus:

* The {u,}mulsidi3 generate two parallel sets which modify the upper and lower
halves of the target register.

* life_analysis() does not track subregister modifications -- if you don't
modify the whole register with a single set, it considers the register
unused.

The simple, if klugy, solution to this is to stick an explicit clobber in.  It
seems to work.

While doing this, I noticed that constant folding was not happening for
32x32->64 multiplies.  This is because the parallel set generated by
{u,}mulsidi3 cannot be folded at all.  To solve this, I first expand to a
normal multiply, and then use a define_insn_and_split to convert it to the
parallel set after CSE and constant folding.

This patch has also been submitted to GCC bugzilla, but who knows if I'll get
a reply to that.
 2003-10-29 21:25:03 +00:00
matt f22eb2930b From gcc-patches: 
* real.c (encode_ieee_single): Ensure proper promotion.
Back out change to i386.c
 2003-10-28 19:08:12 +00:00
mrg 00b48119d0 XXX: build mdsetimage.c / installboot/ffs.c with -O0 on sun2 as GCC3 
barfs on these for now.
 2003-10-28 08:21:26 +00:00
lukem b43426c867 Move the rule to disable .cc -> NULL transforms, so it works again. 2003-10-28 08:13:39 +00:00
lukem b947f38eae don't need .cc->NULL rule here 2003-10-28 08:12:32 +00:00
lukem 5966f94e07 Move the rule to disable .cc -> NULL transforms, so it works again. 
Fixes "includes" when exception.cc is newer than inc/exception.
Noted in private mail by Havard Eidnes.
 2003-10-28 07:38:57 +00:00
lukem ad700b72aa replace $(SHELL) with ${HOST_SH} 2003-10-28 02:51:56 +00:00
uwe 375b04dc77 Default to the little-endian bfd_elf32_shlnbsd_vec. 2003-10-28 01:58:23 +00:00
mrg 45e224e8d4 fix a typo in previous pointed out by lukem 2003-10-28 01:25:33 +00:00
mrg e2c8806fcf compile (but not run-time) tested sh3eb-netbsdelf-gdb support. 2003-10-28 01:21:37 +00:00
uwe 398db2f8fd Item 1 - mention that you can configure and build in a scratch 
directory, thus saving the trouble of doing distclean and cvs update.

Item 4 - it's defs.mk, not defs.h.

New bullet - in defs.mk set G_BFD_CPPFLAGS from `tdefaults' variable
of the generated bfd/Makefile.
 2003-10-28 01:12:11 +00:00
mrg d103a27279 compile-tested sun2 support for GDB 5.3. 2003-10-27 23:56:55 +00:00
mrg 1b3f9cdd10 revert previous 2003-10-27 23:56:14 +00:00
mrg f8fa4ee590 sun2 lib*3 for GCC3. 2003-10-27 17:19:58 +00:00
mrg 42f41e65c7 mostly working configuration updates for sun2. 
XXX - gcc3/backend/ra-colorize.c gets built wrong and the backend programs
XXX   fail to link.
XXX - mdsetimage & installboot/ffs.c get an unrecogised instruction ICE.
 2003-10-27 16:55:32 +00:00
mrg a4697adcfc change "cc -dumpversion" back to "3.3.2", as per old behaviour. 
should fix PR#23280.
 2003-10-27 16:43:29 +00:00
mrg 2a50c63fdf change the "long" version from "3.3.2-nb1" to "3.3.2 (NetBSD nb1 20031026)" 
which is more inline with what we used previously.
 2003-10-27 16:42:33 +00:00
mrg 2f6aa47ade compile (but not runtime) tested support for armeb-netbsdelf-gdb 5.3 2003-10-27 16:36:19 +00:00
thorpej 75516c03f7 When running mkfixinc.sh, explicltly export SHELL, as well, since on 
some platforms (e.g. OS X), we disable exporting make(1) variables in
the environment due to ARG_MAX limitations.
 2003-10-27 16:20:04 +00:00
hannken 0165bf4148 Make it compile again. Blind copy-and-paste from sparc64nbsd-nat.c. 
Approved by: Christos Zoulas <christos@netbsd.org>
 2003-10-27 14:10:54 +00:00
mrg 8d8c032590 add armeb-netbsdelf GCC3 support. (not tested) 2003-10-27 01:37:08 +00:00
wiz 51677832e7 Close parentheses. 2003-10-27 00:57:04 +00:00
cl 44f9690b65 g/c unused variables 2003-10-27 00:41:09 +00:00
lukem 171d653219 Overhaul how `build.sh tools' are used: 
*	Rename "config.h" to "nbtool_config.h" and
	HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
	This makes in more obvious in the source when we're using
	tools/compat/config.h versus "standard autoconf" config.h

    *	Consistently move the inclusion of nbtool_config.h to before
    	<sys/cdefs.h> so that the former can provide __RCSID() (et al),
    	and there's no need to protect those macros any more.

These changes should make it easier to "tool-ify" a program by adding:
	#if HAVE_NBTOOL_CONFIG_H
	#include "nbtool_config.h"
	#endif
to the top of the source files (for the general case).
 2003-10-27 00:12:41 +00:00
christos 94920a3392 Add the declarations in init.c too (mentioned by Christian Limpach) 2003-10-26 17:49:24 +00:00
christos 34e2db4a3c Make this link and work again (from Christian Limpach) 2003-10-26 17:45:13 +00:00
christos f9f34b91b1 Add a guide on how to add new NetBSD platform support to gdb. Requested by 
mrg.
 2003-10-26 17:18:04 +00:00
mrg ce063d020a fix a bad merge: s/ifdef/ifndef/. fixes PR#23275 2003-10-26 09:45:44 +00:00
lukem f85d2d1c14 Use ${HOST_SH} instead of `sh'. 
If necessary, pull in <bsd.sys.mk> to get the definition of HOST_SH;
Makefiles that pull in one of (most of) <bsd.*.mk> will get this anyway.
 2003-10-26 07:25:33 +00:00
lukem 857ab1bdd0 use ${HOST_SH} instead of /bin/sh 2003-10-26 03:50:07 +00:00
lukem 414f1003f0 don't need to set SHELL=/bin/sh 2003-10-26 03:49:39 +00:00
matt 1bafa7eae6 Fix LP64 bug. 2003-10-26 01:32:22 +00:00
chs 9bbee945e7 fix cross-building from linux. this is rev. 1.23 from groff CVS 
(with the definition of SHELL added back).
 2003-10-26 00:39:23 +00:00
simonb 969d07f64f * mips-tdep.c (mips_addr_bits_remove): Don't shift by 32 
From Ian Taylor.
 2003-10-26 00:08:23 +00:00
matt 4be480cd15 Fix 64LPism. Use 1L instead of 1 so that sign extension doesn't happen on 
bit 31.
 2003-10-25 21:55:37 +00:00
matt 6e23d936ae More fixes. 2003-10-25 19:02:07 +00:00
matt 91f24c5034 Try dealing with sigtramps better. 2003-10-25 19:01:45 +00:00
matt 7819efde3c GDB 5.3 support for VAX 2003-10-25 18:26:07 +00:00
matt 17aa9b0c9b Use gregset_t/fpregset_t since that's what is really wanted. 2003-10-25 18:25:42 +00:00
matt 1b7561a047 Add VAX GDB 5.3 support. 2003-10-25 18:24:58 +00:00
matt 3e45a9f0c7 Fix library order so that -static will work. 2003-10-25 18:24:23 +00:00
cl 54641460eb make m68k/gdb53 compile as a crossgdb 
- catch up tm-nbsd.h with gnu/usr.bin/gdb53/arch/m68k/tm.h
- only use nbsd-thread.o in m68k nbsd-host case
 2003-10-25 16:20:58 +00:00
dmcmahill a2cc19e091 avoid trying to do "for f in ; do ..." which breaks while crosscompiling 
on certain non-netbsd host.  Approved by Matthew Green, mrg at netbsd.
 2003-10-25 02:42:00 +00:00
he 31cd1c23b7 More fallout from the NULL change: 
If we HAVE_STRING_H, include it to bring strchr() prototype in scope.
 2003-10-23 16:12:45 +00:00
seb e8b3aed4d7 Update version reported by gcc -dumpversion: it is now `3.3.2-nb1' 
not `3.3.1' as before and neither `3.3.2'.

Approved by mrg@.
 2003-10-23 09:59:25 +00:00
matt b8b1ccbdfd If host is x86-*-netbsd*, don't descend into gcc, gdb, and rcs 2003-10-23 02:58:49 +00:00
skrll 899032e619 Consistent naming. 2003-10-22 20:43:12 +00:00
lukem 5f259b477b add MAKEVERBOSE messages 2003-10-22 06:04:03 +00:00
lukem f6ed8b6422 add MAKEVERBOSE message 2003-10-22 05:58:31 +00:00
skrll a69cb4cc55 ARM thread support. 
- add or expose {supply,fill}_{fp,g}regset functions
	- add lwpid to ptrace calls
 2003-10-21 19:07:36 +00:00
skrll ee632a37d0 more NULL fallout. 2003-10-21 10:47:48 +00:00
lukem 130ab7336b Rework how MAKEVERBOSE operates: 
*	Don't bother prefixing commands with a line of  ${_MKCMD}\
	and instead rely upon "make -s".  This is less intrusive on
	all the Makefiles than the former.  Idea from David Laight.

    *	Rename the variables use to print messages.  The scheme now is:
	    _MKMSG_FOO		Run  _MKMSG 'foo'
	    _MKTARGET_FOO	Run  _MKMSG_FOO ${.TARGET}
	From discussion with Alistair Crooks.
 2003-10-21 10:01:19 +00:00
wiz c8a5e4585e Enable HAVE_MKTIME manually; the configure script fails since it 
checks a weird corner case.
Should fix building platforms without shared libs.
Requested by Chuck Silvers.
 2003-10-20 17:07:38 +00:00
lukem 0aa1a4cb76 revert previous; the items are in tmac.local 2003-10-20 05:26:20 +00:00
lukem 3fc127cd66 Support MAKEVERBOSE 
Use ${HOST_SH}
Use BUILDSYMLINKS
 2003-10-20 01:15:03 +00:00
lukem 46948f220d * Support MAKEVERBOSE 
* Implement GROFFYSRCS, which contain a list of .y files to convert into
  .cpp and _tab.h.  Simplifies .y -> .cpp (et al) .
* Use "realall" target instead of "dependall"/"install" to ensure that
  various files are created correctly.
 2003-10-20 00:54:36 +00:00
lukem 9d55f70359 support MAKEVERBOSE 2003-10-20 00:40:53 +00:00
cl b56d9ca137 add {supply,fill}_{fp,g}regset functions 
remove nbsd_{fp,}reg_to_internal and nbsd_internal_to_{fp,}reg functions
add lwpid to ptrace calls
 2003-10-19 20:11:22 +00:00
jdolecek 47740ad0ef make crc16_table[] static const 2003-10-19 19:30:08 +00:00
lukem d4a55392c2 Support MAKEVERBOSE. 
Use BUILDSYMLINKS instead of home-grown equivalents.
 2003-10-19 14:52:03 +00:00
wiz 5e43ad3add Add 1.4.1 and 1.6.2. 2003-10-19 08:58:52 +00:00
wiz ec09eb497f New sentence, new line. 2003-10-19 08:54:20 +00:00
wiz 8034863446 Add RCS Id. 2003-10-19 08:54:13 +00:00
lukem e08f429064 Add definitions for NetBSD versions: 
1.4.1  1.4.2  1.5.1  1.5.2  1.5.3  1.6.1  1.6.2  2.0
 2003-10-19 06:20:20 +00:00
matt 8d6789fc05 Add ncdcs utility (for ibmnws port). Calculates and writes the NCD checksum 
used in kernel files.  Only built for ibmnws port.
 2003-10-19 03:34:59 +00:00
matt e1c37edb37 Skip build of {postfix,sendmail} if MK{POSTFIX,SENDMAIL} == no, respectively. 
(XXX NO_SENDMAIL= must eventually).
 2003-10-18 23:46:26 +00:00
matt 8df2ba29a1 Skip build of {binutils,cvs,gcc compilers,gdb} if 
MK{BINUTILS,CVS,GCCCMDS,GDB} == no, respectively
 2003-10-18 23:44:46 +00:00
matt b746f8a878 Allow UUCP to not be build when MKUUCP=no 2003-10-18 23:42:31 +00:00
mrg 459d14eddf change the version string to 3.3.2-nb1 so that gcc.c likes it. 2003-10-18 14:47:01 +00:00
mrg 8b9e2c6e0a - call this version "3.3.2nb1" (while some of our changes to 3.3.1 were 
in 3.3.2, many were not.)
- use our bug reporting address, not GCC's.
 2003-10-18 13:15:05 +00:00
mrg 9376f2c5fa merge gcc 3.3.2 2003-10-18 11:54:13 +00:00
mrg eaf2241a0a initial import of GCC 3.3.2. this is a bug-fix release that fixes dozens 
of bugs in both MI and many NetBSD MD parts.

visit http://gcc.gnu.org/gcc-3.3/changes.html#3.3.2 for more details.
 2003-10-18 11:13:12 +00:00
mrg cd84f3bd94 initial import of GCC 3.3.2. this is a bug-fix release that fixes dozens 
of bugs in both MI and many NetBSD MD parts.

visit http://gcc.gnu.org/gcc-3.3/changes.html#3.3.2 for more details.
 2003-10-18 11:03:09 +00:00
lukem cdcb0412a1 Disable make_html and make_install_html; they depend upon pnm tools 
which may not exist, and we don't need them as a host tool.

Don't attempt to rebuild gnu.eps & gnu.png from gnu.xpm because of
timestamp skew.
 2003-10-18 03:01:52 +00:00
nathanw db40e00241 Implement regset routines. 
Pass the LWP of the ptid in PT_{GET,SET}REGS calls.
 2003-10-18 00:27:13 +00:00
nathanw 614f188f9d nbsd_reg_to_internal() and friends are obsolete. 2003-10-18 00:14:09 +00:00
nathanw 37337ad2db Pass the LWP of the ptid in PT_{GET,SET}REGS calls. 2003-10-18 00:13:35 +00:00
nathanw fef0da346a Revert a local change that nbsd-thread.c now handles. 2003-10-18 00:11:51 +00:00
nathanw 804ad776be Pile of cleanups and improvements, including: 
* Use {fill,supply}_{g,fp}regset() routines instead of homegrown
   nbsd_internal_to_reg()/nbsd_reg_to_internal() routines; reduces
   local changes to *-nat files.

 * Use the ptrace(PT_LWPINFO) call to find the set of LWPs in a
   process and map them to threads, and to identify the target of a
   signal event.

 * Consistency sweep on ptid-manipulating macros.

 * More modern GDB integration and removal of redundant functions.
 2003-10-18 00:08:14 +00:00
matt 9d4b91ab87 Keep the unsigned, just add HOST_WIDE_INT. 2003-10-16 22:08:01 +00:00
matt 6bcdd1cd1e Cast using HOST_WIDE_INT, not unsigned. 2003-10-16 22:02:25 +00:00
tron 00c8e095d9 Add "cs-tconfig.h" to "CLEANFILES". 2003-10-16 06:15:20 +00:00
lukem 56528942aa "anti-timestamp-skew" fixes for gengtype-yacc.c and gengtype-lex.c 2003-10-16 04:35:17 +00:00
lukem 9da74b23ca Don't auto-frob .y or .l files. 2003-10-16 03:33:28 +00:00
fvdl 9558d5998a Add NetBSD kcore support. 2003-10-14 22:59:59 +00:00
fvdl 22cd769ea5 Set some better defaults for OSABI_UNKNOWN, and add NetBSD kcore support. 2003-10-14 22:58:38 +00:00
wiz a41ff7dabc Sync with groff repo: 
-j and objdir builds here should now finally work both.
Based on a patch from Chuck Silvers.
 2003-10-14 07:49:24 +00:00
fvdl e67583ab9d NetBSD support for the x86_64 target. 2003-10-14 02:02:17 +00:00
hubertf 83f7cd41fd Prevent funny output when giving a directory name on the command line. 
Before:
	$ dc ..
	miyu% dc ..
	dc: 02 unimplemented
	dc: 0 unimplemented
	dc: 0 unimplemented
	dc: input base must be a number between 2 and 16 (inclusive)
	dc: stack empty
	dc: stack empty
	dc: 'h' (0150) unimplemented
	dc: stack empty
	dc: 'u' (0165) unimplemented
	...
	** get heart attack suspecting major FS corruption **

After:
	$ dc ..
	Cannot use directory as input!
 2003-10-13 21:52:40 +00:00
skrll fd905830f3 Don't pull in libgcc_pic.a for -shared - it was added as a workaround for 
gcc 2.95.3. Removing this fixes EH for gcc 3.3 by using the .eh.section
support in libcrt{begin,end}

Closes PR 22673.
 2003-10-13 13:19:41 +00:00
chs 8868d13163 bring over the rest of the IN_SIGTRAMP() hack from the old gdb so that 
gdb builds on the sun3.  jason sez "check that in, for now".
 2003-10-12 01:05:36 +00:00
enami de00b9ee01 Serialize the build of backend and frontend, since the latter requires 
header files created during the former is built.
 2003-10-11 05:16:18 +00:00
matt 046e9000ba Make etime_.* and dtime_.* explicitly depend on the libU77 version. 
Include *.d so that when make depend emit the dependencies, it picks up
the right file and so that the right file is used when making the objects.
 2003-10-10 23:34:20 +00:00
lukem d47cc49cff Add rules to prevent c-parse.c and objc-parse.c from being rebuilt if the 
respective .y file has a newer timestamp (usually due to cvs checkout issues);
the code isn't .OBJDIR friendly, and it's unnecessary in our build envrionment.
 2003-10-10 15:25:17 +00:00
lukem 83590dd4eb Change method used to disable ``maintainer-mode'' style rules to one 
that is more "[cvs] diff" friendly.
Method suggested by Chris Demetriou on source-changes@ on July 25.
 2003-10-10 15:14:16 +00:00
matt 1a3bbf394a Override TARGET_DEFAULTS to include MASK_STRICT_ALIGN (for IBM403). 
It still includes (MASK_POWERPC | MASK_NEW_MNEMONICS) as well.
Change STACK_BOUNDARY to 128 to make Altivec'ed code easier to deal with.
 2003-10-10 01:58:55 +00:00
christos 77b4aabadd Avoid re-entry in the cleanup signal handlers. 2003-10-09 13:43:00 +00:00
matt 926373d2c9 Turn on HAVE_GAS_HIDDEN. 2003-10-07 02:19:18 +00:00
matt 9fbbfac1b2 Pull in ld_vers / ld_date methods from gcc-current. (avoid GNU sed'isms). 
This allows configure to correctly determine that gas supports .hidden.
 2003-10-07 02:13:49 +00:00
veego 982d95a75b Fix part 3 of toolchain/23067 while adding a depend target to the 
dependall and all targets which build libiberty.
Fix from mrg.
 2003-10-06 21:40:32 +00:00
veego fd5c6da435 Fix part 1 of toolchain/23067 by removing the -nostdinc++, which will be 
added if $DESTDIR is set by the bsd.dep.mk and bsd.prog.mk magic, but caused
problems while not using build.sh.
 2003-10-06 21:36:59 +00:00
wiz da70036ee9 Obsoleted by the update to 3.1.3 (moved to missing_d dir). 2003-10-06 18:27:45 +00:00
wiz 2a3e31828e Add gawk2netbsd, a script to prepare gawk distributions for imports. 2003-10-06 16:55:22 +00:00
wiz d6fe0c8983 Update for 3.1.3. 2003-10-06 16:50:39 +00:00
wiz b3a85f1ea5 Actually, we need these files too -- add them. 2003-10-06 16:44:03 +00:00
wiz 1393b6699e Merge 3.1.3 to HEAD. 2003-10-06 16:33:57 +00:00
wiz 06c09073b3 Initial import of gawk-3.1.3. 
Seven releases since the last import, so the ChangeLog is quite big;
please refer to the NEWS file included in the import for details.
Among others, this also addresses PR 23054.
 2003-10-06 15:55:26 +00:00
wiz b83f0ec829 Initial import of gawk-3.1.3. 
Seven releases since the last import, so the ChangeLog is quite big;
please refer to the NEWS file included in the import for details.
Among others, this also addresses PR 23054.
 2003-10-06 15:43:45 +00:00
wiz 1df948824f Initial import of gawk-3.1.3. 
Seven releases since the last import, so the ChangeLog is quite big;
please refer to the NEWS file included in the import for details.
Among others, this also addresses PR 23054.
 2003-10-06 15:41:38 +00:00
matt 39caecb211 Add in-tree toolchain support for HPPA. 2003-10-06 05:10:13 +00:00
martin b82411d0b6 Ooops, added previous note to the wrong ChangeLog 2003-10-03 09:06:14 +00:00
tron cc5dbcf926 Add dynamically generated sources to "CLEANFILES". 2003-10-01 09:39:50 +00:00
wiz 8d0f21a650 Fix regexo in previous. 2003-09-30 20:11:26 +00:00