Commit Graph

5110 Commits

Author SHA1 Message Date
mrg
173e1af775 generated mknative files for amd64 and GCC4. not tested but a release
build completes.
2006-05-19 18:41:37 +00:00
drochner
7e8ecee6be add alpha native support
(a kernel compiled with gcc4 runs for me, userland untested)
2006-05-19 11:44:24 +00:00
mrg
f97e259a75 add generated mknative files for mipseb platforms. evbmips/mipseb gets
to kernels build now (this breaks due to current -current breakage.)

(i guess christos's csu fixes fixed the builds libgcc_s build, which was
missing symbols that i recall as being weak.)

the mips* binaries are as-yet untested.
2006-05-19 05:44:16 +00:00
mrg
804d35c095 GCC4 mknative files for armeb (big endian) and mipsel (little endian.)
armeb builds but mips fails in libgcc_s.so.
2006-05-18 20:20:49 +00:00
mrg
657e5b49ad GCC4 mknative files for arm (little endian) 2006-05-18 20:18:11 +00:00
mrg
a702cc5589 regen (just rcsid changes) 2006-05-18 18:43:37 +00:00
mrg
93f5fae775 generate arm (little endian) mknative files. "evbarm" release builds,
but this is 100% untested yet.
2006-05-17 05:52:38 +00:00
mrg
e172949eb7 regenerate 2006-05-17 05:33:36 +00:00
mrg
d7a5ec3490 install new header mm_malloc.h that the mmx/sse headers use now.
fixes 3rd party app build problems reported by dogcow@.
2006-05-17 03:29:45 +00:00
mrg
09bf9ad8a0 regenerate with fixed mknative-gcc. 2006-05-15 22:10:05 +00:00
nathanw
c98620e0ca Generated files for sim/ppc. 2006-05-15 19:14:20 +00:00
nathanw
53dc5d083a PowerPC generated and constructed files. 2006-05-15 19:13:22 +00:00
nathanw
b350d93baa Include xmalloc.c for psim. 2006-05-15 19:12:07 +00:00
nathanw
f5c91f0e5d Fix up psim build. 2006-05-15 19:11:51 +00:00
nathanw
092b219bf4 i386 generated and constructed files. 2006-05-15 16:39:11 +00:00
nathanw
95e242dd40 MI generated files. 2006-05-15 16:35:40 +00:00
nathanw
eb212d66cb GDB 6 reachover build machinery. 2006-05-15 16:31:37 +00:00
christos
1e88acc412 Use id casts instead of void * casts. Suggested by thorpej. 2006-05-15 16:31:21 +00:00
nathanw
958e368bed Pull in a change from GDB CVS:
2006-05-01  Nathan J. Williams  <nathanw@wasabisystems.com>

        * ppcnbsd-tdep.c (ppcnbsd_return_value): Change type of last two
        arguments from void * to gdb_byte *.
2006-05-15 16:16:34 +00:00
nathanw
9c496995db Import GDB 6.4 2006-05-15 14:21:09 +00:00
nathanw
76ae6db36e Import GDB 6.4 2006-05-15 14:11:08 +00:00
nathanw
e7275437ab Import GDB 6.4 2006-05-15 14:08:20 +00:00
dogcow
15d9504168 /usr/src/gnu/dist/gcc4 -> ${GNUHOSTDIST} 2006-05-15 13:40:00 +00:00
christos
7a85412dcd We don't need -Wno-error 2006-05-15 04:57:05 +00:00
christos
07252f1516 Casts to appease gcc. 2006-05-14 02:14:33 +00:00
christos
ec71107003 Use __UNCONST() when casting away const. 2006-05-14 00:13:31 +00:00
christos
e9c9374d31 Fix uninitialized variable. 2006-05-13 23:49:02 +00:00
mrg
0e02d282f4 define ENABLE_EXECUTE_STACK to nothing so that _enable_execute_stack()
isn't the empty function in libgcc2.c, but the netbsd one.  powerpc
seems to work somewhat better now...
2006-05-13 10:38:33 +00:00
uwe
e37068ccdf Force building and installing libgcc_pic.a when MKPIC=yes even when
MKPICINSTALL=no, as it is required to build shared libraries.  Move it
to comp-c-lib set to match crtbeginS.o
2006-05-13 01:50:51 +00:00
mrg
6688c1d37d merge the LIBGCC_PICSUFFIX and NETBSD_NATIVE support from GCC3.
now hello.c and hello.cc both work!
2006-05-13 01:49:30 +00:00
mrg
f2a1fc4b01 actually, define STANDARD_LIBEXEC_PREFIX as STANDARD_EXEC_PREFIX in
the NETBSD_NATIVE case, since that's what it really is.
2006-05-13 01:48:14 +00:00
mrg
6e5288e9d6 STANDARD_LIBEXEC_PREFIX needs to end with a /. 2006-05-13 01:43:17 +00:00
mrg
a281677257 properly build the version string with spaces. 2006-05-13 01:18:37 +00:00
christos
fc079150a4 prepend cvs_ to all the x* functions to avoid coverity conflicts with other
programs.
2006-05-12 15:41:53 +00:00
christos
7c8717d509 Revert previous coverity fixes because they were caused by the wrong model
(xstrdup() from xlint was used instead of the locally defined one). Replace
with assertions instead (from Mark D. Baushke)
2006-05-12 15:33:17 +00:00
mrg
3c3bcb101f pull across from GCC3 tree:
- __enable_execute_stack() support
- NetBSD/powerpc specific GCC configury bits.
2006-05-12 01:01:18 +00:00
mrg
014ea08324 - fix #include so that "ppc-asm.h" is found
- make sure tramp.asm is compiled properly
2006-05-12 00:59:32 +00:00
mrg
2d88bcdbb8 - export @enable_shared@ into ENABLE_SHARED (for mknative)
- disable rules for gengtype-yacc.c, gengtype-yacc.h and gengtype-lex.c
- pull across a fix for cygwin building fixincludes from GCC3 tree
- disable %.info rules.
2006-05-12 00:58:25 +00:00
mrg
5cfb97cb39 pull across a patch from GCC3 tree. the comment says it all:
# Use sh specific libgcc-std.ver to avoid exporting some
# lib1func routines which shoud not be called via PLT.
2006-05-12 00:56:35 +00:00
mrg
b41d56701a move #include to the start of the line. 2006-05-12 00:55:59 +00:00
mrg
8a15ddadaa add these generated files that really want bison. 2006-05-12 00:54:29 +00:00
mrg
85a901ce95 pull across from GCC3 tree:
>date: 2004/11/30 01:51:13;  author: jwise;  state: Exp;  lines: +21 -3
>As discussed on tech-toolchain@netbsd.org, make cpp refuse to attempt to
>parse a #include'd file which does not pass S_ISREG() if the environment variable
>CPP_RESTRICTED is set.
>
>This is primarily intended for use by programs such as calendar(1) which
>use cpp to parse untrusted user files -- without this change (and the corresponding
>change to calendar(1)), any user can cause a denial-of-service for the daily
>calendar -a run by #include'ing a named pipe.
>
>Many thanks to christos@netbsd for his help in polishing this.
2006-05-12 00:51:13 +00:00
mrg
871b78e87d switch the default model to medlow, from GCC3 tree. 2006-05-12 00:49:49 +00:00
mrg
8ff761d708 use "bv" branching on NetBSD, like on linux. from the GCC3 tree. 2006-05-12 00:47:50 +00:00
mrg
8ff76fa508 force shift target to "long" with "L" prefix, from GCC3 tree. 2006-05-12 00:47:17 +00:00
mrg
43eb93d243 pull across from the GCC3 tree:
>revision 1.2
>date: 2003/10/29 21:25:03;  author: mycroft;  state: Exp;  lines: +40 -11
>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.
2006-05-12 00:46:33 +00:00
mrg
7e1fdc3c74 default to stabs, from the GCC3 tree 2006-05-12 00:44:17 +00:00
mrg
cc2bf2dacf - define __PIC__ and __pic__ when flag_pic is set.
- add GPLUSPLUS_BACKWARD_INCLUDE_DIR to the INCLUDE_DEFAULTS includes
- define GPLUSPLUS_BACKWARD_INCLUDE_DIR, TOOLDIR_BASE_PREFIX,
  STANDARD_BINDIR_PREFIX and STANDARD_LIBEXEC_PREFIX
- delete NETBSD_LIBGCC_SPEC
- define LIBGCC_PICSUFFIX as "_pic".
2006-05-12 00:43:55 +00:00
mrg
7c427e47b1 pass -Bsymbolic to ld when given -symbolic. 2006-05-12 00:42:13 +00:00
mrg
dfa8a9c641 - call this NetBSD GCC.
- use our send-pr URL.
2006-05-12 00:40:57 +00:00