Commit Graph

5908 Commits

Author SHA1 Message Date
christos
8743a11216 Apply GCC/35998 from FreeBSD http://svnweb.freebsd.org/changeset/base/245024
gcc: avoid generating negative values to DW_AT_byte_size.

There is a bug in gcc (GCC/35998) where dwarf reports
sizes of unsigned -1 (0xffffffff).

On NetBSD this generated a faulty CTF entry which then
caused a segfault in ctfmerge. The issue was worked
around in NetBSD's Dtrace but since the issue originated
in gcc, it seems reasonable to fix it here.

Thanks to Christoph Mallon for pointing out a correct fix.
2013-01-17 20:01:37 +00:00
apb
aa3786b9b6 Change several int variables to size_t, ssize_t, or ptrdiff_t.
This should fix the bug described in CVE-2012-5667 when an input
line is so long that its length cannot be stored in an int
variable.

This change to NetBSD's version of GNU grep 2.5.1 (licenced under
GPLv2) was made without direct reference to any code licenced
under GPLv3.

Thanks to Ignatios Souvatzis for looking at GPLv3-derived
patches and describing the problem in general terms.  Thanks to
pkgsrc/devel/coccinelle for helping me find places where int
variables were used to store the results from pointer arithmetic
or strlen().  Thanks to Martin Husemann for testing.
2013-01-05 09:40:15 +00:00
wiz
6192c2fd3b Set GNATS_ADDR to gnats-bugs@NetBSD.org, like in all other places.
Ok spz for admins.
Requested by Bug Hunting.
2012-12-21 09:40:17 +00:00
apb
ee9a2498cf Adjust everything under src (but outside src/tools) to use
the TOOLDIR version of libnbcompat, associated include files,
and associated defs.mk file, instead of the version from the
.OBJDIR of src/tools/compat.  This should fix PR 47188.
2012-12-02 12:55:27 +00:00
apb
6f1f99ca7a Teach gcc4.1's cpp about the magic __COUNTER__ macro,
which returns a unique integer each time it is expanded.
This code was written without reference to any other
implementation of the same feature.
2012-11-24 09:07:44 +00:00
skrll
042a670178 Remove GDB 6. 2012-11-06 11:51:46 +00:00
joerg
135a988400 Don't link explicitly against libsupc++, just use the normal C++
linkage.
2012-08-10 12:12:18 +00:00
joerg
f184a5c33b Only build gcc here, if MKGCC != no. 2012-08-10 12:11:30 +00:00
christos
edd104c76c PR/46704: Antti Kantee: Recognize tilera in config.guess 2012-07-19 19:34:26 +00:00
matt
0bca5478cf Fix ffssi2 again.
Use a UNSPEC for the internal ffs.
use (match_dup 1) since Z will be set if it is 0 just as if tst:SI was done.
This can let gcc produce better code if you code appropriately.
2012-07-16 17:39:29 +00:00
ragge
f5b90a279f Bugfix builtin ffs, fixes PR port-vax/46677, fix from Paul Koning. 2012-07-11 13:34:45 +00:00
christos
cbd8b26f5f add libexecinfo 2012-06-10 15:04:47 +00:00
christos
2636f12679 revert previous 2012-06-10 15:02:31 +00:00
christos
e848a9e55e add libexecinfo 2012-06-10 00:25:51 +00:00
joerg
12d82dda2b Format strings don't use all arguments in the non-pluralized case. 2012-05-09 07:59:23 +00:00
joerg
b757af438b Disable new -Wstring-plus-int warning where needed for now.
Adjust various places that add GCC-only options to check for the active
compiler first.
2012-04-04 10:59:44 +00:00
christos
0357fcd849 PR/26453: Ken Raeburn: make zero byte allocations return NULL instead of error
out.
2012-03-12 17:41:59 +00:00
plunky
645e112039 format the generated manpage date as per Dd macro, for consistency
(although this field is a free-form footer, mandoc -Tlint warns about it)
2012-03-12 08:19:46 +00:00
wiz
7a488c1ecf Fix typo, now this has a NAME section. 2012-03-11 10:21:25 +00:00
plunky
4dc9c6884a use character escapes for non-ASCII chars, according to mandoc_char(7) 2012-03-11 08:44:45 +00:00
joerg
4acff4c01b Implement sem_timedwait. 2012-03-08 21:59:24 +00:00
christos
540f3eef91 Teach rcs about the new cvs "commitid" keyword, so that we don't bitch each
time we play with an RCS file maintained by CVS.
2012-03-08 21:32:28 +00:00
matt
c706ec8d90 Add bsd_kvm.o 2012-03-02 16:39:26 +00:00
matt
f96b5daea9 target kvm support for mips 2012-03-02 16:38:20 +00:00
apb
0e3dcd74ed Add a definition for ".St -isoC-2011". The equivalent definition
was already in external/bsd/mdocml/dist/st.in
2012-02-21 10:43:46 +00:00
riz
80d0f48794 Add NetBSD 7.0, and make the default 6.0. 2012-02-15 23:06:20 +00:00
matt
703f2e538a Use REG_P, CONST_INT_P, and gcc_unreachable. 2012-02-05 17:34:34 +00:00
matt
a12df59a7e Handle another case the new gcc_assert(0) caught. 2012-02-02 18:33:03 +00:00
matt
b7a0413860 Fix a bug in movmemsi in which couldn't handle sym(reg) expressions properly. 2012-02-02 13:32:17 +00:00
dholland
de37c897b5 Add libquota, as requested by wiz 2012-01-25 22:22:20 +00:00
joerg
a1968c686d Second try to sort out the dangling elses. Just use {} markers.
Produces identical on amd64.
2012-01-12 14:10:56 +00:00
tron
f834547595 Revert last change. The extra while loop intruced by the macro changes
causes an end-less loop in ci(1) which uses "break" inside one of the
macros.

"/etc/security" will now no longer fill up "/" (or "/var" depending on
file-system layout).
2012-01-08 13:57:31 +00:00
joerg
3a4dc84ff4 Don't use dangling elses. 2012-01-06 15:16:02 +00:00
is
1ee229bdd2 Remove pattern for arith_adjacentmem in arm.md, thus fixing
PR toolchain/45576 at a very small run-time cost.

This was suggested by Richard Earnshaw.

A real fix is known to be in gcc 4.5.3 (as in NetBSD-current) and in
gcc-current; don't apply in that tree.
2011-11-07 20:35:39 +00:00
chs
02dd92991a do not build profiling versions of libgcc or libgcc_eh.
they are not used, and they don't build on arm with gcc 4.5.
2011-10-17 14:20:54 +00:00
christos
ecb433c89b undo previous, my sources were wrong. 2011-10-02 15:04:47 +00:00
christos
7cdf1ad1d3 add a missing file, found with the new binutils. No chance we are going
to run mknative again on this, so the manual fix.
2011-10-02 14:24:08 +00:00
apb
b799ebe187 Disable rebuilding files in the source tree. 2011-09-10 20:35:42 +00:00
wiz
7c47a19704 Add libppath. 2011-08-29 08:58:18 +00:00
joerg
4c415c49ae Passes relaxed array boundary checks with clang now. 2011-08-09 13:02:11 +00:00
apb
576d24caba Redo previous change in a better way, as suggested by mrg.
Where the original Makefile.in had this:

	filename: dependencies
		run autosomething to regenerate the file

we now do this:

	filename: dependencies
		@echo "NOT REBUILDING $@"
	NetBSD_DISABLED_filename:
		run autosomething to regenerate the file
2011-07-15 07:04:52 +00:00
apb
a28cf61e18 Sprinkle some "NetBSD_DISABLED_" prefixes in file names in Makefile
targets, to ensure that the associated commands do not run.  This should
prevent source files from being overwritten at build time by autoconf
and friends.  Fixes PR 45132.
2011-07-14 17:29:43 +00:00
martin
7d1d8fc18d replace "(CORE_ADDR) (uintptr_t) NULL" by a simple "0" - avoids the dance
with HAVE_STDINT_H and fixes the tool build for me.
Idea from dsl, OK: mrg.
2011-07-11 18:47:01 +00:00
mrg
fc24f81836 apply a couple of (int) before r_type switch usage. 2011-07-07 07:47:11 +00:00
mrg
55bcf9d074 cast to int as the fake relocs are not in enum elf_ppc_reloc_type, and
GCC 4.5 complains about values not in the enum as part of a switch block.
2011-07-06 09:01:35 +00:00
mrg
fe6ec5094e apply some -O0 for sun2. 2011-07-03 15:26:25 +00:00
mrg
67d3860982 apply some (uintptr_t) between pointers, ints and other casts. 2011-07-03 15:26:06 +00:00
mrg
f15b09a338 use:
extern "C" { stuff-here; }
instead of just
  extern "C" stuff-here;
that GCC 4.5 demands.
2011-06-29 01:59:26 +00:00
mrg
8dbf0d5bb3 prepare to build native gcc 4.5, don't build native gcc 4.1, for HAVE_GCC=45 2011-06-21 04:52:49 +00:00
mrg
0de3f09ec1 remove sh5 code. 2011-06-20 09:27:43 +00:00