Commit Graph

12385 Commits

Author SHA1 Message Date
christos
f1ed7fe14d remove attribute keyword 2014-04-18 02:41:32 +00:00
christos
04a020c93e handle constant attribute 2014-04-18 02:41:17 +00:00
christos
0d1cbc6a00 don't include fmemopen in tools builds. Since tools does not define
_NETBSD_SOURCE, we don't get the fmemopen prototype
2014-04-18 02:17:14 +00:00
christos
c8b1f28fd2 Never return the attribute keywords if we are not in attribute. 2014-04-18 01:39:57 +00:00
christos
1f0ca5f12c builtins only for NetBSD since fmemopen is not portable. 2014-04-18 01:18:54 +00:00
christos
31ce3096fd Add some builtins 2014-04-18 01:15:07 +00:00
christos
687432e7a3 Handle the rest of gcc __attribute__ s. 2014-04-18 00:23:46 +00:00
christos
59bbfbb11d only fill memory with junk if BLKDEBUG 2014-04-18 00:21:14 +00:00
christos
a46cc2f317 handle LDOUBLE 2014-04-18 00:20:37 +00:00
christos
20ba703761 make outname print where it came from in case of error 2014-04-17 18:52:03 +00:00
christos
10e73b5c30 - add a function to dump a node
- better diagnostics on abort
- allow converting a constant
- initialize right node now that we trash memory this makes a difference.
  before it was NULL.
2014-04-17 18:23:18 +00:00
christos
e34b25460d typeof and extension don't need parens 2014-04-17 17:29:24 +00:00
christos
f87049800a Trash freed memory so we catch bad uses. 2014-04-17 16:30:05 +00:00
christos
8a20351779 Don't free memory; it is used later. 2014-04-17 16:29:26 +00:00
ginsbach
42bc9b7b73 Hook getaddrinfo(1) into the build. Hi riastradh! 2014-04-15 22:09:27 +00:00
snj
fc99b6a4d6 Change some standard exit status text to ".Ex -std"
From Eitan Adler in PR bin/47790.
2014-04-13 01:45:34 +00:00
christos
e9d34c2f4d mention -V none 2014-04-11 17:41:09 +00:00
christos
0b1777bf87 recognize -V none 2014-04-11 17:30:03 +00:00
dholland
1c111fd5cf typo in comment 2014-03-23 04:58:16 +00:00
pooka
3f0878753a On glibc systems, issue a warning if LD_DYNAMIC_WEAK is not 2014-03-21 16:26:30 +00:00
riastradh
5c1b7f4542 Clarify some wording in getaddrinfo.1. 2014-03-19 18:21:39 +00:00
ginsbach
aa8e09363a Add missing include files. Don't rely on them bein pulled in as a
side effect of other includes.  Fixes build errors when getaddrinfo(1)
is added to the usr.bin tree.
- need <sys/socket.h> for address family (AF_*) and socket type (SOCK_*)
- need <stdio.h> for printf() and friends
2014-03-19 01:24:32 +00:00
riastradh
6cb10275d0 Merge riastradh-drm2 to HEAD. 2014-03-18 18:20:35 +00:00
dholland
c49e034e83 Fix gcc48 build. No obvious reason why nobody else has hit this... 2014-03-16 01:07:46 +00:00
dholland
19d6c717ae Fix gcc48 build without Kerberos.
While here, rectify related bug where if gethostname() failed, login
would null-terminate uninitialized stack trash and cheerfully pass it
on to kerberos. In this case, revert to "amnesiac" instead.
2014-03-16 00:33:13 +00:00
ginsbach
55791dff41 add missing break 2014-03-14 13:17:18 +00:00
jdc
3db3d1f929 Restore lost break. 2014-03-08 20:51:20 +00:00
riastradh
da030fc8bd `Controller' takes two l's. 2014-03-06 15:00:21 +00:00
matt
b1f346caa8 Use OBJECT_FMTS from <bsd.own.mk> to determine elf32/elf64 needs 2014-03-02 03:55:19 +00:00
matt
cbff955678 Use OBJECT_FMTS from <bsd.own.mk> 2014-03-02 03:42:50 +00:00
ginsbach
28ffd7e64f AF_MAX isn't an address family so don't include it in the generated table. 2014-02-27 01:17:13 +00:00
joerg
a400a0406f struct member can't be null, if the pointer itself is valid. 2014-02-27 00:49:46 +00:00
skrll
6d3ceb1d61 Rename NetBSD/hp700 to NetBSD/hppa.
Unfortunately our VCS isn't very helpful here.
2014-02-24 07:23:38 +00:00
mlelstv
8ab37b8ab4 skip unparseable addresses. 2014-02-23 07:50:01 +00:00
dsl
dc210c7fbd Remove some pointless inclusions os sys/user.h 2014-02-19 20:42:14 +00:00
christos
34fd6c876e add __extension__ and typeof 2014-02-18 22:01:36 +00:00
christos
a675eb32ca __real__ and __imag__ produce lvalues 2014-02-18 20:43:36 +00:00
uwe
28b0656f7c config() in manconf.c now verifies _build (and _crunch) command
templates with fmtcheck(3) so annotate the printf that uses these
commands as safe with a __format_arg wrapper and drop
-Wno-format-nonliteral.

XXX: Using local wrapper for now, solving this in general would be
nice, but it raises namespace pollution issues.

XXX^2: catman(8) also uses manconf.c and uses _build and _crunch so it
can also benefit from this (but see above).
2014-02-17 03:10:12 +00:00
uwe
20ec07777a Check _build and _crunch commands with fmtcheck(3), warn about and
ignore bad ones.
2014-02-17 02:53:48 +00:00
sjg
2adf342aef Put .INCLUDEDFROM* in the right section 2014-02-15 18:55:30 +00:00
christos
7757a9ebde delete the correct variables on eof
fix typo in comment (from sjg)
2014-02-15 00:17:17 +00:00
christos
d2bf840533 fix protocol printing order. 2014-02-14 20:43:34 +00:00
wiz
26dba06cb7 Fix some typos found by Rich Neswold and reported in PR 48597. 2014-02-13 22:36:28 +00:00
sjg
e84a63caab Reduce the calls to realloc from brk_string by basing initial argmax guess
on length of input.

Reviewed by: christos
2014-02-12 01:35:56 +00:00
pooka
9f33b0ead1 Use Var_Value() instead of getenv() to obtain MAKEOBJDIR{,PREFIX}.
Fixes "make -X MAKEOBJDIR=foo".  Purely uncoincidentally, build.sh
creates the nbmake wrapper with MAKEFLAGS -X on FreeBSD.
2014-02-07 17:23:35 +00:00
wiz
47c7c18d7f Correct number of output files in example.
From Jason McIntyre <jmc@OpenBSD>.

While here, also explain about the extra file in the example and
put .Ex into its own section in the correct place. Bump date.
2014-02-04 19:48:48 +00:00
njoly
a935b103a7 Handle another declaration after statement case for lint in c99 mode.
Add the corresponding testcase.
2014-02-04 08:08:59 +00:00
wiz
97cd0ff474 From Igor Sobrado in private email (based on his OpenBSD commit):
improve POSIX compliance by continuing to process the remaining file
    operands after not finding an input file.

    from the IEEE Std 1003.1-2008 (``POSIX.1'') rationale:

    "Unlike other utilities, some historical implementations of cut
    exit after not finding an input file, rather than continuing to
    process the remaining file operands.  This behavior is prohibited
    by this volume of POSIX.1-2008, where only the exit status is
    affected by this problem."

    joint work with jmc@, who identified the compliance issue, and millert@

    ok millert@, jmc@
2014-02-03 20:22:19 +00:00
pooka
dd477c0e3e Clarify that -l params must be given after options. 2014-01-28 14:02:54 +00:00
pooka
7943bb5b88 * make -h actually work
* give usage for invalid usage
2014-01-28 13:58:25 +00:00