f722916cc6
Indent delay slots. Whitespace change only, same object code produced.
2003-09-22 01:57:16 +00:00
4df9d6555a
Bring back previous sparc64 hack, but only #ifdef old gcc.
...
This is used when building libnbcompat for host tools - why it is included
when building on NetBSD is beyound me.
2003-09-21 17:15:36 +00:00
55acdca8c5
No more sparc64 compiler workarounds
2003-09-19 16:30:31 +00:00
57e4de6099
add warning on realloc() size de-synchronization. from openbsd
2003-09-19 05:36:59 +00:00
bb21f0beb4
remove pw_dup()
2003-09-18 09:24:35 +00:00
430f0af1db
remove pw_dup(3), for now at least
...
follows discussion on mailing lists, discussed in private e-mail with Itojun
2003-09-18 09:13:14 +00:00
465ed898bd
bump minor for sa_unblockyield syscall
2003-09-16 13:55:41 +00:00
5c40d56471
fix SA/pthread pagefault failure:
...
- prevent BLOCKED upcalls on double page faults and during upcalls
- make libpthread handle blocked threads which hold locks
- prevent UNBLOCKED upcalls from overtaking their BLOCKED upcall
this adds a new syscall sa_unblockyield
see also http://mail-index.netbsd.org/tech-kern/2003/09/15/0020.html
2003-09-16 13:51:31 +00:00
49b0281b33
Shut lint up.
2003-09-16 07:56:51 +00:00
94da0d16ac
avoid overflow during multiply. David Laight
2003-09-15 23:38:20 +00:00
64408794ea
make popen/pclose thread safe
2003-09-15 22:30:38 +00:00
d46db7b926
In is only to be used in SYNOPSIS; Aq Pa is better for other places.
...
Problem noticed by kleink.
2003-09-14 09:15:23 +00:00
5a12b7062e
Various fixes.
2003-09-14 09:02:49 +00:00
b9f1052a4d
Fix some typos, attempt some markup normalization, align wording a little
...
more with other manual pages.
2003-09-13 23:00:38 +00:00
8694ace9bd
forgot to install timer_settime.
2003-09-13 22:50:21 +00:00
960ffa78a4
Fix markup.
2003-09-13 22:44:23 +00:00
9865e7ff1d
Sprinkle restrict into timer_{create,settime}(2).
2003-09-13 22:31:04 +00:00
f6da4d89f2
weak alias for pw_dup
2003-09-13 22:28:12 +00:00
226b4082f4
Add timer man pages. Hi Nathan!
2003-09-13 22:22:29 +00:00
220a141637
correct setup of ru_b/a for 16/20 bit case
2003-09-13 21:29:57 +00:00
1ecc9b581c
use internal names for randomid()
2003-09-13 21:27:43 +00:00
ab0a78e1a7
- don't define __LIBC12_SOURCE__
...
- need to SIG_IGN SIGSYS for the version check
- simplify code
2003-09-12 20:42:15 +00:00
8adecf9fbe
need __HAVE_SIGINFO to get the right flavor of the sigreturn symbol.
2003-09-12 13:29:46 +00:00
86ad7a15ee
__sigreturn14 -> compat_16___sigreturn14; thanks yamt
2003-09-12 12:42:34 +00:00
5a93912d03
No reason for __ in static variable.
2003-09-11 20:24:39 +00:00
c18cae3e90
For kernels that have COMPAT_16 enabled, detect that and use the old
...
sigcontext trampoline code when SA_SIGINFO is not set.
2003-09-11 20:23:46 +00:00
e4a907f72e
we only have it if __HAVE_SIGINFO is not defined.
2003-09-11 15:29:10 +00:00
07f2a2a309
check malloc failure. yamt
2003-09-11 11:24:33 +00:00
fb114e34c8
Add semicolons when enumerating variables (looks more natural);
...
new sentence, new line; sort SEE ALSO.
2003-09-10 14:27:22 +00:00
a3c9529278
Reflect reality after siginfo changes.
2003-09-10 14:13:07 +00:00
8d66b069c8
Use standard section headers, suggested by jmc@openbsd. Sort sections.
2003-09-10 13:30:16 +00:00
9e4a539564
Pull over some improvements from OpenBSD's v1.26 of this file, including
...
SEE ALSO and RETURN VALUES sections.
2003-09-10 09:35:41 +00:00
0e9a92473d
Fix Nd; use In instead of Fd; grammar fixes.
2003-09-10 09:16:45 +00:00
ef0a2c06dc
Another gcc3 problem: "exp" shadows a builtin. Rename it to "expo".
2003-09-10 07:20:13 +00:00
732fa344d5
This didn't build with gcc3 because it was missing <string.h> for memset.
2003-09-10 07:12:33 +00:00
89d02ff30e
Use In instead of Fd for header files.
2003-09-10 05:51:06 +00:00
0186de22d3
getopt return value has to be held in int, not char.
2003-09-10 00:03:42 +00:00
0257a0c524
bump minor for randomid(3) and pw_dup(3)
2003-09-09 22:18:08 +00:00
cfbab92fdb
use randomid(3). actually generate random ID for DNS queries (previous
...
fix was incomplete)
2003-09-09 22:16:58 +00:00
00552ebbd7
add randomid(3).
2003-09-09 22:10:33 +00:00
24a3a1d964
have pw_dup(3). from openbsd
2003-09-09 22:01:56 +00:00
185463aee4
Fix typos and declare a missing variable in example. From Quentin Garnier in PR 22733.
2003-09-09 19:31:53 +00:00
5c945215d0
__RCSID police
2003-09-09 03:56:39 +00:00
35beb40305
have __rpc_getxid.c
2003-09-09 03:56:23 +00:00
8b08fa0d3b
use niels provos' random sequence number generator to initialize RPC XID.
2003-09-09 00:22:17 +00:00
dca9ccd375
Use St -isoC-9X instead of St -isoC9X, to be in sync with groff.
2003-09-08 17:54:31 +00:00
a2f146d519
Mention RFC 3493; inet_{ntop,pton} are in 1003.1-2001; move XNS5.2 to HISTORY.
...
Okayed by kleink.
2003-09-08 17:21:31 +00:00
c4f582bcc6
Fix typos in comments; from Brian Poole via jmc@openbsd.
2003-09-07 19:06:50 +00:00
47190e80b8
Consistently use 'RFC 1234' instead of 'RFC1234' or 'RFC-1234'.
...
From jmc@openbsd.
2003-09-07 16:22:20 +00:00
ec30b8c224
Add __sigtramp2.S
2003-09-06 22:56:47 +00:00
cdf03491b8
rename exp to expo to avoid global shadowing warning.
2003-09-06 22:47:56 +00:00
28b6d65b33
bump for i386 signal trampoline; not strictly necessary, but convenient
...
for people who have an unsuccessful build.
2003-09-06 22:11:50 +00:00
9c5c2215ea
siginfo signal trampoline code.
2003-09-06 22:10:40 +00:00
9f18c21c61
Add Xref to dirname(1).
2003-09-06 20:58:39 +00:00
d27a834c8e
Add Xref to basename(1).
2003-09-06 20:58:11 +00:00
4bfb9a4429
Add __warn_references for deprecated lastlogxname. Okayed by christos.
2003-09-06 16:42:10 +00:00
08fc9c3284
move provos' copyright closer to res_randomid()
2003-09-06 11:46:32 +00:00
a970cdd909
bring in openbsd lib/libc/net/res_random.c for truely-random res_randomid().
2003-09-06 11:40:52 +00:00
4b3391a2a6
%F is a date conversion, not a time converstion.
...
don't bother to bump Dd
2003-09-03 16:10:56 +00:00
f88c193847
Mark up NULL; bump date for last; drop trailing space.
2003-09-01 15:40:46 +00:00
cf33db0f3a
fix the problem about resetting the state of iconv().
...
iconv() was not conforming to the SUSv3 at this point.
reported by Bruno Haible.
2003-09-01 06:16:13 +00:00
a24c4806d6
Nm -> Fn getnameinfo; grammar improvements; markup improvement;
...
all from jmc@openbsd, thanks.
2003-08-28 12:03:08 +00:00
520d0eab98
Bump date for CAVEATS; fix punctuation usage in one place;
...
drop trailing space; use \*[Am] instead of & for HTML generation.
2003-08-28 09:39:11 +00:00
e3a97c1624
use NI_NUMERICHOST for 2nd invocation
2003-08-28 01:49:30 +00:00
9de573eee1
add int error in CAVEAT
2003-08-28 01:46:45 +00:00
91c72951cb
add CAVEATS section
2003-08-28 01:43:14 +00:00
6333112bb9
plural "characters" for ret of {v}snprintf
2003-08-27 18:47:14 +00:00
2194ee2027
Add a comment that explains that
...
. this function cannot have _PROF_PROLOGUE
. this function cannot be called via PLT
Change ENTRY to NENTRY for the former.
XXX: The latter is important in the gcc3 world, that have shared
libraries. We will need to play tricks with .hidden to make sure
every shared library gets its own private __udivsi3 that it can call
directly, without going through the PLT.
2003-08-27 15:03:06 +00:00
7666801285
Add man page for getlastlogx and friends.
2003-08-26 17:37:48 +00:00
a200eb2c89
Bump to 101 for getlastlogx change, says kleink.
2003-08-26 17:05:13 +00:00
c7025ad491
Add note about rename getlastlogx on next major bump.
2003-08-26 17:03:33 +00:00
31fd31ccf7
Make getlastlogx have the pathname to the lastlogx database as first
...
argument, to be consistent with updlastlogx.
Approved by christos, reviewed by kleink.
[The lastlogxname function should not be used any longer.]
2003-08-26 16:48:32 +00:00
f847c1a7f3
When updating wtmpx, if the write fails make sure to close the file (to
...
release the lock held on it). Use O_SHLOCK when adding just one record
since O_APPEND will enforce the atomicity that is required.
2003-08-25 23:09:37 +00:00
0303d5ba8a
execl, execle and execlp must be weak-aliased.
2003-08-24 21:04:53 +00:00
45cc3e349e
Need namespace.h to generate references to internal names.
2003-08-24 15:14:18 +00:00
f518dc7e04
Sprinkle some restrict, repair a minor nit in iconv(3)'s synopsis.
2003-08-22 14:05:03 +00:00
42a8e50b21
Add missing prototypes to make softfloat compile with -DSOFTFLOAT_NEED_FIXUNS.
2003-08-21 02:12:33 +00:00
79b2073eea
Significantly reduce the non-monotonicity of this function.
...
The way time is apportioned to interrupt/system/user makes it impossible
to guarantee monotonicity except by saving the previous value!
2003-08-19 08:31:18 +00:00
6f517be9eb
Return the right value for modf(). Fixes PR 22503.
2003-08-16 21:45:50 +00:00
ac6cbd4ac3
Pop the caller's PIC prologue and push our own, as we can be called
...
by other shared libraries with different GOTs (e.g. libposix).
2003-08-12 13:28:55 +00:00
d900731978
Resolve dlsym(3) and friends directly so that dlsym(RTLD_NEXT,...) works.
...
Previously dlsym resolved to the version in crt0.o or libc which would
mean that the caller's shared object couldn't be determined correctly
using __builtin_return_address(0).
Mainly from FreeBSD, but adapted by me. Benefits of this solutions are:
- backward comptibility maintained
- existing broken binaries are fixed with a new ld.elf_so
- __mainprog_obj can be removed from crt0.o
- we do the same thing as FreeBSD
Fixes PR 22067.
OKed by Jason and Christos.
2003-08-12 09:18:38 +00:00
bae50cfb57
correct an assertion about minimum stack size: MINSIGSTKSZ is sufficient
2003-08-11 13:25:30 +00:00
4907369a14
export size_t __getcachelinesize(void) for memset()
2003-08-11 02:11:23 +00:00
6ba9f44930
Repeat after me: Thou shalt run the regression tests.
2003-08-10 20:39:15 +00:00
51471a0f3b
Be more explicit about the difference between the err and errx (and
...
similar) functions. Based on wording from Jeremy C. Reed.
Closes PR 22424.
2003-08-10 10:45:33 +00:00
151375e124
avoid some lint errors in the previous
2003-08-10 08:24:52 +00:00
eaa9779f55
Split into a 64 bit version (straight forward shifting and masking) and
...
a 32 bit one with two 32 bit operations with two temporaries later
combined.
This avoids the strict alias violating magic used before, thereby
probably fixing PR bin/22313.
2003-08-09 22:37:49 +00:00
eb7c1594f1
Move UCB-licensed code from 4-clause to 3-clause licence.
...
Patches provided by Joel Baker in PR 22280, verified by myself.
2003-08-07 16:42:00 +00:00
2e48312004
be more pedantic to strlcpy/strcat errors
2003-08-03 04:35:50 +00:00
785d6ef678
use strlcat for safety
2003-08-03 04:14:00 +00:00
d6220d9ca5
fix off-by-one error in realpath()
2003-08-03 04:12:01 +00:00
a93ea220fc
Rework how dependency generation is performed:
...
* DPSRCS contains extra dependencies, but is _NOT_ added to CLEANFILES.
This is a change of behaviour. If a Makefile wants the clean semantics
it must specifically append to CLEANFILES.
Resolves PR toolchain/5204.
* To recap: .d (depend) files are generated for all files in SRCS and DPSRCS
that have a suffix of: .c .m .s .S .C .cc .cpp .cxx
* If YHEADER is set, automatically add the .y->.h to DPSRCS & CLEANFILES
* Ensure that ${OBJS} ${POBJS} ${LOBJS} ${SOBJS} *.d depend upon ${DPSRCS}
* Deprecate the (short lived) DEPENDSRCS
Update the various Makefiles to these new semantics; generally either
adding to CLEANFILES (because DPSRCS doesn't do that anymore), or replacing
specific .o dependencies with DPSRCS entries.
Tested with "make -j 8 distribution" and "make distribution".
2003-08-01 17:03:41 +00:00
05602c761d
add a cast appease gcc3.3
2003-07-27 05:04:02 +00:00
99410184e7
netbsd.org->NetBSD.org
2003-07-26 19:24:24 +00:00
d5fb781871
More properly host-tool-ify stat
2003-07-25 03:21:15 +00:00
6f4e912af4
call malloc_init() on first call to free/realloc, to make malloc.conf to
...
take effect. in response to PR 22223
2003-07-23 08:15:02 +00:00
6d30f1a3db
Disable cancellation on the first lock, too, not just on internal
...
recursive locks (Duh).
Disable cancellation around the cond_wait() call, since that's also a
cancellation point. Arguably, that would be better handled with
pthread_cleanup_*(), but stubbing those for libc is difficult, and the
current non-exception-based implementation of cleanup handlers is
probably no faster than disabling and reenabling cancellation.
Finally, it only happens in the slow path where the thread is going to
sleep anyway...
2003-07-22 00:56:25 +00:00
5a5cd67fff
Set the _LOCKINTERNAL() count on initial locking as well as subsequent
...
recursive locks.
2003-07-21 22:24:47 +00:00
31ede4a9e3
Pull in host-tool config.h, if necessary.
2003-07-21 07:10:28 +00:00
d3ebb21ecd
Use threadlib stub rather than pthread_setcancelstate() directly.
2003-07-19 05:06:16 +00:00
uwe