summaryrefslogtreecommitdiffstats
path: root/lib/libc (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* simpler and saner implementation for tcp_write(). now fully handleseric2013-04-012-40/+16
| | | | | | short writes. prodded by deraadt@
* fix alignment issue in the structure filled by gethostby{name,addr}()eric2013-04-011-10/+23
| | | | spotted by naddy@
* More SYNOPSIS, ERRORS, and STANDARDS updatesguenther2013-03-317-26/+62
|
* Principle of least astonishment: implement nameserver retry/backoff aseric2013-03-312-8/+10
| | | | in the former resolver.
* Add a couple commas, as suggested by jmc@guenther2013-03-311-3/+3
|
* getaddrinfo is now thread-safe.brad2013-03-311-6/+2
| | | | ok eric@
* Document more difference between parent and child.guenther2013-03-311-9/+27
| | | | | | Updates to ERRORS, and STANDARDS ok jmc@
* Updates to RETURN VALUES, ERRORS, and STANDARDSguenther2013-03-311-3/+14
| | | | ok jmc@
* Updates to SYNOPSIS, ERRORS, and STANDARDSguenther2013-03-317-38/+86
| | | | ok jmc@
* Add ERRORS and STANDARDS sections; reorder and reword description toguenther2013-03-311-14/+39
| | | | | | | clarify PROT_NONE vs PROT_* and what protections are actually sure to be supprted. ok jmc@, deraadt@
* Use consistent wording for the EINVAL, EBADF, ENOTDIR, and EACCES casesguenther2013-03-3113-59/+250
| | | | of the various *at() functions.
* allow "0" as service name for raw sockets.eric2013-03-301-2/+2
| | | | ok sthen@ bluhm@
* Do not assume local nameserver if resolv.conf doesn't exist, just useeric2013-03-302-8/+7
| | | | | | /etc/hosts. discussed with deraadt@
* tweak previous;jmc2013-03-302-4/+4
|
* don't ever skip reload the very first time; detective work by rpe@; ok eric@otto2013-03-301-2/+2
|
* Various updates, mostly to SYNOPSIS, ERRORS, and STANDARDS, butguenther2013-03-3020-132/+216
| | | | | | | | some other fixes included... *at() function ERRORS description suggested by matthew@ ok deraadt@
* properly handle scope when parsing IPv6 addresses.eric2013-03-291-2/+32
| | | | ok bluhm@ naddy@ sthen@
* res_querydomain()'s code to terminate the domain with '.' had the assignmentguenther2013-03-291-2/+2
| | | | | | | flipped so that it always used a domain of ".." Heavy lifting by otto@ ok eric@ otto@ miod@
* correctly show the scope for IPv6 addresses in getnameinfo()eric2013-03-291-4/+27
|
* Do not fail in gethostbyname_async() if there are multiple addresses ineric2013-03-291-3/+3
| | | | a DNS packet...
* no Pp before/after Sh;jmc2013-03-281-2/+1
|
* pathconf/fpathconf are now compliant with POSIX-2008.guenther2013-03-281-5/+14
| | | | Adjust the errnos a bit.
* Switch libc and libm to use strong aliases rather than weak aliasesmartynas2013-03-2828-204/+68
| | | | | | | | | | | where appropriate. Among other things makes the symbols consistent across all architectures (notably where ldbl mantissa is 53 bits). While at it, kill unused LINTLIBRARY/PROTOLIB1 cruft which was there to trick lint into recording the right prototypes for aliased functions. Most of the work done at the awesome n2k13 hackathon. Agreed by kettenis@, guenther@, matthew@.
* switch to the new resolver implementation.eric2013-03-281-1/+2
| | | | requested by many.
* Enable fmemopen(3) and open_{w,}memstream(3)eric2013-03-281-12/+13
| | | | commiting on behalf of mpi@
* accept and use any protocol specified by the caller.eric2013-03-281-14/+10
| | | | issue spotted by naddy@
* Correct version numbermpi2013-03-282-6/+6
|
* Add support for _PC_TIMESTAMP_RESOLUTION for ffs/mfs, cd9600, ext2,guenther2013-03-281-2/+4
| | | | | | | msdos, NFS, fifos and devices, plus support for querying it in getconf(2) and the requisite pathconf(2) manpage blurb ok tedu@
* explicitely -> explicitly;jmc2013-03-271-2/+2
|
* Add an open_wmemstream(3) implementation and fix various issues formpi2013-03-273-55/+211
| | | | | | fmemopen(3) and open_memstream(3). With inputs from millert@, stsp@, guenther@, tedu@ and matthew@
* Only use the search domains for DNS lookups, as the current resolver does.eric2013-03-277-85/+63
| | | | | Better not diverge too much in behavior at this point. Typo fix and doc update while there.
* protect the PRG state with a mutex in res_randomid().eric2013-03-261-2/+13
| | | | ok deraadt@ guenther@ djm@
* sysconf(_SC_SEM_VALUE_MAX) should return SEM_VALUE_MAX, not someguenther2013-03-241-7/+3
| | | | | unrelated sysv sem limit. Similarly, sysconf(_SC_SEM_NSEMS_MAX) should return -1 instead of another sysv sem limit.
* dlerror() should not return const char *, as noted by landryderaadt2013-03-241-3/+3
| | | | ok guenther
* Do not need machine/param.hderaadt2013-03-211-2/+1
|
* Use a realloc() loop around the sysctl() for NET_RT_IFLIST, in case anderaadt2013-03-201-9/+21
| | | | | interface is added at just the right... wrong moment. ok millert dlg
* threadsafe random().tedu2013-03-151-8/+43
| | | | reported by ajacoutot and (much earlier, sorry) by Alexey Suslikov
* tedu faith(4), suggested by todd@ some weeks ago after a submission bympi2013-03-141-13/+2
| | | | | | dhill. ok krw@, mikeb@, tedu@ (implicit)
* Return EINVAL if there are fewer than six template Xs in the path.guenther2013-03-122-24/+30
| | | | ok deraadt@ millert@
* Fix kernel profiling on MP systems by using per-CPU buffers and teachmpi2013-03-121-8/+14
| | | | | | | | | | | kgmon(8) to deal with them, this time without public header changes. Previously various CPUs were iterating over the same global buffer at the same time to modify it and never ended. This diff includes some ideas submited by Thor Simon to NetBSD via miod@. ok deraadt@, mikeb@, haesbaert@
* SOCK_SEQPACKET is supported on AF_UNIX too.dlg2013-03-121-5/+6
| | | | ok jmc@ matthew@
* handle ECONNABORTED errors from accept(). In many code blocks they can bederaadt2013-03-111-2/+3
| | | | | ignored silently and without aborting, much like EINTR and EWOULDBLOCK are. ok's from various maintainers of these directories...
* document LOG_ODELAY; from Daniel Dickmanjmc2013-03-111-2/+9
| | | | ok millert
* Describe interaction between write and O_APPEND.tedu2013-03-091-2/+9
| | | | Pointed out by Sachidananda on tech
* Fix return value of wcrtomb() in single-byte locales if the s argument is NULL.stsp2013-03-071-2/+2
| | | | | wcrtomb() must pretend to store one byte (NUL-terminator) in this case. Patch by Vladimir Tamara Patino. ok guenther
* Oops: confstr()'s return value should include the NUL in its countguenther2013-03-071-5/+5
| | | | Problem pointed out by Andres Perera (andres.p (at) zoho.com)
* tweak wording of scanf return value; prompted by Jan Stary; ok jmc@otto2013-03-051-4/+5
|
* Update STANDARDS section as appropriate for C99.brad2013-03-042-7/+9
| | | | ok guenther@
* Fix the combination of 'j' format flag and the XPG "<num>$" modifier.guenther2013-03-021-2/+11
| | | | ok deraadt@ kettenis@ millert@
* zap end of line whitespace;jmc2013-03-021-5/+5
|
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.