summaryrefslogtreecommitdiff
path: root/lib/libc
AgeCommit message (Collapse)Author
2016-09-05correct wording; from rob@2keys.caTheo de Raadt
2016-09-05Delete some abstraction that we don't and won't need.Ingo Schwarze
Declare functions rather than generating declarations with macros. Just call functions rather than mainting function pointer tables. Purge unused arguments. Simplify mbstate_t casting. Garbage collect one empty and one unused function. As a bonus, make mbsinit(3) work at all, it returned garbage in the past due to a missing cast when passing mbstate_t. Apart from that, no functional change. No libc bump needed; only private functions are removed and change prototype and only private structs change size. OK stsp@ mpi@; deraadt@ likes the general direction.
2016-09-04fix Dt;Jason McIntyre
2016-09-04Set errno more consistently, and fix a warning, ok teduNicholas Marriott
2016-09-04rmTed Unangst
2016-09-04oops, name file after main functionTed Unangst
2016-09-04it doesn't say anything yet, but start adding a man pageTed Unangst
2016-09-04Make the key sizes and offsets arrays const, ok teduNicholas Marriott
2016-09-04minor bump for SHA512/256 API additions; ok tedu@ deraadt@Christian Weisgerber
2016-09-04jmc says you can never have too many functions in one man page, so listTed Unangst
the SHA512/256 functions as well.
2016-09-04remove caveats that are either misleading or unhelpful.Ted Unangst
ok jmc naddy
2016-09-03Don't list individual valid CLOCK_* values, instead refer toMichal Mazurek
clock_gettime(2). ok guenther@
2016-09-03Add functions for SHA512/256. The standard says you're supposed to startTed Unangst
with different magic numbers, so we need to add some functions instead of just asking the user to truncate as desired. Sigh. SHA512 is quite a bit faster than SHA256 on 64 bit CPUs, but 256 bit hashes are usually quite sufficient. Best of both. ok deraadt tom
2016-09-03tweak previous;Jason McIntyre
2016-09-03note that getpass is obsolete. removed from posix in 2001.Ted Unangst
2016-09-02fix format stringCharles Longeau
ok guenther@
2016-09-02Make it clear that printable characters of width 0 exist.Ingo Schwarze
While here, polish some wording. OK stsp@
2016-09-02Since we support only one non-ASCII LC_CTYPE locale, we don't needIngo Schwarze
a linked list to store it. No functional change. OK mpi@ guenther@, and stsp@ agrees too that the time is ripe to start such cleanup
2016-09-01bumpOtto Moerbeek
2016-09-01Less lock contention by using more pools for mult-threaded programs.Otto Moerbeek
tested by many (thanks!) ok tedu, guenther@
2016-09-01black magic for sparc page size can goTed Unangst
2016-09-01remove references to sparcTed Unangst
2016-09-01Update list of ioctls for the audio promise. Suggested by jmc@, thanks.Alexandre Ratchov
2016-09-01Mention explicit_bzero(3); ok jmcMartin Natano
2016-09-01retire the sparc library codeTed Unangst
2016-08-29Store the return value of mbrtowc() in a size_t, not int.Todd C. Miller
OK schwarze@
2016-08-28Don't call lstat() before readlink() just to see if it's a symlink,Philip Guenther
as readlink() will tell you that more cheaply. ok millert@
2016-08-27Stop recommending the non-standard and slightly dangerous fgetln(3).Ingo Schwarze
Recommend POSIX getline(3) instead.
2016-08-27When a precision is specified for a string format use strnlen()Todd C. Miller
to determine the length instead of doing it manually. OK schwarze@
2016-08-27improve revision 1.2: in unusual cases, fgetwc(3) can succeedIngo Schwarze
even though ferror(3) is already set; also from Andrey Chernov <ache at freebsd dot org>; OK millert@
2016-08-27Pull in <stdio.h> for NULLPhilip Guenther
ok deraadt@
2016-08-26Pull in <time.h> for clock_gettime()Philip Guenther
ok deraadt@
2016-08-25After read errors, fgetln(3) sometimes succeeded (returning non-NULL)Ingo Schwarze
and failed (setting errno and ferror(3)) both at the same time. That's a bad idea in general, and here in particular since returning partial lines was neither reliable (sometimes, you got NULL anyway) nor predictable (almost always, the line would be truncated long before the actual read error). Instead, on read failure, fail properly and always return NULL. Issue found in a discussion with Andrey Chernov <ache at freebsd dot org> who finally agreed to move FreeBSD into the same direction. The fix is joint work with and OK by millert@.
2016-08-24set the error indicator on malloc(3) failure;Ingo Schwarze
from Andrey Chernov <ache at freebsd dot org>; OK millert@
2016-08-21bugfix: when fgetwc(3) fails, fgetwln(3) must fail as well;Ingo Schwarze
OK jca@ martijn@ millert@
2016-08-21fix obvious typo in the .Dt section numberIngo Schwarze
2016-08-20Sync connect_wait() example with its real usage in ftp(1).Todd C. Miller
2016-08-20Declare all _asr_* debug functions as hidden.Jeremie Courreges-Anglas
Reported by & similar diff by guenther@ some time ago, ok eric@
2016-08-18Add an EXAMPLES section that illustrates how to deal with connect(2)Todd C. Miller
returning EINTR. OK jung@ deraadt@
2016-08-17% is escaped with more %, not backslash.Ted Unangst
2016-08-17Generate syslog warnings for %s fmt strings NULL to "(null)" conversions.Theo de Raadt
Over time we can repair software which performs this non-standard behaviour, and fix bugs along the way. Let's first find out how bad the situation is by deploying this in snapshots. This type of logging is possible because OpenBSD syslog_r(3) -> sendsyslog(2) is side-effect free enough to be used in the bowels of libc. ok tedu
2016-08-17wrterror() is fatal, delete dead code; ok tom@ natano@ tedu@Otto Moerbeek
2016-08-15add a bit of spacing to previous, to keep the notes about deprecatedJason McIntyre
functions out the way of the main body; ok guenther
2016-08-14Reduce qabs() and qdiv() to aliases of llabs() and lldiv().Philip Guenther
Merge the manual pages and call them deprecated there. ok and manpage tweak jmc@, ok natano@
2016-08-14Refer to /etc/passwd consistently as the "legacy password file" andTheo Buehler
remove some references to differences between versions 6 and 7. ok jmc, millert, tedu
2016-08-13modern interfaces should use modern speelings, so spell quad_t as int64_t.Ted Unangst
2016-08-11Between relro and previous RWX work, the __{got,plt}_{start,end} symbolsPhilip Guenther
are no longer needed by ld.so. Move them to the m88k-specific list, as it hasn't made the jump. ok deraadt@
2016-08-09Expand the asynchronous connect information and move it out ofTodd C. Miller
ERRORS and into a new second paragraph. Adapted from NetBSD. OK deraadt@ jmc@
2016-08-09Document that connect goes async when interrupted by signal.Todd C. Miller
2016-08-07Add XOR cookies for lr and sp. Stop saving/restoring r12 to/from the jmpbuf.Philip Guenther
Switch from calling obsolete sig{block,setmask} to directly using the sigprocmask syscall. ok deraadt@ kettenis@