summaryrefslogtreecommitdiffstats
path: root/lib (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* actually, make this identical to what upstream expat ships nowadays by addingjasper2012-07-071-2/+2
| | | | a URL field and removing the empty Requires.
* add expat.pcjasper2012-07-072-1/+82
| | | | | based on a diff by brad ok sthen@
* adapt the NetBSD code to make sigpending work on shjsg2012-07-031-2/+6
| | | | ok miod@ kettenis@
* The kernel returns the pending set; this stub has to save it throughguenther2012-07-011-4/+8
| | | | | | the pointer ok miod@ kettenis@
* make this actually workkettenis2012-07-011-23/+21
| | | | ok miod@
* librthread has eaten libpthread, so build includes via the former insteadguenther2012-06-281-16/+0
| | | | | | | of the latter. Permits the misleading lib/libpthread/Makefile to be removed discussed with deraadt@
* after a talk with ariane, use MAP_FIXED for mquery to avoid the cost oftedu2012-06-261-2/+2
| | | | | | scanning for free space if the hint isn't available. also, on further inspection, this will prevent pmap_prefer from "improving" our hint.
* simplify a little of the posix standards talk. ok deraadt guenther jmctedu2012-06-263-34/+12
|
* Use nl_langinfo(RADIXCHAR) instead of localeconv()->decimal_point inmatthew2012-06-262-4/+10
| | | | | | | | | | | | | | | | printf() and avoid calling it unless needed (i.e., when we have a floating point value to print). This isn't a big concern currently due to our limited locale support and current localeconv() implementation, but it's still technically a data race and implementing POSIX 2008 per-thread locales is likely to make it worse. nl_langinfo() isn't guaranteed by POSIX to be thread-safe either, but at least our current implementation is thread-safe and it's a simpler interface to keep that way. Printing floating point values isn't async-signal-safe anyway due to gdtoa()'s use of malloc(), so that's not an issue. ok deraadt, stsp, millert
* Do not use <machine/float.h>, that is an deformed BSD-ism.deraadt2012-06-259-18/+18
| | | | | It turns out <float.h> is the right file to pull in. ok millert
* In sysconf(), don't bother querying the kernel for userspace-onlymatthew2012-06-241-40/+19
| | | | | | values or features that POSIX now requires to always be provided. From Brad; ok guenther and me; incorporates feedback from millert too
* Change arc4random_uniform() to calculate ``2**32 % upper_bound'' asmatthew2012-06-241-12/+3
| | | | | | | | | | ``-upper_bound % upper_bound''. Simplifies the code and makes it the same on both ILP32 and LP64 architectures, and also slightly faster on LP64 architectures by using a 32-bit remainder instead of a 64-bit remainder. Pointed out by Jorden Verwer on tech@ ok deraadt; no objections from djm or otto
* - document newly imported AF_INET6 support to net_inet_{pton,ntop}gilles2012-06-241-5/+26
| | | | diff from Florian Obser, ok jmc@
* add support for AF_INET6 to inet_net_pton() and inet_net_ntop()gilles2012-06-222-2/+68
| | | | | | using inet_pton() and inet_ntop() as suggested by claudio ok claudio@
* synch build after importespie2012-06-224-273/+72
|
* import 3.7.13espie2012-06-2241-451/+28802
| | | | okay jasper@
* two changes which should improve realloc. first, fix zapcacheregion totedu2012-06-221-10/+19
| | | | | | clear out the entire requested area, not just a perfect fit. second, use mquery to check for room to avoid getting an address we don't like and having to send it back.
* A piece of this is still neededderaadt2012-06-211-0/+16
|
* Update to reflect API changesguenther2012-06-211-17/+26
|
* For now... libpthread is just a container directory, since all the newderaadt2012-06-213-39/+1
| | | | | | | | goop is in librthread. As a result, the top-level Makefile and shlib_version here are simply very confusing and tell lies. Remove them, and update the instructions in libc to not make my mistake again. ok guenther
* __tfork() needs to set the stack address of the new thread in the kernel,guenther2012-06-2117-135/+80
| | | | | | | | | so that it can't get a signal while still running on the parent thread's stack. Also, pass in sizeof(struct __tfork) to provide forward compat when more members are added. This is an ABI change, so switch syscall numbers and bump lib majors this time. ok deraadt@ matthew@
* two small fixes to free page cache. first, we need two nibbles of randomtedu2012-06-201-4/+4
| | | | | | in order to span the the entire cache. second, on free use the same offset to put things in the cache instead of always starting at zero. ok otto
* oops, forgot to install getdtablecount(2); noted by teduderaadt2012-06-191-3/+3
|
* Support larger-than-page-alignment requests in posix_memalign() bymatthew2012-06-182-23/+130
| | | | | | overallocating and then releasing unneeded memory pages. ok otto
* update return values for ibuf_write and msgbuf_writejmatthew2012-06-151-4/+4
| | | | ok gilles@
* Fix typos.matthew2012-06-151-2/+2
| | | | ok pig(6)
* Document mbsnrtowcs() and wcsnrtombs().matthew2012-06-073-117/+150
| | | | ok jmc
* Add support for mbsnrtowcs() and wcsnrtombs() to libc.matthew2012-06-065-175/+204
| | | | Bulk build test by naddy.
* stop editing header by hand, expose script I'm going to use for next versionsespie2012-06-042-1/+46
|
* fix autodetection, somehow I managed to lose that commit earlier ?espie2012-06-041-2/+5
| | | | (brad reminded me we have strerror_r too).
* finish synching libraryespie2012-06-042-3138/+1
|
* minor update to 3.7.12.1espie2012-06-0411-90/+126
|
* crank minor, since a symbol was addedderaadt2012-06-041-1/+1
|
* in imsg_read() avoid calling recvmsg() if we detect that we will be shortgilles2012-06-022-28/+60
| | | | | | | | | | | | | on descriptors, this can be achieved thanks to the new getdtablecount() system call. application may provide a reserve count to ensure that the recvmsg() call is not called when they don't have enough descriptors to work properly. change the API so that transient errors that can be retried immediately are retried within the function right away, whereas transient errors for which the application may want to take action will set errno to EAGAIN. ok deraadt@ and henning@
* Committed the wrong version (an earlier try) of the KERN_FILE2 vsguenther2012-06-021-2/+4
| | | | rthreads fix. Update to the correct one.
* Update STANDARDS section for a few <stdlib.h> functionsguenther2012-06-024-14/+35
| | | | | | Update SYNOPSIS for setkey() to show it's in <stdlib.h> ok jmc@, millert@
* Fix document titles.lteo2012-06-012-6/+6
| | | | ok haesbaert jmc henning
* Improve standards conformance: ecvt(), fcvt(), gcvt(), mktemp(), ttyslot(),guenther2012-06-014-10/+21
| | | | | | | and valloc() are not in the current version, while posix_memalign() mkstemp(), and mkdtemp() are, and setstate()'s argument has lost a bogus 'const'. ok millert@ jmc@ espie@ kettenis@; ports build testing by naddy@
* tweak previous; ok sthen lteojmc2012-05-251-5/+7
|
* Import a number of core functions from libpcap-1.2.0 while preservinglteo2012-05-258-114/+1050
| | | | | | | | | | | | | | | | | | | | | | | | local changes: strncpy() -> strlcpy(), malloc(x * y) -> calloc(x, y), exclude cross-platform cruft, etc. The new functions are pcap_create(), pcap_set_snaplen(), pcap_set_promisc(), pcap_can_set_rfmon(), pcap_set_rfmon(), pcap_set_timeout(), pcap_set_buffer_size(), pcap_activate(), and pcap_statustostr(). This diff was tested on amd64, i386, macppc, and sparc64, where regression tests were done on various pcap-based ports (especially amd64 and i386 where regression tests were run on all pcap-based ports). Testers also tried running pcap-based ports that they are familiar with to ensure that there is no behavioral change. tcpdump and pflogd in base were also tested by different testers. The new pcap_* functions were tested with a proof-of-concept Snort 2.9 port for many months. Thank you to everyone who helped test this diff and provided feedback: haesbaert@, sthen@, matthew@, gonzalo@, brett@, Rodolfo Gouveia, Aaron Bieber, Markus Lude, and Ray Percival. ok haesbaert sthen henning
* some symbols actually changed.espie2012-05-241-3/+3
| | | | | | | | visible interface, not "officially" published, but still... technically visible. Better safe than sorry: major bump. as prodded/discussed with deraadt@
* fuck, forgot we still have !C99 compilersespie2012-05-241-1/+2
| | | | reminded by deraadt@
* Make the "default" string a valid device name that has the same effectratchov2012-05-235-15/+29
| | | | | as NULL has. This will (hopefully) simplify ports where the user passes the device string.
* Don't misleadingly suggest that fpos_t only needs to be a complex typematthew2012-05-221-4/+2
| | | | | on non-UNIX systems: it's a complex type on Linux, and to be C99 compliant we'll probably need it to be a complex type too.
* Document that ftell() may return EOVERFLOW.matthew2012-05-221-6/+31
| | | | Diff from Tim van der Molen.
* those no longer belong, they've been move to the test/ directory upstreamespie2012-05-225-1438/+0
|
* lemon bug: checking lines is not enough, we have to be sure we're at theespie2012-05-221-1/+2
| | | | | end of file to make sure the include file hasn't been altered. problem reported by landry@, deraadt@
* bump version, regen sqlite3.h with our changes (and bump version)espie2012-05-223-9/+22
|
* merge syncespie2012-05-221-18/+16
|
* import sqlite 3.7.12 (tested by landry@)espie2012-05-2268-788/+6752
|
Copyright © 1996 – 2023 Jason A. Donenfeld. All Rights Reverse Engineered.