summaryrefslogtreecommitdiffstats
path: root/lib
Commit message (Collapse)AuthorAgeFilesLines
* Fix the amd64 and pc98 versions of ISC_ATOMIC_ARCH with some helpdougb2007-06-031-5/+3
| | | | | | from ru@. Take a guess at what might work on arm to try and fix the build.
* Update generated files for BIND 9.4.1dougb2007-06-029-220/+512
|
* Update bmake glue for the BIND 9.4.1 import.dougb2007-06-028-18/+64
| | | | | This includes a return to building with threads, since one of the major focuses of the 9.4.x branch is to improve thread performance.
* - Work-around the already partially broken rusage support in kvm byjeff2007-06-011-0/+2
| | | | | | completely disabling it until a full solution is agreed upon. Pointy hat to: me
* Export quad symbols. They were previously commented out. These symbolsdeischen2007-05-311-6/+6
| | | | | | | | | really shouldn't be exported since they should be pulled from libgcc, but the build of some applications is broken and they expect to see them in libc. glibc exports these symbols, although Solaris doesn't appear to, so export them for compatibility's sake. After discussion with: kan
* Some libc symbol map cleanups.deischen2007-05-3111-31/+0
| | | | | | | | | | | | | | | | | | | | | net: endhostdnsent is named _endhostdnsent and is private to netdb family of functions. posix1e: acl_size.c has been never compiled in, so there's no "acl_size". rpc: "getnetid" is a static function. stdtime: "gtime" is #ifdef'ed out in the source. some symbols are specific only to some architectures, e.g., ___tls_get_addr is only defined on i386. __htonl, __htons, __ntohl and __ntohs are no longer functions, they are now (internal) defines in <machine/endian.h>. Submitted by: ru
* Merge the relevant part of rev.1.14 of s_cbrt.c (a micro-optimizationbde2007-05-291-3/+3
| | | | | | involving moving the check for x == 0). The savings in cycles are smaller for cbrtf() than for cbrt(), and positive in all measured cases with gcc-3.4.4, but still very machine/compiler-dependent.
* libarchive 2.2.3kientzle2007-05-2973-1372/+3889
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | * "compression_program" support uses an external program * Portability: no longer uses "struct stat" as a primary data interchange structure internally * Part of the above: refactor archive_entry to separate out copy_stat() and stat() functions * More complete tests for archive_entry * Finish archive_entry_clone() * Isolate major()/minor()/makedev() in archive_entry; remove these from everywhere else. * Bug fix: properly handle decompression look-ahead at end-of-data * Bug fixes to 'ar' support * Fix memory leak in ZIP reader * Portability: better timegm() emulation in iso9660 reader * New write_disk flags to suppress auto dir creation and not overwrite newer files (for future cpio front-end) * Simplify trailing-'/' fixup when writing tar and pax * Test enhancements: fix various compiler warnings, improve portability, add lots of new tests. * Documentation: document new functions, first draft of libarchive_internals.3 MFC after: 14 days Thanks to: Joerg Sonnenberger (compression_program) Thanks to: Kai Wang (ar) Thanks to: Colin Percival (many small fixes) Thanks to: Many others who sent me various patches and problem reports.
* Correct spelling errors in comments.rwatson2007-05-281-2/+2
|
* Precede symbol names consistently with tabs rather than spaces.rwatson2007-05-281-14/+14
|
* Include string.h for memcpy() and memcmp().delphij2007-05-251-0/+1
|
* Const'ify and ANSIfy the internal interfaces of regex(3).delphij2007-05-251-109/+100
| | | | | This is the final change that makes libc to compile with WERROR on my amd64 crashbox.
* ANSIfy function definitions, reduces diff against OpenBSD.delphij2007-05-251-17/+8
|
* Also fix the misspelling of hes_resolve().ru2007-05-251-1/+1
| | | | Submitted by: Danny Braniss
* The usage of "info" in init_hash() is read-only, so constifydelphij2007-05-251-3/+3
| | | | | the internal interface instead of casting away the constant constraint upon calling.
* Re-add support for NIS netgroups (heavily modified from patch in PR)des2007-05-251-4/+22
| | | | | | PR: bin/112955 Submitted by: A. Blake Cooper <blake@cluebie.net> MFC after: 3 weeks
* - When I introduce wide character enabled ncurses into base, all headersrafan2007-05-254-0/+9
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | are installed twice (once in non-widec version, onec in widec version). Headers with widec enabled are compatible with non-widec version for libraries. However, if you do a repeat build/install, the curses.h is always overwritten. The reason is that headers and statics libraries are installed with -S option to preserve their mtime if no actual changes, which saves time when doing incremental builds. The curses.h is installed by non-widec ncurses first, then by widec ncurses. So next time, it happens again. You see something like this: # pwd /usr/src/lib/ncurses # make -s installincludes INSTALL="install -v" ===> ncurses (installincludes) install: curses.h -> /usr/include/curses.h ===> ncursesw (installincludes) install: curses.h -> /usr/include/curses.h # make -s installincludes INSTALL="install -v" ===> ncurses (installincludes) install: curses.h -> /usr/include/curses.h ===> ncursesw (installincludes) install: curses.h -> /usr/include/curses.h The solution is to disable installing headers in non-widec version. Now you see this: # pwd /usr/src/lib/ncurses # make -s installincludes INSTALL="install -v" ===> ncurses (installincludes) ===> ncursesw (installincludes) # make -s installincludes INSTALL="install -v" ===> ncurses (installincludes) ===> ncursesw (installincludes) For form/panel/menu libraries, the headers are the same for both version. To be consistent with ncurses, I also disable the installation in non-widec version. Reported by: des Reviewed by: ru Thanks to: ru Approved by: delphij (mentor) MFC after: 2 weeks
* FreeBSD has <limits.h>.obrien2007-05-241-0/+3
|
* Update for the 'file' 4.21 import.obrien2007-05-241-1/+1
|
* Enable FTP_COMBINE_CWDS.des2007-05-242-10/+2
|
* Update for the 'file' 4.19 import.obrien2007-05-241-1/+10
|
* hesiod_resolv -> hesiod_resolveru2007-05-241-1/+1
| | | | Submitted by: Danny Braniss
* Nit: avoid shadowing truncate(2) with a local variable.des2007-05-231-3/+3
|
* Fix stupid braino in previous commit.des2007-05-231-1/+1
|
* If (flags & O_TRUNC), don't truncate the file until we've successfullydes2007-05-231-1/+11
| | | | | | locked it. MFC after: 3 weeks
* Replace "sizeof(struct bucket[cache_size])" withcperciva2007-05-221-4/+4
| | | | | | | | "cache_size * sizeof(struct bucket)". The former is valid in C99 but can confuse earlier compilers, while the latter is a standard idiom which all C compilers understand. Approved by: kientzle
* Backout 1.5 as requested by deischenjon2007-05-221-1/+0
|
* __cleanup() is needed for ports/devel/valgrind, export it.jon2007-05-221-0/+1
|
* Make pointer argument to kread_string() const since the kernel structurerwatson2007-05-211-3/+4
| | | | | | | field is const, and then employ __DECONST before getting into the kvm code. This eliminates a gcc 4.2 warning about losing constification. __DECONST advice from: sam
* Don't test for NULL when it is both unnecessary (the pointer is checkedcperciva2007-05-211-1/+1
| | | | | | | | against NULL when it is first allocated) and pointless (we've already dereferenced the pointer several times). Found by: Coverity Prevent(tm) CID: 3204
* Remove pointless code: Don't assign a value to a variable when we'recperciva2007-05-211-1/+0
| | | | | | | | | | | going to overwrite it with a new value a few lines later. Visual inspection of the surrounding code indicates that the code does what it's supposed to do; i.e., the pointless code wasn't supposed to be doing something other than what it was doing. CID: 3323 Found by: Coverity Prevent(tm)
* Bump library versions in preparation for 7.0.deischen2007-05-2140-40/+40
| | | | Ok'd by: kan
* Use LDADD to add -lgcc to the end of linker command line. Using LDFLAGSkan2007-05-191-1/+2
| | | | puts it before library's object files, making the whole constuct useless.
* Use __mcount() instead of _mcount() to reduce diffs with NetBSD.cognet2007-05-191-1/+1
|
* Add a level of indirection to the kernel PTE table. The oldmarcel2007-05-191-8/+16
| | | | | | | | | scheme allowed for 1024 PTE pages, each containing 256 PTEs. This yielded 2GB of KVA. This is not enough to boot a kernel on a 16GB box and in general too low for a 64-bit machine. By adding a level of indirection we now have 1024 2nd-level directory pages, each capable of supporting 2GB of KVA. This brings the grand total to 2TB of KVA.
* Don't forget to free a string allocated by strdup where necessary.cperciva2007-05-191-2/+9
| | | | | Found by: Coverity Prevent Coverity ID: CID 3132
* Make sure libc is installed before all other shared libraries. Otherkan2007-05-191-3/+4
| | | | libraries might depend on versioned symbols from libc.
* Use built-in _end symbol insteadof 'end' for consistency with otherkan2007-05-191-2/+2
| | | | architectures. Linker defines end is synonym for _end.
* Do not declare float_detect_tininess as external if it will be re-delaredkan2007-05-191-0/+2
| | | | as static later.
* Make sure GCC will not try to link libc with itself.kan2007-05-191-0/+5
|
* Follow NetBSD, OpenBSD and DragonfyBSD project and add BSD-licensedkan2007-05-193-1/+124
| | | | | SSP functions into FreeBSD libc. Use the same file name and location for consistency with other projects.
* Move nss_compat symbols from r1.11 to private section of symbol map.jon2007-05-181-10/+10
| | | | Suggested by: deischen
* Fix a logic bug I re-introduced in my patch I sent to Danielru2007-05-184-4/+4
| | | | | | | that would cause the selected shared threading library to be overwritten with its 32-bit version on amd64. PR: amd64/112509
* __nss_compat_* from net/nss_compat.c are needed by ports/net/nss_ldap, so ↵jon2007-05-181-0/+10
| | | | export them.
* Allow DEFAULT_THREAD_LIB to be set from /etc/src.conf.deischen2007-05-174-16/+16
| | | | Submitted by: ru
* endrpcent belongs with rpc name service db, not services.jon2007-05-171-1/+1
| | | | MFC after: 1 week
* fix memory leak from reading nsswitch.confjon2007-05-172-0/+7
| | | | MFC after: 1 week
* -Fix so getoptinfo() is in conformance with socket api (itrrs2007-05-161-32/+33
| | | | | | can be used on any read socketopt). -Clean up of sendmsg call and make it conformant when no syscall is available.
* Resolves symbols map so sctp_xxx() functions are included.rrs2007-05-161-0/+14
| | | | | PR: 112704 Submitted by: kou@cozmixng.org
* Do not export 'end' symbol. It is not meant to be visible to outsidekan2007-05-161-1/+1
| | | | | world and it wreak havoc if libc collides with other versioned libraries.
OpenPOWER on IntegriCloud