summaryrefslogtreecommitdiff
path: root/src/lib/libc (follow)
Commit message (Collapse)AuthorAgeFilesLines
...
* libc.so can't be unloaded, so move the hidden atexit() and pthread_atfork()guenther2015-11-101-1/+13
| | | | | | | | | | | stubs for the executable from crtbegin.o into libc, which lets them be excluded from static links that don't use them. For this, drop the normal crt{begin,end}S.o from libc.so: the .init and .fini sections for libc aren't called at the right times anyway, so it's good that they're unused. libc.so just needs __guard_local and the .note.openbsd.ident section, so add them to stack_protector.c for now (this will be improved) "good time" deraadt@
* inet(4), not inet(3);jmc2015-11-081-3/+3
|
* delete old lint ARGSUSED commentsguenther2015-11-011-2/+1
|
* Move the _atfork_list definition to atexit.c so that the fork syscall stubguenther2015-10-251-1/+5
| | | | | | doesn't get pulled into all static executables ok millert@ jca@
* Hide __atexit and __atexit_register_cleanup()guenther2015-10-252-2/+9
| | | | | | | | Wrap __cxa_{atexit,finalize}() so the call from exit() goes direct Switch regress/lib/libc/atexit/ to be built with -static so that it can still access __atexit* ok millert@ jca@
* The only thing that was translated into multiple languages in OpenBSDbluhm2015-10-251-23/+1
| | | | | | | | | are the errno messages and signal names. Everything else is in English. We are not planning to translate more text. Running a mixed system with less than 1% of the text in native language makes no sense. So remove the NLS support from libc messages. The catopen(3) functions stay as they are. OK stsp@ mpi@
* Use sigaction() instead of signal() to avoid pulling in unnecessaryguenther2015-10-251-3/+5
| | | | | | | wrappers. To keep uses from crawling back in, mark signal() as deprecated inside libc. ok deraadt@
* Switch if_nameindex(3) to use the new NET_RT_IFNAMES sysctl to get theclaudio2015-10-233-88/+73
| | | | | | | | list of interface names. At the same time switch if_nametoindex(3) and if_indextoname(3) to use if_nameindex(3) instead of getifaddrs(3). if_nameindex(3) exposes much less then getifaddrs(3) and is allowed by pledge(2). With and OK deraadt@
* Use waitpid() instead of wait() to avoid returning early from another childguenther2015-10-231-2/+3
| | | | | | exiting, and loop the waitpid() on EINTR ok deraadt@ millert@
* Loop the waitpid() on EINTR, and save and restore the disposition ofguenther2015-10-231-9/+12
| | | | | | | SIGINT and SIGQUIT with sigaction() instead of signal() so that all bits are preserved. ok deraadt@ millert@
* Merge the sigaction() and sigprocmask() overloads/wrappers from libpthreadguenther2015-10-232-9/+9
| | | | | | | | | into libc, and move pthread_sigmask() as well (just a trivial wrapper). This provides consistent handling of SIGTHR between single- and multi-threaded programs and is a step in the merge of all the libpthread overloads, providing some ASM and Makefile bits that the other wrappers will need. ok deraadt@ millert@
* Cast ctype function arguments to unsigned char.mmcc2015-10-221-2/+2
| | | | ok guenther@
* In rev 1.15 the sizeof argument was fixed in a strlcat() call butmillert2015-10-131-2/+2
| | | | | | the truncation check immediately following it was not updated to match. Not an issue in practice since the buffers are the same size. OK deraadt@
* Wrap <resolv.h> so that internal calls go directguenther2015-10-053-5/+11
| | | | ok millert@
* wrap __ivaliduser_sa() so the internal call is direct (at least until weguenther2015-10-041-0/+2
| | | | stop exporting it)
* recv() and send() aren't overriden by libpthread (vs recvfrom() and sendto()!)guenther2015-10-042-2/+4
| | | | so wrap them to make internal calls go direct
* Eliminate the last of the LINTEDn and PRINTFLIKEn comments. In oneguenther2015-10-016-15/+7
| | | | | | | case, by deleting some useless '& of an array' we also eliminate the need for the casts which prompted the original lint warnings ok deraadt@
* lint is dead: delete useless LINTLIBRARY commentsguenther2015-09-262-4/+2
| | | | ok millert@
* Don't wrap initialized variables: binutils appears to be mishandling themguenther2015-09-191-1/+3
| | | | | | on arm and m88k problems with optind observed by jsg@
* Only two of the *rand48.c files need <math.h>, so just #include it in themguenther2015-09-143-4/+5
|
* use .Va for global variables, and .Vt where the type is includedschwarze2015-09-141-4/+3
|
* Wrap <ifaddrs.h>, <netinet/in.h>, and <netinet/if_ether.h> so internalguenther2015-09-143-3/+7
| | | | calls go direct and all the symbols are weak
* Wrap <net/if.h> and <net/if_dl.h> so internal calls go direct and all theguenther2015-09-142-2/+4
| | | | symbols are weak
* Remove cast of int* to int*guenther2015-09-141-2/+2
|
* Finish wrapping <netdb.h> so that calls go direct and the symbols are all weakguenther2015-09-149-9/+24
|
* Wrap <arpa/inet.h> and <arpa/nameser.h> so that calls go direct and theguenther2015-09-135-5/+11
| | | | symbols without underbar prefix are all weak
* For now, permit overriding of the malloc family, to make emacs happyguenther2015-09-131-6/+6
|
* Wrap <pwd.h> so that calls go direct and the symbols are all weak.guenther2015-09-132-7/+8
| | | | Hide bcrypt_autorounds(), prefixing with an underbar for static builds.
* Overshot w/PROTO_DEPRECATED: seed48_deterministic() is used internallyguenther2015-09-131-1/+2
|
* The number of rounds is just two digits in the salt. We've alreadymillert2015-09-131-2/+2
| | | | | verified that they are there via isdigit() so we can convert from ASCII to an int without using atoi(). OK guenther@ deraadt@
* Wrap <stdlib.h> so that calls go direct and the symbols not in theguenther2015-09-1322-26/+54
| | | | | | C standard are all weak. Apply __{BEGIN,END}_HIDDEN_DECLS to gdtoa{,imp}.h, hiding the arch-specific __strtorx, __ULtox_D2A, __strtorQ, __ULtoQ_D2A symbols.
* __strsignal() is now declared in hidden/signal.hguenther2015-09-121-2/+0
|
* Wrap <inttypes.h> and finish wrapping of <wchar.h> so that calls go directguenther2015-09-1224-24/+51
| | | | and the symbols not in the C standard are weak
* Wrap <unistd.h> so that internal calls go direct and they're all weak symbolsguenther2015-09-126-4/+12
| | | | Delete unused 'fd' argument from internal function oldttyname()
* Wrap <getopt.h> to make the functions weak and make access to the initializedguenther2015-09-121-1/+6
| | | | variables go direct. (Common variables cannot be aliased.)
* _getnetbyaddr and _getnetbyname appear to be historical accidents inderaadt2015-09-113-103/+2
| | | | our tree. ok guenther miod
* Wrap blowfish, sha*, md5, and rmd160 so that internal calls go directguenther2015-09-112-2/+24
| | | | ok deraadt@
* Add support for building arc4random with MSVC.bcook2015-09-101-7/+8
| | | | | | | By default, MSVC's stdlib.h defines min(), so we need to spell out something less common to avoid picking it up. ok deraadt@ beck@ miod@
* Pull in namespace.h when building all .c files using gcc's -include option,guenther2015-09-102-3/+18
| | | | | | | | so that we can provide asm labels for the memcpy/memset/__stack_smash_handler calls that it generates ab initio. Eliminate direct #includes of it. Make sure it's a dependency of all objects (unnecessary for asm, but close enough). ok deraadt@
* reduce .Nd to one line and kill .Tn while hereschwarze2015-09-105-25/+15
|
* missing commas at the end of .Nm lines in the NAME sectionschwarze2015-09-101-3/+3
|
* tweak Nd after previous; ok mpijmc2015-09-101-2/+2
|
* Remove link_addr(3). A function to encode the name of an interface inmpi2015-09-103-136/+8
| | | | | | | | | | | | a sockaddr_dl is a questionnable interface. But now it makes it harder to properly reference ifp becauses of this. Set sdl_index to the index of the corresponding interface when constructing a routing message. Ridding previous libc crank. ok guenther@, deraadt@, dlg@
* Stop exporting from libc the <mpool.h> and the mpool_* APIguenther2015-09-091-0/+122
| | | | ports scan by sthen@
* Fix aliasing of sys_errlist, sys_nerr, sys_siglist, and sys_signameguenther2015-09-062-15/+4
| | | | | | to eliminate duplicate copies of the tables and get direct access internally ok kettenis@ deraadt@
* Use new framework for wrapping cat{open,gets,close}(), eliminatingguenther2015-09-052-8/+2
| | | | | | _cat* in the process. ok kettenis@
* Add framework for resolving (pun intended) libc namespace issues, usingguenther2015-08-3143-49/+233
| | | | | | | | | | | | wrapper .h files and asm labels to let internal calls resolve directly and not be overridable or use the PLT. Then, apply that framework to most of the functions in stdio.h, string.h, err.h, and wchar.h. Delete the should-have-been-hidden-all-along _v?(err|warn)[cx]? symbols while here. tests clean on i386, amd64, sparc64, powerpc, and mips64 naming feedback from kettenis@ and millert@ ok kettenis@
* Use nanosleep instead of sleep to avoid the extra layer and simplify laterguenther2015-08-301-5/+8
| | | | | | symbol hiding ok w/tweak deraadt@
* Use static and __{BEGIN,ENV}_HIDDEN_DECLS to hide a bunch of internalguenther2015-08-272-4/+7
| | | | | | symbols that are not longer exported. (This improves the generated code.) ok deraadt@
* Move the __rand48_* declarations into rand48.h, and then hide both themguenther2015-08-277-25/+13
| | | | | | and __dorand48() and __rand48_deterministic ok deraadt@
generated by cgit v1.2.3-55-g6feb (git 2.39.1) at 2025-03-22 12:38:15 +0000