about summary refs log tree commit diff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* x86_64 vfork implementationRich Felker2012-02-061-0/+12
| | | | untested; should work.
* run ctors/dtors for shared objects loaded with dlopenRich Felker2012-02-061-0/+1
|
* add support for init/finit (constructors and destructors)Rich Felker2012-02-061-0/+19
| | | | | | | | | | this is mainly in hopes of supporting c++ (not yet possible for other reasons) but will also help applications/libraries which use (and more often, abuse) the gcc __attribute__((__constructor__)) feature in "C" code. x86_64 and arm versions of the new startup asm are untested and may have minor problems.
* add deprecated (removed from posix) [efg]cvt() functionsRich Felker2012-02-063-0/+52
| | | | | these have not been heavily tested, but they should work as described in the old standards. probably broken for non-finite values...
* include dummied-out dlopen and dlsym functions for static binariesRich Felker2012-02-031-1/+10
| | | | | | | | | these don't work (or do anything at all) but at least make it possible to static link programs that insist on "having" dynamic loading support...as long as they don't actually need to use it. adding real support for dlopen/dlsym with static linking is going to be significantly more difficult...
* make stdio open, read, and write operations cancellation pointsRich Felker2012-02-023-5/+28
| | | | | | | | | | it should be noted that only the actual underlying buffer flush and fill operations are cancellable, not reads from or writes to the buffer. this behavior is compatible with POSIX, which makes all cancellation points in stdio optional, and it achieves the goal of allowing cancellation of a thread that's "stuck" on IO (due to a non-responsive socket/pipe peer, slow/stuck hardware, etc.) without imposing any measurable performance cost.
* make passwd/group functions safe against cancellation in stdioRich Felker2012-02-014-6/+35
| | | | these changes are a prerequisite to making stdio cancellable.
* add fgetpwent (nonstandard function)Rich Felker2012-01-291-0/+9
| | | | based on patch by Jeremy Huntwork
* add linux setfs[ug]id syscall wrappersRich Felker2012-01-282-0/+16
| | | | patch by Jeremy Huntwork
* add legacy futimes and lutimes functionsRich Felker2012-01-242-0/+26
| | | | | based on patch by sh4rm4. these functions are deprecated; futimens and utimensat should be used instead in new programs.
* make glob mark symlinks-to-directories with the GLOB_MARK flagRich Felker2012-01-231-1/+1
| | | | | | POSIX is unclear on whether it should, but all historical implementations seem to behave this way, and it seems more useful to applications.
* fix broken copy relocations from dynamic linker cleanupRich Felker2012-01-231-1/+4
| | | | | this issue affected programs which use global variables exported by non-libc libraries.
* dynamic linker support for PIE binaries (position-independent main program)Rich Felker2012-01-231-3/+6
| | | | | even with this change, PIE will not work yet due to deficiencies in the crt1.o startup code.
* cleanup dynamic linker, removing some code duplicationRich Felker2012-01-231-80/+66
|
* fix cancellation failure in single-threaded programsRich Felker2012-01-221-1/+1
| | | | | | | even a single-threaded program can be cancellable, e.g. if it's called pthread_cancel(pthread_self()). the correct predicate to check is not whether multiple threads have been invoked, but whether pthread_self has been invoked.
* support GLOB_PERIOD flag (GNU extension) to glob functionRich Felker2012-01-221-1/+2
| | | | patch by sh4rm4
* use prlimit syscall for getrlimit/setrlimitRich Felker2012-01-202-4/+14
| | | | | | this allows the full range of 64-bit limit arguments even on 32-bit systems. fallback to the old syscalls on old kernels that don't support prlimit.
* add prlimit syscall wrapperRich Felker2012-01-201-0/+8
|
* fix dynamic linker not to depend on DYNAMIC ptr in 0th entry of GOTRich Felker2012-01-201-3/+12
| | | | | | | | | | this fixes an issue using gold instead of gnu ld for linking. it also should eliminate the need of the startup code to even load/pass the got address to the dynamic linker. based on patch submitted by sh4rm4 with minor cosmetic changes. further cleanup will follow.
* alias basename to glibc name for it, to meet abi goalsRich Felker2012-01-181-0/+3
| | | | | | note that regardless of the name used, basename is always conformant. it never takes on the bogus gnu behavior, unlike glibc where basename is nonconformant when declared manually without including libgen.h.
* fix char signedness bug in dynlinker hash functionRich Felker2012-01-171-1/+2
| | | | | this only affects non-ascii symbol names, which are probably not in use anyway..
* fix broken reboot wrapper (syscall needs extra silly magic arguments...)Rich Felker2011-12-231-1/+1
|
* fix issue with excessive mremap syscalls on reallocRich Felker2011-11-161-4/+2
| | | | | | | | | | CHUNK_SIZE macro was defined incorrectly and shaving off at least one significant bit in the size of mmapped chunks, resulting in the test for oldlen==newlen always failing and incurring a syscall. fortunately i don't think this issue caused any other observable behavior; the definition worked correctly for all non-mmapped chunks where its correctness matters more, since their lengths are always multiples of the alignment.
* fix signed overflows at most-negative values in ato(i|l|ll)Rich Felker2011-11-103-6/+9
| | | | patch by Pascal Cuoq (with minor tweaks to comments)
* report sem value overflows in sem_postRich Felker2011-10-261-0/+4
| | | | | this is not required by the standard, but it's nicer than corrupting the state and rather inexpensive.
* support vfork on i386Rich Felker2011-10-141-0/+14
|
* make available a namespace-safe vfork, if supportedRich Felker2011-10-141-1/+4
| | | | this may be useful to posix_spawn..?
* add dummy __cxa_finalizeRich Felker2011-10-141-0/+4
| | | | | | musl's dynamic linker does not support unloading dsos, so there's nothing for this function to do. adding the symbol in case anything depends on its presence..
* support __cxa_atexit, and registering atexit functions from atexit handlersRich Felker2011-10-141-7/+26
| | | | | mildly tested; may have bugs. the locking should be updated not to use spinlocks but that's outside the scope of this one module.
* simplify atexit and fflush-on-exit handlingRich Felker2011-10-143-9/+12
|
* fix F_GETOWN return value handlingRich Felker2011-10-091-0/+1
| | | | | | | | the fcntl syscall can return a negative value when the command is F_GETOWN, and this is not an error code but an actual value. thus we must special-case it and avoid calling __syscall_ret to set errno. this fix is better than the glibc fix (using F_GETOWN_EX) which only works on newer kernels and is more complex.
* fix typo in arm clone() asmRich Felker2011-10-091-1/+1
|
* recovering ownerdead robust mutex must reset recursive lock countRich Felker2011-10-031-0/+1
|
* simplify robust mutex unlock code pathRich Felker2011-10-031-4/+4
| | | | | | | right now it's questionable whether this change is an improvement or not, but if we later want to support priority inheritance mutexes, it will be important to have the code paths unified like this to avoid major code duplication.
* fix crash if pthread_mutex_unlock is called without ever lockingRich Felker2011-10-031-1/+1
| | | | | this is valid for error-checking mutexes; otherwise it invokes UB and would be justified in crashing.
* use count=0 instead of 1 for recursive mutex with only one lock referenceRich Felker2011-10-032-4/+2
| | | | | | | this simplifies the code paths slightly, but perhaps what's nicer is that it makes recursive mutexes fully reentrant, i.e. locking and unlocking from a signal handler works even if the interrupted code was in the middle of locking or unlocking.
* sysconf for PTHREAD_KEYS_MAXRich Felker2011-10-031-1/+1
|
* synchronize cond var destruction with exiting waitsRich Felker2011-10-023-0/+12
|
* fix failure-to-wake in rwlock unlockRich Felker2011-10-011-1/+1
| | | | | | | | | a reader unlocking the lock need only wake one waiter (necessarily a writer, but a writer unlocking the lock must wake all waiters (necessarily readers). if it only wakes one, the remainder can remain blocked indefinitely, or at least until the first reader unlocks (in which case the whole lock becomes serialized and behaves as a mutex rather than a read lock).
* dlsym entry point for armRich Felker2011-10-011-0/+6
|
* dynamic linker entry point for armRich Felker2011-10-011-0/+14
| | | | mildly tested, seems to work
* fix various bugs in path and error handling in execvp/fexecveRich Felker2011-09-292-18/+29
|
* don't crash on null strings in printfRich Felker2011-09-281-1/+1
| | | | passing null pointer for %s is UB but lots of broken programs do it anyway
* fix excessive/insufficient wakes in __vm_unlockRich Felker2011-09-281-3/+3
| | | | | | there is no need to send a wake when the lock count does not hit zero, but when it does, all waiters must be woken (since all with the same sign are eligible to obtain the lock).
* make getmntent_r discard long lines when it returns error, not seek backRich Felker2011-09-281-2/+1
| | | | | seeking back can be performed by the caller, but if the caller doesn't expect it, it will result in an infinite loop of failures.
* improve pshared barriersRich Felker2011-09-282-12/+14
| | | | | | | | | | | | | | eliminate the sequence number field and instead use the counter as the futex because of the way the lock is held, sequence numbers are completely useless, and this frees up a field in the barrier structure to be used as a waiter count for the count futex, which lets us avoid some syscalls in the best case. as of now, self-synchronized destruction and unmapping should be fully safe. before any thread can return from the barrier, all threads in the barrier have obtained the vm lock, and each holds a shared lock on the barrier. the barrier memory is not inspected after the shared lock count reaches 0, nor after the vm lock is released.
* next step making barrier self-sync'd destruction safeRich Felker2011-09-282-6/+18
| | | | i think this works, but it can be simplified. (next step)
* barrier destroy must also wait for threads in other processes exiting barrierRich Felker2011-09-281-0/+2
| | | | | the vm lock only waits for threads in the same process exiting. actually this fix is not enough, but it's a start...
* stupid typo (caused by rather ugly spelling in POSIX..) in aioRich Felker2011-09-281-1/+1
|
* correctly handle the degenerate barrier in the pshared caseRich Felker2011-09-271-1/+1
|