summaryrefslogtreecommitdiff
path: root/src/thread
AgeCommit message (Collapse)AuthorLines
2011-03-19syscall overhaul part two - unify public and internal syscall interfaceRich Felker-15/+9
with this patch, the syscallN() functions are no longer needed; a variadic syscall() macro allows syscalls with anywhere from 0 to 6 arguments to be made with a single macro name. also, manually casting each non-integer argument with (long) is no longer necessary; the casts are hidden in the macros. some source files which depended on being able to define the old macro SYSCALL_RETURNS_ERRNO have been modified to directly use __syscall() instead of syscall(). references to SYSCALL_SIGSET_SIZE and SYSCALL_LL have also been changed. x86_64 has not been tested, and may need a follow-up commit to fix any minor bugs/oversights.
2011-03-19overhaul syscall interfaceRich Felker-2/+2
this commit shuffles around the location of syscall definitions so that we can make a syscall() library function with both SYS_* and __NR_* style syscall names available to user applications, provides the syscall() library function, and optimizes the code that performs the actual inline syscalls in the library itself. previously on i386 when built as PIC (shared library), syscalls were incurring bus lock (lock prefix) overhead at entry and exit, due to the way the ebx register was being loaded (xchg instruction with a memory operand). now the xchg takes place between two registers. further cleanup to arch/$(ARCH)/syscall.h is planned.
2011-03-17implement robust mutexesRich Felker-10/+78
some of this code should be cleaned up, e.g. using macros for some of the bit flags, masks, etc. nonetheless, the code is believed to be working and correct at this point.
2011-03-17avoid function call to pthread_self in mutex unlockRich Felker-1/+1
if the mutex was previously locked, we can assume pthread_self was already called at the time of locking, and thus that the thread pointer is initialized.
2011-03-17unify lock and owner fields of mutex structureRich Felker-7/+6
this change is necessary to free up one slot in the mutex structure so that we can use doubly-linked lists in the implementation of robust mutexes.
2011-03-17optimize contended normal mutex case; add int compare-and-swap atomicRich Felker-1/+1
2011-03-16simplify logic, slightly optimize contended case for non-default mutex typesRich Felker-4/+2
2011-03-16correct error returns for error-checking mutexesRich Felker-2/+6
2011-03-16cut out a syscall on thread creation in the case where guard size is 0Rich Felker-1/+1
2011-03-16don't expose EAGAIN, etc. from timed futex wait to callerRich Felker-1/+4
2011-03-12implement flockfile api, rework stdio lockingRich Felker-0/+1
2011-03-11implement dummy pthread_attr_[gs]etschedparam functionsRich Felker-0/+14
for some reason these functions are not shaded by the PS/TPS option in POSIX, so presumably they are mandatory, even though the functionality they offer is optional. for now, provide them in case any programs depend on their existence, but disallow any priority except the default.
2011-03-11fix pthread_attr_* implementations to match corrected prototypesRich Felker-4/+4
2011-03-10fix failure behavior of sem_open when sem does not existRich Felker-1/+5
2011-03-10fix some semaphore wait semantics (race condition deadlock and error checking)Rich Felker-1/+7
2011-03-10fix sem_open and sem_close to obey posix semanticsRich Felker-33/+80
multiple opens of the same named semaphore must return the same pointer, and only the last close can unmap it. thus the ugly global state keeping track of mappings. the maximum number of distinct named semaphores that can be opened is limited sufficiently small that the linear searches take trivial time, especially compared to the syscall overhead of these functions.
2011-03-10optimize pthread termination in the non-detached caseRich Felker-4/+15
we can avoid blocking signals by simply using a flag to mark that the thread has exited and prevent it from getting counted in the rsyscall signal-pingpong. this restores the original pthread create/join throughput from before the sigprocmask call was added.
2011-03-10security fix: check that cancel/rsyscall signal was sent by the process itselfRich Felker-0/+3
2011-03-09fix error handling for pthread_sigmaskRich Felker-0/+10
it must return errno, not -1, and should reject invalud values for how.
2011-03-09optimize pthread initializationRich Felker-2/+2
the set_tid_address returns the tid (which is also the pid when called from the initial thread) so there is no need to make a separate syscall to get pid/tid.
2011-03-08simplify and optimize pthread_mutex_trylockRich Felker-17/+16
2011-03-08rwlock trylock functions were wrongly returning EAGAIN instead of EBUSYRich Felker-3/+3
2011-03-08fix major breakage in pthread_once (it was always deadlocking)Rich Felker-8/+7
the issue was a break statement that was breaking only from the switch, not the enclosing for loop, and a failure to set the final success state.
2011-03-08fix and optimize non-default-type mutex behaviorRich Felker-25/+18
problem 1: mutex type from the attribute was being ignored by pthread_mutex_init, so recursive/errorchecking mutexes were never being used at all. problem 2: ownership of recursive mutexes was not being enforced at unlock time.
2011-03-08implement pthread_mutexattr_[gs]etpshared functionsRich Felker-0/+16
2011-03-08disallow cpu time clocks as condattr clock valuesRich Felker-1/+1
2011-03-07fix off-by-one error in sem_(timed)wait (using old sem value instead of new)Rich Felker-1/+1
2011-03-07use the selected clock from the condattr for pthread_cond_timedwaitRich Felker-1/+2
2011-03-07implement pthread_rwlockattr_* (essentially no-ops)Rich Felker-0/+28
2011-03-07implement pthread_condattr_* interfacesRich Felker-0/+45
note that, while the attributes are stored, they are not used in pthread_cond_init yet.
2011-03-07reject invalid attribute settingsRich Felker-0/+2
note that this is a pedantic conformance issue and waste of code. it only affects broken code or code that is probing for conformance.
2011-03-07implement barrier attribute functions (essentially no-ops)Rich Felker-0/+27
2011-03-04enforce stack size min in pthread_attr_setstacksizeRich Felker-1/+1
2011-03-04implement POSIX semaphoresRich Felker-0/+201
2011-03-03optimize POSIX TSD for fast pthread_getspecificRich Felker-9/+14
2011-02-19use rt_sigprocmask, not legacy sigprocmask, syscall in pthread exit codeRich Felker-1/+1
2011-02-19race condition fix: block all signals before decrementing thread countRich Felker-22/+2
the existence of a (kernelspace) thread must never have observable effects after the thread count is decremented. if signals are not blocked, it could end up handling the signal for rsyscall and contributing towards the count of threads which have changed ids, causing a thread to be missed. this could lead to one thread retaining unwanted privilege level. this change may also address other subtle race conditions in application code that uses signals.
2011-02-19make pthread_exit run dtors for last thread, wait to decrement thread countRich Felker-3/+3
2011-02-18add pthread_atfork interfaceRich Felker-0/+48
note that this presently does not handle consistency of the libc's own global state during forking. as per POSIX 2008, if the parent process was threaded, the child process may only call async-signal-safe functions until one of the exec-family functions is called, so the current behavior is believed to be conformant even if non-ideal. it may be improved at some later time.
2011-02-17reorganize pthread data structures and move the definitions to alltypes.hRich Felker-83/+83
this allows sys/types.h to provide the pthread types, as required by POSIX. this design also facilitates forcing ABI-compatible sizes in the arch-specific alltypes.h, while eliminating the need for developers changing the internals of the pthread types to poke around with arch-specific headers they may not be able to test.
2011-02-16patch by njk, simplifies thread register initialization for x86_64Rich Felker-4/+0
It's not necessary to save any registers on the stack across syscall in x86_64 __set_thread_area. Don't waste cycles or bytes on it.
2011-02-16patch from njk: make x86_64 __uniclone branchless.Rich Felker-6/+6
2011-02-15Port musl to x86-64. One giant commit!Nicholas J. Kain-0/+61
2011-02-15finish unifying thread register handling in preparation for portingRich Felker-44/+25
2011-02-15begin unifying clone/thread management interface in preparation for portingRich Felker-29/+19
2011-02-15make pthread_create return EAGAIN on resource failure, as required by POSIXRich Felker-1/+1
2011-02-13reorganize thread exit code, make pthread_exit call cancellation handlers (pt2)Rich Felker-22/+51
2011-02-13reorganize thread exit code, make pthread_exit call cancellation handlersRich Felker-25/+0
2011-02-12initial check-in, version 0.5.0v0.5.0Rich Felker-0/+1030