summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)AuthorLines
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
2011-03-16optimize file locking: avoid cache-polluting writes to global storageRich Felker-3/+5
2011-03-14partially-written draft of fmemopen, still in #if 0Rich Felker-3/+48
2011-03-12misplaced & in times() made it fail to work, and clobber the stackRich Felker-1/+1
2011-03-12implement flockfile api, rework stdio lockingRich Felker-15/+66
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-38/+85
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/+16
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.
2011-03-10fix errors in sigqueue (potential information leak, wrong behavior)Rich Felker-5/+7
1. any padding in the siginfo struct was not necessarily zero-filled, so it might have contained private data off the caller's stack. 2. the uid and pid must be filled in from userspace. the previous rsyscall fix broke rsyscalls because the values were always incorrect.
2011-03-10security fix: check that cancel/rsyscall signal was sent by the process itselfRich Felker-0/+3
2011-03-10more cancellation points: tcdrain, clock_nanosleepRich Felker-2/+12
2011-03-10remove useless return value checks for functions that cannot failRich Felker-3/+2
2011-03-10make sigsuspend a cancellation pointRich Felker-1/+6
2011-03-10make sigtimedwait a cancellation pointRich Felker-0/+3
2011-03-10don't fail with EINTR in sigtimedwaitRich Felker-6/+6
POSIX allows either behavior, but sigwait is not allowed to fail with EINTR, so the retry loop would have to be in one or the other anyway.
2011-03-10fix sigsuspend syscallRich Felker-1/+1
2011-03-10make sigaltstack work (missing macros in signal.h, error conditions)Rich Felker-1/+11
2011-03-10fix errno behavior in clock_* functionsRich Felker-3/+0
these functions are specified inconsistent in whether they're specified to return an error value, or return -1 and set errno. hopefully now they all match what POSIX requires.
2011-03-09fix error handling for pthread_sigmaskRich Felker-1/+15
it must return errno, not -1, and should reject invalud values for how.
2011-03-09make fork properly initialize the main thread in the child processRich Felker-0/+7
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.
2011-03-09fix race condition in raise - just mask signalsRich Felker-9/+9
a signal handler could fork after the pid/tid were read, causing the wrong process to be signalled. i'm not sure if this is supposed to have UB or not, but raise is async-signal-safe, so it probably is allowed. the current solution is slightly expensive so this implementation is likely to be changed in the future.
2011-03-09fix raise semantics with threads.Rich Felker-1/+12
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/+19
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/+3
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
2011-03-03implement POSIX shared memoryRich Felker-0/+42
2011-03-01depends on settimeofday which needs _GNU_SOURCE feature testRich Felker-0/+1
2011-02-27implement futimens and utimensatRich Felker-0/+13
2011-02-27implement fexecveRich Felker-0/+10
2011-02-27cleanup utf-8 multibyte code, use visibility if possibleRich Felker-84/+5
this code was written independently of musl, with support for a the backwards, nonstandard "31-bit unicode" some libraries/apps might want. unfortunately the extra code (inside #ifdef) makes the source harder to read and makes code that should be simple look complex, so i'm removing it. anyone who wants to use the old code can find it in the history or from elsewhere. also, change the visibility of the __fsmu8 state machine table to hidden, if supported. this should improve performance slightly in shared-library builds.
2011-02-26fix missing prototype for strsignalRich Felker-0/+1