summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorFilesLines
2011-09-28improve pshared barriersRich Felker2-12/+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.
2011-09-28next step making barrier self-sync'd destruction safeRich Felker2-6/+18
i think this works, but it can be simplified. (next step)
2011-09-28barrier destroy must also wait for threads in other processes exiting barrierRich Felker1-0/+2
the vm lock only waits for threads in the same process exiting. actually this fix is not enough, but it's a start...
2011-09-28stupid typo (caused by rather ugly spelling in POSIX..) in aioRich Felker2-2/+2
2011-09-27correctly handle the degenerate barrier in the pshared caseRich Felker1-1/+1
2011-09-27fix crash in pthread_cond_wait mutex-locked checkRich Felker1-1/+1
it was assuming the result of the condition it was supposed to be checking for, i.e. that the thread ptr had already been initialized by pthread_mutex_lock. use the slower call to be safe.
2011-09-27fix crash in pthread_testcancel if pthread_self has not been calledRich Felker1-1/+1
2011-09-27improve/debloat mutex unlock error checking in pthread_cond_waitRich Felker1-3/+3
we're not required to check this except for error-checking mutexes, but it doesn't hurt. the new test is actually simpler/lighter, and it also eliminates the need to later check that pthread_mutex_unlock succeeds.
2011-09-27check mutex owner in pthread_cond_waitRich Felker1-0/+3
when used with error-checking mutexes, pthread_cond_wait is required to fail with EPERM if the mutex is not locked by the caller. previously we relied on pthread_mutex_unlock to generate the error, but this is not valid, since in the case of such invalid usage the internal state of the cond variable has already been potentially corrupted (due to access outside the control of the mutex). thus, we have to check first.
2011-09-27fix pshared barrier wrong return value.Rich Felker1-1/+1
i set the return value but then never used it... oops!
2011-09-27convert the barrier pshared option back to 0/1 values when getting itRich Felker1-1/+1
2011-09-27improve error handling in mntent stuff...Rich Felker1-2/+9
not sure if this is correct/ideal. it needs further attention.
2011-09-27fix missing va_end in prctl syscall wrapperRich Felker1-0/+1
2011-09-27fix clobbering of errno in get(pw|gr)([ug]id|nam) by fcloseRich Felker2-0/+12
2011-09-27process-shared barrier support, based on discussion with bdonlanRich Felker7-16/+102
this implementation is rather heavy-weight, but it's the first solution i've found that's actually correct. all waiters actually wait twice at the barrier so that they can synchronize exit, and they hold a "vm lock" that prevents changes to virtual memory mappings (and blocks pthread_barrier_destroy) until all waiters are finished inspecting the barrier. thus, it is safe for any thread to destroy and/or unmap the barrier's memory as soon as pthread_barrier_wait returns, without further synchronization.
2011-09-27fix incorrect allocation failure check in pthread_createRich Felker1-1/+1
mmap returns MAP_FAILED not 0 because some idiot thought the ability to mmap the null pointer page would be a good idea...
2011-09-26fix ctype macros to cast argument to (unsigned) firstRich Felker2-7/+7
issue reported by nsz, but it's actually not just pedantic. the functions can take input of any arithmetic type, including floating point, and the behavior needs to be as if the conversion implicit in the function call took place.
2011-09-26another cond var fix: requeue count race conditionRich Felker2-15/+13
lock out new waiters during the broadcast. otherwise the wait count added to the mutex might be lower than the actual number of waiters moved, and wakeups may be lost. this issue could also be solved by temporarily setting the mutex waiter count higher than any possible real count, then relying on the kernel to tell us how many waiters were requeued, and updating the counts afterwards. however the logic is more complex, and i don't really trust the kernel. the solution here is also nice in that it replaces some atomic cas loops with simple non-atomic ops under lock.
2011-09-26fix lost signals in cond varsRich Felker3-14/+17
due to moving waiters from the cond var to the mutex in bcast, these waiters upon wakeup would steal slots in the count from newer waiters that had not yet been signaled, preventing the signal function from taking any action. to solve the problem, we simply use two separate waiter counts, and so that the original "total" waiters count is undisturbed by broadcast and still available for signal.
2011-09-26cleanup various minor issues reported by nszRich Felker6-9/+9
the changes to syscall_ret are mostly no-ops in the generated code, just cleanup of type issues and removal of some implementation-defined behavior. the one exception is the change in the comparison value, which is fixed so that 0xf...f000 (which in principle could be a valid return value for mmap, although probably never in reality) is not treated as an error return.
2011-09-26redo cond vars again, use sequence numbersRich Felker4-48/+52
testing revealed that the old implementation, while correct, was giving way too many spurious wakeups due to races changing the value of the condition futex. in a test program with 5 threads receiving broadcast signals, the number of returns from pthread_cond_wait was roughly 3 times what it should have been (2 spurious wakeups for every legitimate wakeup). moreover, the magnitude of this effect seems to grow with the number of threads. the old implementation may also have had some nasty race conditions with reuse of the cond var with a new mutex. the new implementation is based on incrementing a sequence number with each signal event. this sequence number has nothing to do with the number of threads intended to be woken; it's only used to provide a value for the futex wait to avoid deadlock. in theory there is a danger of race conditions due to the value wrapping around after 2^32 signals. it would be nice to eliminate that, if there's a way. testing showed no spurious wakeups (though they are of course possible) with the new implementation, as well as slightly improved performance.
2011-09-25revert previous change in cond var waiter moveRich Felker1-2/+6
using swap has a race condition: the waiters must be added to the mutex waiter count *before* they are taken off the cond var waiter count, or wake events can be lost.
2011-09-25optimize cond waiter move using atomic swap instead of cas loopRich Felker1-6/+2
2011-09-25fix logic for when wakeup is not desired on cond bcastRich Felker1-3/+4
somehow i forgot that normal-type mutexes don't store the owner tid.
2011-09-25new futex-requeue-based pthread_cond_broadcast implementationRich Felker4-7/+69
this avoids the "stampede effect" where pthread_cond_broadcast would result in all waiters waking up simultaneously, only to immediately contend for the mutex and go back to sleep.
2011-09-23fix ABA race in cond vars, improve them overallRich Felker3-11/+12
previously, a waiter could miss the 1->0 transition of block if another thread set block to 1 again after the signal function set block to 0. we now use the caller's thread id as a unique token to store in block, which no other thread will ever write there. this ensures that if block still contains the tid, no signal has occurred. spurious wakeups will of course occur whenever there is a spurious return from the futex wait and another thread has begun waiting on the cond var. this should be a rare occurrence except perhaps in the presence of interrupting signal handlers. signal/bcast operations have been improved by noting that they need not avoid inspecting the cond var's memory after changing the futex value. because the standard allows spurious wakeups, there is no way for an application to distinguish between a spurious wakeup just before another thread called signal/bcast, and the deliberate wakeup resulting from the signal/bcast call. thus the woken thread must assume that the signalling thread may still be waiting to act on the cond var, and therefore it cannot destroy/unmap the cond var.
2011-09-23FD_ISSET must return an int. this is the easiest way.Rich Felker1-1/+1
casting to int would not be correct because high bits could be lost. mapping the high bits down onto low bits would be costlier in the common case where the result is just used in a conditional. changing the type of the bit array elements to int would permute the order of the bit array on 64-bit big endian systems, so that's not an option either.
2011-09-23sys/user.h may need stdint.hRich Felker1-0/+1
2011-09-22"optimize" arm __pthread_selfRich Felker1-4/+1
actually this is just to avoid gcc being stupid and refusing to inline the function version, even when the size cost is essentially identical whether it's inlined or not.
2011-09-22fix deadlock in condition wait whenever there are multiple waitersRich Felker4-5/+18
it's amazing none of the conformance tests i've run even bothered to check whether something so basic works...
2011-09-22fix x86_64 user.h (previously was just a copy of i386)Rich Felker1-48/+13
2011-09-22cleanup/debloat i386 user.hRich Felker1-43/+11
2011-09-22wrong __WORDSIZE in x86_64 headerRich Felker1-1/+1
2011-09-21protect against/handle cancellation reading shadow passwordsRich Felker1-1/+11
2011-09-21make dns lookups (and thus getaddrinfo) cancellableRich Felker1-4/+11
2011-09-21use poll rather than select in dns lookups (also clock_gettime)Rich Felker1-12/+10
if the file descriptor resource limit has been increased past FD_SETSIZE, this is actually a security issue; we could write past the end of the fd_set object. using poll makes it a non-issue, and simplifies the code at the same time. also, use clock_gettime instead of gettimeofday, for reduced bloat and better entropy.
2011-09-21fix missing SIG_DFL, SIG_IGN, SIG_ERR without posix in signal.hRich Felker1-5/+6
2011-09-21avoid setting FILE lock count when not using flockfileRich Felker1-1/+1
for now this is just a tiny optimization, but later if we support cancellation from __stdio_read and __stdio_write, it will be necessary for the recusrive lock count to be zero in order for these functions to know they are responsible for unlocking the FILE on cancellation.
2011-09-21gnu search.h has struct qelem...Rich Felker1-0/+7
2011-09-21release notes for 0.8.3v0.8.3Rich Felker1-0/+13
2011-09-21update syscalls with off_t arguments to handle argument alignment, if neededRich Felker9-11/+14
the arm syscall abi requires 64-bit arguments to be aligned on an even register boundary. these new macros facilitate meeting the abi requirement without imposing significant ugliness on the code.
2011-09-21new gcc wrapper now supports pre-4.4 gcc versions and is more robustRich Felker1-17/+40
2011-09-21new installation guide with alternate and primary libc setupsRich Felker1-37/+107
2011-09-20fix some header typosRich Felker2-2/+2
2011-09-20make stdbool.h conform to c99Rich Felker1-1/+2
2011-09-20fix incorrect include guard in mqueue.hRich Felker1-2/+2
2011-09-20fix statvfs.c to match new fsid_t definitionRich Felker1-1/+1
2011-09-20fix the fsid_t structure to match name of __valRich Felker1-1/+1
this is a case of poorly written man pages not matching the actual implementation, and why i hate implementing nonstandard interfaces with no actual documentation of how they're intended to work.
2011-09-20fix broken siginfo_t with _GNU_SOURCE definedRich Felker1-6/+5
this bug was introduced in a recent patch. the problem we're working around is that broken GNU software wants to use "struct siginfo" rather than "siginfo_t", but "siginfo" is not in the reserved namespace and thus not legal for the standard header to use.
2011-09-19fix the definition of struct statvfs to match lsb abiRich Felker7-70/+88
at the same time, make struct statfs match the traditional definition and make it more useful, especially the fsid_t stuff.