about summary refs log tree commit diff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* remove everything related to forkallRich Felker2012-05-223-77/+0
| | | | | | | | | | i made a best attempt, but the intended semantics of this function are fundamentally contradictory. there is no consistent way to handle ownership of locks when forking a multi-threaded process. the code could have worked by accident for programs that only used normal mutexes and nothing else (since they don't actually store or care about their owner), but that's about it. broken-by-design interfaces that aren't even in glibc (only solaris) don't belong in musl.
* fix typo in utimes function that made it mess up file timesRich Felker2012-05-211-1/+1
|
* fix out-of-bounds array access in pthread barriers on 64-bitRich Felker2012-05-211-1/+1
| | | | | it's ok to overlap with integer slot 3 on 32-bit because only slots 0-2 are used on process-local barriers.
* fix misplaced semicolon in preprocessor directive (#undef h_errno)Rich Felker2012-05-161-1/+1
|
* remove some no-op end of string tests from regex parserRich Felker2012-05-131-4/+0
| | | | | | | | these are cruft from the original code which used an explicit string length rather than null termination. i blindly converted all the checks to null terminator checks, without noticing that in several cases, the subsequent switch statement would automatically handle the null byte correctly.
* another BRE fix: in ^*, * is literalRich Felker2012-05-131-0/+2
| | | | | | i don't understand why this has to be conditional on being in BRE mode, but enabling this code unconditionally breaks a huge number of ERE test cases.
* let sysconf accurately report # of cpus availableRich Felker2012-05-131-2/+10
| | | | | i've been trying out openmp and it seems like it won't be much use without this...
* use __h_errno_location for h_errnoRich Felker2012-05-122-0/+9
| | | | | | we do not bother making h_errno thread-local since the only interfaces that use it are inherently non-thread-safe. but still use the potentially-thread-local ABI to access it just to avoid lock-in.
* search: add comments to tsearch_avl.cnsz2012-05-131-0/+6
|
* search: add tdestroy (gnu extension)nsz2012-05-131-0/+21
|
* Merge remote-tracking branch 'nsz/master'Rich Felker2012-05-103-5/+8
|\
| * math: fix remquo.c when x==-y and a subnormal remainder bug as wellnsz2012-05-083-5/+8
| | | | | | | | | | backported fix from freebsd: http://svnweb.FreeBSD.org/base?view=revision&revision=233973
* | remove __lock dependency from exitRich Felker2012-05-101-2/+4
| | | | | | | | | | | | | | there's no sense in using a powerful lock in exit, because it will never be unlocked. a thread that arrives at exit while exit is already in progress just needs to hang forever. use the pause syscall for this because it's cheap and easy and universally available.
* | fix missing static (namespace clash)Rich Felker2012-05-101-1/+1
|/
* some assemblers don't like fistpq; use the alt. mnemonic fistpllRich Felker2012-05-073-3/+3
|
* fix error checking for \ at end of regex (this was broken previously)Rich Felker2012-05-071-1/+1
|
* fix copy and paste error in regex code causing mishandling of \) in BRERich Felker2012-05-071-1/+1
|
* fix regex breakage in last commit (failure to handle empty regex, etc.)Rich Felker2012-05-071-4/+1
|
* fix ugly bugs in TRE regex parserRich Felker2012-05-071-60/+31
| | | | | | | | | | | | | | | | | | | | | | 1. * in BRE is not special at the beginning of the regex or a subexpression. this broke ncurses' build scripts. 2. \\( in BRE is a literal \ followed by a literal (, not a literal \ followed by a subexpression opener. 3. the ^ in \\(^ in BRE is a literal ^ only at the beginning of the entire BRE. POSIX allows treating it as an anchor at the beginning of a subexpression, but TRE's code for checking if it was at the beginning of a subexpression was wrong, and fixing it for the sake of supporting a non-portable usage was too much trouble when just removing this non-portable behavior was much easier. this patch also moved lots of the ugly logic for empty atom checking out of the default/literal case and into new cases for the relevant characters. this should make parsing faster and make the code smaller. if nothing else it's a lot more readable/logical. at some point i'd like to revisit and overhaul lots of this code...
* add FORCE_EVAL macro to evaluate float expr for their side effectnsz2012-05-066-24/+25
| | | | | updated nextafter* to use FORCE_EVAL, it can be used in many other places in the math code to improve readability.
* fix unused variable warnings in new nextafter/nexttoward codeRich Felker2012-05-065-6/+12
| | | | | | | apparently initializing a variable is not "using" it but assigning to it is "using" it. i don't really like this fix, but it's better than trying to make a bigger cleanup just before a release, and it should work fine (tested against nsz's math tests).
* Merge remote-tracking branch 'nsz/master'Rich Felker2012-05-065-291/+176
|\
| * math: nextafter and nexttoward cleanupnsz2012-05-065-291/+176
| | | | | | | | | | make nexttoward, nexttowardf independent of long double representation. fix nextafterl: it did not raise underflow flag when the result was 0.
* | add isastream (obsolete STREAMS junk)Rich Felker2012-05-061-0/+7
|/ | | | | | apparently some packages see stropts.h and want to be able to use this. the implementation checks that the file descriptor is valid by using fcntl/F_GETFD so it can report an error if not (as specified).
* update license of njk contributed code (x86_64 asm)Rich Felker2012-05-055-5/+5
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | these changes are based on the following communication via email: "I hereby grant that all of the code I have contributed to musl on or before April 23, 2012 may be licensed under the terms of the following MIT license: Copyright (c) 2011-2012 Nicholas J. Kain Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the "Software"), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions: The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software. THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE."
* make pthread stacks non-executableRich Felker2012-05-041-1/+1
| | | | | | | | | | | | | this change is necessary or pthread_create will always fail on security-hardened kernels. i considered first trying to make the stack executable and simply retrying without execute permissions when the first try fails, but (1) this would incur a serious performance penalty on hardened systems, and (2) having the stack be executable is just a bad idea from a security standpoint. if there is real-world "GNU C" code that uses nested functions with threads, and it can't be fixed, we'll have to consider other ways of solving the problem, but for now this seems like the best fix.
* Merge remote-tracking branch 'nsz/master'Rich Felker2012-05-042-20/+16
|\
| * math: change the formula used for acos.snsz2012-05-052-20/+16
| | | | | | | | | | | | | | | | | | | | old: 2*atan2(sqrt(1-x),sqrt(1+x)) new: atan2(fabs(sqrt((1-x)*(1+x))),x) improvements: * all edge cases are fixed (sign of zero in downward rounding) * a bit faster (here a single call is about 131ns vs 162ns) * a bit more precise (at most 1ulp error on 1M uniform random samples in [0,1), the old formula gave some 2ulp errors as well)
* | fix error reporting for dlsym with global symbolsRich Felker2012-05-041-1/+2
|/
* fix uninitialized var in vfwprintf printing 0-prec stringRich Felker2012-05-041-1/+1
| | | | this could lead to spurious failures of wide printf functions
* avoid setting nondefault scheduler tooRich Felker2012-05-031-1/+1
|
* implement stub versions of sched_*Rich Felker2012-05-037-0/+49
| | | | | these actually work, but for now they prohibit actually setting priority levels and report min/max priority as 0.
* overhaul SSP support to use a real canaryRich Felker2012-05-036-12/+25
| | | | | | | | | | | | | pthread structure has been adjusted to match the glibc/GCC abi for where the canary is stored on i386 and x86_64. it will need variants for other archs to provide the added security of the canary's entropy, but even without that it still works as well as the old "minimal" ssp support. eventually such changes will be made anyway, since they are also needed for GCC/C11 thread-local storage support (not yet implemented). care is taken not to attempt initializing the thread pointer unless the program actually uses SSP (by reference to __stack_chk_fail).
* fix longstanding exit logic bugs in mbsnrtowcs and wcsnrtombsRich Felker2012-05-022-4/+9
| | | | | | | | these are POSIX 2008 (previously GNU extension) functions that are rarely used. apparently they had never been tested before, since the end-of-string logic was completely missing. mbsnrtowcs is used by modern versions of bash for its glob implementation, and and this bug was causing tab completion to hang in an infinite loop.
* support alternate glibc name pow10 for exp10Rich Felker2012-05-013-0/+9
|
* add C stub for sqrtl (already implemented in asm on i386 and x86_64)Rich Felker2012-04-301-0/+9
|
* first try at writing an efficient and "correct" exp10Rich Felker2012-04-303-0/+55
| | | | | | | | | this is a nonstandard function so it's not clear what conditions it should satisfy. my intent is that it be fast and exact for positive integral exponents when the result fits in the destination type, and fast and correctly rounded for small negative integral exponents. otherwise we aim for at most 1ulp error; it seems to differ from pow by at most 1ulp and it's often 2-5 times faster than pow.
* make stack protector work with gcc configured for non-tls canaryRich Felker2012-04-301-0/+2
|
* fix off-by-one error that caused uninitialized memory read in floatscanRich Felker2012-04-301-1/+1
| | | | | | this caused misreading of certain floating point values that are exact multiples of large powers of ten, unpredictable depending on prior stack contents.
* fix typo in the x86_64 rounding asmRich Felker2012-04-294-4/+4
|
* new math asm (abs/rounding) for x86_64Rich Felker2012-04-296-0/+36
| | | | untested
* add linux-specific unshare syscall wrapperRich Felker2012-04-291-0/+8
|
* fix longstanding missing static in mq_notify (namespace pollution)Rich Felker2012-04-291-1/+1
|
* new fnmatch implementationRich Felker2012-04-281-131/+273
| | | | | | | | | | unlike the old one, this one's algorithm does not suffer from potential stack overflow issues or pathologically bad performance on certain patterns. instead of backtracking, it uses a matching algorithm which I have not seen before (unsure whether I invented or re-invented it) that runs in O(1) space and O(nm) time. it may be possible to improve the time to O(n), but not without significantly greater complexity.
* update fnmatch to POSIX 2008 semanticsRich Felker2012-04-261-4/+11
| | | | | | | an invalid bracket expression must be treated as if the opening bracket were just a literal character. this is to fix a bug whereby POSIX left the behavior of the "[" shell command undefined due to it being an invalid bracket expression.
* gdb shared library debugging supportRich Felker2012-04-251-5/+32
| | | | | provide the minimal level of dynamic linker-to-debugger glue needed to let gdb find loaded libraries and load their symbols.
* first attempt at enabling stack protector supportRich Felker2012-04-243-0/+28
| | | | | | | | the code is written to pre-init the thread pointer in static linked programs that pull in __stack_chk_fail or dynamic-linked programs that lookup the symbol. no explicit canary is set; the canary will be whatever happens to be in the thread structure at the offset gcc hard-coded. this can be improved later.
* ditch the priority inheritance locks; use malloc's version of lockRich Felker2012-04-2415-77/+54
| | | | | | | | | | | | | | | | | | | i did some testing trying to switch malloc to use the new internal lock with priority inheritance, and my malloc contention test got 20-100 times slower. if priority inheritance futexes are this slow, it's simply too high a price to pay for avoiding priority inversion. maybe we can consider them somewhere down the road once the kernel folks get their act together on this (and perferably don't link it to glibc's inefficient lock API)... as such, i've switch __lock to use malloc's implementation of lightweight locks, and updated all the users of the code to use an array with a waiter count for their locks. this should give optimal performance in the vast majority of cases, and it's simple. malloc is still using its own internal copy of the lock code because it seems to yield measurably better performance with -O3 when it's inlined (20% or more difference in the contention stress test).
* internal locks: new owner of contended lock must set waiters flagRich Felker2012-04-241-1/+1
| | | | | | | | | | | | | | | | | | | | | | this bug probably would have gone unnoticed since it's only used in the fallback code for systems where priority-inheritance locking fails. unfortunately this approach results in one spurious wake syscall on the final unlock, when there are no waiters remaining. the alternative (possibly better) would be to use broadcast wakes instead of reflagging the waiter unconditionally, and let each waiter reflag itself; this saves one syscall at the expense of invoking the "thundering herd" effect (worse performance degredation) when there are many waiters. ideally we would be able to update all of our locks to use an array of two ints rather than a single int, and use a separate counter system like proper mutexes use; then we could avoid all spurious wake calls without resorting to broadcasts. however, it's not clear to me that priority inheritance futexes support this usage. the kernel sets the waiters flag for them (just like we're doing now) and i can't tell if it's safe to bypass the kernel when unlocking just because we know (from private data, the waiter count) that there are no waiters. this is something that could be explored in the future.
* new internal locking primitive; drop spinlocksRich Felker2012-04-242-7/+29
| | | | | | we use priority inheritance futexes if possible so that the library cannot hit internal priority inversion deadlocks in the presence of realtime priority scheduling (full support to be added later).