summaryrefslogtreecommitdiff
AgeCommit message (Collapse)AuthorLines
2012-03-17minor 387 fenv optimizationsRich Felker-6/+5
2012-03-17Merge remote branch 'nsz/master'Rich Felker-10/+10
2012-03-17fix i386 fegetround and make fesetround fasternsz-10/+10
Note that the new fesetround has slightly different semantics: Storing the floating-point environment with fnstenv makes the next fldenv (or fldcw) "non-signaling", so unmasked and pending exceptions does not invoke the exception handler. (These are rare since exceptions are handled immediately and by default all exceptions are masked anyway. But if one manually unmasks an exception in the control word then either sets the corresponding exception flag in the status word or the execution of an exception raising floating-point operation gets interrupted then it may happen). So the old implementation did not trap in some rare cases where the new implementation traps. However POSIX does not specify anything like the x87 exception handling traps and the fnstenv/fldenv pair is significantly slower than the fnstcw/fldcw pair (new code is about 5x faster here and it's dominated by the function call overhead).
2012-03-17don't fail on inability to install dynamic linker (e.g. if not root)Rich Felker-1/+1
2012-03-17one more fenv availability issue: lroundRich Felker-0/+2
2012-03-16make fma and lrint functions build without full fenv supportRich Felker-4/+28
this is necessary to support archs where fenv is incomplete or unavailable (presently arm). fma, fmal, and the lrint family should work perfectly fine with this change; fmaf is slightly broken with respect to rounding as it depends on non-default rounding modes to do its work.
2012-03-16other side of the signgam namespace fix: use the internal nameRich Felker-3/+7
2012-03-16make signgam a weak alias for an internal symbolRich Felker-2/+5
otherwise, the standard C lgamma function will clobber a symbol in the namespace reserved for the application.
2012-03-16fix namespace issues for lgamma, etc.Rich Felker-14/+25
standard functions cannot depend on nonstandard symbols
2012-03-16Merge remote branch 'nsz/master'Rich Felker-237/+440
2012-03-16remove junk sincos implementations in preparation to merge nsz's real onesRich Felker-27/+0
2012-03-16revert COPYRIGHT file changes in preparation to merge nsz's math branchRich Felker-4/+5
2012-03-16update copyright status (math library and new year)Rich Felker-6/+5
2012-03-15remove special nan handling from x86 sqrt asmRich Felker-3/+0
a double precision nan, when converted to extended (80-bit) precision, will never end in 0x400, since the corresponding bits do not exist in the original double precision value. thus there's no need to waste time and code size on this check.
2012-03-15simplify nan check in sqrt (x86 asm); result of sqrt is never negativeRich Felker-4/+3
2012-03-15in math.h make lgamma_r and non-double bessel _GNU_SOURCE onlynsz-20/+21
long double and float bessel functions are no longer xsi extensions
2012-03-15efficient sincos based on sin and cosnsz-8/+251
2012-03-15implement sincosf and sincosl functions; add prototypesRich Felker-0/+19
presumably broken gcc may generate calls to these, and it's said that ffmpeg makes use of sincosf.
2012-03-15avoid changing NaNs in sqrt (x86 asm) to satisfy c99 f.9 recommendationRich Felker-0/+4
2012-03-15correctly rounded sqrt() asm for x86 (i387)Rich Felker-0/+16
the fsqrt opcode is correctly rounded, but only in the fpu's selected precision mode, which is 80-bit extended precision. to get a correctly rounded double precision output, we check for the only corner cases where two-step rounding could give different results than one-step (extended-precision mantissa ending in 0x400) and adjust the mantissa slightly in the opposite direction of the rounding which the fpu already did (reported in the c1 flag of the fpu status word). this should have near-zero cost in the non-corner cases and at worst very low cost. note that in order for sqrt() to get used when compiling with gcc, the broken, non-conformant builtin sqrt must be disabled.
2012-03-13correct rounding for i387 sqrtf functionRich Felker-0/+2
2012-03-13math cleanup: use 1.0f instead of 1.0Fnsz-6/+6
2012-03-13math cleanup: use 1.0f instead of (float)1.0nsz-96/+96
2012-03-13remove libm.h includes when math.h and float.h are enoughnsz-30/+47
2012-03-13fix scanf handling of "0" (followed by immediate EOF) with "%x"Rich Felker-11/+6
other cases with %x were probably broken too. I would actually like to go ahead and replace this code in scanf with calls to the new __intparse framework, but for now this calls for a quick and unobtrusive fix without the risk of breaking other things.
2012-03-13clean up __expo2.c, use a slightly better k constantnsz-84/+14
2012-03-13fix copyright notice for the math libnsz-5/+9
2012-03-13implement nan, nanf, nanlRich Felker-0/+18
2012-03-13first commit of the new libm!Rich Felker-7738/+20547
thanks to the hard work of Szabolcs Nagy (nsz), identifying the best (from correctness and license standpoint) implementations from freebsd and openbsd and cleaning them up! musl should now fully support c99 float and long double math functions, and has near-complete complex math support. tgmath should also work (fully on gcc-compatible compilers, and mostly on any c99 compiler). based largely on commit 0376d44a890fea261506f1fc63833e7a686dca19 from nsz's libm git repo, with some additions (dummy versions of a few missing long double complex functions, etc.) by me. various cleanups still need to be made, including re-adding (if they're correct) some asm functions that were dropped.
2012-03-09add .gitignore fileRich Felker-0/+9
I've had this around for a long time but somehow it never got committed.
2012-03-02fix nan/infinity macros in math.h, etc.Rich Felker-6/+8
the previous version not only failed to work in c++, but also failed to produce constant expressions, making the macros useless as initializers for objects of static storage duration. gcc 3.3 and later have builtins for these, which sadly seem to be the most "portable" solution. the alternative definitions produce exceptions (for NAN) and compiler warnings (for INFINITY) on newer versions of gcc.
2012-03-02fix obscure bug in strtoull reading the highest 16 possible valuesRich Felker-1/+1
2012-03-02typo in math.h c version checkRich Felker-1/+1
2012-03-02make math.h compatibe with c89Rich Felker-1/+5
2012-03-02remove debug cruft that was left in getdateRich Felker-2/+0
2012-03-02first try at implementing getdate functionRich Felker-0/+47
2012-03-02fix bugs in strptime handling of string day/month names, literalsRich Felker-0/+2
2012-03-01implement a64l and l64a (legacy xsi stuff)Rich Felker-0/+26
2012-03-01add all missing wchar functions except floating point parsersRich Felker-0/+98
these are mostly untested and adapted directly from corresponding byte string functions and similar.
2012-03-01support null buffer argument to getcwd, auto-allocating behaviorRich Felker-1/+6
this is a popular extension some programs depend on, and by using a temporary buffer and strdup rather than malloc prior to the syscall, i've avoided the dependency on free and thus minimized the bloat cost of supporting this feature.
2012-03-01add memory.h, bogus legacy alias for string.hRich Felker-0/+1
2012-03-01search internal headers firstRich Felker-1/+1
this is necessitated by the ugly <syscall.h> just added
2012-02-29use c++-friendly initializers for pthread initializer definitionsRich Felker-3/+3
these will also avoid obnoxious warnings with gcc -Wbraces.
2012-02-29add <syscall.h> as an alias for <sys/syscall.h>Rich Felker-0/+1
apparently some broken stuff (libstdc++) needs this.
2012-02-28implement wcsftime functionRich Felker-0/+32
2012-02-28release notes for 0.8.6v0.8.6Rich Felker-0/+7
2012-02-28fix pthread_cleanup_pop(1) crash in non-thread-capable, static-linked programsRich Felker-2/+2
2012-02-27update release notes for 0.8.5v0.8.5Rich Felker-0/+35
2012-02-27work around "signal loses thread pointer" issue with "approach 2"Rich Felker-2/+8
this was discussed on the mailing list and no consensus on the preferred solution was reached, so in anticipation of a release, i'm just committing a minimally-invasive solution that avoids the problem by ensuring that multi-threaded-capable programs will always have initialized the thread pointer before any signal handler can run. in the long term we may switch to initializing the thread pointer at program start time whenever the program has the potential to access any per-thread data.
2012-02-25use __attribute__((const)) on arm __pthread_self functionRich Felker-1/+3
generated by cgit v1.2.1 (git 2.18.0) at 2024-04-25 12:40:29 +0000