about summary refs log tree commit diff
path: root/src
Commit message (Collapse)AuthorAgeFilesLines
* asm for log2Rich Felker2012-03-193-0/+21
|
* asm for remquoRich Felker2012-03-193-0/+43
| | | | | this could perhaps use some additional testing for corner cases, but it seems to be correct.
* optimize exponential asm for i386Rich Felker2012-03-192-58/+77
| | | | | | up to 30% faster exp2 by avoiding slow frndint and fscale functions. expm1 also takes a much more direct path for small arguments (the expected usage case).
* Merge remote branch 'nsz/master'Rich Felker2012-03-195-163/+208
|\
| * simplify scalbn*.c implementationsnsz2012-03-193-152/+65
| | | | | | | | | | The old scalbn.c was wrong and slow, the new one is just slow. (scalbn(0x1p+1023,-2097) should give 0x1p-1074, but the old code gave 0)
| * Merge branch 'master' of git://git.etalabs.net/muslnsz2012-03-199-0/+75
| |\
| * \ Merge branch 'master' of git://git.etalabs.net/muslnsz2012-03-1921-0/+157
| |\ \
| * \ \ Merge branch 'master' of git://git.etalabs.net/muslnsz2012-03-1914-1/+109
| |\ \ \
| * \ \ \ Merge branch 'master' of git://git.etalabs.net/muslnsz2012-03-199-0/+89
| |\ \ \ \
| * | | | | remove unnecessary TODO comments from fma.cnsz2012-03-191-5/+1
| | | | | |
| * | | | | fmal bug fix: nan input should not raise exceptionnsz2012-03-191-4/+4
| | | | | |
| * | | | | add fma implementation for x86nsz2012-03-191-7/+143
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | correctly rounded double precision fma using extended precision arithmetics for ld80 systems (x87)
* | | | | | fix broken modf family functionsRich Felker2012-03-193-27/+66
| | | | | |
* | | | | | asm for modf functionsRich Felker2012-03-193-0/+45
| |_|_|_|/ |/| | | |
* | | | | asm for floor/ceil/truncRich Felker2012-03-199-0/+75
| |_|_|/ |/| | |
* | | | asm for scalbn familyRich Felker2012-03-199-0/+64
| | | | | | | | | | | | | | | | | | | | | | | | | | | | unlike some implementations, these functions perform the equivalent of gcc's -ffloat-store on the result before returning. this is necessary to raise underflow/overflow/inexact exceptions, perform the correct rounding with denormals, etc.
* | | | asm for inverse trig functionsRich Felker2012-03-1912-0/+93
| |_|/ |/| | | | | | | | | | | | | | | | | unlike trig functions, these are easy to do in asm because they do not involve (arbitrary-precision) argument reduction. fpatan automatically takes care of domain issues, and in asin and acos, fsqrt takes care of them for us.
* | | asm for log functionsRich Felker2012-03-186-0/+42
| | |
* | | fix broken exponential asmRich Felker2012-03-182-1/+21
| | | | | | | | | | | | | | | | | | | | | | | | | | | infinities were getting converted into nans. the new code simply tests for infinity and replaces it with a large magnitude value of the same sign. also, the fcomi instruction is apparently not part of the i387 instruction set, so avoid using it.
* | | asm for lrint family on i386Rich Felker2012-03-186-0/+46
| |/ |/|
* | asm exponential functions for i386Rich Felker2012-03-189-0/+89
|/
* assembly optimizations for fmod/remainder functionsRich Felker2012-03-188-0/+88
|
* asm versions of some simple math functions for i386 and x86_64Rich Felker2012-03-188-0/+48
| | | | | | | these are functions that have direct fpu approaches to implementation without problematic exception or rounding issues. x86_64 lacks float/double versions because i'm unfamiliar with the necessary sse code for performing these operations.
* simplify lround and llround functionsnsz2012-03-186-112/+20
| | | | | Simple wrappers around round is enough because spurious inexact exception is allowed.
* make lrint and llrint functions work without fenv supportnsz2012-03-186-6/+16
|
* faster lrint and llrint functionsnsz2012-03-186-80/+99
| | | | | | | A faster workaround for spurious inexact exceptions when the result cannot be represented. The old code actually could be wrong, because gcc reordered the integer conversion and the exception check.
* fix loads of missing const in new libm, and some global vars (?!) in powlRich Felker2012-03-1820-51/+51
|
* try fixing/optimizing x86_64 fenv exception codeRich Felker2012-03-171-18/+23
| | | | untested; may need followup-fixes.
* optimize x86 feclearexceptRich Felker2012-03-171-16/+20
| | | | | if all exception flags will be cleared, we can avoid the expensive store/reload of the environment and just use the fnclex instruction.
* fix x86_64 fe[gs]etround, analogous to nsz's x86 changesRich Felker2012-03-171-8/+9
|
* minor 387 fenv optimizationsRich Felker2012-03-171-6/+5
|
* fix i386 fegetround and make fesetround fasternsz2012-03-171-10/+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).
* one more fenv availability issue: lroundRich Felker2012-03-171-0/+2
|
* make fma and lrint functions build without full fenv supportRich Felker2012-03-164-4/+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.
* other side of the signgam namespace fix: use the internal nameRich Felker2012-03-163-3/+7
|
* make signgam a weak alias for an internal symbolRich Felker2012-03-161-2/+5
| | | | | otherwise, the standard C lgamma function will clobber a symbol in the namespace reserved for the application.
* fix namespace issues for lgamma, etc.Rich Felker2012-03-167-14/+25
| | | | standard functions cannot depend on nonstandard symbols
* Merge remote branch 'nsz/master'Rich Felker2012-03-1666-216/+413
|\
| * in math.h make lgamma_r and non-double bessel _GNU_SOURCE onlynsz2012-03-153-0/+3
| | | | | | | | long double and float bessel functions are no longer xsi extensions
| * efficient sincos based on sin and cosnsz2012-03-154-8/+247
| |
| * math cleanup: use 1.0f instead of 1.0Fnsz2012-03-134-6/+6
| |
| * math cleanup: use 1.0f instead of (float)1.0nsz2012-03-1325-96/+96
| |
| * remove libm.h includes when math.h and float.h are enoughnsz2012-03-1331-30/+47
| |
| * clean up __expo2.c, use a slightly better k constantnsz2012-03-132-84/+14
| |
* | remove junk sincos implementations in preparation to merge nsz's real onesRich Felker2012-03-163-24/+0
| |
* | remove special nan handling from x86 sqrt asmRich Felker2012-03-151-3/+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.
* | simplify nan check in sqrt (x86 asm); result of sqrt is never negativeRich Felker2012-03-151-4/+3
| |
* | implement sincosf and sincosl functions; add prototypesRich Felker2012-03-152-0/+16
| | | | | | | | | | presumably broken gcc may generate calls to these, and it's said that ffmpeg makes use of sincosf.
* | avoid changing NaNs in sqrt (x86 asm) to satisfy c99 f.9 recommendationRich Felker2012-03-151-0/+4
| |
* | correctly rounded sqrt() asm for x86 (i387)Rich Felker2012-03-151-0/+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.