Attached a testbed for the QuadMath lib - about 90 functions. Building requires MasmBasic 17 August (http://masm32.com/board/index.php?topic=94.0) and a 32-bit GCC installation.
It seems to work so far (in the sense: with arbitrary numbers, it doesn't throw exceptions), but my math skills are clearly not good enough. Unfortunately, the documentation for that library is lousy. Does anybody know enough about these functions to design meaningful tests?
3.16227766016837933199889354443272e-01
Complex numbers:
-0.300000000000000000000000000000000r 0.100000000000000000000000000000000i
-0.500000000000000000000000000000000r 0.200000000000000000000000000000000i
1 6.79675056344163539816123196439774e-01 Quad$(QuadMath(acosq, f128_1)) arc cosine function
2 -nan Quad$(QuadMath(acoshq, f128_1)) inverse hyperbolic cosine function
3 8.91121270450733079415198495199977e-01 Quad$(QuadMath(asinq, f128_1)) arc sine function
4 7.15220808521240713368180068139420e-01 Quad$(QuadMath(asinhq, f128_1)) inverse hyperbolic sine function
5 6.61042684235119596446884984928197e-01 Quad$(QuadMath(atanq, f128_1)) arc tangent function
6 1.03971880209293260540357689630520e+00 Quad$(QuadMath(atanhq, f128_1)) inverse hyperbolic tangent function
7 7.18829999621624505417014151525905e-01 Quad$(QuadMath(atan2q, f128_1, f128 arc tangent function
8 9.19641085579807331708184954882012e-01 Quad$(QuadMath(cbrtq, f128_1)) cube root function
9 1.00000000000000000000000000000000e+00 Quad$(QuadMath(ceilq, f128_1)) ceiling value function
10 7.77777000000000000000000000000000e-01 Quad$(QuadMath(copysignq, f128_1, f copy sign of a number
11 1.31802721007488863001046816772184e+00 Quad$(QuadMath(coshq, f128_1)) hyperbolic cosine function
12 7.12475170303218178481897562102759e-01 Quad$(QuadMath(cosq, f128_1)) cosine function
13 7.28642947148188881602143870565993e-01 Quad$(QuadMath(erfq, f128_1)) error function
14 2.71357052851811118397856129434007e-01 Quad$(QuadMath(erfcq, f128_1)) complementary error function
15 2.17662823878251521093957473563285e+00 Quad$(QuadMath(expq, f128_1)) exponential function
16 1.17662823878251521093957473563285e+00 Quad$(QuadMath(expm1q, f128_1)) exponential minus 1 function
17 7.77777000000000000000000000000000e-01 Quad$(QuadMath(fabsq, f128_1)) absolute value function
18 0.00000000000000000000000000000000e+00 Quad$(QuadMath(fdimq, f128_1, f128_ positive difference function
19 1 Str$(QuadMath(finiteq, f128_1)) check finiteness of value
20 0.00000000000000000000000000000000e+00 Quad$(QuadMath(floorq, f128_1)) floor value function
21 1.69135564197599999999999999999999e+00 Quad$(QuadMath(fmaq, f128_1, f128_2 fused multiply and add
22 8.88887999999999999999999999999999e-01 Quad$(QuadMath(fmaxq, f128_1, f128_ determine maximum of two values
23 7.77776999999999999999999999999999e-01 Quad$(QuadMath(fminq, f128_1, f128_ determine minimum of two values
24 7.77776999999999999999999999999999e-01 Quad$(QuadMath(fmodq, f128_1, f128_ remainder value function
25 7.77776999999999999999999999999999e-01 Quad$(QuadMath(frexpq, f128_1, &int extract mantissa and exponent
26 1.18112613139875963037219043508971e+00 Quad$(QuadMath(hypotq, f128_1, f128 Eucledian distance function
27 0 Str$(QuadMath(isinfq, f128_1)) check for infinity
28 -1 Str$(QuadMath(ilogbq, f128_1)) get exponent of the value
29 0 Str$(QuadMath(isnanq, f128_1)) check for not a number
30 8.54388504707771712092309739814944e-01 Quad$(QuadMath(j0q, f128_1)) Bessel function of the first kind, first order
31 3.60213810637432487067489691829813e-01 Quad$(QuadMath(j1q, f128_1)) Bessel function of the first kind, second order
33 7.77776999999999999999999999999999e-01 Quad$(QuadMath(ldexpq, f128_1)) load exponent of the value
34 1.74081138059681937201628653041458e-01 Quad$(QuadMath(lgammaq, f128_1)) logarithmic gamma function
35 0 Str$(QuadMath(llrintq, f128_1)) round to nearest integer value
36 1 Str$(QuadMath(llroundq, f128_1)) round to nearest integer value away from zero
37 -2.51315428281406078018471313735404e-01 Quad$(QuadMath(logq, f128_1)) natural logarithm function
38 -1.09144903719767094515383760835130e-01 Quad$(QuadMath(log10q, f128_1)) base 10 logarithm function
39 -3.62571522080470492430259119929199e-01 Quad$(QuadMath(log2q, f128_1)) base 2 logarithm function
40 5.75363707403466151725524591370230e-01 Quad$(QuadMath(log1pq, f128_1)) compute natural logarithm of the value plus one
41 0 Str$(QuadMath(lrintq, f128_1)) round to nearest integer value
42 1 Str$(QuadMath(lroundq, f128_1)) round to nearest integer value away from zero
43 7.77776999999999999999999999999999e-01 Quad$(QuadMath(modfq, f128_1, OWORD decompose the floating-point number
44 nan Quad$(QuadMath(nanq, dword PTR [ebx return quiet NaN
45 0.00000000000000000000000000000000e+00 Quad$(QuadMath(nearbyintq, f128_1)) round to nearest integer
46 7.77776999999999999999999999999999e-01 Quad$(QuadMath(nextafterq, f128_1, next representable floating-point number
47 1.00000000000000000000000000000000e+00 Quad$(QuadMath(powq, f128_1, f128_2 power function
48 -nan Quad$(QuadMath(remainderq, f128_1, remainder function
49 -nan Quad$(QuadMath(remquoq, f128_1, f12 remainder and part of quotient
50 0.00000000000000000000000000000000e+00 Quad$(QuadMath(rintq, f128_1)) round-to-nearest integral value
51 1.00000000000000000000000000000000e+00 Quad$(QuadMath(roundq, f128_1)) round-to-nearest integral value, return __float128
52 7.77776999999999999999999999999999e-01 Quad$(QuadMath(scalblnq, f128_1)) compute exponent using FLT_RADIX
53 7.77776999999999999999999999999999e-01 Quad$(QuadMath(scalbnq, f128_1)) compute exponent using FLT_RADIX
54 0 Str$(QuadMath(signbitq, f128_1)) return sign bit
55_ 7.77776999999999999999999999999999e-01 QuadMath(sincosq, f128_1, OWORD PTR calculate sine and cosine simultaneously
56 8.58601028707626580929106567911009e-01 Quad$(QuadMath(sinhq, f128_1)) hyperbolic sine function
57 7.01697321999592822577494428203881e-01 Quad$(QuadMath(sinq, f128_1)) sine function
58 8.81916662729534779709025956238766e-01 Quad$(QuadMath(sqrtq, f128_1)) square root function
59 9.84872668195526766344208902855793e-01 Quad$(QuadMath(tanq, f128_1)) tangent function
60 6.51428909922764009034212653601591e-01 Quad$(QuadMath(tanhq, f128_1)) hyperbolic tangent function
61 1.19015212853728978240171569191066e+00 Quad$(QuadMath(tgammaq, f128_1)) true gamma function
62 0.00000000000000000000000000000000e+00 Quad$(QuadMath(truncq, f128_1)) round to integer, towards zero
63 -1.08823755104290320639507753297712e-01 Quad$(QuadMath(y0q, f128_1)) Bessel function of the second kind, first order
64 -1.00388964645596038174697269505841e+00 Quad$(QuadMath(y1q, f128_1)) Bessel function of the second kind, second order
65 -inf Quad$(QuadMath(ynq, f128_1)) Bessel function of the second kind, n-th order
66 1.11111100000000000000000000000000e+00 Quad$(QuadMath(cabsq, c128_1)) complex absolute value function
67 0.00000000000000000000000000000000e+00 Quad$(QuadMath(cargq, c128_2)) calculate the argument
68 0.00000000000000000000000000000000e+00 Quad$(QuadMath(cimagq, c128_1)) imaginary part of complex number
69 1.11111100000000000000000000000000e+00 Quad$(QuadMath(crealq, c128_1)) real part of complex number
70 0.00000000000000000000000000000000e+00 Quad$(QuadMath(cacosq, c128_1)) complex arc cosine function
71 4.67145078687467774975308067585712e-01 Quad$(QuadMath(cacoshq, c128_1)) complex arc hyperbolic cosine function
72 1.57079632679489661923132169163975e+00 Quad$(QuadMath(casinq, c128_1)) complex arc sine function
73 9.57800374871255507537319505105416e-01 Quad$(QuadMath(casinhq, c128_1)) complex arc hyperbolic sine function
74 8.37981175284630374344180916658704e-01 Quad$(QuadMath(catanq, c128_1)) complex arc tangent function
75 1.47221996326768006396617005905066e+00 Quad$(QuadMath(catanhq, c128_1)) complex arc hyperbolic tangent function
76 4.43666121279137034963262698715523e-01 Quad$(QuadMath(ccosq, c128_1)) complex cosine function
77 1.68346223218832726875681425264904e+00 Quad$(QuadMath(ccoshq, c128_1)) complex hyperbolic cosine function
78 3.03773143999174827676530639951211e+00 Quad$(QuadMath(cexpq, c128_1)) complex exponential function
79 7.12475170303218178481897562102759e-01 Quad$(QuadMath(cexpiq, f128_1)) computes the exponential function of "i" times a real value
80 1.05360415657821301227167647480979e-01 Quad$(QuadMath(clogq, c128_1)) complex natural logarithm
81 4.57574471312247636122071472795808e-02 Quad$(QuadMath(clog10q, c128_1)) complex base 10 logarithm
82 1.11111100000000000000000000000000e+00 Quad$(QuadMath(conjq, c128_1)) complex conjugate function
83 1.00000000000000000000000000000000e+00 Quad$(QuadMath(cpowq, c128_1, c128_ complex power function
84 1.11111100000000000000000000000000e+00 Quad$(QuadMath(cprojq, c128_1)) project into Riemann Sphere
85 8.96192151733726168465708950029809e-01 Quad$(QuadMath(csinq, c128_1)) complex sine function
86 1.35426920780342100800849214686307e+00 Quad$(QuadMath(csinhq, c128_1)) complex hyperbolic sine function
87 1.05409250068483079024421803054925e+00 Quad$(QuadMath(csqrtq, c128_1)) complex square root
88 2.01996976724278164327568071148291e+00 Quad$(QuadMath(ctanq, c128_1)) complex tangent function
89 8.04454761092567379921780988485073e-01 Quad$(QuadMath(ctanhq, c128_1)) complex hyperbolic tangent function
------------- all done -------------
deleted
__float128 q = 3.1415926535897932384626433832795028q;
Quote from: TWell on August 17, 2017, 02:55:51 PM
__float128 q = 3.1415926535897932384626433832795028q;
Indeed :t
include \masm32\MasmBasic\MasmBasic.inc ; download (http://masm32.com/board/index.php?topic=94.0)
include \masm32\MasmBasic\Res\QuadMath.inc
SetGlobals MyPI:REAL16
Init quad
PrintLine "digits", Tb$, "1.23456789012345678901234567890123"
PrintLine "PI", Tb$, "3.14159265358979323846264338327950", Tb$, "(as text)"
movups MyPI, Quad(3.1415926535897932384626433832795028841971)
PrintLine "MyPI", Tb$, Quad$(MyPI, g), Tb$, "Real16 precision"
movups MyPI, Quad(FP10(3.1415926535897932384626433832795028841971))
Inkey "MyPI", Tb$, Quad$(MyPI, g), Tb$, "Real10 precision"
EndOfCodedigits 1.23456789012345678901234567890123
PI 3.14159265358979323846264338327950 (as text)
MyPI 3.14159265358979323846264338327950 Real16 precision
MyPI 3.14159265358979323851280895940619 Real10 precision
But that is under control, thanks. I am more interested in the more complicated math stuff...
deleted
Quote from: jj2007 on August 17, 2017, 10:28:30 AM
Does anybody know enough about these functions to design meaningful tests?
Most of the functions can be used in high precision Digital Signal Processing.
Ask Raistlin, AFAIK he does calculations on pulsar data.
Thanks, Marinus.
Just discovered that C:\Windows\System32\calc.exe uses quad precision. But PEView doesn't give any hint to an undocumented Windows quadmath DLL :(
So they do it probably with a tiny part of the 776,192 bytes of calc.exe...
QuoteIn Windows 98 and later, it uses an arbitrary-precision arithmetic library
https://en.wikipedia.org/wiki/Windows_Calculator
Yes, but they don't tell you which one :(
The Wiki has a link to a defunct "calc plus" page:
Quick Details
File Name: CalcPlus.msi
...
Download Size: 476 KB
Estimated Download Time: 2 min 56K
Good ol' times :P
QuotePOLINK: fatal error: File not found: 'C:\TDM-GCC-32\lib\gcc\mingw32\4.9.2\libquadmath.a'.
I've installed the version tdm-gcc-5.1.0
Try Init dynaquad (that forces using the DLLs, which should always be in the same location).
If that doesn't work, try inserting this line before Init:
QuadLibStaticP$ equ <C:\TDM-GCC-32\lib\gcc\mingw32\5.1.0\>
There are more constants that can be adapted, but normally you shouldn't touch them:
QmStatic$ equ <libquadmath.a>
GccStatic$ equ <libgcc.a>
QuadLibDynaP$ equ <C:\TDM-GCC-32\bin\>
QmDll$ equ <libquadmath-0>
GccDll$ equ <libgcc_s_dw2-1>
Quote from: jj2007 on October 21, 2017, 03:10:38 AM
Try Init dynaquad (that forces using the DLLs, which should always be in the same location).
It works, thanks :t
The static version should work, too. It will give you plenty of warnings of the type "POLINK: warning: Section name '.debug_info' is longer than 8 characters; truncated." but they can be ignored.
In RichMasm, File/New Masm source opens a list of templates; towards the bottom is a QuadMath example; I just remembered that Quad(immediate) is no longer necessary, i.e. you can use the immediate REAL16 directly:
movups PiBy180, QuadMath(__divtf3, MyPI, Quad(180.0))
PrintLine "PI/180", Tb$, Quad$(PiBy180)
movups PiBy180, QuadMath(__divtf3, MyPI, 180.0)
PrintLine "PI/180", Tb$, Quad$(PiBy180)
Hi,
here is the latest version libquadmath_x32_x64_gcc-7_2_0 https://mega.co.nz/#!VxQhGYiZ!AAAAAAAAAACUIvLONoOxdwAAAAAAAAAAlCLyzjaDsXc (https://mega.co.nz/#!VxQhGYiZ!AAAAAAAAAACUIvLONoOxdwAAAAAAAAAAlCLyzjaDsXc) from the msys package, with headers :eusa_boohoo:
QuadLibDynaP$ equ <\Masm32\MasmBasic\Members\LiaoMi\x32\> ; path to the DLLs
Init dynaquad ; use dynaquad to link the DLLs dynamically
Works fine, thanks. Any big difference to the previous version? The new ones are a few kBytes shorter.
Quote from: jj2007 on October 22, 2017, 01:50:23 AM
QuadLibDynaP$ equ <\Masm32\MasmBasic\Members\LiaoMi\x32\> ; path to the DLLs
Init dynaquad ; use dynaquad to link the DLLs dynamically
Works fine, thanks. Any big difference to the previous version? The new ones are a few kBytes shorter.
There are almost no changes, here you can see the latest changes https://github.com/gcc-mirror/gcc/commits/master/libquadmath (https://github.com/gcc-mirror/gcc/commits/master/libquadmath). I installed the msys64 update and decided at that moment to take the actual libraries for projects, although the old libraries converge with the new ones :biggrin: