[fpc-devel] Broken frac function in FPC3.1.1 / Windows x86_64
If you compile and run this 64-bit program on Win 64 you get a crash
program tfrac;
var
x,y: double;
begin
x := 1e20;
writeln('x=',x);
y := frac(x);
writeln('frac(x)=',y);
end.
D:\Work\DAMath>D:\FPC311\bin\i386-win32\ppcrossx64.exe tfrac.pas
Free Pascal Compiler version 3.1.1-r20:38794 [2018/04/22] for x86_64
Copyright (c) 1993-2018 by Florian Klaempfl and others
Target OS: Win64 for x64
Compiling tfrac.pas
Linking tfrac.exe
9 lines compiled, 0.1 sec, 37472 bytes code, 2036 bytes data
D:\Work\DAMath>tfrac
x= 1.E+020
Runtime error 207 at $00011BD0
$00011BD0
$00011614
$00011696
$000195D0
$00011580
$773D59CD
$7763383D
I recently posted an issue to the bugtracker
https://bugs.freepascal.org/view.php?id=33635
But the only responses are by a reporter named Thaddy de Koning, which
are for me totally useless and IMO he did miss the point.
Are there changes to the frac function (or its documentation) in the
newer 64-bit windows cross-compiler? Especially for the allowed ranges?
This would not be intuitive because the corresponding int function works as
expected, and meanwhile I use the work-around frac(x) := int(x) -x
Wolfgang Ehrhardt
___
fpc-devel maillist - fpc-devel@lists.freepascal.org
http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel
[fpc-devel] Panel Top Negative Limitation.
Hi I have a problem using a panel, that is large say 7. If I try to bring areas of this into the view area by using panel.top:=-avalue, this works fine until you hit –32769; at which point the panel vanishes. How can this be corrected; or is there a workaround? I have a forum thread; where I have attached simple project demonstrating the problem. http://forum.lazarus.freepascal.org/index.php/topic,31355.msg200797.html#msg200797 Any thoughts appreciated. Josh___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel
Re: [fpc-devel] FPC 3.0.0 on OSX 10.6.8 32bit Compile problem
After setting the -Aaa-darwin flag It progresses further, but now stops here 1010) Writing Resource String Table file: uPSRuntime.rsj :3687:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :8518:2: error: invalid instruction mnemonic 'fistpq' fistpq -12(%ebp) ^~ :8936:2: error: invalid instruction mnemonic 'fistpq' fistpq -12(%ebp) ^~ :10178:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :10189:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :10200:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :10211:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :10223:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :10232:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :10241:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :10249:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :10257:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :10266:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :10280:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :11146:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :11372:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :11549:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :11983:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :15014:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :20147:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :20162:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :20177:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :20192:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :20206:2: error: invalid instruction mnemonic 'fistpq' fistpq (%ecx) ^~ :20219:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :20230:2: error: invalid instruction mnemonic 'fistpq' fistpq (%ecx) ^~ :20242:2: error: invalid instruction mnemonic 'fistpq' fistpq (%ecx) ^~ :20254:2: error: invalid instruction mnemonic 'fistpq' fistpq (%ebx) ^~ :20267:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :20277:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :20295:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :21287:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :21302:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :21317:2: error: invalid instruction mnemonic 'fistpq' fistpq (%ecx) ^~ :21332:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :21346:2: error: invalid instruction mnemonic 'fistpq' fistpq (%ecx) ^~ :21359:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :21370:2: error: invalid instruction mnemonic 'fistpq' fistpq (%ecx) ^~ :21382:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :21394:2: error: invalid instruction mnemonic 'fistpq' fistpq (%edx) ^~ :21407:2: error: invalid instruction mnemonic 'fistpq' fistpq (%ecx) ^~ :21417:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :21435:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :22402:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :22420:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :22438:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :22456:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :22470:2: error: invalid instruction mnemonic 'fistpq' fistpq (%ecx) ^~ :22486:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :22497:2: error: invalid instruction mnemonic 'fistpq' fistpq (%ecx) ^~ :22512:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :22527:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~ :22543:2: error: invalid instruction mnemonic 'fistpq' fistpq (%eax) ^~
[fpc-devel] FPC 3.0.0 on OSX 10.6.8 32bit Compile problem
Hi I have been trying to get laz 1.6 rc2 on mac osx 10.6.8, I can instal 1.4.4 FPC 2.6.4; without problem. I have run the uninstall script and attempted 1.6rc2, when I try to rebuild Lazarus, or add a component that requires a rebuild I get the below error I have googled fistpq, and found that this is 64bit related. Any ideas Josh Hint: (11030) Start of reading config file /etc/fpc.cfg Hint: (11031) End of reading config file /etc/fpc.cfg Free Pascal Compiler version 3.0.0 [2015/11/14] for i386 Copyright (c) 1993-2015 by Florian Klaempfl and others (1002) Target OS: Darwin for i386 Free Pascal Compiler version 3.0.0 [2015/11/14] for i386 Copyright (c) 1993-2015 by Florian Klaempfl and others (1002) Target OS: Darwin for i386 (3104) Compiling alllclunits.pp (3104) Compiling lclproc.pas :4284:2: error: invalid instruction mnemonic 'fistpq' fistpq -12(%ebp) ^~ :4303:2: error: invalid instruction mnemonic 'fistpq' fistpq -12(%ebp) ^~ :4326:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ :4350:2: error: invalid instruction mnemonic 'fistpq' fistpq -16(%ebp) ^~ (9001) Assembling (pipe) /Developer/lazarus/lcl/units/i386-darwin/lclproc.s lclproc.pas(3221) Fatal: (10026) There were 1 errors compiling module, stopping Fatal: (1018) Compilation aborted ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel
Re: [fpc-devel] FPC 3.0.0 on OSX 10.6.8 32bit Compile problem
Hi Jonas Maebe, Yes using the latest xcode for 10.6.8 Sorry to sound stupid, how do I add -Aas-darwin to the compiler for lazarus so it can rebuild ok. Josh Jonas Maebe wrote: i...@voiceliveeditor.com wrote: I have been trying to get laz 1.6 rc2 on mac osx 10.6.8, I can instal 1.4.4 FPC 2.6.4; without problem. I have run the uninstall script and attempted 1.6rc2, when I try to rebuild Lazarus, or add a component that requires a rebuild I get the below error I have googled fistpq, and found that this is 64bit related. FPC 3.0 switched to using Clang by default as assembler, because the old assembler is no longer supported with newer Xcode versions. Normally, that should work fine under Mac OS X 10.6.8 too, as I tested that as well. In case I missed something, you can switch back to using the old assembler with the FPC command line parameter -Aas-darwin. Also, which version of Xcode do you have installed? Have you tried with the latest version of Xcode available for Mac OS X 10.6.8? (Xcode 3.2.6, IIRC) Jonas ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel ___ fpc-devel maillist - fpc-devel@lists.freepascal.org http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel
[fpc-devel] Changed Math functions in FPC development trunk 2.7
A year ago there was a post from Jonas Maebe on this list
(http://www.mail-archive.com/fpc-devel@lists.freepascal.org/msg28483.html)
asking about the possibility to reuse parts of AMath/DAMath in FPC.
Today I downloaded the development source files from
(ftp://ftp.freepascal.org/pub/fpc/snapshot/trunk/source/fpc.zip)
and found very close source code similarity between the changed
(compared with FPC264) math.pp routines and AMath routines.
See the list below.
Although, as Jonas wrote, zlib is compatible with FPC, there is the
need to give references and/or mark the changes. Unfortunately this
is not done.
I am quite sure that these routines are 'inspired' by AMath.
If this is the case, I would appreciate reference to 'AMath by W.Ehrhardt'
or so, as it is done in the FPC264 source for Arjan van Dijk.
Best wishes
Wolfgang Ehrhardt
hypot
---
AMATH
function hypot(x,y: extended): extended;
{-Return sqrt(x*x + y*y)}
begin
x := abs(x);
y := abs(y);
if xy then hypot := x*sqrt(1.0+sqr(y/x))
else if x0.0 then hypot := y*sqrt(1.0+sqr(x/y)) {here y = x 0}
else hypot := y; {here x=0}
end;
FPC-Dev
function hypot(x,y : float) : float;
begin
x:=abs(x);
y:=abs(y);
if (xy) then
hypot:=x*sqrt(1.0+sqr(y/x))
else if (x0.0) then
hypot:=y*sqrt(1.0+sqr(x/y))
else
hypot:=y;
end;
FPC 264
function hypot(x,y : float) : float;
begin
hypot:=Sqrt(x*x+y*y)
end;
arcsin and arccos
---
AMATH
function arcsin(x: extended): extended;
{-Return the inverse circular sine of x, |x| = 1}
begin
{basic formula arcsin(x) = arctan(x/sqrt(1-x^2))}
arcsin := arctan2(x, sqrt((1.0-x)*(1.0+x)))
end;
function arccos(x: extended): extended;
{-Return the inverse circular cosine of x, |x| = 1}
begin
{basic formula arccos(x) = arctan(sqrt(1-x^2)/x))}
if abs(x)=1.0 then begin
if x0.0 then arccos := Pi else arccos := 0.0;
end
else arccos := arctan2(sqrt((1.0-x)*(1.0+x)),x)
end;
FPC-Dev
function arcsin(x : float) : float;
begin
arcsin:=arctan2(x,sqrt((1.0-x)*(1.0+x)));
end;
function Arccos(x : Float) : Float;
begin
if abs(x)=1.0 then
if x0.0 then
arccos:=Pi
else
arccos:=0
else
arccos:=arctan2(sqrt((1.0-x)*(1.0+x)),x);
end;
FPC 264
{ ArcSin and ArcCos from Arjan van Dijk (arjan.vand...@user.metair.wau.nl) }
function arcsin(x : float) : float;
begin
if abs(x) 1 then InvalidArgument
else if abs(x) 0.5 then
arcsin := arctan(x/sqrt(1-sqr(x)))
else
arcsin := sign(x) * (pi*0.5 - arctan(sqrt(1 / sqr(x) - 1)));
end;
function Arccos(x : Float) : Float;
begin
arccos := pi*0.5 - arcsin(x);
end;
lnxp1 = ln1p
---
AMath (up to V1.74 of 04.Oct.2013, V1.75 introduces an additional branch)
function ln1p(x: extended): extended;
{-Return ln(1+x), accurate even for x near 0}
var
y: extended;
const
x0 = 4.0;
begin
if x = x0 then ln1p := ln(1.0 + x)
else begin
y := 1.0 + x;
{The following formula is more accurate than Goldberg [3], Theorem 4. The}
{Taylor series f(x) = f(x0) + (x-x0)*f'(x0) + .. for f(x) = ln(1+x) gives}
{ln1p(x) = ln(1+x0) + (x-x0)/(1+x0) = ln(y) + (x-(y-1))/y, with y = 1+x0.}
if y=1.0 then ln1p := x
else ln1p := ln(y) + (x-(y-1.0))/y;
end
end;
FPC-Dev
function lnxp1(x : float) : float;
var
y: float;
begin
if (x=4.0) then
lnxp1:=ln(1.0+x)
else
begin
y:=1.0+x;
if (y=1.0) then
lnxp1:=x
else
lnxp1:=ln(y)+(x-(y-1.0))/y;
end;
end;
FPC 264
function lnxp1(x : float) : float;
begin
if x-1 then
InvalidArgument;
lnxp1:=ln(1+x);
end;
___
fpc-devel maillist - fpc-devel@lists.freepascal.org
http://lists.freepascal.org/cgi-bin/mailman/listinfo/fpc-devel
