[fpc-devel] compilation error of h2pas.pas on Mac OS X in the course of "make all" of the SVN sources
Florian Klämpfl
florian at freepascal.org
Sat Sep 17 11:34:10 CEST 2005
Dr. Rolf Jansen wrote:
> since some days I am no more able to "make all" the latest and greatest
> fpc from SVN sources, because there are compilation errors for h2pas.pas.
>
> Here comes a log of a session which reproduces the error:
>
> PBRJ:~/Programmieren/Xcode/Pascal/fpc Rolf$ svn update
> At revision 1091.
>
> PBRJ:~/Programmieren/Xcode/Pascal/fpc Rolf$ make all
> make compiler_cycle RELEASE=1
> ...
> ...
> ...
> make -C h2pas all
> /Users/Rolf/Programmieren/Xcode/Pascal/fpc/compiler/ppcppc -Ur -Xs - O1
> -n -Sg -Fu/Users/Rolf/Programmieren/Xcode/Pascal/fpc/rtl/units/
> powerpc-darwin -FE. -FUunits/powerpc-darwin -dpowerpc -dRELEASE h2pas.pas
> h2pas.pas:5526:Fixup of 34988 too large for field width of 16 bits
> h2pas.pas:5502:Fixup of 35076 too large for field width of 16 bits
> h2pas.pas:5484:Fixup of 35140 too large for field width of 16 bits
> h2pas.pas:5444:Fixup of 35288 too large for field width of 16 bits
> h2pas.pas:5365:Fixup of 35580 too large for field width of 16 bits
> h2pas.pas:5363:Fixup of 35588 too large for field width of 16 bits
> h2pas.pas:5360:Fixup of 35600 too large for field width of 16 bits
> h2pas.pas(2879,15) Error: Error while assembling exitcode 1
> h2pas.pas(2879,15) Fatal: There were 2 errors compiling module, stopping
> h2pas.pas(2879,15) Fatal: Compilation aborted
> make[3]: *** [h2pas] Error 1
> make[2]: *** [h2pas_all] Error 2
> make[1]: *** [utils_all] Error 2
> make: *** [build-stamp.powerpc-darwin] Error 2
>
> I had a look at the source of h2pas.pas, but I found no obvious errors
> at the reported lines. Has anybody else experienced this problem, or
> might there be something wrong with my set up.
>
> Any suggestions are welcome.
This is a known problem, it's a bug in the powerpc code generator, Jonas
is already working on it afaik.
Background: h2pas.pas is generated by yacc from h2pas.y and generates a
huge procedure which a real programmer would never create, that's why it
poped up so late.
More information about the fpc-devel
mailing list