https://gcc.gnu.org/bugzilla/show_bug.cgi?id=123976

--- Comment #232 from Peter Dyballa <Peter_Dyballa at web dot de> ---
I saw these messages:

checking linker version... 85.2.1
checking dsymutil version "@(#)PROGRAM:dsymutil PROJECT:dwarfutils-70"... 70 :
DWARFUTILS 70 0 0  

and also that Python 3 would be nice to have. The last build log also exactly
tells (a few times):

checking whether ln -s works... yes
checking whether ln works... yes


I also tried a bit of further quality assurance checking the last build log. In
lines like

checking for powerpc-apple-darwin9.8.0-ranlib... ranlib

the final word "ranlib" is not "ranlib" but "ranlib␣", i.e. the word "ranlib"
followed by a SPACE character. Except when "Configuring stage 1 in
powerpc-apple-darwin9.8.0/libgcc", it's two SPACE characters.

Similarly when "Configuring stage 1 in powerpc-apple-darwin9.8.0/libgcc" I get:

checking for powerpc-apple-darwin9.8.0-ar... ar␣

"Configuring stage 1 in ./gcc" reports:

checking for BSD- or MS-compatible name lister (nm)... no

All other configure scripts have found that "nm" exists. But OK, could be it's
neither BSD- nor MS-compatible…

Once, in "Configuring stage 1 in ./gcc", it's found

checking for a BSD compatible install... /usr/bin/install -c

Eleven other times it's found

checking for a BSD-compatible install... /usr/bin/install -c

with a "-" between "BSD" and "compatible". When "Configuring stage 1 in ./gmp"
it's found these two insights:

checking for none-apple-darwin9.8.0-dlltool... dlltool
checking for none-apple-darwin9.8.0-objdump... objdump

Correct would be what other configure script find:

checking what objdump to use... not found
checking what readelf to use... not found

And the same for objdump, I think. Similarly when "Configuring stage 1 in"
./mpc & ./mpfr:

checking for powerpc-apple-darwin9.8.0-dlltool... dlltool

Can this be true? Or has someone hallucinations? Most configure scripts find:

checking whether we are cross compiling... no

but when "Configuring stage 1 in ./gmp" I get:

checking whether we are cross compiling... yes

Reply via email to