This is a regression from 2.59 to current CVS, I assume from the 2004-04-20 change:
touch a.c cat >configure.ac <<EOF AC_INIT AC_PROG_CXX EOF autoconf ./configure ls -l a.c # ls: a.c: No such file or directory This part seems to remove too much, for the sake of the IRIX linker: | # List of possible output files, starting from the most likely. | # The algorithm is not robust to junk in `.', hence go to wildcards (a.*) | # only as a last resort. b.out is created by i960 compilers. | ac_files='a_out.exe a.exe conftest.exe a.out conftest a.* conftest.* b.out' | # | # The IRIX 6 linker writes into existing files which may not be | # executable, retaining their permissions. Remove them first so a | # subsequent execution test works. | ac_rmfiles= | for ac_file in $ac_files | do | case $ac_file in | _AC_COMPILER_EXEEXT_REJECT ) ;; | * ) ac_rmfiles="$ac_rmfiles $ac_file";; | esac | done | rm -f $ac_rmfiles How about we remove only the output file that the IRIX linker will actually create (iff that's the only linker with this "feature")? I like writing test cases with files like a.c, b.c, etc., and I don't see where the Autoconf manual forbids `a.*', au contraire, a.c is even used in one example. Cheers, Ralf
