https://gcc.gnu.org/bugzilla/show_bug.cgi?id=109460
Eric Botcazou <ebotcazou at gcc dot gnu.org> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |WAITING
--- Comment #4 from Eric Botcazou <ebotcazou at gcc dot gnu.org> ---
> Not to me because this has been built successfully before. If this was
> the problem then it would never build, right?
This looks like a bug in any case and thus needs to be fixed.
> What happened in this case was that sym-mingw32.o was treated as an input to
> the compiler (see "No such file or directory" error) rather than as an
> output which it really is.
Yes, because -o is missing.
> I suspect this had to do with -fno-PIE triggering building an executable,
> which I still don't know where it is coming from and whether it should be
> used for compiling this file, as this has nothing to do with an executable -
> we are only compiling the .cc file to a .o file.
-fno-PIE is just a standard -f compilation option, nothing more.
> Still I think it is important to have details of the overall build process
> followed to get some context.
No disagreement.
