[Bug bootstrap/52947] [4.7/4.8 Regression] bootstrap fails due to wrong include search path composition

["rainer at emrich-ebersheim dot de" <gcc-bugzilla at gcc dot gnu dot org>]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=52947

--- Comment #2 from Rainer Emrich <rainer@emrich-ebersheim.de> 2012-04-12 15:36:11 UTC ---
(In reply to comment #1)
> I remember Kai did surgery in this place.  Did you identify a patch that caused
> this regression?  My bet would be
> 
> 2011-03-25  Kai Tietz  <ktietz@redhat.com>
> 
>         * collect2.c (write_c_file_stat): Handle backslash
>         as right-hand directory separator.
>         (resolve_lib_name): Use IS_DIR_SEPARATOR instead of
>         checking just for slash.
>         * coverage.c (coverage_init): Use IS_ABSOLUTE_PATH
>         instead of checking for trailing slash.
>         * gcc.c (record_temp_file): Use filename_cmp instead
>         of strcmp.
>         (do_spec_1): Likewise.
> ...

To be honest, I don't know. I tried native bootstrapping on x86_64-w64-mingw32
at the beginning of last year but gave up due to the issues found. Lately I
managed to bootstrap the 4.6 series with some manual intervention even ada
included.
There are still issues, one of them is really fundamental, but that's a
different story.

Btw. the PATH composition for --with-local-prefix is messed up the same way.