This is the mail archive of the crossgcc@cygnus.com mailing list for the crossgcc project.
Index Nav: | [Date Index] [Subject Index] [Author Index] [Thread Index] | |
---|---|---|
Message Nav: | [Date Prev] [Date Next] | [Thread Prev] [Thread Next] |
Jx Saarinen wrote: > Actually I managed to do this OK. But when I tested the mingw32-h8300 > cross toolset, h8300-hms-gcc test.c somehow managed to call i386-as > claiming 'c:\tmp\tjaetj.s this is not a legal i386 instruction'. The '<GCC> -print-search-dirs' (GCC = h8300-hms-gcc now) is the normal way to check or debug the installation. If it shows quite wrong search paths for programs (cpp, cc1, as, ld), something went wrong, or you have to use the directories it uses, or have to use the GCC_EXEC_PREFIX, or to do something else. > building binutils-2.9.1 (the newest?) to --target=h8300-hms is > OK when --host=i586-linux but when i586-cygwin32/mingw32 is involved > (ie. compiling mingw/cygwin cross h8300-hms in linux or compiling > target=h8300-hms in cygwin fails at srconv.c)? In Linux I can > build binutils-2.9.1 without problems at all. There is a hassle with the 'EXEEXT' in 'SRCONV_PROG'... As you can see, in the Makefile the 'SRCONV_PROG' is defined to be three programs, but it is used with as '$(SRCONV_PROG)$(EXEEXT)', just as it was a single program... So you have to edit the Makefile and add the '$(EXEEXT)' after the three program names in the 'SRCONV_PROG' definition, and remove all the '$(EXEEXT)'s after the '$(SRCONV_PROG)'s... I think this being fixed in the current egcs-snapshots, i.e. these things edited in the Makefile.in. If you think you will reconfigure, as is the case with the two hosts you need -- edit the 'Makefile.in' instead... Cheers, Kai _______________________________________________ New CrossGCC FAQ: http://www.objsw.com/CrossGCC _______________________________________________ To remove yourself from the crossgcc list, send mail to crossgcc-request@cygnus.com with the text 'unsubscribe' (without the quotes) in the body of the message.