This is the mail archive of the
gcc-patches@gcc.gnu.org
mailing list for the GCC project.
Re: Cygwin breakage caused by 2000-01-05 varasm.c change
- To: Chris Faylor <cgf at redhat dot com>
- Subject: Re: Cygwin breakage caused by 2000-01-05 varasm.c change
- From: Jeffrey A Law <law at redhat dot com>
- Date: Fri, 12 Jan 2001 10:41:08 -0700
- cc: gcc-patches at gcc dot gnu dot org, mrs at wrs dot com
- Reply-To: law at redhat dot com
In message <20010111234437.A1347@redhat.com>you write:
> Hello,
> This recent change to varasm.c has caused some breakage in cygwin
> compilers.
>
> 2001-01-05 Mike Stump <mrs@wrs.com>
>
> * varasm.c (assemble_name): Ensure we output the stripped name.
>
> This change strips the '@' from a symbol before writing it, which means
> that "stdcall" function lables are not properly adorned. stdcall functions
> in Windows are suffixed with the number of bytes in the argument list so
> stripping the @n causes subsequent linker problems.
>
> I don't know if this change is wrong or if Cygwin's STRIP_NAME_ENCODING
> is wrong but, for now, I've reverted this change in my sandbox so that
> I can continue to use the CVS version of gcc.
>
> I'd appreciate knowing what the real resolution to this problem might
> be.
I believe Mike's patch is wrong; furthermore it should not have been
installed.
If one looks at how ports define ASM_OUTPUT_LABELREF it's obvious the
patch Mike installed is wrong as ports do their own stripping in
ASM_OUTPUT_LABELREF.
I'm reverting Mike's patch.
Mike, try to be more careful in the future.
jeff