This is the mail archive of the
gcc-bugs@gcc.gnu.org
mailing list for the GCC project.
Re: gcc.dg/20001013-1.c
- To: Mark Mitchell <mark at codesourcery dot com>
- Subject: Re: gcc.dg/20001013-1.c
- From: Jakub Jelinek <jakub at redhat dot com>
- Date: Mon, 14 May 2001 05:19:08 -0400
- Cc: rth at redhat dot com, davem at redhat dot com, gcc-bugs at gcc dot gnu dot org
- References: <20010514011432J.mitchell@codesourcery.com>
- Reply-To: Jakub Jelinek <jakub at redhat dot com>
On Mon, May 14, 2001 at 01:14:32AM -0700, Mark Mitchell wrote:
>
> This is a SPARC-specific test that uses `-O2 -mvis' to compile. The
> -mvis option is undocumented.
>
> On Solaris 2.8, I get errors like:
>
> /usr/ccs/bin/as: "/var/tmp/ccQ94Ir5.s", line 33: error: cannot use
> v8plus instructions in a non-v8plus target binary
>
> from the Solaris assembler.
>
> To get the file to compile, the assembler needs to be passed
> `-xarch=v9a'. I assume you tested with GAS?
I've tested it on sparc64-linux if I remember right, there is -Av9a passed
to GAS automatically.
Anyway, -mvis should not turn -xarch=v9a, but v8plusa or v9a depending on
whether -m32 or -m64 (or what's the default).
This patch should take care of it, though I'm not able to test it on
Solaris (no access to it), will test it later on sparc*-linux.
While adding invoke.texi descriptions for -mvis and -mv8plus I took the
liberty of removing the descriptions of long time deprecated options, they
are deprecated since at least egcs 1.1.x and IMHO there is no need to even
tell anyone they exists still. I'll post a separate patch to remove them for
trunk.
Jakub