This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
Re: PATCH: Support "-march=armv6j", not "-march=armv6"
- From: Richard Earnshaw <rearnsha at arm dot com>
- To: mark at codesourcery dot com
- Cc: binutils at sources dot redhat dot com, rearnsha at arm dot com
- Date: Fri, 19 Dec 2003 10:14:24 +0000
- Subject: Re: PATCH: Support "-march=armv6j", not "-march=armv6"
- Organization: ARM Ltd.
- Reply-to: Richard dot Earnshaw at arm dot com
>
> This patch makes GAS match GCC with respect to the V6 "-march="
> option. GCC spells it "-march=armv6j", so now so does GAS.
>
> It's my understand that all V6 cores have the Jazelle extensions.
> Still, It's possible that we should add "-march=armv6" as a synonym to
> both GCC and GAS. Richard, do you have thoughts about that?
No, this isn't quite accurate. All cores will have the ARM-state
extensions required to support Jazelle (the bxj instruction). But not all
cores are required to support Jazelle itself (direct interpretation of the
java byte-codes). Indeed, ARM has already announced arm11 cores that do
not contain Jazelle.
So there really is a distinction between a v6 and a v6j implementation --
it just so happens that the distinctions are irrelevant in the assembler
(and the compiler).
I think the above is sufficient justification, however, for us to support
both v6 and v6j options in the tools.
R.