This is the mail archive of the
mailing list for the binutils project.
Re: Partial WebAssembly backend
- From: David Edelsohn <dje dot gcc at gmail dot com>
- To: Florian Weimer <fw at deneb dot enyo dot de>
- Cc: Pip Cet <pipcet at gmail dot com>, Binutils Development <binutils at sourceware dot org>
- Date: Mon, 6 Mar 2017 18:59:24 -0500
- Subject: Re: Partial WebAssembly backend
- Authentication-results: sourceware.org; auth=none
- References: <CAGWvnymEedcq2cekks6tFyFp-d9rkA19ok=eDu67JkWVLyQQBQ@mail.gmail.com> <email@example.com>
On Mon, Mar 6, 2017 at 5:07 PM, Florian Weimer <firstname.lastname@example.org> wrote:
> * David Edelsohn:
>> You have a copyright assignment on file for GCC and Emacs, but not for
>> Binutils and GLIBC. You will need to assign the copyright for those
>> components to the FSF as well.
> The GCC bits also need review from the FSF how they interact with the
> GCC Runtime Library Exception. If I'm not mistaken, the exception was
> designed to prevent something like that from happening (using the GCC
> front ends with proprietary backends such as the compilers in
> Microsoft Edge or Google Chrome).
As has been discussed numerous times and I alluded to in the reply,
GCC already supports targets whose ISAs are not directly implemented
in the processor, namely GPU ISA. In addition, GCC code can be run on
software simulators that optimize code. Architectures implemented
with dynamic binary translators, such as Transmeta. Architectures
implemented with microcode. Even x86 processors no longer directly
execute the user ISA, but translate to micro-ops, including
optimizations such as fusing instructions.
WebAssembly stack machine is the target user ISA for this environment.
Please stop re-arguing this issue. It only serves to create FUD and to
No, we don't need to review this yet again.