From: Laurent DESNOGUES <laurent.desnogues@wanadoo.fr>
To: qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] I want to add the ARMv6 instructions, who can give some advices?
Date: Wed, 7 Jun 2006 18:34:13 +0200 (CEST) [thread overview]
Message-ID: <13441846.1149698053893.JavaMail.www@wwinf1507> (raw)
> > All the information used to implement the current qemu Arm support is
> > available from other sources not covered by this licence. I'm confident I
> > could prove this if necessary.
True for ARMv6, probably not for ARMv7, be it -A, -R or -M.
> In the thread you cited earlier, Wolfgang Schildbach refers to "ARM
> System Developer's Guide", by Sloss, Symes, and Wright, Elsevier 2004
> as documentation that may be sufficient for Qemu ARMv6 support, and
> which is not covered by the license.
Yes. But you would need instruction encoding that are not in the
book. You will find them in binutils (opcode directory).
> If ARMv6 support were developed by someone else, using only that book,
> and maybe looking at other code (GCC, Linux etc.), that would prove
> that all the information used is from other sources not covered by the
> license, wouldn't it?
Linux uses a small subset of the System Coprocessor. If you want
information about ARMv6 cp15, look in ARM TRM (arm11), I don't think
they have any restriction such as ARMv7-M.
> The support might be missing a few features, if the other information
> sources are incomplete, but even incomplete support that can be
> deduced from those sources would be good enough for most purposes.
Certainly yes. But that's not a small effort :)
But bear in mind a few things: for instance supersection,
that are ARMv6 specific, were added to Linux kernel in 2.6.13
which is kind of recent; so if you omit some features, you
might have to add them in one day.
> If that were done, you could prove that the resulting feature in Qemu
> was written using sources not covered by the license.
>
> So would you have any problem contributing to Qemu after ARMv6 support
> was integrated, if the person who contributes ARMv6 support states
> that they have never seen the ARM document and refers to the sources
> they have used instead?
I think that it could be very difficult to prove you did not
use any official NDA document. I am doing simulator development
at ARM and would certainly never approach anything v6 or v7
related until information is publically available with no
restriction.
Anyway I am not speaking for ARM :)
Laurent
next reply other threads:[~2006-06-07 16:34 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2006-06-07 16:34 Laurent DESNOGUES [this message]
-- strict thread matches above, loose matches on Subject: below --
2006-06-07 14:58 [Qemu-devel] I want to add the ARMv6 instructions, who can give some advices? Laurent DESNOGUES
2006-06-07 7:11 wang lianwei
2006-06-07 13:35 ` Paul Brook
2006-06-07 14:38 ` John R.
2006-06-07 14:46 ` Paul Brook
2006-06-07 14:53 ` Jamie Lokier
2006-06-07 15:07 ` Paul Brook
2006-06-07 15:48 ` Jamie Lokier
2006-06-07 15:58 ` Paul Brook
2006-06-07 16:18 ` Jamie Lokier
2006-06-07 16:34 ` Paul Brook
2006-06-07 17:21 ` Jamie Lokier
2006-06-07 17:25 ` Paul Brook
2006-06-07 17:42 ` John R.
2006-06-07 18:36 ` Jamie Lokier
2006-06-07 14:46 ` Jamie Lokier
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=13441846.1149698053893.JavaMail.www@wwinf1507 \
--to=laurent.desnogues@wanadoo.fr \
--cc=qemu-devel@nongnu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).