Date: Mon, 17 Jan 2005 20:39:47 -0800 From: Ade Lovett <ade@FreeBSD.org> To: John Baldwin <jhb@FreeBSD.org> Cc: freebsd-amd64@FreeBSD.org Subject: Re: [RFC] what to name linux 32-bit compat Message-ID: <41EC9313.90008@FreeBSD.org> In-Reply-To: <200501172235.34509.jhb@FreeBSD.org> References: <20050117203818.GA29131@dragon.nuxi.com> <41EC7D01.2070107@freebsd.org> <20050118032436.GA5325@odin.ac.hmc.edu> <200501172235.34509.jhb@FreeBSD.org>
next in thread | previous in thread | raw e-mail | index | archive | help
John Baldwin wrote:
> Actually, I think going the non-ambiguous route and using the fuller names
> like that (now that I see it) is probably the best bet when there is more
> than one possibility.
If you're (plural) going to go for the fuller names, given that /compat,
(or more likely /usr/compat), could be shared in a large environment, it
may be wise to revisit completely the naming:
/usr/compat/linux/<fbsdarch>-<linuxarch>/...
eg: linux/amd64-i386, linux/alpha-alpha
{/usr}/compat/linux/proc could still easily exist for linprocfs, if
required.
hell, even linux-pc98-i386 should something like that ever exist.
I'm purposely not extending things to differentiate between redhat,
suse, etc., since I'm a firm believer that we should be choosing one,
and exactly one, "linux distribution" for the linuxulator.
The key point I guess I'm trying to make is that this particular problem
is not really limited to amd64 vs. i386 (or even 64- vs. 32-bit), and
probably needs a bit more widespread thinking before going with a
solution. If things are going to change, let's at least get them right
for all possible architectures we know of (and support) now, rather than
specific hacks for one particular instantiation of FreeBSD.
-aDe
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?41EC9313.90008>