sm-discuss AT lists.ibiblio.org
Subject: Public SourceMage Discussion List
List archive
- From: Karsten Behrmann <BearPerson AT gmx.net>
- To: sm-discuss AT lists.ibiblio.org
- Subject: Re: [SM-Discuss] State of x86_64
- Date: Wed, 30 Nov 2005 17:54:24 +0100
On Tue, 29 Nov 2005 23:08:26 -0500, Sergey A. Lipnevich wrote:
> Hi Jaanus,
>
> I believe that the brief status is that it mostly Just Works(tm).
Well, as far as I know, its mostly "'everybody' wants it, nobody tests it,
nobody writes code for it.", though I've heard of several working boxes
on x86_64, so it must be reasonably working.
> I think I used one of the fresh "testing" ISOs to make my latest amd64
> install this Summer.
Well, the x86_64's are not quite so "fresh", they are made by Benoit
PAPILLAULT using the old generation of scripts, which have been replaced
by what made 0.9.5 and will be used from this point. I don't know how well
they work, though, I haven't used them.
As for new x86_64 ISOs (for example, one for 0.9.6), I remember the official
policy being "getting to 1.0 is more important", so it is a bit on the back-
burner.
As the total lack of testing the ppc ISOs are getting shows (The first
person to run through a total install with them was a new user), making not-
so-common architectures work is often quite hard simply because "you have no
clue if the thing actually works".
That being said, while it's not quite a priority for me to get x86_64
working and released (ATM, ppc is more important to me), I would probably
not mind "testing the waters" a bit by generating an x86_64 ISO off the
current x86 line.
However, there is one important hurdle to master: I need an ISO factory host
before I am able to do anything. This means that I need a box that is
running SMGL on x86_64, with ssh access to it, and root access (usually via
su and the root password).
It also needs >1GB spare hard disk space, the sorcery on the host system
might be unavailable (e.g. for casting) for an hour or more maybe every
month, and we'll need plenty of processor time.
Having a good bit of _free_ RAM (300MB) for HD caching helps things. Also,
experience shows that these boxes have a rather high tendency to get memory
leaks in the kernel (or other problems) and crash.
If anyone wants to volunteer for one, feel free to drop me or dkowis a line,
or drop by #sourcemage-cauldron.
So Far,
Karsten Behrmann
--
Karsten Behrmann | Source Mage GNU/Linux developer, ISO team
BearPerson AT gmx.net | http://www.sourcemage.org
Attachment:
signature.asc
Description: PGP signature
-
[SM-Discuss] State of x64,
Jaanus Heeringson, 11/29/2005
-
Re: [SM-Discuss] State of x64,
Sergey A. Lipnevich, 11/29/2005
-
Re: [SM-Discuss] State of x86_64,
Karsten Behrmann, 11/30/2005
-
Re: [SM-Discuss] State of x86_64,
Sergey A. Lipnevich, 11/30/2005
- Re: [SM-Discuss] State of x86_64, David Kowis, 11/30/2005
-
Re: [SM-Discuss] State of x86_64,
Jeremy Blosser (emrys), 11/30/2005
-
Re: [SM-Discuss] State of x86_64,
Eric Sandall, 11/30/2005
- Re: [SM-Discuss] State of x86_64, Andrew, 11/30/2005
-
Re: [SM-Discuss] State of x86_64,
Eric Sandall, 11/30/2005
-
Re: [SM-Discuss] State of x86_64,
Sergey A. Lipnevich, 11/30/2005
-
Re: [SM-Discuss] State of x86_64,
Karsten Behrmann, 11/30/2005
- <Possible follow-up(s)>
-
[SM-Discuss] State of x64,
Janus Heeringson, 11/30/2005
- Re: [SM-Discuss] State of x64, Unet, 11/30/2005
-
Re: [SM-Discuss] State of x64,
Sergey A. Lipnevich, 11/29/2005
Archive powered by MHonArc 2.6.24.