[Top] [All Lists]

Re: [PATCH -queue v0 1/6] [loongson] add basic loongson-2f support

To: Wu Zhangjin <>
Subject: Re: [PATCH -queue v0 1/6] [loongson] add basic loongson-2f support
From: Ralf Baechle <>
Date: Wed, 4 Nov 2009 12:19:57 +0100
Cc: Arnaud Patard <>,, LKML <>,,, Zhang Le <>, Thomas Gleixner <>, Nicholas Mc Guire <>,,
In-reply-to: <>
Original-recipient: rfc822;
References: <> <> <> <>
User-agent: Mutt/1.5.19 (2009-01-05)
On Wed, Nov 04, 2009 at 07:04:12PM +0800, Wu Zhangjin wrote:

> > Small question : Why don't you restrict to 64bit kernels only ? From
> > what I remember from some discussions with ST, trying to use a 32-bit
> > kernel on 2f is a nice way to get troubles. It would be better imho to
> > forbid such a configuration. As a side effect, this will remove all
> > 'defined(CONFIG_64BIT)' parts of your #ifdef tests. 
> > 
> It's hard to make such a decision ;) Perhaps some guys want to play with
> the 32bit version.

We have other systems where 32-bit kernel support is just remarkably ugly.
We've dropped 32-bit support for the SGI IP32 aka O2 - nobody seems to even
have really noticed that.  The Sibyte systems would be good candidates to do
the same as accesses to outside the 32-bit address space are needed very


<Prev in Thread] Current Thread [Next in Thread>