linux-mips-fnet
[Top] [All Lists]

Re: Location of init_task_union

To: "Bradley D. LaRonde" <brad@ltc.com>
Subject: Re: Location of init_task_union
From: Ralf Baechle <ralf@oss.sgi.com>
Date: Wed, 24 Nov 1999 00:29:42 +0100
Cc: Ralf Baechle <ralf@oss.sgi.com>, linux@engr.sgi.com, linux-mips@fnet.fr, linux-mips@vger.rutgers.edu
In-reply-to: <023301bf3606$338ee650$b8119526@ltc.com>
References: <007801bf3500$90edf810$b8119526@ltc.com> <19991123230824.E16508@uni-koblenz.de> <01f101bf3601$0cf42fa0$b8119526@ltc.com> <19991123232110.H16508@uni-koblenz.de> <021f01bf3602$97f7b0d0$b8119526@ltc.com> <19991123235314.I16508@uni-koblenz.de> <023301bf3606$338ee650$b8119526@ltc.com>
On Tue, Nov 23, 1999 at 05:58:05PM -0500, Bradley D. LaRonde wrote:

> > Btw, take a look at arch/mips64/kernel/head.S which looks much cleaner
> than
> > what we currently have for the 32-bit kernel.  That head.S was definately
> > the single most ugly thing we had in the kernel.  It's all gone now; the
> > TLB exception handlers and the CPU detection code is now in other files;
> > the later has been rewritten to C.
> 
> Oooh, cool.  Yes, that was an eyesore.  In fact was toying with the idea of
> hacking it up (to not include exception handlers we will never need).

> So when do we get this for mips32?  :-)

Lemme hack some more on the MIPS64 stuff, I want it get stable first before
I put the bits into MIPS32.  Right now I'm working on stabilizing the
binary compatibility stuff for MIPS32.  We neither have a 64-bit userland
nor want one for almost all cases except large memory apps, so right now I'm
putting my efforts behind the compatibility stuff.  Oh, unlike the 32-bit
kernel the 64-bit kernel allows the full use of all MIPS III and MIPS IV
features including 64-bit registers even for 32-bit apps.  This opens some
possibilities for optimization of performance critical code like in libc.

  Ralf

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