linux-mips
[Top] [All Lists]

Re: Floating point questions

To: "Andrew R. Baker" <andrewb@uab.edu>
Subject: Re: Floating point questions
From: Ralf Baechle <ralf@oss.sgi.com>
Date: Tue, 18 Jan 2000 04:53:34 +0100
Cc: Ralf Baechle <ralf@oss.sgi.com>, Linux SGI <linux@cthulhu.engr.sgi.com>
In-reply-to: <Pine.LNX.3.96.1000117213351.28191G-100000@mdk187.tucc.uab.edu>
References: <20000118042655.C10759@uni-koblenz.de> <Pine.LNX.3.96.1000117213351.28191G-100000@mdk187.tucc.uab.edu>
Sender: owner-linuxmips@oss.sgi.com
On Mon, Jan 17, 2000 at 09:41:01PM -0600, Andrew R. Baker wrote:

> > correct one just leaves all exceptions disabled.
> > 
> > Btw, C9x provides interfaces to manipulate the exception bits.
> 
> OK, I'll just set my test progs to clear the enable bits.

It may be hairy but I think you'll have to deal with exception sooner
or later ...

> I was basing my assumptions on what the sparc fpu system looked like.  The
> code seems to infer that, in SMP, the fp emulator can be run for a process
> that is not 'current'.  I'll stick with the direct access method than (I
> like it better anyway).
> 
> I am going to try and add the rest of the ops this week and put something
> out for people to look at.  Then I will clean it up some.  Is there any
> disadvantage to leaving this as a module?

At least as in CVS the module doesn't make sense except for debugging -
we don't have a mechanism for loading it on demand.

  Ralf

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