[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] CVS notes...

From: Chad Page
Subject: Re: [Qemu-devel] CVS notes...
Date: Wed, 8 Oct 2003 15:59:35 -0700 (PDT)

        Cool!  I'd definitely like to look at this sometime.  I don't have
much IEEE fp experience but I'd still like to do whatever I can to move it
along.  I know some common code uses the registers as 64-bit registers for
data moving, so at least that should be implemented.

        I don't think supervisor mode is needed yet for user emulation. 
If the same approach as qemu-i386 was taken to system emulation (i.e. 
running dosemu first), then the first step would be to simulate the MOL
kernel module's handling of things, and then full supervisor mode would
come later.

        Specific chip emulation would be good too... if it had to support
only one chip I'd vote for the G3.

        Hoping to see this in qemu soon...,

        - Chad

On 8 Oct 2003, Jocelyn Mayer wrote:

> Hi,
> I already have an user-mode PPC emulation, with no float able to launch
> static PPC binaries. I have no supervisor mode nor 64 bit PPC support,
> for now, and I only emulate the generic 32 bits PPC, according to the
> specification. I need to resync with latest architectural changes
> and it would be great that if I could fix dynamic binaries loading,
> which used to work, some times ago, then I will be able to send my (big)
> patch to Fabrice. If you'd like to help me and have some notions about
> IEEE floats, maybe you could make some generic helpers for float ops
> emulation ? This would be great !
> Another thing to do is to point differences between common PPC chips and
> the spec, so qemu would be able to emulate "real" PPC chips. I think PPC
> 603 and G3 would be a good start.
> Regards.
> -- 
> Jocelyn Mayer <address@hidden>
> Never organized
> _______________________________________________
> Qemu-devel mailing list
> address@hidden
> http://mail.nongnu.org/mailman/listinfo/qemu-devel

reply via email to

[Prev in Thread] Current Thread [Next in Thread]