[Qemu-devel] Re: Inquiry, speed comparison on OS X, QEMU vs Virual PC

[Qemu-devel] Re: Inquiry, speed comparison on OS X, QEMU vs Virual PC

[Fred Hope]

 >> I think that including support for the PowerPC swapping instructions 
in QEMU
 >> will break compatibility with host PowerPCs before G3, so that 
instructions
 >> should be used in a run-time capability detection scheme.

 >It also dosen't work on the 970 (G5) - one of my friends has been
 >waiting for QEMU for OS X for that reason ;)

About breaking support for PowerPCs before G3, I wouldn't think that 
matters, since OS X only runs on G3 and later unless you use a special 
hack or something.

But if it doesn't work on the G5, then yeah hopefully it can be 
controlled at runtime and still let G3/G4 users have this stuff.

Anyone care to explain to me what this stuff we're talking about 
actually does?  :) Sorry but I dunno much about processor instructions 
and stuff.  I assume it would make QEMU faster...would it make any 
substantial difference, or add any other features?

Fred Hope

Re: [Qemu-devel] Re: Inquiry, speed comparison on OS X, QEMU vs Virual PC

[Sander Nagtegaal]

Are you talking about pseudo little-endian mode on a G3/G4? Make no mistakes 
that would speed up the emulation by about 30/50% ( Atleast it did on Virtual 
PC )
I've been asking about this before but got no replies from here at all.
Also..........since QEMU-fast only works on a modified Linux kernel and makes 
QEMU about 50% faster ( atleast that's what the website says ) is there a way 
to make QEMU-fast work on WINDOWS/95/98/2000/NT/XP to? I suppose Virtual PC 
does this somehow. No idea how though..........

Op zaterdag 10 juli 2004 6:41 am, schreef Fred Hope:
>  >> I think that including support for the PowerPC swapping instructions
>
> in QEMU
>
>  >> will break compatibility with host PowerPCs before G3, so that
>
> instructions
>
>  >> should be used in a run-time capability detection scheme.
>  >
>  >It also dosen't work on the 970 (G5) - one of my friends has been
>  >waiting for QEMU for OS X for that reason ;)
>
> About breaking support for PowerPCs before G3, I wouldn't think that
> matters, since OS X only runs on G3 and later unless you use a special
> hack or something.
>
> But if it doesn't work on the G5, then yeah hopefully it can be
> controlled at runtime and still let G3/G4 users have this stuff.
>
> Anyone care to explain to me what this stuff we're talking about
> actually does?  :) Sorry but I dunno much about processor instructions
> and stuff.  I assume it would make QEMU faster...would it make any
> substantial difference, or add any other features?
>
> Fred Hope
>
>
>
> _______________________________________________
> Qemu-devel mailing list
> Qemu-devel@nongnu.org
> http://lists.nongnu.org/mailman/listinfo/qemu-devel

Re: [Qemu-devel] Re: Inquiry, speed comparison on OS X, QEMU vs Virual PC

[John R. Hogerhuis]

The code primitives output by the dynamic translator are actually just
the result of compiling short swaths of C code.

Whether an PowerPC little-endian mode can be used would depend only on:

1) Whether the compiler (gcc) supports such a mode
2) If there is an option for it, how that option is enabled, and whether
that option is enabled. Is it just a command line option? Do you have to
use pragma's in the code?

Neither of which I know the answer to, but that should give you
somewhere to look.

-- John.

Re: [Qemu-devel] Re: Inquiry, speed comparison on OS X, =?iso-8859-1?q? QEMU=20vs=20Virual?= PC

[Colin Watson]

On Mon, Jul 12, 2004 at 09:33:10AM -0700, John R. Hogerhuis wrote:
> The code primitives output by the dynamic translator are actually just
> the result of compiling short swaths of C code.
> 
> Whether an PowerPC little-endian mode can be used would depend only on:
> 
> 1) Whether the compiler (gcc) supports such a mode

`-mlittle'
`-mlittle-endian'
     On System V.4 and embedded PowerPC systems compile code for the
     processor in little endian mode.  The `-mlittle-endian' option is
     the same as `-mlittle'.

-- 
Colin Watson                                  [cjwatson@flatline.org.uk]