[kaffe] JamVM on ARM (XScale)

Robert Lougher rob.lougher at gmail.com
Tue Oct 12 11:27:18 PDT 2004


If anybody is interested I've hacked the calling-convention for ARM on
JamVM to assume soft-fp.  Note, this is a pure guess, as I don't have
access to an ARM machine until the weekend!



P.S.  The attached files go into src/arm...

P.P.S.  For anybody who's interested in BE machines, I believe doubles
are "pure-big-endian", unlike on LE ARMs (each word is BE, but the
words are in LE order).  Copying the class_md.h file from src/powerpc
to src/arm MAY work, if my guess is again correct.  I can test some of
this on the weekend, but not before :(

On Tue, 12 Oct 2004 09:57:02 +0100, S. Meslin-Weber
<steph at tangency.co.uk> wrote:
> Hi Rob,
> On Mon, Oct 11, 2004 at 12:46:02PM +0100, Robert Lougher wrote:
> > The problems that could arise are that I assume floating-point
> > emulation is enabled in the kernel (affects calling-convention).  I
> Ah, this could cause issues with some linux distributions for ARM
> (familiar and OpenZaurus) which are switching to soft-float and not
> relying on the kernel to do floating point handling. Note that hybrid
> systems with both types requires a set of core libraries for each
> *-float so on embedded platforms only one should really be there.
> The rationale for this is convincing: 1000% benchmarked improvement in
> float speed. I've had this explained to me (in simple terms):
>  "with hard float, a kernel exception gets launched ... which needs
>  saving status registers and changing privilege rings ... then jump
>  into kernel ... then handle that ... then jump out of kernel ...
>  that's slooooooooooow"
> It'd be nice to be able to use JamVM (and other jvms) on these embedded
> distributions too :)
> I've verified that SableVM works fine using soft-float too :)
> Cheers,
> Steph
> --
> ================================================================
> Stephane Meslin-Weber         Email: steph at tangency.co.uk
> Software Engineer             Web: http://odonata.tangency.co.uk
> ================================================================

More information about the kaffe mailing list