[comp.sys.transputer] real arithmetic on transputers in tah mode.

HALLAM@physics.oxford.ac.uk ("Phillip M. Hallam-Baker") (04/30/91)

Dear net,

	It appears from the toolset manual that there is no way of using the
floating point libraries with code compiled for arbitary (ie tah) transputers.

I understand why inmos did not allow this originaly but fail to see why it
has not been fixed since. After all if there is code to emulat the floating
point on t414/t425 then why not have code doing emulation on t800 as well?

***Yes I know that that is inefficient ****

But I don't much care for speed at the moment - I am producing a set of
example programs for a callaboration using assorted t4/t8 hardware and
would like to have a single set of code for every transputer.

Any ideas ?

	Phillip M. Hallam-Baker