Ivan Boldyrev <[EMAIL PROTECTED]> writes:

> CMU CL has non-standard operations like kernel:32bit-logical-xor.
> What about creating functions for 32-bit or fixnum arithmetics?  I
> would even try to participate :)

AFAIH, Raymond has ported (UNSIGNED-BYTE 32)[*] arithmetic,
automatically generating efficient assembler for a code like
(LDB (BYTE 32 0) (LOGNOT (* (LOGNOR X Y) Z))). A similar
optimization could be done for (SIGNED-BYTE 32) and (SIGNED-BYTE 30),
if there is an interest in it (but we probably need to add EXT:SLDB).

[*] In fact, (UNSIGNED-BYTE word-size).

-- 
Regards,
Alexey Dejneka

"Alas, the spheres of truth are less transparent than those of
illusion." -- L.E.J. Brouwer

Reply via email to