- Jul 03, 1990
-
-
ram authored
and do an immediate load. We still allocate the indirect constant, but don't waste time indirecting twice.
-
ram authored
comparison stuff to use the constant arg stuff.
-
ram authored
It is admittedly somewhat bogus to allocate a indirect constant and then access it as an immediate, but less bogus than indirecting for recreational value.
-
wlott authored
-
wlott authored
-
wlott authored
-
wlott authored
-
wlott authored
-
- Jul 02, 1990
-
-
ram authored
-
ram authored
-
ram authored
-
ram authored
-
ram authored
-
ram authored
since the compare doesn't work for floats or complexes.
-
wlott authored
-
wlott authored
-
wlott authored
-
wlott authored
get-type :translate functions of the same name.
-
wlott authored
non-other-pointer things, and extended it to return NIL if the header-data is zero.
-
wlott authored
symbols function slot to be the undefined function.
-
- Jun 29, 1990
-
-
wlott authored
-
- Jun 27, 1990
-
-
wlott authored
compute the code object from the LRA a second time when a single value was returned.
-
- Jun 26, 1990
- Jun 25, 1990
-
-
wlott authored
-
ram authored
-
wlott authored
dynamic-space-free-pointer, etc, to return the type system-area-pointer not (system-area-pointer).
-
wlott authored
SC for the data in set-header-data. In compute-function, force the offset to be in either a signed-reg or unsigned-reg, so we don't have to shift it yourself.
-
wlott authored
-
wlott authored
-
wlott authored
thing and use a longer code sequence when necessary.
-
ram authored
for moving to/from odd FP registers. Generalized memory reference definition macro to be usable to define FP load/store. Ripped stuff out of MOVE for doing integer/FP register moves, since it can't work with odd FP registers. Instead defined mtc1, mtc1-odd, etc. Defined the FLOAT-OP instruction for +,-,*,/ and flushed the separate dyadic float operation instructions.
-
- Jun 23, 1990
- Jun 22, 1990
- Jun 21, 1990
-
-
ram authored
continuation is constant.
-
- Jun 20, 1990