- Apr 23, 1990
-
-
ram authored
-
ram authored
-
ram authored
In DEBUG-INFO-FOR-COMPONENT, give the main entry to an optional-dispatch the original name and NIL kind. Fixed debug location dumping to dump locations in overflow blocks. Split COMPUTE-DEBUG-BLOCKS up into several functions for readability.
-
ram authored
always do the FORCE-TN-TO-STACK unless it is :COMPONENT. Formerly, any TNs that weren't :NORMAL weren't forced, including :ENVIRONMENT arguments.
-
ram authored
-
ram authored
assembly level code considers to be the stack top.
-
ram authored
-
ram authored
-
ram authored
-
- Apr 22, 1990
-
-
wlott authored
of a node.
-
wlott authored
Therefore, :component tns will not be removed.
-
wlott authored
of just spinning forever marking the same tn as being on the number stack.
-
wlott authored
move function to get ahold of the current-nfp-tn, and vop-node can be used if the node is really needed.
-
wlott authored
single element of element size. Why bother wrapping, when it's never going to fit?
-
wlott authored
instead of the used templates note being NIL. Also, if the used templates note is NIL, use it's name instead.
-
wlott authored
non :normal tns are still around andhave not reads or writes. Also in check-tn-conflicts, pass ir2-component-component-tns the component-info of component, not component itself.
-
- Apr 21, 1990
-
-
ram authored
the TN-REF-LOAD-TN. Changed EMIT-SAVES to run before PACK-LOAD-TNs. This should have no effect on semantics, but lets the saved registers be used for load TNs without having to spill them again. Probably not a real efficiency issue, but saving the same register twice looks silly.
-
wlott authored
number of blocks. Even if its not NIL, it will be wrong. Use ir2-block-count instead.
-
ram authored
-
ram authored
undefined. Removed a duplicate version of the code that added onto the DD-INCLUDED-BY.
-
- Apr 20, 1990
-
-
ram authored
-
ram authored
-
ram authored
-
ram authored
-
ram authored
-
ram authored
-
chiles authored
-
ram authored
within the enclosed form. This was causing uses of (TRULY-THE NIL ...) in XEPs to have truly bizarre effects (we "proved" that the XEP calls returned NIL, wich allowed us to assume that all non-fixed-arg functions return 1 value.)
-
wlott authored
which included commits to RCS files with non-trunk default branches.
-
wlott authored
set-symbol-function-carefully so that we can use them before we get all the trace stuff up and running.
-
- Apr 19, 1990
-
-
wlott authored
during the instruction you branch to...
-
ram authored
potentially conditionally allocate the NFP.
-
ram authored
NFP.
-
ram authored
-
ram authored
allocated (initial size of 0).
-
ram authored
-
ram authored
return PC. Fixed some lifetimes in DEFINED-FULL-CALL, primarily affecting tail call.
-
ram authored
as an info arg. This way, there is a general handle for getting at info about the callee (such as the NFP.)
-
ram authored
Added CALLEE-NFP-TN and CALLEE-RETURN-PC-TN.
-
ram authored
detecting argument/result references while scanning the NEXT-REF thread had become untenable now that times can explicitly be specified for operands. Fortunately solving this problem simplified the code. Also, changed TARGET-IF-DESIRABLE to always target so that load-TN targeting will work better. Changed COMPUTE-LIVE-TNS to kill any result TNs before returning. Because I stopped advancing to the previous VOP before packing arg load TNs, results appeared to conflict with argument load TNs. We don't need this, since all conflicts for TNs used by the VOP are explicitly tested for in the other code.
-