- 12 Dec, 1991 1 commit
-
-
wlott authored
-
- 11 Dec, 1991 21 commits
-
-
ram authored
-
ram authored
type of the result continuation, since the RETURN of functions that only do tail local calls will be deleted.
-
ram authored
-
ram authored
fact, why not always assign one, and just guard the return hackery with a test for whether this function has a return.
-
ram authored
edit.
-
ram authored
-
ram authored
:TYPE BOOLEAN.
-
ram authored
that all blocks have a NUMBER.
-
ram authored
Added an assertion that when we see a tail local call, the caller and callee tail sets had better be the same.
-
ram authored
MAYBE-CONVERT-TO-ASSIGNMENT and MAYBE-CONVERT-TAIL-LOCAL-CALL. Generalized let conversion to handle both assignments and normal LETs. Now local calls that we can prove will be TR are converted to jump directly to the function. We set NODE-TAIL-P to indicate this. RETURN nodes may be deleted because of this. We have to handle let conversion where the caller's RETURN was deleted due to tail recursion. Changed MERGE-LETS to remove the lambda from the tail-set and set the LAMBDA-TAIL-SET to NIL.
-
ram authored
Updated stuff to know that NODE-TAIL-P is no longer a tail set.
-
ram authored
new block which needs a DFN assigned. Added :ASSIGNMENT to various dispatches off of FUNCTIONAL-KIND. In DELETE-LAMBDA, remove the lambda from its tail-set. Whenever we try let conversion, try assignment conversion if it fails. Changed DELETE-RETURN not to delete the function from the tail set. Fixed DELETE-COMPONENT to set FUNCTIONAL-KIND to NIL before calling DELETE-LAMBDA so that we can delete random kinds of lambdas.
-
ram authored
is probably always a bug, but I don't want to use ASSERT until all the problems are fixed. Changed RETURN optimization to call MAYBE-CONVERT-TAIL-CALL. Updated for new tail-set semantics. Changed VALUES IR1 transform and SUBSTITUTE-SINGLE-USE-CONTINUATION to clear out the old NODE-DERIVED-TYPE so that the type doesn't spuriously go to NIL.
-
ram authored
but some may have a return type of NIL.
-
ram authored
:OPTIONAL entries, etc.) Changed to set NODE-TAIL-P to T instead of a tail set.
-
ram authored
out that return nodes can be deleted (by FIND-INITIAL-DFO).
-
ram authored
:ASSIGNMENT to the recognized function kinds.
-
ram authored
to functions in the same environment. Also, eliminated the assertion that NLX entry code doesn't reach the function's bind node. Instead, we just move the bind node to be first if it was reached.
-
ram authored
that is conditionally present, and not the LAMBDA-TAIL-SET.
-
ram authored
-
ram authored
-
- 10 Dec, 1991 1 commit
-
-
ram authored
-
- 09 Dec, 1991 1 commit
-
-
ram authored
'df' to determine whether you have a tmpfs filesystem.
-
- 08 Dec, 1991 7 commits
-
-
ram authored
prevent type inference anomalies.
-
ram authored
to prevent type inference anomalies.
-
ram authored
of trying to "optimize" the case where the destination is an IF. This was gratuitous, and confused type inference.
-
wlott authored
-
wlott authored
Changed DI:DO-BLOCKS to DI:DO-DEBUG-FUNCTION-BLOCKS.
-
wlott authored
-
wlott authored
-
- 07 Dec, 1991 4 commits
- 06 Dec, 1991 5 commits
-
-
wlott authored
-
wlott authored
the list of sections is reversed. Also, check to see if zero is less than or equal the index, not the index less than or equal zero.
-
wlott authored
which included commits to RCS files with non-trunk default branches.
-
wlott authored
so that char-pos will be reset to zero whenever the user presses enter.
-
wlott authored
(cons symbol) to pprint-function-call. This way the results of (list-all-packages) won't show up as a function call. Added TAB support.
-