ANDxx after DOUxx, and two held-out fixed-form probes #677
Loading…
Reference in a new issue
No description provided.
Delete branch "rpg-hx-and-dou-20261031"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
ANDxx after DOUxx, and two held-out fixed-form probes
The last ANDxx of a DOUxx compares and jumps on true like the one of an IFxx,
and then leaves the loop with one more unconditional jump (
17 L; 18 E). TheC line rule took only the four operations of the comparison, so
N DOUGT 50/N ANDLT 100failed with "no rule for a C line". The ANDxxform now accepts the exit jump. EXPERIMENTS/hx-cas-sub (CASGE, CASLE, CASNE,
SUB, MULT, Z-ADD in subroutines, DOUGT with ANDLT) emits 19 of 19 statements
and its text matches the object.
EXPERIMENTS/hx-tags-order is the second probe: TAGs and subroutines whose
labels interleave in name order, a CABGT forward to the last TAG and a GOTO
back to the first. It does not recover yet (the statement of the last TAG
before SETON LR has the operations
27 K; 35 Sand no role); it is aknown_gap row of the loop, not a rule of this change.
Gates