GOTO, and CABxx before a TAG #672
Loading…
Reference in a new issue
No description provided.
Delete branch "rpg-l8-goto-20261029"
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?
GOTO, and CABxx before a TAG
FLOW/goto-subroutine has
N CABLT 3 AGAINback to a TAG andGOTO DONEoutof a subroutine. A CABxx statement may end with the labels of the statement
after it (
DONE TAGstarts there), and its target must start a TAG. GOTO isone jump to a TAG label, or, out of a subroutine, the reset of the return
address of that subroutine (a constant of kind 0x18 and class 0x12 stored into
its slot) before the jump:
27 K; 35 S254; 18 L. A TAG is the target ofeither, and its label is no other kind: no conditional jump takes it (the end
of an IF or a loop), no subroutine starts there (a BEGSR has no code either),
the statement there is a source statement, and the jump does not follow a
reset of the statement status (an ENDIF that carries the jump of an ELSE,
FLOW/nested-exsr statement 10). The probes that move a jump to another label
(rpg_flow_probe_test, rpg_exits_test) keep refusing it. The emitter writes
GOTO Tnnnnnas a C line, and a CABxx or GOTO lineonly when a TAG line has its name. The text of FLOW/goto-subroutine matched its
object on the host (calibration J, captures/calJ-20261029).
Gates