Repository navigation
Ladder climb L1->L4: 74 burned programs below L4 by estate and rung (coverage backlog after the translate stage) #6072
Description
Why
The translate stage is done: the 6 programs not translated whole are all held for non-translator reasons (#5866 source defect, #5859 assembler, #5106/#5867 owner ruling or z/OS calibration). The evidence ladder's backlog is now coverage: 80 burned programs sit below L4, and nothing in the churn queue targets most of them. #5864 covers only the 6 programs one branch short.
Numbers from the generated evidence reports (main 64a92cfd8, 2026-10-22). Bar: L3 = 100% paragraph coverage, L4 = 100% net branch coverage (a reviewed infeasible outcome in tests/equivalence/infeasible_outcomes.json, #4602, counts as covered).
L1 -> L2 (5): an equivalence case equal on every scenario
- CardDemo COPAUS1C, DBUNLDGS, PAUDBLOD: no case runs them (IMS/DLI: with Siblings of #4893: DL/I calls beyond the first slice (EXEC DLI, UOW calls, GSAM, I/O PCB, CICS) #5233) (CardDemo PAUDBLOD 100% net branches (hex input rule), PAUDBUNL 22/22 (Part of #6072) #6226)
- DBB epscmort, epscsmrt: det port ledgered as differing in det_sweep (mortgage-cmort / mortgage-csmrt): triage the difference first (det sweep: pin the DBB mortgage baseline cause, fix stale EPSMPMT hole text (Part of #6072) #6253)
L2 -> L3 (31): scenarios that reach every paragraph
- CardDemo (8): CBPAUP0C 93.8, PAUDBUNL 90.0, COTRTLIC 89.8, COTRTUPC 98.4, CBSTM03A 92.0, COACTUPC 93.7, COACTVWC 96.9, COBSWAIT (not measured: no fresh det-sweep entry)
- CBSA (12): BNK1DCS 90.2, BNKMENU 76.0, CRDTAGY1-5 57.1 each (one shape: do together), DBCRFUN 72.7, INQACC 88.0, INQACCCU 72.7, INQCUST 88.0, XFRFUN 96.4
- GenApp (6): lgapdb01 85.7, lgtestc1 90.0, lgtestp1-3 70.0, lgtestp4 77.8 (ladder(cics-genapp): lgacdb01-generated 314 true stated infeasible; GenApp already L4 on main (Part of #6072) #6795)
- zECS (3): ZECS000 78.9, ZECS001 31.1, ZECS003 90.5
- DBB (1): epsmlist 83.3
- Out of scope: CBSA CRECUST 35.1 (stays refused: CEEDAYS owner ruling 2026-10-17)
L3 -> L4 (38): scenarios that take every net branch, or a reviewed infeasible outcome
- one branch short: L4 lift: L3 programs one branch short (scenario or reviewed infeasible outcome) #5864 (CBSA DELACC, CUSTCTRL; GenApp lgacdb01, lgipdb01, lgucdb01). CBSA BANKDATA 97.9 is on hold (owner ruling, PR equivalence: CBSA CUSTCTRL 4/4, DELACC and BANKDATA net (Part of #5864) #5958).
- CardDemo (7): COCRDLIC 92.5 net, COPAUS0C 86.0, COPAUA0C 84.1, COBTUPDT 80.0, CSUTLDTC 40.0, CODATE01 15.4, COACCT01 12.9 (ladder(carddemo-copaua0c): 501-request scenario takes the request-limit exit, net 66/71 (Part of #6072) #6791)
- CBSA (9): BNK1UAC 85.2, DELCUS 81.2, BNK1CRA 77.6, BNK1CCS 76.2, BNK1TFN 73.6, BNK1DAC 73.3, BNK1CCA 70.7, CREACC 68.4, BNK1CAC 57.7
- GenApp (15): lgicdb01 91.7, lgucdb01 90.0, lgicvs01 83.3, lgupol01 81.2, lgupdb01 68.4, lgdpol01 66.7, lgipvs01 66.7, lgapol01 62.5, lgucus01 62.5, lgapvs01 55.6, lgupvs01 55.6, lgacus01 50.0, lgicus01 50.0, lgipol01 50.0, lgstsq 50.0, lgdpvs01 0.0 (ladder(cics-genapp): lgacdb01-generated 314 true stated infeasible; GenApp already L4 on main (Part of #6072) #6795)
- zECS (1): ECS001 83.3 (zECS equivalence slice: ZECS003 L4, ECS001 infeasible outcomes, ZECS000 partial (Part of #6072) #6149)
How to slice
One PR per estate per rung, highest current coverage first (cheap lifts before deep ones). Each PR adds scenarios to the program's equivalence case (or reviewed infeasible outcomes, each with its reason), re-runs the det sweep, and states the before/after coverage. Programs sharing a shape (CRDTAGY1-5, lgtestp1-3, the GenApp *vs01 family) go in one slice. Low-coverage programs (CODATE01, COACCT01, lgdpvs01 0.0, ZECS001) may be scenario-generation gaps: if one needs harness work, file it as a sibling issue and move on, don't stall the slice.
Burned estates only. Evidence levels, not "proven"; oracle is GnuCOBOL + IBM documentation.
Activity
squid-protocol commentedon Oct 22, 2026
Churn run
6072-20261022T2207Z(host melek) ended without a PR. plan approved; overlap is regenerated evidence.jsonsquid-protocol commentedon Oct 22, 2026
Churn run
6072-20261022T2217Z(host melek) ended without a PR. superseded: sent back with pin approval; new run 6072-20261022T2250Z runningacctview port: ABEND-DATA SEND is a plain SEND, not SEND TEXT (ERASE …
acctview: re-run the COACTVWC mutation score on the new port; proof_r…
det-sweep coverage ledger: refresh cbstm03a and acctview-generated af…
CardDemo L2->L3 slice 1: cover unrun live paragraphs (Part of #6072) (#…
COCRDLIC: six PF7/PF8 scenarios reach 11 of its 13 uncovered branch o…
squid-protocol commentedon Oct 23, 2026
#6144 merged: a partial slice of [c:3d9f3444] "CardDemo (7): COCRDLIC 92.5 net, COPAUS0C 86.0, COPAUA0C 84.1, COBTUPDT 80.0, CSUTLDTC 40.0, CODATE01 15.4, COACCT01 12.9"; left: CSUTLDTC / CODATE01 / COACCT01 not measured; remaining ENTER-form scenarios stay X8-refused. (promoter-bot)
GenApp L2->L3 slice 3: LGAPDB01 endowment add and short-COMMAREA scen…
squid-protocol commentedon Oct 23, 2026
#6142 merged: a partial slice of [c:a51f514c] "GenApp (6): lgapdb01 85.7, lgtestc1 90.0, lgtestp1-3 70.0, lgtestp4 77.8"; left: lgtest* arms land with #5900; re-measure after it merges (promoter-bot)
CBSA L3->L4 slice 3: DELCUS, BNK1UAC, BNK1CRA fault and input scenari…
squid-protocol commentedon Oct 23, 2026
#6143 merged: a partial slice of [c:3c7856f6] "CBSA (9): BNK1UAC 85.2, DELCUS 81.2, BNK1CRA 77.6, BNK1CCS 76.2, BNK1TFN 73.6, BNK1DAC 73.3, BNK1CCA 70.7, CREACC 68.4, BNK1CAC 57.7"; left: DELCUS done in its main program; BNK1UAC and BNK1CRA main programs have open arms listed above; BNK1CCS, BNK1TFN, BNK1DAC, BNK1CCA, CREACC, BNK1CAC not started. (promoter-bot)
230 remaining items
squid-protocol commentedon Oct 27, 2026
#6789 merged: a partial slice of [c:3d9f3444] "CardDemo (7): COCRDLIC 92.5 net, COPAUS0C 86.0, COPAUA0C 84.1, COBTUPDT 80.0, CSUTLDTC 40.0, CODATE01 15.4, COACCT01 12.9"; left: COPAUA0C 339:true still needs the 501-request scenario (this PR is the runtime fix only) (promoter-bot)
ladder(carddemo-copaua0c): 501-request scenario takes the request-lim…
Oracle stub: WEB OPEN sessions and stated responses grow past 64 / 25…
squid-protocol commentedon Oct 27, 2026
Churn run
6072-20261027T1251Z(host melek) ended without a PR. Regraft DONE: #6777 merged to main (72af6bf) while regraft ran; duplicate #6794 closed. zECS [c:6fa7e863] still partial; #6780 unblocked by #6777 merge.squid-protocol commentedon Oct 27, 2026
Churn run
6072-20261027T1322Z(host melek) ended without a PR. plan-held: real overlap on carddemo-copaua0c/case.json with live worker 5278; not approved, ladder items remain queued, retry after 5278's PRsquid-protocol commentedon Oct 27, 2026
Churn run
6072-20261027T1330Z(host melek) ended without a PR. No PR: CBSA slice premise false; all remaining open outcomes blocked by harness/oracle limits (abend injection, UCTRAN, X54, DEEDIT, X8, P118). Sibling gaps listed in report.squid-protocol commentedon Oct 27, 2026
Churn run
6072-20261027T1337Z(host melek) ended without a PR. cics-crucible not a burned estate; no PR. Owner decision: burn it as a case-bearing estate or drop from #6072 list; tool route would be a #6271 siblingsquid-protocol commentedon Oct 27, 2026
Churn run
6072-20261027T1339Z(host melek) ended without a PR. DBB mortgage slice blocked on #5867 (EPSMPMT hole); no PRladder(cics-genapp): lgacdb01-generated states the 314 true outcome i…
ladder:zecs - refresh ZECS000 coverage ledger after merging main (#6777…
ladder:zecs - ZECS000 expired-501-peer-201 scenario, 48/48 net branch…
squid-protocol commentedon Oct 27, 2026
#6780 merged: a partial slice of [c:6fa7e863] "zECS (3): ZECS000 78.9, ZECS001 31.1, ZECS003 90.5"; left: ZECS000 now complete; ZECS001 31.1 still blocked by P117 / #6271 (promoter-bot)
ladder(cics-genapp): ledger lgicdb01's three facade failures under #6692
ladder(cics-genapp): drop lgicdb01 from this slice (its model port ne…
ladder(cics-genapp): three Db2 twins reach every net branch outcome o…