[Imprint][Data Privacy Statement][E-Mail]



View long theorem info. View the proof. Back to main index. View the specification. View theorem base summary, View theorem base listing, the used properties by spec, the dependency list, the symbol table,

Proof step 1 from proof Crash-Theorem (conclusion) (back to tree)

DMSQGInv(gs), pcf = pcf0, lsf = lsf0, gs = gs0 ⊦ ⟪ps := gs.ps, vs := gs.vs, phead := gs.phead, vhead := gs.vhead, vtail := gs.vtail, vretRefs := gs.vretRefs, pretRefs := gs.pretRefs, vrvHp := gs.vrvHp, prvHp := gs.prvHp, owns := gs.owns; CRASH(ps, phead, pretRefs, prvHp; vs, vhead, vtail, vretRefs, vrvHp, owns, pcf); gs := (((((gs.vs:= vs).vhead:= vhead).vtail:= vtail).vretRefs:= vretRefs).vrvHp:= vrvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs0.ps ∧ gs.vs = gs0.ps ∧ gs.phead = gs0.phead ∧ gs.vretRefs = gs0.pretRefs ∧ gs.vrvHp = gs0.prvHp ∧ gs.qrl = gs0.qrl ∧ gs.orl = gs0.orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. pcf(tid) = (pcf0(tid) = idle ⊃ idle;crashed) ∧ lsf(tid).tid = lsf0(tid).tid))

System applied rule normalize

And obtained the following new goals

Rule argument

Additional information

Proof step 2 from proof Crash-Theorem (previous step) (back to tree)

gs = gs0, lsf = lsf0, pcf = pcf0, DMSQGInv(gs) ⊦ ⟪ps := gs.ps, vs := gs.vs, phead := gs.phead, vhead := gs.vhead, vtail := gs.vtail, vretRefs := gs.vretRefs, pretRefs := gs.pretRefs, vrvHp := gs.vrvHp, prvHp := gs.prvHp, owns := gs.owns; CRASH(ps, phead, pretRefs, prvHp; vs, vhead, vtail, vretRefs, vrvHp, owns, pcf); gs := (((((gs.vs:= vs).vhead:= vhead).vtail:= vtail).vretRefs:= vretRefs).vrvHp:= vrvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs0.ps ∧ gs.vs = gs0.ps ∧ gs.phead = gs0.phead ∧ gs.vretRefs = gs0.pretRefs ∧ gs.vrvHp = gs0.prvHp ∧ gs.qrl = gs0.qrl ∧ gs.orl = gs0.orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. pcf(tid) = (pcf0(tid) = idle ⊃ idle;crashed) ∧ lsf(tid).tid = lsf0(tid).tid))

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 3 from proof Crash-Theorem (previous step) (back to tree)

gs = gs0, pcf = pcf0, DMSQGInv(gs0) ⊦ ⟪ps := gs.ps, vs := gs.vs, phead := gs.phead, vhead := gs.vhead, vtail := gs.vtail, vretRefs := gs.vretRefs, pretRefs := gs.pretRefs, vrvHp := gs.vrvHp, prvHp := gs.prvHp, owns := gs.owns; CRASH(ps, phead, pretRefs, prvHp; vs, vhead, vtail, vretRefs, vrvHp, owns, pcf); gs := (((((gs.vs:= vs).vhead:= vhead).vtail:= vtail).vretRefs:= vretRefs).vrvHp:= vrvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs0.ps ∧ gs.vs = gs0.ps ∧ gs.phead = gs0.phead ∧ gs.vretRefs = gs0.pretRefs ∧ gs.vrvHp = gs0.prvHp ∧ gs.qrl = gs0.qrl ∧ gs.orl = gs0.orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. pcf(tid) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 4 from proof Crash-Theorem (previous step) (back to tree)

gs = gs0, owns = gs.owns, pcf = pcf0, phead = gs.phead, pretRefs = gs.pretRefs, prvHp = gs.prvHp, ps = gs.ps, vhead = gs.vhead, vretRefs = gs.vretRefs, vrvHp = gs.vrvHp, vs = gs.vs, vtail = gs.vtail, DMSQGInv(gs0) ⊦ ⟪CRASH(ps, phead, pretRefs, prvHp; vs, vhead, vtail, vretRefs, vrvHp, owns, pcf); gs := (((((gs.vs:= vs).vhead:= vhead).vtail:= vtail).vretRefs:= vretRefs).vrvHp:= vrvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs0.ps ∧ gs.vs = gs0.ps ∧ gs.phead = gs0.phead ∧ gs.vretRefs = gs0.pretRefs ∧ gs.vrvHp = gs0.prvHp ∧ gs.qrl = gs0.qrl ∧ gs.orl = gs0.orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. pcf(tid) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 5 from proof Crash-Theorem (previous step) (back to tree)

gs = gs0, owns = gs0.owns, pcf = pcf0, vhead = gs0.vhead, vretRefs = gs0.vretRefs, vrvHp = gs0.vrvHp, vs = gs0.vs, vtail = gs0.vtail, DMSQGInv(gs0) ⊦ ⟪CRASH(gs0.ps, gs0.phead, gs0.pretRefs, gs0.prvHp; vs, vhead, vtail, vretRefs, vrvHp, owns, pcf); gs := (((((gs.vs:= vs).vhead:= vhead).vtail:= vtail).vretRefs:= vretRefs).vrvHp:= vrvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs0.ps ∧ gs.vs = gs0.ps ∧ gs.phead = gs0.phead ∧ gs.vretRefs = gs0.pretRefs ∧ gs.vrvHp = gs0.prvHp ∧ gs.qrl = gs0.qrl ∧ gs.orl = gs0.orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. pcf(tid) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic elimination applied rule apply elim lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:elim-mkgs, seq:⊦ ( ( ( ( ( ( ( ( ( ( (ps = gs.ps ∧ vs = gs.vs) ∧ phead = gs.phead) ∧ vhead = gs.vhead) ∧ vtail = gs.vtail) ∧ vretRefs = gs.vretRefs) ∧ pretRefs = gs.pretRefs) ∧ vrvHp = gs.vrvHp) ∧ prvHp = gs.prvHp) ∧ qrl = gs.qrl) ∧ orl = gs.orl) ∧ owns = gs.owns) ∧ rvowns = gs.rvowns ↔ gs = mkgs (ps, vs, phead, vhead, vtail, vretRefs, pretRefs, vrvHp, prvHp, qrl, orl, owns, rvowns), subst:[qrl, ps, phead, vrvHp, vretRefs, vhead, vtail, gs, owns, rvowns, orl, prvHp, pretRefs, vs] → [qrl, ps, phead, vrvHp0, vretRefs0, vhead0, vtail0, gs0, owns0, rvowns, orl, prvHp, pretRefs, vs0]

Additional information

Used simplifier rules: ⊦ ( ( ( ( ( ( ( ( ( ((ps = gs.ps ∧ vs = gs.vs) ∧ phead = gs.phead) ∧ vhead = gs.vhead) ∧ vtail = gs.vtail) ∧ vretRefs = gs.vretRefs) ∧ pretRefs = gs.pretRefs) ∧ vrvHp = gs.vrvHp) ∧ prvHp = gs.prvHp) ∧ qrl = gs.qrl) ∧ orl = gs.orl) ∧ owns = gs.owns) ∧ rvowns = gs.rvowns ↔ gs = mkgs (ps, vs, phead, vhead, vtail, vretRefs, pretRefs, vrvHp, prvHp, qrl, orl, owns, rvowns); (elim) Used terms: qrl, ps, phead, vrvHp0, vretRefs0, vhead0, vtail0, gs0, owns0, rvowns, orl, prvHp, pretRefs, vs0

Proof step 6 from proof Crash-Theorem (previous step) (back to tree)

gs = gs0, gs0 = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), owns = owns0, pcf = pcf0, vhead = vhead0, vretRefs = vretRefs0, vrvHp = vrvHp0, vs = vs0, vtail = vtail0, DMSQGInv(gs0) ⊦ ⟪CRASH(ps, phead, pretRefs, prvHp; vs, vhead, vtail, vretRefs, vrvHp, owns, pcf); gs := (((((gs.vs:= vs).vhead:= vhead).vtail:= vtail).vretRefs:= vretRefs).vrvHp:= vrvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = ps ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. pcf(tid) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 7 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), pcf = pcf0, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ ⟪CRASH(ps, phead, pretRefs, prvHp; vs0, vhead0, vtail0, vretRefs0, vrvHp0, owns0, pcf); gs := (((((gs.vs:= vs0).vhead:= vhead0).vtail:= vtail0).vretRefs:= vretRefs0).vrvHp:= vrvHp0).owns:= owns0⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. pcf(tid) = (pcf0(tid) = idle ⊃ idle;crashed)))

User applied rule call right

And obtained the following new goals

Rule argument

pos: right 1

Additional information

Used simplifier rules: CRASH(ps, phead, pretRefs, prvHp; vs, vhead, vtail, vretRefs, vrvHp, owns, pcf) {pcf := λ tid. (pcf(tid) = idle ⊃ idle;crashed); vretRefs := pretRefs; vrvHp := prvHp; vs := ps; vhead := phead; owns := λ r. None; while ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None do vhead := ps[vhead].next; vtail := vhead; while ps[vtail].next ≠ null do vtail := ps[vtail].next };

Proof step 8 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), owns = owns0, pcf = pcf0, vhead = vhead0, vretRefs = vretRefs0, vrvHp = vrvHp0, vs = vs0, vtail = vtail0, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ ⟪pcf := λ tid. (pcf(tid) = idle ⊃ idle;crashed); vretRefs := pretRefs; vrvHp := prvHp; vs := ps; vhead := phead; owns := λ r. None; while ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None do vhead := ps[vhead].next; vtail := vhead; while ps[vtail].next ≠ null do vtail := ps[vtail].next; gs := (((((gs.vs:= vs).vhead:= vhead).vtail:= vtail).vretRefs:= vretRefs).vrvHp:= vrvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. pcf(tid) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 9 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), owns = owns0, pcf = pcf0, vhead = vhead0, vretRefs = vretRefs0, vrvHp = vrvHp0, vs = vs0, vtail = vtail0, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ ⟪vretRefs := pretRefs; vrvHp := prvHp; vs := ps; vhead := phead; owns := λ r. None; while ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None do vhead := ps[vhead].next; vtail := vhead; while ps[vtail].next ≠ null do vtail := ps[vtail].next; gs := (((((gs.vs:= vs).vhead:= vhead).vtail:= vtail).vretRefs:= vretRefs).vrvHp:= vrvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. (pcf(tid) = idle ⊃ idle;crashed) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 10 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), owns = owns0, pcf = pcf0, vhead = vhead0, vretRefs = vretRefs0, vrvHp = vrvHp0, vs = vs0, vtail = vtail0, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ ⟪vrvHp := prvHp; vs := ps; vhead := phead; owns := λ r. None; while ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None do vhead := ps[vhead].next; vtail := vhead; while ps[vtail].next ≠ null do vtail := ps[vtail].next; gs := (((((gs.vs:= vs).vhead:= vhead).vtail:= vtail).vretRefs:= pretRefs).vrvHp:= vrvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. (pcf(tid) = idle ⊃ idle;crashed) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 11 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), owns = owns0, pcf = pcf0, vhead = vhead0, vretRefs = vretRefs0, vrvHp = vrvHp0, vs = vs0, vtail = vtail0, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ ⟪vs := ps; vhead := phead; owns := λ r. None; while ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None do vhead := ps[vhead].next; vtail := vhead; while ps[vtail].next ≠ null do vtail := ps[vtail].next; gs := (((((gs.vs:= vs).vhead:= vhead).vtail:= vtail).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. (pcf(tid) = idle ⊃ idle;crashed) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 12 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), owns = owns0, pcf = pcf0, vhead = vhead0, vretRefs = vretRefs0, vrvHp = vrvHp0, vs = vs0, vtail = vtail0, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ ⟪vhead := phead; owns := λ r. None; while ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None do vhead := ps[vhead].next; vtail := vhead; while ps[vtail].next ≠ null do vtail := ps[vtail].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. (pcf(tid) = idle ⊃ idle;crashed) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 13 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), owns = owns0, pcf = pcf0, vhead = phead, vhead1 = vhead0, vretRefs = vretRefs0, vrvHp = vrvHp0, vs = vs0, vtail = vtail0, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ ⟪owns := λ r. None; while ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None do vhead := ps[vhead].next; vtail := vhead; while ps[vtail].next ≠ null do vtail := ps[vtail].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= owns⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. (pcf(tid) = idle ⊃ idle;crashed) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 14 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), owns = owns0, pcf = pcf0, vhead = phead, vhead1 = vhead0, vretRefs = vretRefs0, vrvHp = vrvHp0, vs = vs0, vtail = vtail0, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ ⟪while ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None do vhead := ps[vhead].next; vtail := vhead; while ps[vtail].next ≠ null do vtail := ps[vtail].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last ∧ (∀ tid. (pcf(tid) = idle ⊃ idle;crashed) = (pcf0(tid) = idle ⊃ idle;crashed)))

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 15 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead = phead, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ ⟪while ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None do vhead := ps[vhead].next; vtail0 := vhead; while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last)

User applied rule invariant right

And obtained the following new goals

Rule argument

inv: vhead ∈ orl ∧ vhead ∈ ps ∧ vhead ≠ null ∧ (ps[vhead].next ≠ null → ps[vhead].next ∈ ps) fmapos: right 1 bound: #(orl from vhead)

Additional information

Used formulas: vhead ∈ orl ∧ vhead ∈ ps ∧ vhead ≠ null ∧ (ps[vhead].next ≠ null → ps[vhead].next ∈ ps), #(orl from vhead) = #(orl from vhead)

Proof step 16 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead = phead, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ vhead ∈ orl ∧ vhead ∈ ps ∧ vhead ≠ null ∧ (ps[vhead].next ≠ null → ps[vhead].next ∈ ps)

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 17 from proof Crash-Theorem (previous step) (back to tree)

DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)) ⊦ phead ∈ orl ∧ phead ∈ ps ∧ phead ≠ null ∧ (ps[phead].next ≠ null → ps[phead].next ∈ ps)

User applied rule apply rewrite lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:DMSQGInv, seq:⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed (gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null), subst:[gs] → [mkgs (ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)], <all paths>

Additional information

Used terms: mkgs (ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)

Proof step 18 from proof Crash-Theorem (previous step) (back to tree)

⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed(gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null)

A spec-lemma

Proof step 19 from proof Crash-Theorem (previous step) (back to tree)

dom mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp = dom mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl ≠ [], allocrv (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp), allocrv (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp), listp ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).owns), markedp (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps), markedp (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs), nonullRetRefs ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs), rvownerok (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns), rvownerok (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns), unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps), mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).phead ∈ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, ¬ dups ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl), mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ≠ null ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID ≠ None → mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID .tid)) = Claimed (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID .tid, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .val), mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.last = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ∧ ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ≠ [] → unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.tail, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs)) ∨ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.butlast ≠ [] ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.butlast.last = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).phead ≠ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.last ∧ unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs), mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vtail = ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ).last ∨ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ≠ [] ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vtail = ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.butlast ).last ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.last ] .deqID = None ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ).last ] .next = null ⊦ phead ∈ orl ∧ phead ∈ ps ∧ phead ≠ null ∧ (ps[phead].next ≠ null → ps[phead].next ∈ ps)

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] list-elem-in-app11: ⊦ listp(orl + qrl, vs, ps, owns) ∧ node ∈ orl → node ∈ vs; (forward) from specification DMSQ list-elem-in-app12: ⊦ listp(orl + qrl, vs, ps, owns) ∧ node ∈ orl → node ∈ ps; (forward) from specification DMSQ ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vhead = phead0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vs = ps0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .rvowns = rvowns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vretRefs = vretRefs; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .qrl = qrl; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .orl = qrl0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .phead = phead; (s) ⊦ y ≠ [] → (x + y).last = y.last; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vtail = phead1; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vrvHp = vrvHp; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .prvHp = vrvHp0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .pretRefs = vretRefs0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .owns = owns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .ps = ps; (s)

Proof step 20 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, orl ≠ [], allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(orl + qrl, vs0, ps, owns0), markedp(orl, ps), markedp(orl, vs0), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, phead ∈ orl, ¬ dups(qrl), ¬ dups(orl), ¬ dups(orl + qrl), vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0), vtail0 = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ phead ≠ null ∧ (ps[phead].next ≠ null → ps[phead].next ∈ ps)

User applied rule apply lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:listp-in-postfix1, seq:⊦ listp(rl + rl0, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2 + rl0, vs, ps, owns)), subst:[ps, rl, owns, r, rl0, vs] → [ps, orl, owns0, phead, qrl, vs0]

Additional information

Used terms: ps, orl, owns0, phead, qrl, vs0

Proof step 21 from proof Crash-Theorem (previous step) (back to tree)

⊦ listp(rl + rl0, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2 + rl0, vs, ps, owns))

A spec-lemma

Proof step 22 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, orl ≠ [], allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(orl + qrl, vs0, ps, owns0), markedp(orl, vs0), markedp(orl, ps), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), unmarkedp(qrl, ps), phead ∈ orl, phead ∈ vs0, phead ∈ ps, ¬ dups(qrl), ¬ dups(orl), ¬ dups(orl + qrl), vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), listp(orl + qrl, vs0, ps, owns0) ∧ phead ∈ orl → (∃ rl1, rl2. orl = rl1 + phead ' + rl2 ∧ listp(phead ' + rl2 + qrl, vs0, ps, owns0)), vtail0 = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0) ⊦ phead ≠ null ∧ (ps[phead].next ≠ null → ps[phead].next ∈ ps)

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ ¬ listp(rl + null ' + rl0, vs, ps, owns); (s) ⊦ (x + a ' + y).last = (a ' + y).last; (s) ⊦ (x + a ' + y).butlast = x + (a ' + y).butlast; (s) ⊦ ¬ a ∈ x ∧ ¬ a ∈ y → (dups(x + a ' + y) ↔ dups(x + y)); (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ x + a ' + y ≠ []; (s) ⊦ x = y + a ' + z → a ∈ x; (s)

Proof step 23 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, ps[phead].next ≠ null, phead ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((rl1 + phead ' + rl2) + qrl, vs0, ps, owns0), listp(phead ' + rl2 + qrl, vs0, ps, owns0), markedp(rl2, ps), markedp(rl1, ps), markedp(rl2, vs0), markedp(rl1, vs0), nonullRetRefs((rl1 + phead ' + rl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[phead].deqID), some?(vs0[phead].deqID), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, ¬ dups(rl1 + rl2), ¬ dups(rl2 + qrl), ¬ dups(rl1), ¬ dups(rl1 + rl2 + qrl), ¬ dups(rl2), ¬ dups(qrl), ¬ ps[phead].next ∈ ps, ¬ phead ∈ qrl, ¬ phead ∈ rl2, ¬ phead ∈ rl1, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), (phead ' + rl2).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ rl1 + (phead ' + rl2).butlast ≠ [] ∧ (rl1 + (phead ' + rl2).butlast).last = vhead0 ∧ phead ≠ (phead ' + rl2).last ∧ unmarkedp(qrl, vs0), vtail0 = (phead ' + rl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (phead ' + rl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

User applied rule constructor cut

And obtained the following new goals

Rule argument

var: rl2

Additional information

Used simplifier rules: reflist freely generated by [], +;;

Proof step 24 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, rl2 = [], ps[phead].next ≠ null, phead ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(phead ' + rl2 + qrl, vs0, ps, owns0), listp((rl1 + phead ' + rl2) + qrl, vs0, ps, owns0), markedp(rl1, vs0), markedp(rl2, vs0), markedp(rl1, ps), markedp(rl2, ps), nonullRetRefs((rl1 + phead ' + rl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[phead].deqID), some?(ps[phead].deqID), unmarkedp(qrl, ps), phead ∈ vs0, phead ∈ ps, ¬ dups(rl1 + rl2), ¬ dups(rl2 + qrl), ¬ dups(rl1), ¬ dups(rl1 + rl2 + qrl), ¬ dups(rl2), ¬ dups(qrl), ¬ ps[phead].next ∈ ps, ¬ phead ∈ qrl, ¬ phead ∈ rl2, ¬ phead ∈ rl1, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail0 = (phead ' + rl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (phead ' + rl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, (phead ' + rl2).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ rl1 + (phead ' + rl2).butlast ≠ [] ∧ (rl1 + (phead ' + rl2).butlast).last = vhead0 ∧ phead ≠ (phead ' + rl2).last ∧ unmarkedp(qrl, vs0) ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ x ≠ [] → (a ' + x).last = x.last; (s) ⊦ some?(opttid) → opttid ≠ None; (as) ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s) ⊦ x + [] = x; (s) ⊦ unmarkedp([], vs); (s) a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ [] + x = x; (s) ⊦ a '.last = a; (s) ⊦ a '.butlast = []; (s) ⊦ ¬ a ∈ []; (s) ⊦ ¬ dups([]); (s) ⊦ markedp([], vs); (s)

Proof step 25 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, ps[vhead0].next ≠ null, qrl ≠ [], vhead0 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vhead0 ' + qrl, vs0, ps, owns0), listp(rl1 + vhead0 ' + qrl, vs0, ps, owns0), markedp(rl1, vs0), markedp(rl1, ps), nonullRetRefs(rl1 + vhead0 ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead0].deqID), some?(ps[vhead0].deqID), unmarkedp(qrl, ps), unmarkedp(qrl.tail, vs0), vhead0 ∈ vs0, vhead0 ∈ ps, ¬ dups(qrl), ¬ dups(rl1), ¬ dups(rl1 + qrl), ¬ vhead0 ∈ rl1, ¬ vhead0 ∈ qrl, ¬ ps[vhead0].next ∈ ps, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail0 = qrl.last ∨ vtail0 = (vhead0 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic elimination applied rule apply elim lemma

And obtained the following new goals

Rule argument

spec: list-data[reflist], name:elim-add, seq: ⊦ x ≠ [] → (head = x.head ∧ tail = x.tail ↔ x = head + tail), subst:[x, head, tail] → [qrl, vtail, rl]

Additional information

Used simplifier rules: ⊦ x ≠ [] → (head = x.head ∧ tail = x.tail ↔ x = head + tail); (elim) Used terms: qrl, head, rl

Proof step 26 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, qrl = vtail + rl, ps[vhead0].next ≠ null, qrl ≠ [], vhead0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(rl1 + vhead0 ' + qrl, vs0, ps, owns0), listp(vhead0 ' + qrl, vs0, ps, owns0), markedp(rl1, ps), markedp(rl1, vs0), nonullRetRefs(rl1 + vhead0 ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead0].deqID), some?(vs0[vhead0].deqID), unmarkedp(rl, vs0), unmarkedp(qrl, ps), vhead0 ∈ ps, vhead0 ∈ vs0, ¬ dups(qrl), ¬ dups(rl1), ¬ dups(rl1 + qrl), ¬ vhead0 ∈ rl1, ¬ vhead0 ∈ qrl, ¬ ps[vhead0].next ∈ ps, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail0 = qrl.last ∨ vtail0 = (vhead0 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ ¬ a ∈ st → st[a] = ∅[a]; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] listp-twoin: ⊦ listp(rl + r ' + r0 ' + rl0, vs, ps, owns) → ps[r].next = r0; (forward) from specification DMSQ a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a + x = a ' + x; (s) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ some?(opttid) → opttid ≠ None; (as) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ a ' + x ≠ []; (s)

Proof step 27 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead0].deqID = vs0[vhead0].deqID, ∅[vs0[vhead0].next].deqID = None, ps[vhead0].next = vs0[vhead0].next, ps[vhead0].val = vs0[vhead0].val, dom vrvHp0 = dom prvHp, owns0(vhead0) = None, vs0[vhead0].next ≠ null, vhead0 ≠ null, vhead0 ≠ vs0[vhead0].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vhead0].next ' + rl, vs0, ps, owns0), listp(rl1 + vhead0 ' + vs0[vhead0].next ' + rl, vs0, ps, owns0), markedp(rl1, vs0), markedp(rl1, ps), nonullRetRefs(rl1 + vhead0 ' + vs0[vhead0].next ' + rl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead0].deqID), unmarkedp(rl, ps), unmarkedp(rl, vs0), vhead0 ∈ vs0, vhead0 ∈ ps, ¬ dups(rl1 + rl), ¬ dups(rl1), ¬ dups(rl), ¬ weak?(vs0[vhead0].deqID), ¬ vs0[vhead0].next ∈ ps, ¬ vs0[vhead0].next ∈ rl, ¬ vs0[vhead0].next ∈ rl1, ¬ vhead0 ∈ rl1, ¬ vhead0 ∈ rl, vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail0 = (vs0[vhead0].next ' + rl).last ∨ vtail0 = (vhead0 ' + (vs0[vhead0].next ' + rl).butlast).last ∧ vs0[(vs0[vhead0].next ' + rl).last].deqID = None ∧ vs0[(vs0[vhead0].next ' + rl).last].next = null ⊦

User applied rule constructor cut

And closed the goal!

Rule argument

var: rl

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ unmarkedp([], vs); (s) ⊦ x + [] = x; (s) ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s)

Proof step 28 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, rl2 = phead0 + rl, ps[phead].next ≠ null, phead ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(phead ' + rl2 + qrl, vs0, ps, owns0), listp((rl1 + phead ' + rl2) + qrl, vs0, ps, owns0), markedp(rl1, vs0), markedp(rl2, vs0), markedp(rl1, ps), markedp(rl2, ps), nonullRetRefs((rl1 + phead ' + rl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[phead].deqID), some?(ps[phead].deqID), unmarkedp(qrl, ps), phead ∈ vs0, phead ∈ ps, ¬ dups(rl1 + rl2), ¬ dups(rl2 + qrl), ¬ dups(rl1), ¬ dups(rl1 + rl2 + qrl), ¬ dups(rl2), ¬ dups(qrl), ¬ ps[phead].next ∈ ps, ¬ phead ∈ qrl, ¬ phead ∈ rl2, ¬ phead ∈ rl1, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail0 = (phead ' + rl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (phead ' + rl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, (phead ' + rl2).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ rl1 + (phead ' + rl2).butlast ≠ [] ∧ (rl1 + (phead ' + rl2).butlast).last = vhead0 ∧ phead ≠ (phead ' + rl2).last ∧ unmarkedp(qrl, vs0) ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ ¬ a ∈ st → st[a] = ∅[a]; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] listp-twoin: ⊦ listp(rl + r ' + r0 ' + rl0, vs, ps, owns) → ps[r].next = r0; (forward) from specification DMSQ a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ (b ' + a ' + x).last = (a ' + x).last; (s) ⊦ a + x = a ' + x; (s) ⊦ (x + a ' + y).last = (a ' + y).last; (s) ⊦ (a ' + b ' + x).butlast = a ' + (b ' + x).butlast; (s) ⊦ x + a ' + y ≠ []; (s) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ some?(opttid) → opttid ≠ None; (as) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s)

Proof step 29 from proof Crash-Theorem (previous step) (back to tree)

ps[phead].deqID = vs0[phead].deqID, ps[phead].next = vs0[phead].next, ps[phead].val = vs0[phead].val, dom vrvHp0 = dom prvHp, owns0(phead) = None, vs0[phead].next ≠ null, phead ≠ null, phead ≠ vs0[phead].next, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(rl1 + phead ' + (vs0[phead].next ' + rl) + qrl, vs0, ps, owns0), listp(vs0[phead].next ' + rl + qrl, vs0, ps, owns0), markedp(rl, ps), markedp(rl1, ps), markedp(rl, vs0), markedp(rl1, vs0), nonullRetRefs(rl1 + phead ' + (vs0[phead].next ' + rl) + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vs0[phead].next].deqID), some?(vs0[phead].deqID), some?(∅[vs0[phead].next].deqID), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, ¬ dups(rl1 + rl + qrl), ¬ dups(rl1), ¬ dups(rl + qrl), ¬ dups(rl1 + rl), ¬ dups(qrl), ¬ dups(rl), ¬ weak?(vs0[phead].deqID), ¬ phead ∈ rl1, ¬ vs0[phead].next ∈ rl1, ¬ vs0[phead].next ∈ rl, ¬ vs0[phead].next ∈ qrl, ¬ vs0[phead].next ∈ ps, ¬ phead ∈ qrl, ¬ phead ∈ rl, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), (vs0[phead].next ' + rl).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ (phead ' + (vs0[phead].next ' + rl).butlast).last = vhead0 ∧ phead ≠ (vs0[phead].next ' + rl).last ∧ unmarkedp(qrl, vs0), vtail0 = (vs0[phead].next ' + rl + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[phead].next ' + rl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

User applied rule constructor cut

And obtained the following new goals

Rule argument

var: rl

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist]

Proof step 30 from proof Crash-Theorem (previous step) (back to tree)

ps[phead].deqID = vs0[phead].deqID, ps[phead].next = vs0[phead].next, ps[phead].val = vs0[phead].val, dom vrvHp0 = dom prvHp, owns0(phead) = None, rl = [], vs0[phead].next ≠ null, phead ≠ null, phead ≠ vs0[phead].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[phead].next ' + rl + qrl, vs0, ps, owns0), listp(rl1 + phead ' + (vs0[phead].next ' + rl) + qrl, vs0, ps, owns0), markedp(rl1, vs0), markedp(rl, vs0), markedp(rl1, ps), markedp(rl, ps), nonullRetRefs(rl1 + phead ' + (vs0[phead].next ' + rl) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(∅[vs0[phead].next].deqID), some?(vs0[phead].deqID), some?(vs0[vs0[phead].next].deqID), unmarkedp(qrl, ps), phead ∈ vs0, phead ∈ ps, ¬ dups(rl1 + rl + qrl), ¬ dups(rl1), ¬ dups(rl + qrl), ¬ dups(rl1 + rl), ¬ dups(qrl), ¬ dups(rl), ¬ weak?(vs0[phead].deqID), ¬ phead ∈ rl1, ¬ vs0[phead].next ∈ rl1, ¬ vs0[phead].next ∈ rl, ¬ vs0[phead].next ∈ qrl, ¬ vs0[phead].next ∈ ps, ¬ phead ∈ qrl, ¬ phead ∈ rl, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail0 = (vs0[phead].next ' + rl + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[phead].next ' + rl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, (vs0[phead].next ' + rl).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ (phead ' + (vs0[phead].next ' + rl).butlast).last = vhead0 ∧ phead ≠ (vs0[phead].next ' + rl).last ∧ unmarkedp(qrl, vs0) ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ [] + x = x; (s) ⊦ a '.last = a; (s) ⊦ x + [] = x; (s) ⊦ a '.butlast = []; (s) ⊦ ¬ a ∈ []; (s) ⊦ ¬ dups([]); (s) ⊦ markedp([], vs); (s)

Proof step 31 from proof Crash-Theorem (previous step) (back to tree)

ps[phead].deqID = vs0[phead].deqID, ps[phead].next = vs0[phead].next, ps[phead].val = vs0[phead].val, dom vrvHp0 = dom prvHp, owns0(phead) = None, vs0[phead].next ≠ null, phead ≠ null, phead ≠ vs0[phead].next, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(rl1 + phead ' + vs0[phead].next ' + qrl, vs0, ps, owns0), listp(vs0[phead].next ' + qrl, vs0, ps, owns0), markedp(rl1, ps), markedp(rl1, vs0), nonullRetRefs(rl1 + phead ' + vs0[phead].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vs0[phead].next].deqID), some?(vs0[phead].deqID), some?(∅[vs0[phead].next].deqID), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, ¬ dups(rl1 + qrl), ¬ dups(rl1), ¬ dups(qrl), ¬ weak?(vs0[phead].deqID), ¬ phead ∈ rl1, ¬ vs0[phead].next ∈ rl1, ¬ vs0[phead].next ∈ qrl, ¬ vs0[phead].next ∈ ps, ¬ phead ∈ qrl, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vs0[phead].next = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ phead = vhead0 ∧ unmarkedp(qrl, vs0), vtail0 = (vs0[phead].next ' + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[phead].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic elimination applied rule apply elim lemma

And closed the goal!

Rule argument

spec: list-data[reflist], name:elim-list-c, seq:x ≠ [] ⊦ a = x.head ∧ y = x.tail ↔ x = a + y, subst:[x, a, y] → [qrl, vtail, qrl0]

Additional information

Used simplifier rules: ⊦ a ' + x ≠ []; (s) ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ unmarkedp([], vs); (s) ⊦ x + [] = x; (s) ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s) x ≠ [] ⊦ a = x.head ∧ y = x.tail ↔ x = a + y; (elim) Used terms: qrl, vtail, y

Proof step 32 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead = vhead1, vhead2 = phead, ps[ps[vhead].next].deqID ≠ None, ps[vhead].next ≠ null, vhead ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), vhead ∈ orl, vhead ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ vhead ∈ ps ⊦ ⟪vhead := ps[vhead].next⟫ ((vhead ∈ orl ∧ vhead ∈ ps ∧ vhead ≠ null ∧ (ps[vhead].next ≠ null → ps[vhead].next ∈ ps)) ∧ #(orl from vhead) < #(orl from vhead1))

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 33 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead = vhead1, vhead2 = phead, ps[ps[vhead].next].deqID ≠ None, ps[vhead].next ≠ null, vhead ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), vhead ∈ ps, vhead ∈ orl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ vhead ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ⊦ (ps[vhead].next ∈ orl ∧ ps[vhead].next ∈ ps ∧ ps[vhead].next ≠ null ∧ (ps[ps[vhead].next].next ≠ null → ps[ps[vhead].next].next ∈ ps)) ∧ #(orl from ps[vhead].next) < #(orl from vhead1)

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 34 from proof Crash-Theorem (previous step) (back to tree)

ps[ps[vhead1].next].deqID ≠ None, ps[vhead1].next ≠ null, vhead1 ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), vhead1 ∈ orl, vhead1 ∈ ps, ps[vhead1].next ∈ ps ⊦ ps[vhead1].next ∈ orl ∧ (ps[ps[vhead1].next].next ≠ null → ps[ps[vhead1].next].next ∈ ps) ∧ #(orl from ps[vhead1].next) < #(orl from vhead1)

User applied rule apply rewrite lemma

And obtained the following new goals

Rule argument

spec: list[reflist], name:in-def, seq: ⊦ a ∈ x ↔ (∃ y, z. x = y + (a + z)), subst:[a, x] → [vhead1, orl], <all paths>

Additional information

Used terms: vhead1, orl

Proof step 35 from proof Crash-Theorem (previous step) (back to tree)

⊦ a ∈ x ↔ (∃ y, z. x = y + (a + z))

A spec-lemma

Proof step 36 from proof Crash-Theorem (previous step) (back to tree)

ps[ps[vhead1].next].deqID ≠ None, ps[vhead1].next ≠ null, vhead1 ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), ps[vhead1].next ∈ ps, vhead1 ∈ ps, ∃ y, z. orl = y + (vhead1 + z) ⊦ ps[vhead1].next ∈ orl ∧ (ps[ps[vhead1].next].next ≠ null → ps[ps[vhead1].next].next ∈ ps) ∧ #(orl from ps[vhead1].next) < #(orl from vhead1)

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ a + x = a ' + x; (s) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s)

Proof step 37 from proof Crash-Theorem (previous step) (back to tree)

ps[ps[vhead1].next].deqID ≠ None, ps[vhead1].next ≠ null, vhead1 ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns)), vhead1 ∈ ps, ps[vhead1].next ∈ ps ⊦ (ps[vhead1].next ∈ qrl0 ∨ ps[vhead1].next = vhead1 ∨ ps[vhead1].next ∈ qrl1) ∧ (ps[ps[vhead1].next].next ≠ null → ps[ps[vhead1].next].next ∈ ps) ∧ #(qrl0 + vhead1 ' + qrl1 from ps[vhead1].next) < #(qrl0 + vhead1 ' + qrl1 from vhead1)

User applied rule apply rewrite lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:DMSQGInv, seq:⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed (gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null), subst:[gs] → [mkgs (ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns)], <all paths>

Additional information

Used terms: mkgs (ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns)

Proof step 38 from proof Crash-Theorem (previous step) (back to tree)

⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed(gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null)

A spec-lemma

Proof step 39 from proof Crash-Theorem (previous step) (back to tree)

dom mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vrvHp = dom mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).prvHp, ps[ps[vhead1].next].deqID ≠ None, ps[vhead1].next ≠ null, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl ≠ [], vhead1 ≠ null, allocrv (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vrvHp), allocrv (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).pretRefs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).prvHp), listp ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).ps, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).owns), markedp (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).ps), markedp (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs), nonullRetRefs ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).pretRefs), rvownerok (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vrvHp, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).rvowns), rvownerok (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).pretRefs, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).prvHp, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).rvowns), unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).ps), ps[vhead1].next ∈ ps, vhead1 ∈ ps, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).phead ∈ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl, ¬ dups ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl), mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vhead ] .next ≠ null ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vhead ] .next ] .deqID ≠ None → mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).rvowns (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vretRefs (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vhead ] .next ] .deqID .tid)) = Claimed (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vhead ] .next ] .deqID .tid, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vhead ] .next ] .val), mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl.last = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vhead ∧ ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl ≠ [] → unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl.tail, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs)) ∨ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl.butlast ≠ [] ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl.butlast.last = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vhead ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).phead ≠ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl.last ∧ unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs), mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vtail = ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl ).last ∨ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl ≠ [] ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vtail = ( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl.butlast ).last ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl.last ] .deqID = None ∧ mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).vs [( mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, qrl0 + vhead1 ' + qrl1, owns0, rvowns).qrl ).last ] .next = null ⊦ (ps[vhead1].next ∈ qrl0 ∨ ps[vhead1].next = vhead1 ∨ ps[vhead1].next ∈ qrl1) ∧ (ps[ps[vhead1].next].next ≠ null → ps[ps[vhead1].next].next ∈ ps) ∧ #(qrl0 + vhead1 ' + qrl1 from ps[vhead1].next) < #(qrl0 + vhead1 ' + qrl1 from vhead1)

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] c: ⊦ m + n = n + m; from specification nat-basic a: ⊦ (m + n0) + n1 = m + n0 + n1; from specification nat-basic ⊦ m < n + 1 ↔ ¬ n < m; (s) ⊦ m +1 = m + 1; (s) ⊦ #(a ' + x) = (# x)+1; (s) ⊦ r ' + rl from r = r ' + rl; (s) ⊦ ¬ r ∈ orl → orl + qrl from r = qrl from r; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vhead = phead0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vs = ps0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .rvowns = rvowns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vretRefs = vretRefs; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .qrl = qrl; (s) ⊦ (x + a ' + y).last = (a ' + y).last; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .orl = qrl0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .phead = phead; (s) ⊦ (x + a ' + y).butlast = x + (a ' + y).butlast; (s) ⊦ y ≠ [] → (x + y).last = y.last; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vtail = phead1; (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vrvHp = vrvHp; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .prvHp = vrvHp0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .pretRefs = vretRefs0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .owns = owns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .ps = ps; (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ x + a ' + y ≠ []; (s)

Proof step 40 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, ps[ps[vhead1].next].deqID ≠ None, ps[vhead1].next ≠ null, vhead1 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((qrl0 + vhead1 ' + qrl1) + qrl, vs0, ps, owns0), markedp(qrl1, ps), markedp(qrl0, ps), markedp(qrl1, vs0), markedp(qrl0, vs0), nonullRetRefs((qrl0 + vhead1 ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead1].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), ps[vhead1].next ∈ ps, vhead1 ∈ ps, ¬ dups(qrl0 + qrl1), ¬ dups(qrl1 + qrl), ¬ dups(qrl0 + qrl1 + qrl), ¬ dups(qrl1), ¬ dups(qrl), ¬ dups(qrl0), ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl1, ¬ vhead1 ∈ qrl0, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), (vhead1 ' + qrl1).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 + (vhead1 ' + qrl1).butlast ≠ [] ∧ (qrl0 + (vhead1 ' + qrl1).butlast).last = vhead0 ∧ phead ≠ (vhead1 ' + qrl1).last ∧ unmarkedp(qrl, vs0), vtail0 = (vhead1 ' + qrl1 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vhead1 ' + qrl1 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead ∈ qrl1 ⊦ (ps[vhead1].next ∈ qrl0 ∨ ps[vhead1].next = vhead1 ∨ ps[vhead1].next ∈ qrl1) ∧ (ps[ps[vhead1].next].next ≠ null → ps[ps[vhead1].next].next ∈ ps) ∧ ¬ # qrl1 < #(qrl0 + vhead1 ' + qrl1 from ps[vhead1].next)

User applied rule apply lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:listp-postfix, seq: ⊦ listp(rl + rl0, vs, ps, owns) → (listp(rl0, vs, ps, owns) ↔ rl0 ≠ []), subst:[ps, rl, owns, rl0, vs] → [ps, qrl0, owns0, vhead1 ' + qrl1 + qrl, vs0]

Additional information

Used terms: ps, qrl0, owns0, vhead1 ' + qrl1 + qrl, vs0

Proof step 41 from proof Crash-Theorem (previous step) (back to tree)

⊦ listp(rl + rl0, vs, ps, owns) → (listp(rl0, vs, ps, owns) ↔ rl0 ≠ [])

A spec-lemma

Proof step 42 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, ps[ps[vhead1].next].deqID ≠ None, ps[vhead1].next ≠ null, vhead1 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp((qrl0 + vhead1 ' + qrl1) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl1, vs0), markedp(qrl0, ps), markedp(qrl1, ps), nonullRetRefs((qrl0 + vhead1 ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead1].deqID), some?(ps[vhead1].deqID), unmarkedp(qrl, ps), vhead1 ∈ ps, ps[vhead1].next ∈ ps, ¬ dups(qrl0 + qrl1), ¬ dups(qrl1 + qrl), ¬ dups(qrl0 + qrl1 + qrl), ¬ dups(qrl1), ¬ dups(qrl), ¬ dups(qrl0), ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl1, ¬ vhead1 ∈ qrl0, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), listp(qrl0 + vhead1 ' + qrl1 + qrl, vs0, ps, owns0) → (listp(vhead1 ' + qrl1 + qrl, vs0, ps, owns0) ↔ vhead1 ' + qrl1 + qrl ≠ []), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead ∈ qrl1, vtail0 = (vhead1 ' + qrl1 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vhead1 ' + qrl1 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, (vhead1 ' + qrl1).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 + (vhead1 ' + qrl1).butlast ≠ [] ∧ (qrl0 + (vhead1 ' + qrl1).butlast).last = vhead0 ∧ phead ≠ (vhead1 ' + qrl1).last ∧ unmarkedp(qrl, vs0) ⊦ (ps[vhead1].next ∈ qrl0 ∨ ps[vhead1].next = vhead1 ∨ ps[vhead1].next ∈ qrl1) ∧ (ps[ps[vhead1].next].next ≠ null → ps[ps[vhead1].next].next ∈ ps) ∧ ¬ # qrl1 < #(qrl0 + vhead1 ' + qrl1 from ps[vhead1].next)

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ a ' + x ≠ []; (s)

Proof step 43 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, ps[ps[vhead1].next].deqID ≠ None, ps[vhead1].next ≠ null, vhead1 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((qrl0 + vhead1 ' + qrl1) + qrl, vs0, ps, owns0), listp(vhead1 ' + qrl1 + qrl, vs0, ps, owns0), markedp(qrl1, ps), markedp(qrl0, ps), markedp(qrl1, vs0), markedp(qrl0, vs0), nonullRetRefs((qrl0 + vhead1 ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead1].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), ps[vhead1].next ∈ ps, vhead1 ∈ ps, ¬ dups(qrl0 + qrl1), ¬ dups(qrl1 + qrl), ¬ dups(qrl0 + qrl1 + qrl), ¬ dups(qrl1), ¬ dups(qrl), ¬ dups(qrl0), ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl1, ¬ vhead1 ∈ qrl0, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), (vhead1 ' + qrl1).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 + (vhead1 ' + qrl1).butlast ≠ [] ∧ (qrl0 + (vhead1 ' + qrl1).butlast).last = vhead0 ∧ phead ≠ (vhead1 ' + qrl1).last ∧ unmarkedp(qrl, vs0), vtail0 = (vhead1 ' + qrl1 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vhead1 ' + qrl1 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead ∈ qrl1 ⊦ (ps[vhead1].next ∈ qrl0 ∨ ps[vhead1].next = vhead1 ∨ ps[vhead1].next ∈ qrl1) ∧ (ps[ps[vhead1].next].next ≠ null → ps[ps[vhead1].next].next ∈ ps) ∧ ¬ # qrl1 < #(qrl0 + vhead1 ' + qrl1 from ps[vhead1].next)

User applied rule constructor cut

And obtained the following new goals

Rule argument

var: qrl1

Additional information

Used simplifier rules: reflist freely generated by [], +;;

Proof step 44 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, qrl1 = [], ps[ps[vhead1].next].deqID ≠ None, ps[vhead1].next ≠ null, vhead1 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vhead1 ' + qrl1 + qrl, vs0, ps, owns0), listp((qrl0 + vhead1 ' + qrl1) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl1, vs0), markedp(qrl0, ps), markedp(qrl1, ps), nonullRetRefs((qrl0 + vhead1 ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead1].deqID), some?(ps[vhead1].deqID), unmarkedp(qrl, ps), vhead1 ∈ ps, ps[vhead1].next ∈ ps, ¬ dups(qrl0 + qrl1), ¬ dups(qrl1 + qrl), ¬ dups(qrl0 + qrl1 + qrl), ¬ dups(qrl1), ¬ dups(qrl), ¬ dups(qrl0), ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl1, ¬ vhead1 ∈ qrl0, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead ∈ qrl1, vtail0 = (vhead1 ' + qrl1 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vhead1 ' + qrl1 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, (vhead1 ' + qrl1).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 + (vhead1 ' + qrl1).butlast ≠ [] ∧ (qrl0 + (vhead1 ' + qrl1).butlast).last = vhead0 ∧ phead ≠ (vhead1 ' + qrl1).last ∧ unmarkedp(qrl, vs0) ⊦ (ps[vhead1].next ∈ qrl0 ∨ ps[vhead1].next = vhead1 ∨ ps[vhead1].next ∈ qrl1) ∧ (ps[ps[vhead1].next].next ≠ null → ps[ps[vhead1].next].next ∈ ps) ∧ ¬ # qrl1 < #(qrl0 + vhead1 ' + qrl1 from ps[vhead1].next)

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ # x = 0 ↔ x = []; (s) ⊦ 0 < m ↔ m ≠ 0; (s) ⊦ x + [] = x; (s) ⊦ # [] = 0; (s) ⊦ ¬ a ∈ []; (s) ⊦ [] + x = x; (s) ⊦ a '.last = a; (s) ⊦ a '.butlast = []; (s) ⊦ ¬ dups([]); (s) ⊦ markedp([], vs); (s)

Proof step 45 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, ps[ps[vhead1].next].deqID ≠ None, ps[vhead1].next ≠ null, vhead1 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(qrl0 + vhead1 ' + qrl, vs0, ps, owns0), listp(vhead1 ' + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs(qrl0 + vhead1 ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead1].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), ps[vhead1].next ∈ ps, vhead1 ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl0, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vhead1 = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead1 ∧ unmarkedp(qrl, vs0), vtail0 = (vhead1 ' + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vhead1 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ⊦ (ps[vhead1].next ∈ qrl0 ∨ ps[vhead1].next = vhead1) ∧ (ps[ps[vhead1].next].next ≠ null → ps[ps[vhead1].next].next ∈ ps) ∧ qrl0 + vhead1 ' from ps[vhead1].next = []

Heuristic elimination applied rule apply elim lemma

And closed the goal!

Rule argument

spec: list-data[reflist], name:elim-list-c, seq:x ≠ [] ⊦ a = x.head ∧ y = x.tail ↔ x = a + y, subst:[x, a, y] → [qrl, vtail, qrl1]

Additional information

Used simplifier rules: ⊦ a ' + x ≠ []; (s) ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ unmarkedp([], vs); (s) ⊦ x + [] = x; (s) ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s) x ≠ [] ⊦ a = x.head ∧ y = x.tail ↔ x = a + y; (elim) Used terms: qrl, vtail, y

Proof step 46 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, qrl1 = vhead + qrl2, ps[ps[vhead1].next].deqID ≠ None, ps[vhead1].next ≠ null, vhead1 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vhead1 ' + qrl1 + qrl, vs0, ps, owns0), listp((qrl0 + vhead1 ' + qrl1) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl1, vs0), markedp(qrl0, ps), markedp(qrl1, ps), nonullRetRefs((qrl0 + vhead1 ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead1].deqID), some?(ps[vhead1].deqID), unmarkedp(qrl, ps), vhead1 ∈ ps, ps[vhead1].next ∈ ps, ¬ dups(qrl0 + qrl1), ¬ dups(qrl1 + qrl), ¬ dups(qrl0 + qrl1 + qrl), ¬ dups(qrl1), ¬ dups(qrl), ¬ dups(qrl0), ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl1, ¬ vhead1 ∈ qrl0, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead ∈ qrl1, vtail0 = (vhead1 ' + qrl1 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vhead1 ' + qrl1 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, (vhead1 ' + qrl1).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 + (vhead1 ' + qrl1).butlast ≠ [] ∧ (qrl0 + (vhead1 ' + qrl1).butlast).last = vhead0 ∧ phead ≠ (vhead1 ' + qrl1).last ∧ unmarkedp(qrl, vs0) ⊦ (ps[vhead1].next ∈ qrl0 ∨ ps[vhead1].next = vhead1 ∨ ps[vhead1].next ∈ qrl1) ∧ (ps[ps[vhead1].next].next ≠ null → ps[ps[vhead1].next].next ∈ ps) ∧ ¬ # qrl1 < #(qrl0 + vhead1 ' + qrl1 from ps[vhead1].next)

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: c: ⊦ m + n = n + m; from specification nat-basic a: ⊦ (m + n0) + n1 = m + n0 + n1; from specification nat-basic ⊦ m + n < m + n0 ↔ n < n0; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ m +1 = m + 1; (s) ⊦ #(a ' + x) = (# x)+1; (s) ⊦ r ' + rl from r = r ' + rl; (s) ⊦ r0 ≠ r → r0 ' + rl from r = rl from r; (s) ⊦ ¬ r ∈ orl → orl + qrl from r = qrl from r; (s) listp-twoin: ⊦ listp(rl + r ' + r0 ' + rl0, vs, ps, owns) → ps[r].next = r0; (forward) from specification DMSQ a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a + x = a ' + x; (s) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ (b ' + a ' + x).last = (a ' + x).last; (s) ⊦ (x + a ' + y).last = (a ' + y).last; (s) ⊦ (a ' + b ' + x).butlast = a ' + (b ' + x).butlast; (s) ⊦ x + a ' + y ≠ []; (s) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ some?(opttid) → opttid ≠ None; (as) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s)

Proof step 47 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, ps, owns0), listp(vs0[vhead1].next ' + qrl2 + qrl, vs0, ps, owns0), markedp(qrl2, ps), markedp(qrl0, ps), markedp(qrl2, vs0), markedp(qrl0, vs0), nonullRetRefs(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vhead1].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vs0[vhead1].next].deqID), unmarkedp(qrl, ps), vs0[vhead1].next ∈ ps, vhead1 ∈ ps, vhead1 ∈ vs0, ¬ dups(qrl0 + qrl2 + qrl), ¬ dups(qrl0), ¬ dups(qrl2 + qrl), ¬ dups(qrl0 + qrl2), ¬ dups(qrl), ¬ dups(qrl2), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl2, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl2, ¬ ps[vs0[vhead1].next].next ∈ ps, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), (vs0[vhead1].next ' + qrl2).last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ (vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last = vhead0 ∧ phead ≠ (vs0[vhead1].next ' + qrl2).last ∧ unmarkedp(qrl, vs0), vtail0 = (vs0[vhead1].next ' + qrl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead ∈ qrl2 ⊦

User applied rule case distinction

And obtained the following new goals

Rule argument

pos: left 44

Additional information

Proof step 48 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, (vs0[vhead1].next ' + qrl2).last = vhead0, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vhead1].next ' + qrl2 + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl2, vs0), markedp(qrl0, ps), markedp(qrl2, ps), nonullRetRefs(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vs0[vhead1].next].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, ¬ dups(qrl0 + qrl2 + qrl), ¬ dups(qrl0), ¬ dups(qrl2 + qrl), ¬ dups(qrl0 + qrl2), ¬ dups(qrl), ¬ dups(qrl2), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl2, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl2, ¬ ps[vs0[vhead1].next].next ∈ ps, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), qrl ≠ [] → unmarkedp(qrl.tail, vs0), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead ∈ qrl2, vtail0 = (vs0[vhead1].next ' + qrl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 49 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, ps, owns0), listp(vs0[vhead1].next ' + qrl2 + qrl, vs0, ps, owns0), markedp(qrl2, ps), markedp(qrl0, ps), markedp(qrl2, vs0), markedp(qrl0, vs0), nonullRetRefs(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vhead1].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vs0[vhead1].next].deqID), unmarkedp(qrl, ps), vs0[vhead1].next ∈ ps, vhead1 ∈ ps, vhead1 ∈ vs0, ¬ dups(qrl0 + qrl2 + qrl), ¬ dups(qrl0), ¬ dups(qrl2 + qrl), ¬ dups(qrl0 + qrl2), ¬ dups(qrl), ¬ dups(qrl2), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl2, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl2, ¬ ps[vs0[vhead1].next].next ∈ ps, qrl ≠ [] → unmarkedp(qrl.tail, vs0), vs0[(vs0[vhead1].next ' + qrl2).last].next ≠ null ∧ vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].deqID.tid)) = Claimed(vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].deqID.tid, vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].val), vtail0 = (vs0[vhead1].next ' + qrl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead ∈ qrl2 ⊦

User applied rule constructor cut

And obtained the following new goals

Rule argument

var: qrl2

Additional information

Used simplifier rules: reflist freely generated by [], +;;

Proof step 50 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, qrl2 = [], ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vhead1].next ' + qrl2 + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl2, vs0), markedp(qrl0, ps), markedp(qrl2, ps), nonullRetRefs(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vs0[vhead1].next].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, ¬ dups(qrl0 + qrl2 + qrl), ¬ dups(qrl0), ¬ dups(qrl2 + qrl), ¬ dups(qrl0 + qrl2), ¬ dups(qrl), ¬ dups(qrl2), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl2, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl2, ¬ ps[vs0[vhead1].next].next ∈ ps, vs0[(vs0[vhead1].next ' + qrl2).last].next ≠ null ∧ vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].deqID.tid)) = Claimed(vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].deqID.tid, vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].val), qrl ≠ [] → unmarkedp(qrl.tail, vs0), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead ∈ qrl2, vtail0 = (vs0[vhead1].next ' + qrl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ x ≠ [] → (a ' + x).last = x.last; (s) ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s) ⊦ x + [] = x; (s) ⊦ unmarkedp([], vs); (s) a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a '.last = a; (s) ⊦ ¬ a ∈ []; (s) ⊦ [] + x = x; (s) ⊦ ¬ dups([]); (s) ⊦ markedp([], vs); (s)

Proof step 51 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, ps[vs0[vhead1].next].deqID ≠ None, ps[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ null, qrl ≠ [], vhead1 ≠ vs0[vhead1].next, vhead1 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vhead1].next ' + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vs0[vhead1].next].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), unmarkedp(qrl.tail, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, ¬ dups(qrl), ¬ dups(qrl0), ¬ dups(qrl0 + qrl), ¬ weak?(vs0[vhead1].deqID), ¬ ps[vs0[vhead1].next].next ∈ ps, ¬ vhead1 ∈ qrl, ¬ vs0[vhead1].next ∈ qrl, ¬ vs0[vhead1].next ∈ qrl0, ¬ vhead1 ∈ qrl0, vs0[vs0[vhead1].next].next ≠ null ∧ vs0[vs0[vs0[vhead1].next].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vs0[vhead1].next].next].deqID.tid)) = Claimed(vs0[vs0[vs0[vhead1].next].next].deqID.tid, vs0[vs0[vs0[vhead1].next].next].val), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next, vtail0 = qrl.last ∨ vtail0 = (vs0[vhead1].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic elimination applied rule apply elim lemma

And obtained the following new goals

Rule argument

spec: list-data[reflist], name:elim-add, seq: ⊦ x ≠ [] → (head = x.head ∧ tail = x.tail ↔ x = head + tail), subst:[x, head, tail] → [qrl, vtail, qrl1]

Additional information

Used simplifier rules: ⊦ x ≠ [] → (head = x.head ∧ tail = x.tail ↔ x = head + tail); (elim) Used terms: qrl, head, qrl1

Proof step 52 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, qrl = vtail + qrl1, ps[vs0[vhead1].next].deqID ≠ None, ps[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ null, qrl ≠ [], vhead1 ≠ vs0[vhead1].next, vhead1 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, ps, owns0), listp(vs0[vhead1].next ' + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vhead1].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vs0[vhead1].next].deqID), unmarkedp(qrl1, vs0), unmarkedp(qrl, ps), vs0[vhead1].next ∈ ps, vhead1 ∈ ps, vhead1 ∈ vs0, ¬ dups(qrl), ¬ dups(qrl0), ¬ dups(qrl0 + qrl), ¬ weak?(vs0[vhead1].deqID), ¬ ps[vs0[vhead1].next].next ∈ ps, ¬ vhead1 ∈ qrl, ¬ vs0[vhead1].next ∈ qrl, ¬ vs0[vhead1].next ∈ qrl0, ¬ vhead1 ∈ qrl0, vs0[vs0[vhead1].next].next ≠ null ∧ vs0[vs0[vs0[vhead1].next].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vs0[vhead1].next].next].deqID.tid)) = Claimed(vs0[vs0[vs0[vhead1].next].next].deqID.tid, vs0[vs0[vs0[vhead1].next].next].val), vtail0 = qrl.last ∨ vtail0 = (vs0[vhead1].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ ¬ a ∈ st → st[a] = ∅[a]; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] listp-twoin: ⊦ listp(rl + r ' + r0 ' + rl0, vs, ps, owns) → ps[r].next = r0; (forward) from specification DMSQ a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a + x = a ' + x; (s) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ a ' + x ≠ []; (s)

Proof step 53 from proof Crash-Theorem (previous step) (back to tree)

ps[vs0[vhead1].next].deqID = vs0[vs0[vhead1].next].deqID, ps[vhead1].deqID = vs0[vhead1].deqID, ∅[vs0[vs0[vhead1].next].next].deqID = None, ps[vs0[vhead1].next].next = vs0[vs0[vhead1].next].next, ps[vhead1].next = vs0[vhead1].next, ps[vs0[vhead1].next].val = vs0[vs0[vhead1].next].val, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vs0[vhead1].next) = None, owns0(vhead1) = None, vs0[vs0[vhead1].next].deqID ≠ None, vs0[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ null, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ null, vhead1 ≠ vs0[vs0[vhead1].next].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vs0[vhead1].next].next ' + qrl1, vs0, ps, owns0), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl1, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl1, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl1, ps), unmarkedp(qrl1, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, vs0[vhead1].next ∈ vs0, ¬ dups(qrl0 + qrl1), ¬ dups(qrl0), ¬ dups(qrl1), ¬ weak?(vs0[vs0[vhead1].next].deqID), ¬ weak?(vs0[vhead1].deqID), ¬ vs0[vhead1].next ∈ qrl0, ¬ vhead1 ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ qrl1, ¬ vs0[vs0[vhead1].next].next ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ ps, ¬ vhead1 ∈ qrl1, ¬ vs0[vhead1].next ∈ qrl1, vs0[vs0[vs0[vhead1].next].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vs0[vhead1].next].next].deqID.tid)) = Claimed(vs0[vs0[vs0[vhead1].next].next].deqID.tid, vs0[vs0[vs0[vhead1].next].next].val), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next, vtail0 = (vs0[vs0[vhead1].next].next ' + qrl1).last ∨ vtail0 = (vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last ∧ vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].deqID = None ∧ vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next = null ⊦

User applied rule constructor cut

And closed the goal!

Rule argument

var: qrl1

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ unmarkedp([], vs); (s) ⊦ x + [] = x; (s) ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s)

Proof step 54 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, qrl2 = vhead + qrl1, ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vhead1].next ' + qrl2 + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl2, vs0), markedp(qrl0, ps), markedp(qrl2, ps), nonullRetRefs(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vs0[vhead1].next].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, ¬ dups(qrl0 + qrl2 + qrl), ¬ dups(qrl0), ¬ dups(qrl2 + qrl), ¬ dups(qrl0 + qrl2), ¬ dups(qrl), ¬ dups(qrl2), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl2, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl2, ¬ ps[vs0[vhead1].next].next ∈ ps, vs0[(vs0[vhead1].next ' + qrl2).last].next ≠ null ∧ vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].deqID.tid)) = Claimed(vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].deqID.tid, vs0[vs0[(vs0[vhead1].next ' + qrl2).last].next].val), qrl ≠ [] → unmarkedp(qrl.tail, vs0), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead ∈ qrl2, vtail0 = (vs0[vhead1].next ' + qrl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ ¬ a ∈ st → st[a] = ∅[a]; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] listp-twoin: ⊦ listp(rl + r ' + r0 ' + rl0, vs, ps, owns) → ps[r].next = r0; (forward) from specification DMSQ a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ (b ' + a ' + x).last = (a ' + x).last; (s) ⊦ a + x = a ' + x; (s) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ ¬ a ∈ x ∧ ¬ a ∈ y → (dups(x + a ' + y) ↔ dups(x + y)); (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s)

Proof step 55 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vs0[vhead1].next].deqID = vs0[vs0[vhead1].next].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vs0[vhead1].next].next = vs0[vs0[vhead1].next].next, ps[vhead1].val = vs0[vhead1].val, ps[vs0[vhead1].next].val = vs0[vs0[vhead1].next].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, owns0(vs0[vhead1].next) = None, vs0[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, vs0[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ vs0[vs0[vhead1].next].next, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ vs0[vs0[vhead1].next].next, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1) + qrl, vs0, ps, owns0), listp(vs0[vs0[vhead1].next].next ' + qrl1 + qrl, vs0, ps, owns0), markedp(qrl1, ps), markedp(qrl0, ps), markedp(qrl1, vs0), markedp(qrl0, vs0), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vs0[vs0[vhead1].next].next].deqID), some?(vs0[vhead1].deqID), some?(vs0[vs0[vhead1].next].deqID), some?(∅[vs0[vs0[vhead1].next].next].deqID), unmarkedp(qrl, ps), vs0[vhead1].next ∈ vs0, vs0[vhead1].next ∈ ps, vhead1 ∈ ps, vhead1 ∈ vs0, ¬ dups(qrl0 + qrl1 + qrl), ¬ dups(qrl0), ¬ dups(qrl1), ¬ dups(qrl), ¬ dups(qrl1 + qrl), ¬ dups(qrl0 + qrl1), ¬ weak?(vs0[vs0[vhead1].next].deqID), ¬ weak?(vs0[vhead1].deqID), ¬ vs0[vs0[vhead1].next].next ∈ ps, ¬ vs0[vs0[vhead1].next].next ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ qrl1, ¬ vs0[vs0[vhead1].next].next ∈ qrl, ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl1, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl1, qrl ≠ [] → unmarkedp(qrl.tail, vs0), vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next ≠ null ∧ vs0[vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next].deqID.tid)) = Claimed(vs0[vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next].deqID.tid, vs0[vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next].val), vtail0 = (vs0[vs0[vhead1].next].next ' + qrl1 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vs0[vhead1].next].next ' + qrl1 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead = vs0[vs0[vhead1].next].next ∨ phead ∈ qrl1 ⊦

User applied rule constructor cut

And obtained the following new goals

Rule argument

var: qrl1

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist]

Proof step 56 from proof Crash-Theorem (previous step) (back to tree)

ps[vs0[vhead1].next].deqID = vs0[vs0[vhead1].next].deqID, ps[vhead1].deqID = vs0[vhead1].deqID, ps[vs0[vhead1].next].next = vs0[vs0[vhead1].next].next, ps[vhead1].next = vs0[vhead1].next, ps[vs0[vhead1].next].val = vs0[vs0[vhead1].next].val, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vs0[vhead1].next) = None, owns0(vhead1) = None, qrl1 = [], vs0[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, vs0[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ vs0[vs0[vhead1].next].next, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ vs0[vs0[vhead1].next].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vs0[vhead1].next].next ' + qrl1 + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl1, vs0), markedp(qrl0, ps), markedp(qrl1, ps), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(∅[vs0[vs0[vhead1].next].next].deqID), some?(vs0[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), some?(vs0[vs0[vs0[vhead1].next].next].deqID), unmarkedp(qrl, ps), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, vs0[vhead1].next ∈ vs0, ¬ dups(qrl0 + qrl1 + qrl), ¬ dups(qrl0), ¬ dups(qrl1), ¬ dups(qrl), ¬ dups(qrl1 + qrl), ¬ dups(qrl0 + qrl1), ¬ weak?(vs0[vs0[vhead1].next].deqID), ¬ weak?(vs0[vhead1].deqID), ¬ vs0[vs0[vhead1].next].next ∈ ps, ¬ vs0[vs0[vhead1].next].next ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ qrl1, ¬ vs0[vs0[vhead1].next].next ∈ qrl, ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl1, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl1, vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next ≠ null ∧ vs0[vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next].deqID.tid)) = Claimed(vs0[vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next].deqID.tid, vs0[vs0[(vs0[vs0[vhead1].next].next ' + qrl1).last].next].val), qrl ≠ [] → unmarkedp(qrl.tail, vs0), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead = vs0[vs0[vhead1].next].next ∨ phead ∈ qrl1, vtail0 = (vs0[vs0[vhead1].next].next ' + qrl1 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vs0[vhead1].next].next ' + qrl1 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ x ≠ [] → (a ' + x).last = x.last; (s) ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s) ⊦ x + [] = x; (s) ⊦ unmarkedp([], vs); (s) a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a '.last = a; (s) ⊦ ¬ a ∈ []; (s) ⊦ [] + x = x; (s) ⊦ ¬ dups([]); (s) ⊦ markedp([], vs); (s)

Proof step 57 from proof Crash-Theorem (previous step) (back to tree)

ps[vs0[vhead1].next].deqID = vs0[vs0[vhead1].next].deqID, ps[vhead1].deqID = vs0[vhead1].deqID, ps[vs0[vhead1].next].next = vs0[vs0[vhead1].next].next, ps[vhead1].next = vs0[vhead1].next, ps[vs0[vhead1].next].val = vs0[vs0[vhead1].next].val, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vs0[vhead1].next) = None, owns0(vhead1) = None, vs0[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ vs0[vs0[vhead1].next].next, vs0[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ null, qrl ≠ [], vhead1 ≠ vs0[vs0[vhead1].next].next, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vs0[vhead1].next].next ' + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(∅[vs0[vs0[vhead1].next].next].deqID), some?(vs0[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), some?(vs0[vs0[vs0[vhead1].next].next].deqID), unmarkedp(qrl, ps), unmarkedp(qrl.tail, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, vs0[vhead1].next ∈ vs0, ¬ dups(qrl), ¬ dups(qrl0), ¬ dups(qrl0 + qrl), ¬ weak?(vs0[vhead1].deqID), ¬ weak?(vs0[vs0[vhead1].next].deqID), ¬ vhead1 ∈ qrl, ¬ vs0[vhead1].next ∈ qrl, ¬ vs0[vhead1].next ∈ qrl0, ¬ vhead1 ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ qrl, ¬ vs0[vs0[vhead1].next].next ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ ps, vs0[vs0[vs0[vhead1].next].next].next ≠ null ∧ vs0[vs0[vs0[vs0[vhead1].next].next].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vs0[vs0[vhead1].next].next].next].deqID.tid)) = Claimed(vs0[vs0[vs0[vs0[vhead1].next].next].next].deqID.tid, vs0[vs0[vs0[vs0[vhead1].next].next].next].val), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead = vs0[vs0[vhead1].next].next, vtail0 = qrl.last ∨ vtail0 = (vs0[vs0[vhead1].next].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic elimination applied rule apply elim lemma

And closed the goal!

Rule argument

spec: list-data[reflist], name:elim-add, seq: ⊦ x ≠ [] → (head = x.head ∧ tail = x.tail ↔ x = head + tail), subst:[x, head, tail] → [qrl, vtail, qrl1]

Additional information

Used simplifier rules: ⊦ a ' + x ≠ []; (s) ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ x ≠ [] → (head = x.head ∧ tail = x.tail ↔ x = head + tail); (elim) Used terms: qrl, head, qrl1

Proof step 58 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, (vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last = vhead0, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, phead ≠ (vs0[vhead1].next ' + qrl2).last, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vhead1].next ' + qrl2 + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl2, vs0), markedp(qrl0, ps), markedp(qrl2, ps), nonullRetRefs(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vs0[vhead1].next].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), unmarkedp(qrl, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, ¬ dups(qrl0 + qrl2 + qrl), ¬ dups(qrl0), ¬ dups(qrl2 + qrl), ¬ dups(qrl0 + qrl2), ¬ dups(qrl), ¬ dups(qrl2), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl2, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl2, ¬ ps[vs0[vhead1].next].next ∈ ps, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead ∈ qrl2, vtail0 = (vs0[vhead1].next ' + qrl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 59 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, phead ≠ (vs0[vhead1].next ' + qrl2).last, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, ps, owns0), listp(vs0[vhead1].next ' + qrl2 + qrl, vs0, ps, owns0), markedp(qrl2, ps), markedp(qrl0, ps), markedp(qrl2, vs0), markedp(qrl0, vs0), nonullRetRefs(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vhead1].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vs0[vhead1].next].deqID), unmarkedp(qrl, vs0), unmarkedp(qrl, ps), vs0[vhead1].next ∈ ps, vhead1 ∈ ps, vhead1 ∈ vs0, ¬ dups(qrl0 + qrl2 + qrl), ¬ dups(qrl0), ¬ dups(qrl2 + qrl), ¬ dups(qrl0 + qrl2), ¬ dups(qrl), ¬ dups(qrl2), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl2, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl2, ¬ ps[vs0[vhead1].next].next ∈ ps, vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next ≠ null ∧ vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].deqID.tid)) = Claimed (vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].deqID.tid, vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].val), vtail0 = (vs0[vhead1].next ' + qrl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead ∈ qrl2 ⊦

User applied rule constructor cut

And obtained the following new goals

Rule argument

var: qrl2

Additional information

Used simplifier rules: reflist freely generated by [], +;;

Proof step 60 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, qrl2 = [], ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, phead ≠ (vs0[vhead1].next ' + qrl2).last, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vhead1].next ' + qrl2 + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl2, vs0), markedp(qrl0, ps), markedp(qrl2, ps), nonullRetRefs(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vs0[vhead1].next].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), unmarkedp(qrl, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, ¬ dups(qrl0 + qrl2 + qrl), ¬ dups(qrl0), ¬ dups(qrl2 + qrl), ¬ dups(qrl0 + qrl2), ¬ dups(qrl), ¬ dups(qrl2), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl2, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl2, ¬ ps[vs0[vhead1].next].next ∈ ps, vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next ≠ null ∧ vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].deqID.tid)) = Claimed (vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].deqID.tid, vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].val), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead ∈ qrl2, vtail0 = (vs0[vhead1].next ' + qrl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a '.last = a; (s) ⊦ x + [] = x; (s) ⊦ a '.butlast = []; (s) ⊦ some?(opttid) → opttid ≠ None; (as) ⊦ ¬ a ∈ []; (s) ⊦ [] + x = x; (s) ⊦ ¬ dups([]); (s) ⊦ markedp([], vs); (s)

Proof step 61 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, rvowns(vretRefs0(vs0[vs0[vhead1].next].deqID.tid)) = Claimed(vs0[vs0[vhead1].next].deqID.tid, vs0[vs0[vhead1].next].val), ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, phead ≠ vs0[vhead1].next, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, ps, owns0), listp(vs0[vhead1].next ' + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vhead1].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vs0[vhead1].next].deqID), unmarkedp(qrl, vs0), unmarkedp(qrl, ps), vs0[vhead1].next ∈ ps, vhead1 ∈ ps, vhead1 ∈ vs0, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ ps[vs0[vhead1].next].next ∈ ps, vtail0 = (vs0[vhead1].next ' + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ⊦

Heuristic elimination applied rule apply elim lemma

And obtained the following new goals

Rule argument

spec: opttid, name:elim-Some, seq: ⊦ some?(opttid) → (tid = opttid.tid ↔ opttid = Some(tid)), subst:[opttid, tid] → [vs0[vs0[vhead1].next].deqID, tid]

Additional information

Used simplifier rules: ⊦ some?(opttid) → (tid = opttid.tid ↔ opttid = Some(tid)); (elim) Used terms: vs0[vs0[vhead1].next].deqID, tid

Proof step 62 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, vs0[vs0[vhead1].next].deqID = Some(tid), ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, rvowns(vretRefs0(tid)) = Claimed(tid, vs0[vs0[vhead1].next].val), ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, phead ≠ vs0[vhead1].next, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vhead1].next ' + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vs0[vhead1].next].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), unmarkedp(qrl, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ ps[vs0[vhead1].next].next ∈ ps, phead ∈ qrl0 ∨ phead = vhead1, vtail0 = (vs0[vhead1].next ' + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ some?(Some(tid)); (s)

Proof step 63 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, vs0[vs0[vhead1].next].deqID = Some(tid), ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, rvowns(vretRefs0(tid)) = Claimed(tid, vs0[vs0[vhead1].next].val), ps[vs0[vhead1].next].deqID ≠ None, ps[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ null, phead ≠ vs0[vhead1].next, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vhead1].next ' + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), unmarkedp(qrl, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, ¬ dups(qrl), ¬ dups(qrl0), ¬ dups(qrl0 + qrl), ¬ weak?(vs0[vhead1].deqID), ¬ ps[vs0[vhead1].next].next ∈ ps, ¬ vhead1 ∈ qrl, ¬ vs0[vhead1].next ∈ qrl, ¬ vs0[vhead1].next ∈ qrl0, ¬ vhead1 ∈ qrl0, phead ∈ qrl0 ∨ phead = vhead1, vtail0 = (vs0[vhead1].next ' + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

User applied rule constructor cut

And obtained the following new goals

Rule argument

var: qrl

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ unmarkedp([], vs); (s) ⊦ x + [] = x; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s)

Proof step 64 from proof Crash-Theorem (previous step) (back to tree)

vs0[vs0[vhead1].next].deqID = Some(tid), ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, qrl = vhead + qrl3, rvowns(vretRefs0(tid)) = Claimed(tid, vs0[vs0[vhead1].next].val), ps[vs0[vhead1].next].deqID ≠ None, ps[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ null, phead ≠ vs0[vhead1].next, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, ps, owns0), listp(vs0[vhead1].next ' + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vhead1].deqID), some?(ps[vs0[vhead1].next].deqID), unmarkedp(qrl, vs0), unmarkedp(qrl, ps), vs0[vhead1].next ∈ ps, vhead1 ∈ ps, vhead1 ∈ vs0, ¬ dups(qrl), ¬ dups(qrl0), ¬ dups(qrl0 + qrl), ¬ weak?(vs0[vhead1].deqID), ¬ ps[vs0[vhead1].next].next ∈ ps, ¬ vhead1 ∈ qrl, ¬ vs0[vhead1].next ∈ qrl, ¬ vs0[vhead1].next ∈ qrl0, ¬ vhead1 ∈ qrl0, vtail0 = (vs0[vhead1].next ' + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ ¬ a ∈ st → st[a] = ∅[a]; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] listp-twoin: ⊦ listp(rl + r ' + r0 ' + rl0, vs, ps, owns) → ps[r].next = r0; (forward) from specification DMSQ ⊦ ¬ weak?(Some(tid)); (s) ⊦ some?(Some(tid)); (s) ⊦ None ≠ Some(tid); (s) a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a + x = a ' + x; (s) ⊦ a ' + x ≠ []; (s) ⊦ (b ' + a ' + x).last = (a ' + x).last; (s) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s)

Proof step 65 from proof Crash-Theorem (previous step) (back to tree)

vs0[vs0[vs0[vhead1].next].next].deqID = None, ps[vs0[vhead1].next].deqID = Some(tid), ps[vhead1].deqID = vs0[vhead1].deqID, vs0[vs0[vhead1].next].deqID = Some(tid), ∅[vs0[vs0[vhead1].next].next].deqID = None, ps[vs0[vhead1].next].next = vs0[vs0[vhead1].next].next, ps[vhead1].next = vs0[vhead1].next, ps[vs0[vhead1].next].val = vs0[vs0[vhead1].next].val, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vs0[vhead1].next) = None, owns0(vhead1) = None, rvowns(vretRefs0(tid)) = Claimed(tid, vs0[vs0[vhead1].next].val), vs0[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ null, phead ≠ vs0[vhead1].next, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ null, vhead1 ≠ vs0[vs0[vhead1].next].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vs0[vhead1].next].next ' + qrl3, vs0, ps, owns0), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl3, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl3, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead1].deqID), unmarkedp(qrl3, ps), unmarkedp(qrl3, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, vs0[vhead1].next ∈ vs0, ¬ dups(qrl0 + qrl3), ¬ dups(qrl0), ¬ dups(qrl3), ¬ weak?(vs0[vhead1].deqID), ¬ vs0[vhead1].next ∈ qrl0, ¬ vhead1 ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ qrl3, ¬ vs0[vs0[vhead1].next].next ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ ps, ¬ vhead1 ∈ qrl3, ¬ vs0[vhead1].next ∈ qrl3, phead ∈ qrl0 ∨ phead = vhead1, vtail0 = (vs0[vs0[vhead1].next].next ' + qrl3).last ∨ vtail0 = (vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl3).butlast).last ∧ vs0[(vs0[vs0[vhead1].next].next ' + qrl3).last].deqID = None ∧ vs0[(vs0[vs0[vhead1].next].next ' + qrl3).last].next = null ⊦

User applied rule constructor cut

And closed the goal!

Rule argument

var: qrl3

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ unmarkedp([], vs); (s) ⊦ x + [] = x; (s) ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s)

Proof step 66 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, qrl2 = vhead + qrl1, ps[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, ps[vs0[vhead1].next].next ≠ null, phead ≠ (vs0[vhead1].next ' + qrl2).last, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vhead1].next ' + qrl2 + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl2, vs0), markedp(qrl0, ps), markedp(qrl2, ps), nonullRetRefs(qrl0 + vhead1 ' + (vs0[vhead1].next ' + qrl2) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vs0[vhead1].next].deqID), some?(ps[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), unmarkedp(qrl, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, ¬ dups(qrl0 + qrl2 + qrl), ¬ dups(qrl0), ¬ dups(qrl2 + qrl), ¬ dups(qrl0 + qrl2), ¬ dups(qrl), ¬ dups(qrl2), ¬ weak?(vs0[vhead1].deqID), ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl2, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl2, ¬ ps[vs0[vhead1].next].next ∈ ps, vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next ≠ null ∧ vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].deqID.tid)) = Claimed (vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].deqID.tid, vs0[vs0[(vhead1 ' + (vs0[vhead1].next ' + qrl2).butlast).last].next].val), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead ∈ qrl2, vtail0 = (vs0[vhead1].next ' + qrl2 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vhead1].next ' + qrl2 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ ¬ a ∈ st → st[a] = ∅[a]; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] listp-twoin: ⊦ listp(rl + r ' + r0 ' + rl0, vs, ps, owns) → ps[r].next = r0; (forward) from specification DMSQ a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ (b ' + a ' + x).last = (a ' + x).last; (s) ⊦ (a ' + b ' + x).butlast = a ' + (b ' + x).butlast; (s) ⊦ a + x = a ' + x; (s) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ ¬ a ∈ x ∧ ¬ a ∈ y → (dups(x + a ' + y) ↔ dups(x + y)); (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s)

Proof step 67 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vs0[vhead1].next].deqID = vs0[vs0[vhead1].next].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vs0[vhead1].next].next = vs0[vs0[vhead1].next].next, ps[vhead1].val = vs0[vhead1].val, ps[vs0[vhead1].next].val = vs0[vs0[vhead1].next].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, owns0(vs0[vhead1].next) = None, vs0[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, vs0[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ vs0[vs0[vhead1].next].next, phead ≠ (vs0[vs0[vhead1].next].next ' + qrl1).last, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ vs0[vs0[vhead1].next].next, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1) + qrl, vs0, ps, owns0), listp(vs0[vs0[vhead1].next].next ' + qrl1 + qrl, vs0, ps, owns0), markedp(qrl1, ps), markedp(qrl0, ps), markedp(qrl1, vs0), markedp(qrl0, vs0), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vs0[vs0[vhead1].next].next].deqID), some?(vs0[vhead1].deqID), some?(vs0[vs0[vhead1].next].deqID), some?(∅[vs0[vs0[vhead1].next].next].deqID), unmarkedp(qrl, vs0), unmarkedp(qrl, ps), vs0[vhead1].next ∈ vs0, vs0[vhead1].next ∈ ps, vhead1 ∈ ps, vhead1 ∈ vs0, ¬ dups(qrl0 + qrl1 + qrl), ¬ dups(qrl0), ¬ dups(qrl1), ¬ dups(qrl), ¬ dups(qrl1 + qrl), ¬ dups(qrl0 + qrl1), ¬ weak?(vs0[vs0[vhead1].next].deqID), ¬ weak?(vs0[vhead1].deqID), ¬ vs0[vs0[vhead1].next].next ∈ ps, ¬ vs0[vs0[vhead1].next].next ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ qrl1, ¬ vs0[vs0[vhead1].next].next ∈ qrl, ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl1, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl1, vs0[(vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last].next ≠ null ∧ vs0[vs0[(vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last].next].deqID.tid)) = Claimed (vs0[vs0[(vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last].next].deqID.tid, vs0[vs0[(vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last].next].val), vtail0 = (vs0[vs0[vhead1].next].next ' + qrl1 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vs0[vhead1].next].next ' + qrl1 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead = vs0[vs0[vhead1].next].next ∨ phead ∈ qrl1 ⊦

User applied rule constructor cut

And obtained the following new goals

Rule argument

var: qrl1

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ (b ' + a ' + x).last = (a ' + x).last; (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist]

Proof step 68 from proof Crash-Theorem (previous step) (back to tree)

ps[vs0[vhead1].next].deqID = vs0[vs0[vhead1].next].deqID, ps[vhead1].deqID = vs0[vhead1].deqID, ps[vs0[vhead1].next].next = vs0[vs0[vhead1].next].next, ps[vhead1].next = vs0[vhead1].next, ps[vs0[vhead1].next].val = vs0[vs0[vhead1].next].val, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vs0[vhead1].next) = None, owns0(vhead1) = None, qrl1 = [], vs0[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, vs0[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ vs0[vs0[vhead1].next].next, phead ≠ (vs0[vs0[vhead1].next].next ' + qrl1).last, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ vs0[vs0[vhead1].next].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vs0[vhead1].next].next ' + qrl1 + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl1, vs0), markedp(qrl0, ps), markedp(qrl1, ps), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(∅[vs0[vs0[vhead1].next].next].deqID), some?(vs0[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), some?(vs0[vs0[vs0[vhead1].next].next].deqID), unmarkedp(qrl, ps), unmarkedp(qrl, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, vs0[vhead1].next ∈ vs0, ¬ dups(qrl0 + qrl1 + qrl), ¬ dups(qrl0), ¬ dups(qrl1), ¬ dups(qrl), ¬ dups(qrl1 + qrl), ¬ dups(qrl0 + qrl1), ¬ weak?(vs0[vs0[vhead1].next].deqID), ¬ weak?(vs0[vhead1].deqID), ¬ vs0[vs0[vhead1].next].next ∈ ps, ¬ vs0[vs0[vhead1].next].next ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ qrl1, ¬ vs0[vs0[vhead1].next].next ∈ qrl, ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl1, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, ¬ vhead1 ∈ qrl1, vs0[(vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last].next ≠ null ∧ vs0[vs0[(vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last].next].deqID.tid)) = Claimed (vs0[vs0[(vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last].next].deqID.tid, vs0[vs0[(vs0[vhead1].next ' + (vs0[vs0[vhead1].next].next ' + qrl1).butlast).last].next].val), phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ∨ phead = vs0[vs0[vhead1].next].next ∨ phead ∈ qrl1, vtail0 = (vs0[vs0[vhead1].next].next ' + qrl1 + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vs0[vhead1].next].next ' + qrl1 + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a '.last = a; (s) ⊦ x + [] = x; (s) ⊦ a '.butlast = []; (s) ⊦ some?(opttid) → opttid ≠ None; (as) ⊦ ¬ a ∈ []; (s) ⊦ [] + x = x; (s) ⊦ ¬ dups([]); (s) ⊦ markedp([], vs); (s)

Proof step 69 from proof Crash-Theorem (previous step) (back to tree)

ps[vhead1].deqID = vs0[vhead1].deqID, ps[vs0[vhead1].next].deqID = vs0[vs0[vhead1].next].deqID, ps[vhead1].next = vs0[vhead1].next, ps[vs0[vhead1].next].next = vs0[vs0[vhead1].next].next, ps[vhead1].val = vs0[vhead1].val, ps[vs0[vhead1].next].val = vs0[vs0[vhead1].next].val, dom vrvHp0 = dom prvHp, owns0(vhead1) = None, owns0(vs0[vhead1].next) = None, rvowns(vretRefs0(vs0[vs0[vs0[vhead1].next].next].deqID.tid)) = Claimed(vs0[vs0[vs0[vhead1].next].next].deqID.tid, vs0[vs0[vs0[vhead1].next].next].val), vs0[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, vs0[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ vs0[vs0[vhead1].next].next, phead ≠ vs0[vs0[vhead1].next].next, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ vs0[vs0[vhead1].next].next, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl, vs0, ps, owns0), listp(vs0[vs0[vhead1].next].next ' + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(vs0[vs0[vs0[vhead1].next].next].deqID), some?(vs0[vhead1].deqID), some?(vs0[vs0[vhead1].next].deqID), some?(∅[vs0[vs0[vhead1].next].next].deqID), unmarkedp(qrl, vs0), unmarkedp(qrl, ps), vs0[vhead1].next ∈ vs0, vs0[vhead1].next ∈ ps, vhead1 ∈ ps, vhead1 ∈ vs0, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ weak?(vs0[vs0[vhead1].next].deqID), ¬ weak?(vs0[vhead1].deqID), ¬ vs0[vs0[vhead1].next].next ∈ ps, ¬ vs0[vs0[vhead1].next].next ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ qrl, ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, vtail0 = (vs0[vs0[vhead1].next].next ' + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vs0[vhead1].next].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next ⊦

Heuristic elimination applied rule apply elim lemma

And obtained the following new goals

Rule argument

spec: opttid, name:elim-Some, seq: ⊦ some?(opttid) → (tid = opttid.tid ↔ opttid = Some(tid)), subst:[opttid, tid] → [vs0[vs0[vs0[vhead1].next].next].deqID, tid]

Additional information

Used simplifier rules: ⊦ some?(opttid) → (tid = opttid.tid ↔ opttid = Some(tid)); (elim) Used terms: vs0[vs0[vs0[vhead1].next].next].deqID, tid

Proof step 70 from proof Crash-Theorem (previous step) (back to tree)

ps[vs0[vhead1].next].deqID = vs0[vs0[vhead1].next].deqID, ps[vhead1].deqID = vs0[vhead1].deqID, vs0[vs0[vs0[vhead1].next].next].deqID = Some(tid), ps[vs0[vhead1].next].next = vs0[vs0[vhead1].next].next, ps[vhead1].next = vs0[vhead1].next, ps[vs0[vhead1].next].val = vs0[vs0[vhead1].next].val, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vs0[vhead1].next) = None, owns0(vhead1) = None, rvowns(vretRefs0(tid)) = Claimed(tid, vs0[vs0[vs0[vhead1].next].next].val), vs0[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ null, vs0[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ vs0[vs0[vhead1].next].next, phead ≠ vs0[vs0[vhead1].next].next, vhead1 ≠ null, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ vs0[vs0[vhead1].next].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vs0[vhead1].next].next ' + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(∅[vs0[vs0[vhead1].next].next].deqID), some?(vs0[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), some?(vs0[vs0[vs0[vhead1].next].next].deqID), unmarkedp(qrl, ps), unmarkedp(qrl, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, vs0[vhead1].next ∈ vs0, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ weak?(vs0[vs0[vhead1].next].deqID), ¬ weak?(vs0[vhead1].deqID), ¬ vs0[vs0[vhead1].next].next ∈ ps, ¬ vs0[vs0[vhead1].next].next ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ qrl, ¬ vhead1 ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl0, ¬ vs0[vhead1].next ∈ qrl, ¬ vhead1 ∈ qrl, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next, vtail0 = (vs0[vs0[vhead1].next].next ' + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vs0[vhead1].next].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ some?(Some(tid)); (s)

Proof step 71 from proof Crash-Theorem (previous step) (back to tree)

ps[vs0[vhead1].next].deqID = vs0[vs0[vhead1].next].deqID, ps[vhead1].deqID = vs0[vhead1].deqID, vs0[vs0[vs0[vhead1].next].next].deqID = Some(tid), ps[vs0[vhead1].next].next = vs0[vs0[vhead1].next].next, ps[vhead1].next = vs0[vhead1].next, ps[vs0[vhead1].next].val = vs0[vs0[vhead1].next].val, ps[vhead1].val = vs0[vhead1].val, dom vrvHp0 = dom prvHp, owns0(vs0[vhead1].next) = None, owns0(vhead1) = None, rvowns(vretRefs0(tid)) = Claimed(tid, vs0[vs0[vs0[vhead1].next].next].val), vs0[vs0[vhead1].next].deqID ≠ None, vs0[vhead1].next ≠ vs0[vs0[vhead1].next].next, vs0[vs0[vhead1].next].next ≠ null, vs0[vhead1].next ≠ null, phead ≠ vs0[vs0[vhead1].next].next, vhead1 ≠ vs0[vs0[vhead1].next].next, vhead1 ≠ vs0[vhead1].next, vhead1 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vs0[vs0[vhead1].next].next ' + qrl, vs0, ps, owns0), listp(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs(qrl0 + vhead1 ' + vs0[vhead1].next ' + vs0[vs0[vhead1].next].next ' + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(∅[vs0[vs0[vhead1].next].next].deqID), some?(vs0[vs0[vhead1].next].deqID), some?(vs0[vhead1].deqID), unmarkedp(qrl, ps), unmarkedp(qrl, vs0), vhead1 ∈ vs0, vhead1 ∈ ps, vs0[vhead1].next ∈ ps, vs0[vhead1].next ∈ vs0, ¬ dups(qrl), ¬ dups(qrl0), ¬ dups(qrl0 + qrl), ¬ weak?(vs0[vhead1].deqID), ¬ weak?(vs0[vs0[vhead1].next].deqID), ¬ vhead1 ∈ qrl, ¬ vs0[vhead1].next ∈ qrl, ¬ vs0[vhead1].next ∈ qrl0, ¬ vhead1 ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ qrl, ¬ vs0[vs0[vhead1].next].next ∈ qrl0, ¬ vs0[vs0[vhead1].next].next ∈ ps, phead ∈ qrl0 ∨ phead = vhead1 ∨ phead = vs0[vhead1].next, vtail0 = (vs0[vs0[vhead1].next].next ' + qrl).last ∨ qrl ≠ [] ∧ vtail0 = (vs0[vs0[vhead1].next].next ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦

User applied rule constructor cut

And closed the goal!

Rule argument

var: qrl

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ unmarkedp([], vs); (s) ⊦ x + [] = x; (s) ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s)

Proof step 72 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead2 = phead, vhead ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail0, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), vhead ∈ orl, vhead ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ vhead ∈ ps ⊦ ⟪vtail0 := vhead; while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 73 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead2 = phead, vtail0 = vhead, vhead ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), vhead ∈ ps, vhead ∈ orl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ vhead ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 74 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vtail0 = vhead, vhead ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), vhead ∈ orl, vhead ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule cut formula

And obtained the following new goals

Rule argument

fma: vhead = orl.last

Additional information

Used formulas: vhead = orl.last

Proof step 75 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead = orl.last, vtail0 = vhead, vhead ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), vhead ∈ ps, vhead ∈ orl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ x.last ∈ x ↔ x ≠ []; (s)

Proof step 76 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead = orl.last, vtail0 = orl.last, orl.last ≠ null, orl ≠ [], DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), orl.last ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

User applied rule invariant right

And obtained the following new goals

Rule argument

inv: vtail0 ∈ orl + qrl ∧ vtail0 ∈ ps ∧ vtail0 ≠ null ∧ vhead = orl.last ∧ vhead ∈ ps ∧ vhead ≠ null ∧ ( ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None) fmapos: right 1 bound: #(orl + qrl from vtail0)

Additional information

Used simplifier rules: ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ x.last ∈ x ↔ x ≠ []; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] Used formulas: vtail0 ∈ orl + qrl ∧ vtail0 ∈ ps ∧ vtail0 ≠ null ∧ vhead = orl.last ∧ vhead ∈ ps ∧ vhead ≠ null ∧ (ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None), #(orl + qrl from vtail0) = #(orl + qrl from vtail0)

Proof step 77 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead = orl.last, vtail0 = vtail1, vtail2 = orl.last, orl.last ≠ null, ps[vtail0].next ≠ null, orl ≠ [], vhead ≠ null, vtail0 ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), orl.last ∈ ps, vtail0 ∈ orl + qrl, vhead ∈ ps, vtail0 ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None ⊦ ⟪vtail0 := ps[vtail0].next⟫ ( ( vtail0 ∈ orl + qrl ∧ vtail0 ∈ ps ∧ vtail0 ≠ null ∧ vhead = orl.last ∧ vhead ∈ ps ∧ vhead ≠ null ∧ (ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None)) ∧ #(orl + qrl from vtail0) < #(orl + qrl from vtail1)), ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 78 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead = orl.last, vtail0 = vtail1, vtail2 = orl.last, orl.last ≠ null, ps[vtail0].next ≠ null, orl ≠ [], vhead ≠ null, vtail0 ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), vtail0 ∈ ps, vhead ∈ ps, vtail0 ∈ orl + qrl, orl.last ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, ( ps[vtail0].next ∈ orl + qrl ∧ ps[vtail0].next ∈ ps ∧ ps[vtail0].next ≠ null ∧ vhead = orl.last ∧ vhead ∈ ps ∧ vhead ≠ null ∧ (ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None)) ∧ #(orl + qrl from ps[vtail0].next) < #(orl + qrl from vtail1)

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s)

Proof step 79 from proof Crash-Theorem (previous step) (back to tree)

orl.last ≠ null, ps[vtail1].next ≠ null, orl ≠ [], vtail1 ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), orl.last ∈ ps, vtail1 ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, vtail1 ∈ orl ∨ vtail1 ∈ qrl ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, (ps[vtail1].next ∈ orl ∨ ps[vtail1].next ∈ qrl) ∧ ps[vtail1].next ∈ ps ∧ #(orl + qrl from ps[vtail1].next) < #(orl + qrl from vtail1)

User applied rule apply rewrite lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:DMSQGInv, seq:⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed (gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null), subst:[gs] → [mkgs (ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)], <all paths>

Additional information

Used terms: mkgs (ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)

Proof step 80 from proof Crash-Theorem (previous step) (back to tree)

⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed(gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null)

A spec-lemma

Proof step 81 from proof Crash-Theorem (previous step) (back to tree)

dom mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp = dom mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp, orl.last ≠ null, ps[vtail1].next ≠ null, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl ≠ [], orl ≠ [], vtail1 ≠ null, allocrv (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp), allocrv (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp), listp ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).owns), markedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps), markedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs), nonullRetRefs ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs), rvownerok (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns), rvownerok (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns), unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps), vtail1 ∈ ps, orl.last ∈ ps, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).phead ∈ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, ¬ dups ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ≠ null ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID ≠ None → mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID .tid)) = Claimed (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID .tid, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .val), vtail1 ∈ orl ∨ vtail1 ∈ qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.last = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ∧ ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ≠ [] → unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.tail, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs)) ∨ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.butlast ≠ [] ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.butlast.last = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).phead ≠ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.last ∧ unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs), mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vtail = ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ).last ∨ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ≠ [] ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vtail = ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.butlast ).last ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.last ] .deqID = None ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ).last ] .next = null ⊦ (ps[vtail1].next ∈ orl ∨ ps[vtail1].next ∈ qrl) ∧ ps[vtail1].next ∈ ps ∧ #(orl + qrl from ps[vtail1].next) < #(orl + qrl from vtail1), ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] list-elem-in-app11: ⊦ listp(orl + qrl, vs, ps, owns) ∧ node ∈ orl → node ∈ vs; (forward) from specification DMSQ list-elem-in-app12: ⊦ listp(orl + qrl, vs, ps, owns) ∧ node ∈ orl → node ∈ ps; (forward) from specification DMSQ ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vhead = phead0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vs = ps0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .rvowns = rvowns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vretRefs = vretRefs; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .qrl = qrl; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .orl = qrl0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .phead = phead; (s) ⊦ y ≠ [] → (x + y).last = y.last; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vtail = phead1; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vrvHp = vrvHp; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .prvHp = vrvHp0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .pretRefs = vretRefs0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .owns = owns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .ps = ps; (s)

Proof step 82 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, orl.last ≠ null, ps[vtail1].next ≠ null, orl ≠ [], vtail1 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(orl + qrl, vs0, ps, owns0), markedp(orl, vs0), markedp(orl, ps), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), unmarkedp(qrl, ps), phead ∈ orl, orl.last ∈ ps, vtail1 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail1 ∈ orl ∨ vtail1 ∈ qrl, vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0) ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, (ps[vtail1].next ∈ orl ∨ ps[vtail1].next ∈ qrl) ∧ ps[vtail1].next ∈ ps ∧ #(orl + qrl from ps[vtail1].next) < #(orl + qrl from vtail1)

User applied rule apply lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:listp-in-postfix, seq:⊦ listp(rl, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2, vs, ps, owns)), subst:[ps, rl, owns, r, vs] → [ps, orl + qrl, owns0, vtail1, vs0]

Additional information

Used terms: ps, orl + qrl, owns0, vtail1, vs0

Proof step 83 from proof Crash-Theorem (previous step) (back to tree)

⊦ listp(rl, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2, vs, ps, owns))

A spec-lemma

Proof step 84 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, orl.last ≠ null, ps[vtail1].next ≠ null, orl ≠ [], vtail1 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(orl + qrl, vs0, ps, owns0), markedp(orl, ps), markedp(orl, vs0), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail1 ∈ ps, orl.last ∈ ps, phead ∈ orl, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, listp(orl + qrl, vs0, ps, owns0) ∧ vtail1 ∈ orl + qrl → (∃ rl1, rl2. orl + qrl = rl1 + vtail1 ' + rl2 ∧ listp(vtail1 ' + rl2, vs0, ps, owns0)), orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0), vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail1 ∈ orl ∨ vtail1 ∈ qrl ⊦ (ps[vtail1].next ∈ orl ∨ ps[vtail1].next ∈ qrl) ∧ ps[vtail1].next ∈ ps ∧ #(orl + qrl from ps[vtail1].next) < #(orl + qrl from vtail1), ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s)

Proof step 85 from proof Crash-Theorem (previous step) (back to tree)

orl + qrl = rl1 + vtail1 ' + rl2, dom vrvHp0 = dom prvHp, orl.last ≠ null, ps[vtail1].next ≠ null, orl ≠ [], vtail1 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(orl + qrl, vs0, ps, owns0), listp(vtail1 ' + rl2, vs0, ps, owns0), markedp(orl, vs0), markedp(orl, ps), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), unmarkedp(qrl, ps), phead ∈ orl, orl.last ∈ ps, vtail1 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail1 ∈ orl ∨ vtail1 ∈ qrl, vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0) ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, (ps[vtail1].next ∈ orl ∨ ps[vtail1].next ∈ qrl) ∧ ps[vtail1].next ∈ ps ∧ #(orl + qrl from ps[vtail1].next) < #(orl + qrl from vtail1)

User applied rule constructor cut

And obtained the following new goals

Rule argument

var: rl2

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ x + [] = x; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s)

Proof step 86 from proof Crash-Theorem (previous step) (back to tree)

orl + qrl = rl1 + vtail1 ' + rl2, dom vrvHp0 = dom prvHp, rl2 = vtail0 + rl, orl.last ≠ null, ps[vtail1].next ≠ null, orl ≠ [], vtail1 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(vtail1 ' + rl2, vs0, ps, owns0), listp(orl + qrl, vs0, ps, owns0), markedp(orl, ps), markedp(orl, vs0), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail1 ∈ ps, orl.last ∈ ps, phead ∈ orl, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0), vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail1 ∈ orl ∨ vtail1 ∈ qrl ⊦ (ps[vtail1].next ∈ orl ∨ ps[vtail1].next ∈ qrl) ∧ ps[vtail1].next ∈ ps ∧ #(orl + qrl from ps[vtail1].next) < #(orl + qrl from vtail1), ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ a + x = a ' + x; (s)

Proof step 87 from proof Crash-Theorem (previous step) (back to tree)

orl + qrl = rl1 + vtail1 ' + vs0[vtail1].next ' + rl, ps[vtail1].next = vs0[vtail1].next, ps[vtail1].val = vs0[vtail1].val, dom vrvHp0 = dom prvHp, owns0(vtail1) = None, orl.last ≠ null, vs0[vtail1].next ≠ null, orl ≠ [], vtail1 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(vs0[vtail1].next ' + rl, vs0, ps, owns0), listp(orl + qrl, vs0, ps, owns0), markedp(orl, ps), markedp(orl, vs0), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), unmarkedp(qrl, ps), vtail1 ∈ vs0, phead ∈ ps, phead ∈ vs0, vtail1 ∈ ps, orl.last ∈ ps, phead ∈ orl, ¬ dups(qrl), ¬ dups(orl), ¬ dups(orl + qrl), ¬ weak?(vs0[vtail1].deqID), vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[vtail1].deqID = vs0[vtail1].deqID ∨ ps[vtail1].deqID = None, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0), vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail1 ∈ orl ∨ vtail1 ∈ qrl ⊦ (vs0[vtail1].next ∈ orl ∨ vs0[vtail1].next ∈ qrl) ∧ vs0[vtail1].next ∈ ps ∧ #(orl + qrl from vs0[vtail1].next) < #(orl + qrl from vtail1), ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

User applied rule subst equation

And obtained the following new goals

Rule argument

pos: left 1, flags:

Additional information

Proof step 88 from proof Crash-Theorem (previous step) (back to tree)

orl + qrl = rl1 + vtail1 ' + vs0[vtail1].next ' + rl, ps[vtail1].next = vs0[vtail1].next, ps[vtail1].val = vs0[vtail1].val, dom vrvHp0 = dom prvHp, owns0(vtail1) = None, orl.last ≠ null, vs0[vtail1].next ≠ null, orl ≠ [], vtail1 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(rl1 + vtail1 ' + vs0[vtail1].next ' + rl, vs0, ps, owns0), listp(vs0[vtail1].next ' + rl, vs0, ps, owns0), markedp(orl, vs0), markedp(orl, ps), nonullRetRefs(rl1 + vtail1 ' + vs0[vtail1].next ' + rl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), unmarkedp(qrl, ps), phead ∈ orl, orl.last ∈ ps, vtail1 ∈ ps, phead ∈ vs0, phead ∈ ps, vtail1 ∈ vs0, ¬ dups(qrl), ¬ dups(orl), ¬ dups(rl1 + vtail1 ' + vs0[vtail1].next ' + rl), ¬ weak?(vs0[vtail1].deqID), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail1 ∈ orl ∨ vtail1 ∈ qrl, vtail = (rl1 + vtail1 ' + vs0[vtail1].next ' + rl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0), ps[vtail1].deqID = vs0[vtail1].deqID ∨ ps[vtail1].deqID = None ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, (vs0[vtail1].next ∈ orl ∨ vs0[vtail1].next ∈ qrl) ∧ vs0[vtail1].next ∈ ps ∧ #(rl1 + vtail1 ' + vs0[vtail1].next ' + rl from vs0[vtail1].next) < #(rl1 + vtail1 ' + vs0[vtail1].next ' + rl from vtail1)

User applied rule apply rewrite lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:from-append-snd, seq: ⊦ ¬ r ∈ orl → orl + qrl from r = qrl from r, subst:[orl, qrl, r] → [rl1, vtail1 ' + vs0[vtail1].next ' + rl, vs0[vtail1].next], <all paths>

Additional information

Used simplifier rules: ⊦ a ∈ x → dups(x + b ' + a ' + y); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] Used terms: rl1, vtail1 ' + vs0[vtail1].next ' + rl, vs0[vtail1].next

Proof step 89 from proof Crash-Theorem (previous step) (back to tree)

⊦ ¬ r ∈ orl → orl + qrl from r = qrl from r

A spec-lemma

Proof step 90 from proof Crash-Theorem (previous step) (back to tree)

orl + qrl = rl1 + vtail1 ' + vs0[vtail1].next ' + rl, ps[vtail1].next = vs0[vtail1].next, ps[vtail1].val = vs0[vtail1].val, dom vrvHp0 = dom prvHp, owns0(vtail1) = None, orl.last ≠ null, vs0[vtail1].next ≠ null, orl ≠ [], vtail1 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(vs0[vtail1].next ' + rl, vs0, ps, owns0), listp(rl1 + vtail1 ' + vs0[vtail1].next ' + rl, vs0, ps, owns0), markedp(orl, ps), markedp(orl, vs0), nonullRetRefs(rl1 + vtail1 ' + vs0[vtail1].next ' + rl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), unmarkedp(qrl, ps), vtail1 ∈ vs0, phead ∈ ps, phead ∈ vs0, vtail1 ∈ ps, orl.last ∈ ps, phead ∈ orl, ¬ dups(qrl), ¬ dups(orl), ¬ dups(rl1 + vtail1 ' + vs0[vtail1].next ' + rl), ¬ weak?(vs0[vtail1].deqID), ¬ vs0[vtail1].next ∈ rl1, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[vtail1].deqID = vs0[vtail1].deqID ∨ ps[vtail1].deqID = None, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0), vtail = (rl1 + vtail1 ' + vs0[vtail1].next ' + rl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail1 ∈ orl ∨ vtail1 ∈ qrl ⊦ (vs0[vtail1].next ∈ orl ∨ vs0[vtail1].next ∈ qrl) ∧ vs0[vtail1].next ∈ ps ∧ #(vtail1 ' + vs0[vtail1].next ' + rl from vs0[vtail1].next) < #(rl1 + vtail1 ' + vs0[vtail1].next ' + rl from vtail1), ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] c: ⊦ m + n = n + m; from specification nat-basic a: ⊦ (m + n0) + n1 = m + n0 + n1; from specification nat-basic ⊦ m + n < m + n0 ↔ n < n0; (s) ⊦ m +1 = m + 1; (s) ⊦ #(a ' + x) = (# x)+1; (s) ⊦ r ' + rl from r = r ' + rl; (s) ⊦ ¬ r ∈ orl → orl + qrl from r = qrl from r; (s) ⊦ r0 ≠ r → r0 ' + rl from r = rl from r; (s) ⊦ (b ' + a ' + x).last = (a ' + x).last; (s) ⊦ (x + a ' + y).last = (a ' + y).last; (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s)

Proof step 91 from proof Crash-Theorem (previous step) (back to tree)

orl + qrl = rl1 + vtail1 ' + vs0[vtail1].next ' + rl, ps[vtail1].next = vs0[vtail1].next, ps[vtail1].val = vs0[vtail1].val, dom vrvHp0 = dom prvHp, owns0(vtail1) = None, orl.last ≠ null, vs0[vtail1].next ≠ null, orl ≠ [], vtail1 ≠ null, vtail1 ≠ vs0[vtail1].next, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(orl + qrl, vs0, ps, owns0), listp(vs0[vtail1].next ' + rl, vs0, ps, owns0), markedp(orl, vs0), markedp(orl, ps), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), unmarkedp(qrl, ps), phead ∈ orl, orl.last ∈ ps, vtail1 ∈ ps, phead ∈ vs0, phead ∈ ps, vtail1 ∈ vs0, ¬ dups(qrl), ¬ dups(orl), ¬ dups(rl1 + rl), ¬ dups(rl1), ¬ dups(rl), ¬ weak?(vs0[vtail1].deqID), ¬ vs0[vtail1].next ∈ rl1, ¬ vtail1 ∈ rl1, ¬ vtail1 ∈ rl, ¬ vs0[vtail1].next ∈ rl, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail1 ∈ orl ∨ vtail1 ∈ qrl, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0), ps[vtail1].deqID = vs0[vtail1].deqID ∨ ps[vtail1].deqID = None, vtail = (vs0[vtail1].next ' + rl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, (vs0[vtail1].next ∈ orl ∨ vs0[vtail1].next ∈ qrl) ∧ vs0[vtail1].next ∈ ps

User applied rule constructor cut

And closed the goal!

Rule argument

var: rl

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ a + x = a ' + x; (s) ⊦ (b ' + a ' + x).last = (a ' + x).last; (s) notin-forward: ⊦ rl + rl0 = rl1 + rl2 + rl3 ∧ ¬ r ∈ rl ∧ ¬ r ∈ rl0 → ¬ r ∈ rl1 ∧ ¬ r ∈ rl2 ∧ ¬ r ∈ rl3; (forward) from specification DMSQ a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ a ≠ b → (a ∈ b ' + x ↔ a ∈ x); (s) ⊦ a ∈ b ' ↔ a = b; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s) ⊦ ¬ dups([]); (s) ⊦ ¬ a ∈ []; (s) ⊦ a '.last = a; (s) ni: ⊦ x + a ' = y + z ∧ ¬ a ∈ z → z = []; (forward) from specification list[reflist] in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist] ⊦ x + [] = x; (s)

Proof step 92 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead = orl.last, vtail2 = orl.last, orl.last ≠ null, orl ≠ [], vhead ≠ null, vtail0 ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), orl.last ∈ ps, vtail0 ∈ orl + qrl, vhead ∈ ps, vtail0 ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None ⊦ ⟪gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

Heuristic symbolic execution applied rule assign right

And obtained the following new goals

Rule argument

Additional information

Proof step 93 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vhead = orl.last, vtail2 = orl.last, orl.last ≠ null, orl ≠ [], vhead ≠ null, vtail0 ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), vtail0 ∈ ps, vhead ∈ ps, vtail0 ∈ orl + qrl, orl.last ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, DMSQGInv((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)) ∧ ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).ps = ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vs ∧ ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vs = ps ∧ ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).phead = phead ∧ ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vretRefs = pretRefs ∧ ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vrvHp = prvHp ∧ ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).qrl = qrl ∧ ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).orl = orl ∧ ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).owns = (λ r. None) ∧ ( ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vs [((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vhead ] .next ≠ null → ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vs [((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vs [((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vhead ] .next ] .deqID = None) ∧ ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vhead = ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).orl.last ∧ ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).vtail = ( ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).orl + ((((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)).qrl ).last

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .qrl = qrl; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .owns:= owns0 = mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns0, rvowns); (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vrvHp:= vrvHp1 = mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp1, vrvHp0, qrl, qrl0, owns, rvowns); (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vretRefs:= vretRefs1 = mkgs (ps, ps0, phead, phead0, phead1, vretRefs1, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns); (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vtail:= phead4 = mkgs (ps, ps0, phead, phead0, phead4, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns); (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vhead:= phead3 = mkgs (ps, ps0, phead, phead3, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns); (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vs:= ps2 = mkgs (ps, ps2, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns); (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .orl = qrl0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vtail = phead1; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vhead = phead0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vs = ps0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .owns = owns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vrvHp = vrvHp; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vretRefs = vretRefs; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .phead = phead; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .ps = ps; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s)

Proof step 94 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, orl.last ≠ null, orl ≠ [], vtail0 ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), orl.last ∈ ps, vtail0 ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, vtail0 ∈ orl ∨ vtail0 ∈ qrl ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, DMSQGInv(mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns)) ∧ (ps[orl.last].next ≠ null → ps[ps[orl.last].next].deqID = None) ∧ vtail0 = (orl + qrl).last

User applied rule apply rewrite lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:DMSQGInv, seq:⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed (gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null), subst:[gs] → [mkgs (ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)], <all paths>

Additional information

Used terms: mkgs (ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)

Proof step 95 from proof Crash-Theorem (previous step) (back to tree)

⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed(gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null)

A spec-lemma

Proof step 96 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp = dom mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp, orl.last ≠ null, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl ≠ [], orl ≠ [], vtail0 ≠ null, allocrv (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp), allocrv (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp), listp ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).owns), markedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps), markedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs), nonullRetRefs ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs), rvownerok (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns), rvownerok (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns), unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps), vtail0 ∈ ps, orl.last ∈ ps, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).phead ∈ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, ¬ dups ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ≠ null ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID ≠ None → mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID .tid)) = Claimed (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID .tid, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .val), vtail0 ∈ orl ∨ vtail0 ∈ qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.last = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ∧ ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ≠ [] → unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.tail, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs)) ∨ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.butlast ≠ [] ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.butlast.last = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).phead ≠ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.last ∧ unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs), mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vtail = ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ).last ∨ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ≠ [] ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vtail = ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.butlast ).last ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.last ] .deqID = None ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ).last ] .next = null ⊦ DMSQGInv(mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns)) ∧ (ps[orl.last].next ≠ null → ps[ps[orl.last].next].deqID = None) ∧ vtail0 = (orl + qrl).last, ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] list-elem-in-app11: ⊦ listp(orl + qrl, vs, ps, owns) ∧ node ∈ orl → node ∈ vs; (forward) from specification DMSQ list-elem-in-app12: ⊦ listp(orl + qrl, vs, ps, owns) ∧ node ∈ orl → node ∈ ps; (forward) from specification DMSQ ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vhead = phead0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vs = ps0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .rvowns = rvowns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vretRefs = vretRefs; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .qrl = qrl; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .orl = qrl0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .phead = phead; (s) ⊦ y ≠ [] → (x + y).last = y.last; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vtail = phead1; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vrvHp = vrvHp; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .prvHp = vrvHp0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .pretRefs = vretRefs0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .owns = owns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .ps = ps; (s)

Proof step 97 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, orl.last ≠ null, orl ≠ [], vtail0 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(orl + qrl, vs0, ps, owns0), markedp(orl, vs0), markedp(orl, ps), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), unmarkedp(qrl, ps), phead ∈ orl, orl.last ∈ ps, vtail0 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail0 ∈ orl ∨ vtail0 ∈ qrl, vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0) ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, DMSQGInv(mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns)) ∧ (ps[orl.last].next ≠ null → ps[ps[orl.last].next].deqID = None) ∧ vtail0 = (orl + qrl).last

User applied rule apply rewrite lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:DMSQGInv, seq:⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed (gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null), subst:[gs] → [mkgs (ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns)], <all paths>

Additional information

Used terms: mkgs (ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns)

Proof step 98 from proof Crash-Theorem (previous step) (back to tree)

⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed(gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null)

A spec-lemma

Proof step 99 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, orl.last ≠ null, orl ≠ [], vtail0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(orl + qrl, vs0, ps, owns0), markedp(orl, ps), markedp(orl, vs0), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail0 ∈ ps, orl.last ∈ ps, phead ∈ orl, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0), vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ orl ∨ vtail0 ∈ qrl ⊦ ( listp ( mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl + mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).ps, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).owns) ∧ ¬ dups ( mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl + mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl) ∧ markedp (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).ps) ∧ markedp (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs) ∧ unmarkedp (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).ps) ∧ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl ≠ [] ∧ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).phead ∈ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl ∧ rvownerok (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vretRefs, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vrvHp, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).rvowns) ∧ allocrv (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vretRefs, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vrvHp) ∧ nonullRetRefs ( mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl + mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vretRefs, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).pretRefs) ∧ rvownerok (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).pretRefs, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).prvHp, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).rvowns) ∧ allocrv (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).pretRefs, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).prvHp) ∧ dom mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vrvHp = dom mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).prvHp ∧ ( mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vhead ] .next ≠ null ∧ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vhead ] .next ] .deqID ≠ None → mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).rvowns (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vretRefs (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vhead ] .next ] .deqID .tid)) = Claimed (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vhead ] .next ] .deqID .tid, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vhead ] .next ] .val)) ∧ ( mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl.last = mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vhead ∧ ( mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl ≠ [] → unmarkedp (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl.tail, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs)) ∨ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl.butlast ≠ [] ∧ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl.butlast.last = mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vhead ∧ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).phead ≠ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl.last ∧ unmarkedp (mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl, mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs)) ∧ ( mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vtail = ( mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl + mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl ).last ∨ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl ≠ [] ∧ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vtail = ( mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl + mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl.butlast ).last ∧ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl.last ] .deqID = None ∧ mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).vs [( mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).orl + mkgs(ps, ps, phead, orl.last, (orl + qrl).last, pretRefs, pretRefs, prvHp, prvHp, qrl, orl, λ r. None, rvowns).qrl ).last ] .next = null)) ∧ (ps[orl.last].next ≠ null → ps[ps[orl.last].next].deqID = None) ∧ vtail0 = (orl + qrl).last, ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .qrl = qrl; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .orl = qrl0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vtail = phead1; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vs = ps0; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ y ≠ [] → (x + y).last = y.last; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .phead = phead; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vhead = phead0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .rvowns = rvowns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vretRefs = vretRefs; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .prvHp = vrvHp0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vrvHp = vrvHp; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .pretRefs = vretRefs0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .ps = ps; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .owns = owns; (s)

Proof step 100 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, orl.last ≠ null, orl ≠ [], vtail0 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(orl + qrl, vs0, ps, owns0), markedp(orl, vs0), markedp(orl, ps), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), unmarkedp(qrl, ps), phead ∈ orl, orl.last ∈ ps, vtail0 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail0 ∈ orl ∨ vtail0 ∈ qrl, vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0) ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, listp(orl + qrl, ps, ps, λ r. None) ∧ nonullRetRefs(orl + qrl, ps, pretRefs, pretRefs) ∧ ((qrl ≠ [] → unmarkedp(qrl.tail, ps)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = orl.last ∧ phead ≠ orl.last) ∧ (ps[orl.last].next ≠ null → ps[ps[orl.last].next].deqID = None) ∧ vtail0 = (orl + qrl).last

User applied rule apply rewrite lemma

And obtained the following new goals

Rule argument

spec: <toplevel>, name:listp-crash, seq: ⊦ listp(rl, vs, ps, owns) → listp(rl, ps, ps, λ r. None), subst:[vs, owns, rl, ps] → [vs0, owns0, orl + qrl, ps], <all paths>

Additional information

Used terms: vs0, owns0, orl + qrl, ps

Proof step 101 from proof Crash-Theorem (previous step) (back to tree)

⊦ listp(rl, vs, ps, owns) → listp(rl, ps, ps, λ r. None)

Lemma listp-crash

Proof step 102 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, orl.last ≠ null, orl ≠ [], vtail0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(orl + qrl, vs0, ps, owns0), markedp(orl, ps), markedp(orl, vs0), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail0 ∈ ps, orl.last ∈ ps, phead ∈ orl, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0), vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ orl ∨ vtail0 ∈ qrl ⊦ true ∧ nonullRetRefs(orl + qrl, ps, pretRefs, pretRefs) ∧ ((qrl ≠ [] → unmarkedp(qrl.tail, ps)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = orl.last ∧ phead ≠ orl.last) ∧ (ps[orl.last].next ≠ null → ps[ps[orl.last].next].deqID = None) ∧ vtail0 = (orl + qrl).last, ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 103 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, orl.last ≠ null, orl ≠ [], vtail0 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(orl + qrl, vs0, ps, owns0), markedp(orl, vs0), markedp(orl, ps), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), unmarkedp(qrl, ps), phead ∈ orl, orl.last ∈ ps, vtail0 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps, ps[orl.last].next ≠ null → ps[orl.last].next ∈ ps ∧ ps[ps[orl.last].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail0 ∈ orl ∨ vtail0 ∈ qrl, vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0) ⊦ ps[orl.last].next ≠ null ∧ ps[ps[orl.last].next].deqID ≠ None, nonullRetRefs(orl + qrl, ps, pretRefs, pretRefs) ∧ ((qrl ≠ [] → unmarkedp(qrl.tail, ps)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = orl.last ∧ phead ≠ orl.last) ∧ (ps[orl.last].next ≠ null → ps[ps[orl.last].next].deqID = None) ∧ vtail0 = (orl + qrl).last

User applied rule apply elim lemma

And obtained the following new goals

Rule argument

spec: list-last[reflist], name:last-elim, seq:x ≠ [] ⊦ a = x.last ∧ y = x.butlast ↔ x = y + a ', subst:[x, a, y] → [orl, vhead, qrl0]

Additional information

Used simplifier rules: x ≠ [] ⊦ a = x.last ∧ y = x.butlast ↔ x = y + a '; (elim) Used terms: orl, vhead, y

Proof step 104 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, orl = qrl0 + vhead ', orl ≠ [], vhead ≠ null, vtail0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(orl + qrl, vs0, ps, owns0), markedp(orl, ps), markedp(orl, vs0), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail0 ∈ ps, vhead ∈ ps, phead ∈ orl, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0), vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ orl ∨ vtail0 ∈ qrl ⊦ nonullRetRefs(orl + qrl, ps, pretRefs, pretRefs) ∧ ((qrl ≠ [] → unmarkedp(qrl.tail, ps)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead ∧ phead ≠ vhead) ∧ (ps[vhead].next ≠ null → ps[ps[vhead].next].deqID = None) ∧ vtail0 = (orl + qrl).last, ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ (x + a ' + y).last = (a ' + y).last; (s) ⊦ a ∈ b ' ↔ a = b; (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ ¬ b ∈ x → (dups(x + b ') ↔ dups(x)); (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ x + a ' ≠ []; (s)

Proof step 105 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, vhead ≠ null, vtail0 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp((qrl0 + vhead ') + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs((qrl0 + vhead ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead].deqID), some?(ps[vhead].deqID), unmarkedp(qrl, ps), vhead ∈ ps, vtail0 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ vhead ∈ qrl0, ¬ vhead ∈ qrl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0), phead ∈ qrl0 ∨ phead = vhead, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None, nonullRetRefs((qrl0 + vhead ') + qrl, ps, pretRefs, pretRefs) ∧ ((qrl ≠ [] → unmarkedp(qrl.tail, ps)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead ∧ phead ≠ qrl0.last) ∧ (ps[vhead].next ≠ null → ps[ps[vhead].next].deqID = None) ∧ vtail0 = (vhead ' + qrl).last

User applied rule case distinction

And obtained the following new goals

Rule argument

pos: right 2

Additional information

Proof step 106 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, vhead ≠ null, vtail0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((qrl0 + vhead ') + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs((qrl0 + vhead ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead].deqID), some?(vs0[vhead].deqID), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail0 ∈ ps, vhead ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ nonullRetRefs((qrl0 + vhead ') + qrl, ps, pretRefs, pretRefs), ¬ vhead ∈ qrl0, ¬ vhead ∈ qrl, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, phead ∈ qrl0 ∨ phead = vhead, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0) ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ (x + a ' + y).head = (x + a ').head; (s) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ nonullRetRefs(rl, vs, vretRefs, pretRefs) → ( nonullRetRefs(rl0, vs0, vretRefs0, pretRefs0) ↔ (∀ r. ( r ∈ rl ∧ r ≠ rl.head ∧ vs[r].deqID ≠ None → vretRefs(vs[r].deqID.tid) ≠ null ∧ pretRefs(vs[r].deqID.tid) ≠ null) → r ∈ rl0 ∧ r ≠ rl0.head ∧ vs0[r].deqID ≠ None → vretRefs0(vs0[r].deqID.tid) ≠ null ∧ pretRefs0(vs0[r].deqID.tid) ≠ null)); (ss)

Proof step 107 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, pretRefs(ps[r].deqID.tid) = null, ps[r].deqID ≠ None, r ≠ (qrl0 + vhead ').head, vhead ≠ null, vtail0 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp((qrl0 + vhead ') + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs((qrl0 + vhead ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead].deqID), some?(ps[vhead].deqID), unmarkedp(qrl, ps), vhead ∈ ps, vtail0 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ vhead ∈ qrl0, ¬ vhead ∈ qrl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vs0[r].deqID ≠ None → vretRefs0(vs0[r].deqID.tid) ≠ null ∧ pretRefs(vs0[r].deqID.tid) ≠ null, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0), phead ∈ qrl0 ∨ phead = vhead, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, r ∈ qrl0 ∨ r = vhead ∨ r ∈ qrl ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic elimination applied rule apply elim lemma

And obtained the following new goals

Rule argument

spec: list-data[reflist], name:elim-list-c, seq:x ≠ [] ⊦ a = x.head ∧ y = x.tail ↔ x = a + y, subst:[x, a, y] → [qrl0 + vhead ', phead0, qrl1]

Additional information

Used simplifier rules: ⊦ x + y = [] ↔ x = [] ∧ y = []; (as) ⊦ a ' ≠ []; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] x ≠ [] ⊦ a = x.head ∧ y = x.tail ↔ x = a + y; (elim) Used terms: qrl0 + vhead ', phead0, y

Proof step 108 from proof Crash-Theorem (previous step) (back to tree)

qrl0 + vhead ' = phead0 + qrl1, ps[vtail0].next = null, dom vrvHp0 = dom prvHp, pretRefs(ps[r].deqID.tid) = null, qrl0 + vhead ' ≠ [], ps[r].deqID ≠ None, r ≠ phead0, vhead ≠ null, vtail0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((qrl0 + vhead ') + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs((qrl0 + vhead ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead].deqID), some?(vs0[vhead].deqID), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail0 ∈ ps, vhead ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ vhead ∈ qrl0, ¬ vhead ∈ qrl, vs0[r].deqID ≠ None → vretRefs0(vs0[r].deqID.tid) ≠ null ∧ pretRefs(vs0[r].deqID.tid) ≠ null, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, r ∈ qrl0 ∨ r = vhead ∨ r ∈ qrl, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, phead ∈ qrl0 ∨ phead = vhead, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0) ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ a ∈ b ' ↔ a = b; (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) in-14: ⊦ a ' + x = y → a ∈ y; (forward) from specification list[reflist] in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist] ⊦ x + a ' ≠ []; (s) ⊦ a + x = a ' + x; (s)

Proof step 109 from proof Crash-Theorem (previous step) (back to tree)

qrl0 + vhead ' = phead0 ' + qrl1, ps[vtail0].next = null, dom vrvHp0 = dom prvHp, pretRefs(ps[r].deqID.tid) = null, ps[r].deqID ≠ None, r ≠ phead0, vhead ≠ null, vtail0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((phead0 ' + qrl1) + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs((phead0 ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead].deqID), some?(vs0[vhead].deqID), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail0 ∈ ps, vhead ∈ ps, ¬ dups(qrl), ¬ dups(qrl0), ¬ dups(qrl0 + qrl), ¬ vhead ∈ qrl, ¬ vhead ∈ qrl0, vs0[r].deqID ≠ None → vretRefs0(vs0[r].deqID.tid) ≠ null ∧ pretRefs(vs0[r].deqID.tid) ≠ null, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, r ∈ qrl0 ∨ r = vhead ∨ r ∈ qrl, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, phead ∈ qrl0 ∨ phead = vhead, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0), vhead = phead0 ∨ vhead ∈ qrl1, phead0 ∈ qrl0 ∨ phead0 = vhead ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule apply lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:listp-in-postfix, seq:⊦ listp(rl, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2, vs, ps, owns)), subst:[ps, rl, owns, r, vs] → [ps, (qrl0 + vhead ') + qrl, owns0, r, vs0]

Additional information

Used terms: ps, (qrl0 + vhead ') + qrl, owns0, r, vs0

Proof step 110 from proof Crash-Theorem (previous step) (back to tree)

⊦ listp(rl, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2, vs, ps, owns))

A spec-lemma

Proof step 111 from proof Crash-Theorem (previous step) (back to tree)

qrl0 + vhead ' = phead0 ' + qrl1, ps[vtail0].next = null, dom vrvHp0 = dom prvHp, pretRefs(ps[r].deqID.tid) = null, ps[r].deqID ≠ None, r ≠ phead0, vhead ≠ null, vtail0 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp((phead0 ' + qrl1) + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs((phead0 ' + qrl1) + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead].deqID), some?(ps[vhead].deqID), unmarkedp(qrl, ps), vhead ∈ ps, vtail0 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(qrl), ¬ dups(qrl0), ¬ dups(qrl0 + qrl), ¬ vhead ∈ qrl, ¬ vhead ∈ qrl0, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vs0[r].deqID ≠ None → vretRefs0(vs0[r].deqID.tid) ≠ null ∧ pretRefs(vs0[r].deqID.tid) ≠ null, listp((qrl0 + vhead ') + qrl, vs0, ps, owns0) ∧ r ∈ (qrl0 + vhead ') + qrl → (∃ rl1, rl2. (qrl0 + vhead ') + qrl = rl1 + r ' + rl2 ∧ listp(r ' + rl2, vs0, ps, owns0)), phead0 ∈ qrl0 ∨ phead0 = vhead, vhead = phead0 ∨ vhead ∈ qrl1, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0), phead ∈ qrl0 ∨ phead = vhead, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, r ∈ qrl0 ∨ r = vhead ∨ r ∈ qrl ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] list-elem-in-app11: ⊦ listp(orl + qrl, vs, ps, owns) ∧ node ∈ orl → node ∈ vs; (forward) from specification DMSQ list-elem-in-app12: ⊦ listp(orl + qrl, vs, ps, owns) ∧ node ∈ orl → node ∈ ps; (forward) from specification DMSQ ⊦ ¬ a ∈ y → (y + c ' + z = a ' + z0 ↔ y = [] ∧ c = a ∧ z = z0); (ws) ⊦ a ≠ b → (a ∈ b ' + x ↔ a ∈ x); (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) in-14: ⊦ a ' + x = y → a ∈ y; (forward) from specification list[reflist] ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s)

Proof step 112 from proof Crash-Theorem (previous step) (back to tree)

(phead0 ' + qrl1) + qrl = rl1 + r ' + rl2, qrl0 + vhead ' = phead0 ' + qrl1, ps[vtail0].next = null, dom vrvHp0 = dom prvHp, pretRefs(ps[r].deqID.tid) = null, ps[r].deqID ≠ None, r ≠ phead0, vhead ≠ null, vtail0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(r ' + rl2, vs0, ps, owns0), listp(rl1 + r ' + rl2, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs(rl1 + r ' + rl2, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead].deqID), some?(vs0[vhead].deqID), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail0 ∈ ps, vhead ∈ ps, phead0 ∈ rl1, phead0 ∈ vs0, phead0 ∈ ps, ¬ dups(qrl), ¬ dups(qrl0), ¬ dups(qrl0 + qrl), ¬ vhead ∈ qrl, ¬ vhead ∈ qrl0, vs0[r].deqID ≠ None → vretRefs0(vs0[r].deqID.tid) ≠ null ∧ pretRefs(vs0[r].deqID.tid) ≠ null, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, r ∈ qrl0 ∨ r = vhead ∨ r ∈ qrl, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, phead ∈ qrl0 ∨ phead = vhead, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0), vhead = phead0 ∨ vhead ∈ qrl1, phead0 ∈ qrl0 ∨ phead0 = vhead ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule constructor cut

And closed the goal!

Rule argument

var: rl2

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ a + x = a ' + x; (s) in-14: ⊦ a ' + x = y → a ∈ y; (forward) from specification list[reflist] ⊦ a ∈ x → a ∈ x + y; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s) ⊦ x + [] = x; (s) lapp-04: ⊦ a ' + x = y + b ' ∧ a ≠ b → x.last = b; (forward) from specification list-last[reflist] inapp-10: ⊦ a ' + x = y + b ' ∧ a ≠ b → a ∈ y ∧ b ∈ x; (forward) from specification list[reflist] in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist] ⊦ a ≠ b → (a ∈ b ' + x ↔ a ∈ x); (s) ⊦ x.last ∈ x ↔ x ≠ []; (s)

Proof step 113 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, qrl ≠ [], vhead ≠ null, vtail0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((qrl0 + vhead ') + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs((qrl0 + vhead ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead].deqID), some?(vs0[vhead].deqID), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail0 ∈ ps, vhead ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ unmarkedp(qrl.tail, ps), ¬ vhead ∈ qrl0, ¬ vhead ∈ qrl, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, phead ∈ qrl0 ∨ phead = vhead, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0) ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None, qrl0 ≠ [] ∧ qrl0.last = vhead ∧ phead ≠ qrl0.last

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ x ≠ [] → (a ' + x).last = x.last; (s)

Proof step 114 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, qrl ≠ [], vhead ≠ null, vtail0 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp((qrl0 + vhead ') + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs((qrl0 + vhead ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead].deqID), some?(ps[vhead].deqID), unmarkedp(qrl, ps), vhead ∈ ps, vtail0 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ unmarkedp(qrl.tail, ps), ¬ vhead ∈ qrl0, ¬ vhead ∈ qrl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), phead ∈ qrl0 ∨ phead = vhead, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, vhead = vhead0 ∧ unmarkedp(qrl.tail, vs0) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0), vtail = qrl.last ∨ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ qrl0 ≠ [] ∧ qrl0.last = vhead ∧ phead ≠ qrl0.last, ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic elimination applied rule apply elim lemma

And closed the goal!

Rule argument

spec: list-data[reflist], name:elim-add, seq: ⊦ x ≠ [] → (head = x.head ∧ tail = x.tail ↔ x = head + tail), subst:[x, head, tail] → [qrl, phead0, qrl1]

Additional information

Used simplifier rules: ⊦ a ' + x ≠ []; (s) ⊦ a + x = a ' + x; (s) ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ x ≠ [] → (head = x.head ∧ tail = x.tail ↔ x = head + tail); (elim) Used terms: qrl, head, qrl1

Proof step 115 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, vhead ≠ null, vtail0 ≠ (vhead ' + qrl).last, vtail0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((qrl0 + vhead ') + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs((qrl0 + vhead ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead].deqID), some?(vs0[vhead].deqID), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail0 ∈ ps, vhead ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ vhead ∈ qrl0, ¬ vhead ∈ qrl, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, phead ∈ qrl0 ∨ phead = vhead, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0) ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule apply lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:listp-in-postfix, seq:⊦ listp(rl, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2, vs, ps, owns)), subst:[ps, rl, owns, r, vs] → [ps, (qrl0 + vhead ') + qrl, owns0, vtail0, vs0]

Additional information

Used terms: ps, (qrl0 + vhead ') + qrl, owns0, vtail0, vs0

Proof step 116 from proof Crash-Theorem (previous step) (back to tree)

⊦ listp(rl, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2, vs, ps, owns))

A spec-lemma

Proof step 117 from proof Crash-Theorem (previous step) (back to tree)

ps[vtail0].next = null, dom vrvHp0 = dom prvHp, vhead ≠ null, vtail0 ≠ (vhead ' + qrl).last, vtail0 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp((qrl0 + vhead ') + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs((qrl0 + vhead ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead].deqID), some?(ps[vhead].deqID), unmarkedp(qrl, ps), vhead ∈ ps, vtail0 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ vhead ∈ qrl0, ¬ vhead ∈ qrl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), listp((qrl0 + vhead ') + qrl, vs0, ps, owns0) ∧ vtail0 ∈ (qrl0 + vhead ') + qrl → (∃ rl1, rl2. (qrl0 + vhead ') + qrl = rl1 + vtail0 ' + rl2 ∧ listp(vtail0 ' + rl2, vs0, ps, owns0)), vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0), phead ∈ qrl0 ∨ phead = vhead, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s)

Proof step 118 from proof Crash-Theorem (previous step) (back to tree)

(qrl0 + vhead ') + qrl = rl1 + vtail0 ' + rl2, ps[vtail0].next = null, dom vrvHp0 = dom prvHp, vhead ≠ null, vtail0 ≠ (vhead ' + qrl).last, vtail0 ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(rl1 + vtail0 ' + rl2, vs0, ps, owns0), listp(vtail0 ' + rl2, vs0, ps, owns0), markedp(qrl0, ps), markedp(qrl0, vs0), nonullRetRefs(rl1 + vtail0 ' + rl2, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead].deqID), some?(vs0[vhead].deqID), unmarkedp(qrl, ps), phead ∈ ps, phead ∈ vs0, vtail0 ∈ ps, vhead ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ vhead ∈ qrl0, ¬ vhead ∈ qrl, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, phead ∈ qrl0 ∨ phead = vhead, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0) ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule constructor cut

And obtained the following new goals

Rule argument

var: rl2

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ a + x = a ' + x; (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ ¬ listp(null ' + rl0, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist]

Proof step 119 from proof Crash-Theorem (previous step) (back to tree)

(qrl0 + vhead ') + qrl = rl1 + vtail0 ' + rl2, ps[vtail0].next = null, dom vrvHp0 = dom prvHp, rl2 = [], vhead ≠ null, vtail0 ≠ (vhead ' + qrl).last, vtail0 ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(vtail0 ' + rl2, vs0, ps, owns0), listp(rl1 + vtail0 ' + rl2, vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs(rl1 + vtail0 ' + rl2, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead].deqID), some?(ps[vhead].deqID), unmarkedp(qrl, ps), vhead ∈ ps, vtail0 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ vhead ∈ qrl0, ¬ vhead ∈ qrl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0), phead ∈ qrl0 ∨ phead = vhead, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist] ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s) ⊦ x + [] = x; (s)

Proof step 120 from proof Crash-Theorem (previous step) (back to tree)

(qrl0 + vhead ') + qrl = rl1 + vtail0 ', ps[vtail0].next = null, ps[vtail0].val = vs0[vtail0].val, dom vrvHp0 = dom prvHp, owns0(vtail0) = None, vhead ≠ null, vtail0 ≠ null, vtail0 ≠ (vhead ' + qrl).last, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(rl1 + vtail0 ', vs0, ps, owns0), markedp(qrl0, vs0), markedp(qrl0, ps), nonullRetRefs(rl1 + vtail0 ', vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead].deqID), some?(ps[vhead].deqID), unmarkedp(qrl, ps), vtail0 ∈ vs0, vhead ∈ ps, vtail0 ∈ ps, phead ∈ vs0, phead ∈ ps, ¬ dups(qrl), ¬ dups(qrl0), ¬ dups(qrl0 + qrl), ¬ weak?(vs0[vtail0].deqID), ¬ vhead ∈ qrl, ¬ vhead ∈ qrl0, vs0[vtail0].next ≠ null → vs0[vtail0].next ∈ vs0 ∧ vs0[vtail0].next ∈ ps ∧ ps[vs0[vtail0].next] = vs0[vs0[vtail0].next] ∧ ps[vs0[vtail0].next].deqID = None ∧ ps[vs0[vtail0].next].next = null ∧ weak?(owns0(vs0[vtail0].next)), ps[vhead].next ≠ null → ps[vhead].next ∈ ps, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ∧ ps[ps[vhead].next].deqID = None, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vtail0].deqID = vs0[vtail0].deqID ∨ ps[vtail0].deqID = None, vhead = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ qrl0 ≠ [] ∧ qrl0.last = vhead0 ∧ phead ≠ vhead ∧ unmarkedp(qrl, vs0), phead ∈ qrl0 ∨ phead = vhead, vtail = (vhead ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, vtail0 ∈ qrl0 ∨ vtail0 = vhead ∨ vtail0 ∈ qrl ⊦ ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic elimination applied rule apply elim lemma

And closed the goal!

Rule argument

spec: list-data[reflist], name:elim-list-c, seq:x ≠ [] ⊦ a = x.head ∧ y = x.tail ↔ x = a + y, subst:[x, a, y] → [qrl, phead0, rl]

Additional information

Used simplifier rules: ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ unmarkedp(rl + rl0, vs) ↔ unmarkedp(rl, vs) ∧ unmarkedp(rl0, vs); (s) ⊦ unmarkedp(r ', vs) ↔ vs[r].deqID = None; (s) ⊦ (b ' + a ' + x).last = (a ' + x).last; (s) ⊦ a ' + x ≠ []; (s) ⊦ a + x = a ' + x; (s) a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ni: ⊦ x + a ' = y + z ∧ ¬ a ∈ z → z = []; (forward) from specification list[reflist] ⊦ a ∈ y → a ∈ x + y; (s) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) lapp-04: ⊦ a ' + x = y + b ' ∧ a ≠ b → x.last = b; (forward) from specification list-last[reflist] inapp-10: ⊦ a ' + x = y + b ' ∧ a ≠ b → a ∈ y ∧ b ∈ x; (forward) from specification list[reflist] ⊦ a ∈ x → (b ' + x).last = x.last; (s) in-14: ⊦ a ' + x = y → a ∈ y; (forward) from specification list[reflist] in-16: ⊦ x + a ' = y + b ' + z ∧ a ≠ b → a ∈ z; (forward) from specification list[reflist] inex: ⊦ x + a ' = y + b ' + z ∧ a ≠ b → (∃ z0. z = z0 + a '); (forward) from specification list[reflist] in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist] ⊦ # y = # y0 → (x + y = z + y0 ↔ x = z ∧ y = y0); (ws) ⊦ (x + a ').last = a; (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ x + a ' = y + b ' ↔ x = y ∧ a = b; (ws) ⊦ x + [] = x; (s) ⊦ a '.last = a; (s) x ≠ [] ⊦ a = x.head ∧ y = x.tail ↔ x = a + y; (elim) Used terms: qrl, phead0, y

Proof step 121 from proof Crash-Theorem (previous step) (back to tree)

gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vtail0 = vhead, vhead ≠ orl.last, vhead ≠ null, DMSQGInv(mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)), vhead ∈ ps, vhead ∈ orl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule apply rewrite lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:DMSQGInv, seq:⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed (gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null), subst:[gs] → [mkgs (ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)], <all paths>

Additional information

Used terms: mkgs (ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns)

Proof step 122 from proof Crash-Theorem (previous step) (back to tree)

⊦ DMSQGInv(gs) ↔ listp(gs.orl + gs.qrl, gs.vs, gs.ps, gs.owns) ∧ ¬ dups(gs.orl + gs.qrl) ∧ markedp(gs.orl, gs.ps) ∧ markedp(gs.orl, gs.vs) ∧ unmarkedp(gs.qrl, gs.ps) ∧ gs.orl ≠ [] ∧ gs.phead ∈ gs.orl ∧ rvownerok(gs.vretRefs, gs.vrvHp, gs.rvowns) ∧ allocrv(gs.vretRefs, gs.vrvHp) ∧ nonullRetRefs(gs.orl + gs.qrl, gs.vs, gs.vretRefs, gs.pretRefs) ∧ rvownerok(gs.pretRefs, gs.prvHp, gs.rvowns) ∧ allocrv(gs.pretRefs, gs.prvHp) ∧ dom gs.vrvHp = dom gs.prvHp ∧ ( gs.vs[gs.vhead].next ≠ null ∧ gs.vs[gs.vs[gs.vhead].next].deqID ≠ None → gs.rvowns(gs.vretRefs(gs.vs[gs.vs[gs.vhead].next].deqID.tid)) = Claimed(gs.vs[gs.vs[gs.vhead].next].deqID.tid, gs.vs[gs.vs[gs.vhead].next].val)) ∧ ( gs.orl.last = gs.vhead ∧ (gs.qrl ≠ [] → unmarkedp(gs.qrl.tail, gs.vs)) ∨ gs.orl.butlast ≠ [] ∧ gs.orl.butlast.last = gs.vhead ∧ gs.phead ≠ gs.orl.last ∧ unmarkedp(gs.qrl, gs.vs)) ∧ ( gs.vtail = (gs.orl + gs.qrl).last ∨ gs.qrl ≠ [] ∧ gs.vtail = (gs.orl + gs.qrl.butlast).last ∧ gs.vs[gs.qrl.last].deqID = None ∧ gs.vs[(gs.orl + gs.qrl).last].next = null)

A spec-lemma

Proof step 123 from proof Crash-Theorem (previous step) (back to tree)

dom mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp = dom mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vtail0 = vhead, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl ≠ [], vhead ≠ orl.last, vhead ≠ null, allocrv (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp), allocrv (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp), listp ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).owns), markedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps), markedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs), nonullRetRefs ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs), rvownerok (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vrvHp, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns), rvownerok (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).pretRefs, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).prvHp, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns), unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).ps), vhead ∈ orl, vhead ∈ ps, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).phead ∈ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl, ¬ dups ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl), ps[vhead].next ≠ null → ps[vhead].next ∈ ps, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ≠ null ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID ≠ None → mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).rvowns (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vretRefs (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID .tid)) = Claimed (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .deqID .tid, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ] .next ] .val), mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.last = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ∧ ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ≠ [] → unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.tail, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs)) ∨ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.butlast ≠ [] ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.butlast.last = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vhead ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).phead ≠ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl.last ∧ unmarkedp (mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl, mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs), mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vtail = ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ).last ∨ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ≠ [] ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vtail = ( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.butlast ).last ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl.last ] .deqID = None ∧ mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).vs [( mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).orl + mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns).qrl ).last ] .next = null ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] list-elem-in-app11: ⊦ listp(orl + qrl, vs, ps, owns) ∧ node ∈ orl → node ∈ vs; (forward) from specification DMSQ list-elem-in-app12: ⊦ listp(orl + qrl, vs, ps, owns) ∧ node ∈ orl → node ∈ ps; (forward) from specification DMSQ ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vhead = phead0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vs = ps0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .rvowns = rvowns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vretRefs = vretRefs; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .qrl = qrl; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .orl = qrl0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .phead = phead; (s) ⊦ y ≠ [] → (x + y).last = y.last; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vtail = phead1; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .vrvHp = vrvHp; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .prvHp = vrvHp0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .pretRefs = vretRefs0; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .owns = owns; (s) ⊦ mkgs (ps, ps0, phead, phead0, phead1, vretRefs, vretRefs0, vrvHp, vrvHp0, qrl, qrl0, owns, rvowns) .ps = ps; (s)

Proof step 124 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), vtail0 = vhead, orl ≠ [], vhead ≠ orl.last, vhead ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(orl + qrl, vs0, ps, owns0), markedp(orl, vs0), markedp(orl, ps), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), unmarkedp(qrl, ps), phead ∈ orl, vhead ∈ orl, phead ∈ vs0, vhead ∈ vs0, phead ∈ ps, vhead ∈ ps, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, orl.last = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl.butlast ≠ [] ∧ orl.butlast.last = vhead0 ∧ phead ≠ orl.last ∧ unmarkedp(qrl, vs0) ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule apply elim lemma

And obtained the following new goals

Rule argument

spec: list-last[reflist], name:last-elim, seq:x ≠ [] ⊦ a = x.last ∧ y = x.butlast ↔ x = y + a ', subst:[x, a, y] → [orl, vtail1, orl0]

Additional information

Used simplifier rules: x ≠ [] ⊦ a = x.last ∧ y = x.butlast ↔ x = y + a '; (elim) Used terms: orl, vtail1, y

Proof step 125 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl, owns0, rvowns), orl = orl0 + vtail1 ', vtail0 = vhead, orl ≠ [], vhead ≠ vtail1, vhead ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(orl + qrl, vs0, ps, owns0), markedp(orl, ps), markedp(orl, vs0), nonullRetRefs(orl + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), unmarkedp(qrl, ps), vhead ∈ ps, phead ∈ ps, vhead ∈ vs0, phead ∈ vs0, vhead ∈ orl, phead ∈ orl, ¬ dups(orl + qrl), ¬ dups(orl), ¬ dups(qrl), vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vtail1 = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl0 ≠ [] ∧ orl0.last = vhead0 ∧ phead ≠ vtail1 ∧ unmarkedp(qrl, vs0), vtail = (orl + qrl).last ∨ qrl ≠ [] ∧ vtail = (orl + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ a ∈ b ' ↔ a = b; (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist] ⊦ (x + a ' + y).last = (a ' + y).last; (s) ⊦ ¬ b ∈ x → (dups(x + b ') ↔ dups(x)); (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ x + a ' ≠ []; (s)

Proof step 126 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl0 + vtail1 ', owns0, rvowns), orl = orl0 + vtail1 ', vtail0 = vhead, vhead ≠ null, vhead ≠ vtail1, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((orl0 + vtail1 ') + qrl, vs0, ps, owns0), markedp(orl0, ps), markedp(orl0, vs0), nonullRetRefs((orl0 + vtail1 ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vtail1].deqID), some?(vs0[vtail1].deqID), unmarkedp(qrl, ps), vhead ∈ ps, phead ∈ ps, vhead ∈ vs0, phead ∈ vs0, vhead ∈ orl0, ¬ dups(qrl), ¬ dups(orl0), ¬ dups(orl0 + qrl), ¬ vtail1 ∈ qrl, ¬ vtail1 ∈ orl0, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vtail = (vtail1 ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vtail1 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ orl0 ∨ phead = vtail1, vtail1 = vhead0 ∧ (qrl ≠ [] → unmarkedp(qrl.tail, vs0)) ∨ orl0 ≠ [] ∧ orl0.last = vhead0 ∧ phead ≠ vtail1 ∧ unmarkedp(qrl, vs0) ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule case distinction

And obtained the following new goals

Rule argument

pos: left 32

Additional information

Proof step 127 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl0 + vtail1 ', owns0, rvowns), orl = orl0 + vtail1 ', vtail0 = vhead, vtail1 = vhead0, vhead ≠ null, vhead ≠ vtail1, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp((orl0 + vtail1 ') + qrl, vs0, ps, owns0), markedp(orl0, vs0), markedp(orl0, ps), nonullRetRefs((orl0 + vtail1 ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vtail1].deqID), some?(ps[vtail1].deqID), unmarkedp(qrl, ps), vhead ∈ orl0, phead ∈ vs0, vhead ∈ vs0, phead ∈ ps, vhead ∈ ps, ¬ dups(qrl), ¬ dups(orl0), ¬ dups(orl0 + qrl), ¬ vtail1 ∈ qrl, ¬ vtail1 ∈ orl0, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), qrl ≠ [] → unmarkedp(qrl.tail, vs0), phead ∈ orl0 ∨ phead = vtail1, vtail = (vtail1 ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vtail1 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ a ∈ b ' ↔ a = b; (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist]

Proof step 128 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl0 + vhead0 ', owns0, rvowns), orl = orl0 + vhead0 ', vtail0 = vhead, vhead ≠ vhead0, vhead ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp((orl0 + vhead0 ') + qrl, vs0, ps, owns0), markedp(orl0, vs0), markedp(orl0, ps), nonullRetRefs((orl0 + vhead0 ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead0].deqID), some?(ps[vhead0].deqID), unmarkedp(qrl, ps), vhead ∈ orl0, phead ∈ vs0, vhead ∈ vs0, phead ∈ ps, vhead ∈ ps, ¬ dups(orl0 + qrl), ¬ dups(orl0), ¬ dups(qrl), ¬ vhead0 ∈ orl0, ¬ vhead0 ∈ qrl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), qrl ≠ [] → unmarkedp(qrl.tail, vs0), phead ∈ orl0 ∨ phead = vhead0, vtail = (vhead0 ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead0 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule apply lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:listp-in-postfix1, seq:⊦ listp(rl + rl0, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2 + rl0, vs, ps, owns)), subst:[ps, rl, owns, r, rl0, vs] → [ps, orl0 + vhead0 ', owns0, vhead, qrl, vs0]

Additional information

Used terms: ps, orl0 + vhead0 ', owns0, vhead, qrl, vs0

Proof step 129 from proof Crash-Theorem (previous step) (back to tree)

⊦ listp(rl + rl0, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2 + rl0, vs, ps, owns))

A spec-lemma

Proof step 130 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl0 + vhead0 ', owns0, rvowns), orl = orl0 + vhead0 ', vtail0 = vhead, vhead ≠ vhead0, vhead ≠ null, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((orl0 + vhead0 ') + qrl, vs0, ps, owns0), markedp(orl0, ps), markedp(orl0, vs0), nonullRetRefs((orl0 + vhead0 ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead0].deqID), some?(vs0[vhead0].deqID), unmarkedp(qrl, ps), vhead ∈ ps, phead ∈ ps, vhead ∈ vs0, phead ∈ vs0, vhead ∈ orl0, ¬ dups(orl0 + qrl), ¬ dups(orl0), ¬ dups(qrl), ¬ vhead0 ∈ orl0, ¬ vhead0 ∈ qrl, qrl ≠ [] → unmarkedp(qrl.tail, vs0), vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps, listp((orl0 + vhead0 ') + qrl, vs0, ps, owns0) ∧ vhead ∈ orl0 + vhead0 ' → (∃ rl1, rl2. orl0 + vhead0 ' = rl1 + vhead ' + rl2 ∧ listp(vhead ' + rl2 + qrl, vs0, ps, owns0)), vtail = (vhead0 ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead0 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null, phead ∈ orl0 ∨ phead = vhead0 ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ a ≠ b → (a ∈ b ' + x ↔ a ∈ x); (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ a ∈ b ' ↔ a = b; (s) ⊦ x = y + a ' + z → a ∈ x; (s) ⊦ # y = # y0 → (x + y = z + y0 ↔ x = z ∧ y = y0); (ws) ⊦ a ∈ y → a ∈ x + y; (s) in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist] inex: ⊦ x + a ' = y + b ' + z ∧ a ≠ b → (∃ z0. z = z0 + a '); (forward) from specification list[reflist] in-16: ⊦ x + a ' = y + b ' + z ∧ a ≠ b → a ∈ z; (forward) from specification list[reflist] ⊦ a ∈ x → a ∈ x + y; (s)

Proof step 131 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, (rl1 + vhead ' + qrl0) + vhead0 ', owns0, rvowns), orl = (rl1 + vhead ' + qrl0) + vhead0 ', vtail0 = vhead, vhead ≠ vhead0, vhead ≠ null, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp(((rl1 + vhead ' + qrl0) + vhead0 ') + qrl, vs0, ps, owns0), listp(vhead ' + (qrl0 + vhead0 ') + qrl, vs0, ps, owns0), markedp(qrl0, vs0), markedp(rl1, vs0), markedp(qrl0, ps), markedp(rl1, ps), nonullRetRefs(((rl1 + vhead ' + qrl0) + vhead0 ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vhead].deqID), some?(ps[vhead].deqID), some?(vs0[vhead0].deqID), some?(ps[vhead0].deqID), unmarkedp(qrl, ps), phead ∈ vs0, vhead ∈ vs0, phead ∈ ps, vhead ∈ ps, ¬ dups(rl1 + qrl0), ¬ dups(qrl0 + qrl), ¬ dups(rl1), ¬ dups(rl1 + qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ vhead ∈ qrl, ¬ vhead ∈ qrl0, ¬ vhead ∈ rl1, ¬ vhead0 ∈ qrl0, ¬ vhead0 ∈ rl1, ¬ vhead0 ∈ qrl, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), qrl ≠ [] → unmarkedp(qrl.tail, vs0), phead ∈ rl1 ∨ phead = vhead ∨ phead ∈ qrl0 ∨ phead = vhead0, vtail = (vhead0 ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vhead0 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule constructor cut

And closed the goal!

Rule argument

var: qrl0

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ a + x = a ' + x; (s) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ some?(opttid) → opttid ≠ None; (as) listp-twoin: ⊦ listp(rl + r ' + r0 ' + rl0, vs, ps, owns) → ps[r].next = r0; (forward) from specification DMSQ ⊦ ¬ listp(null ' + rl0, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ markedp([], vs); (s) ⊦ [] + x = x; (s) ⊦ ¬ dups([]); (s) ⊦ ¬ a ∈ []; (s) ⊦ x + [] = x; (s) ⊦ listp(r ', vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = null ∧ ( vs[r].next ≠ null → vs[r].next ∈ vs ∧ vs[r].next ∈ ps ∧ ps[vs[r].next] = vs[vs[r].next] ∧ ps[vs[r].next].deqID = None ∧ ps[vs[r].next].next = null ∧ weak?(owns(vs[r].next))); (s) ⊦ x ≠ [] → (a ' + x).last = x.last; (s) ⊦ y ≠ [] → (x + y).last = y.last; (s) in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist] ⊦ a ≠ b → (a ∈ b ' + x ↔ a ∈ x); (s) ⊦ a ∈ b ' ↔ a = b; (s)

Proof step 132 from proof Crash-Theorem (previous step) (back to tree)

orl0.last = vhead0, dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, vhead0, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl0 + vtail1 ', owns0, rvowns), orl = orl0 + vtail1 ', vtail0 = vhead, orl0 ≠ [], phead ≠ vtail1, vhead ≠ null, vhead ≠ vtail1, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp((orl0 + vtail1 ') + qrl, vs0, ps, owns0), markedp(orl0, vs0), markedp(orl0, ps), nonullRetRefs((orl0 + vtail1 ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vtail1].deqID), some?(ps[vtail1].deqID), unmarkedp(qrl, ps), unmarkedp(qrl, vs0), vhead ∈ orl0, phead ∈ vs0, vhead ∈ vs0, phead ∈ ps, vhead ∈ ps, ¬ dups(qrl), ¬ dups(orl0), ¬ dups(orl0 + qrl), ¬ vtail1 ∈ qrl, ¬ vtail1 ∈ orl0, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vs0[vhead0].next ≠ null ∧ vs0[vs0[vhead0].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[vhead0].next].deqID.tid)) = Claimed(vs0[vs0[vhead0].next].deqID.tid, vs0[vs0[vhead0].next].val), phead ∈ orl0 ∨ phead = vtail1, vtail = (vtail1 ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vtail1 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Proof step 133 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, orl0.last, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl0 + vtail1 ', owns0, rvowns), orl = orl0 + vtail1 ', vtail0 = vhead, orl0 ≠ [], phead ≠ vtail1, vhead ≠ null, vhead ≠ vtail1, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp((orl0 + vtail1 ') + qrl, vs0, ps, owns0), markedp(orl0, ps), markedp(orl0, vs0), nonullRetRefs((orl0 + vtail1 ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vtail1].deqID), some?(vs0[vtail1].deqID), unmarkedp(qrl, vs0), unmarkedp(qrl, ps), vhead ∈ ps, phead ∈ ps, vhead ∈ vs0, phead ∈ vs0, vhead ∈ orl0, phead ∈ orl0, ¬ dups(qrl), ¬ dups(orl0), ¬ dups(orl0 + qrl), ¬ vtail1 ∈ qrl, ¬ vtail1 ∈ orl0, vs0[orl0.last].next ≠ null ∧ vs0[vs0[orl0.last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[orl0.last].next].deqID.tid)) = Claimed(vs0[vs0[orl0.last].next].deqID.tid, vs0[vs0[orl0.last].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vtail = (vtail1 ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vtail1 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule apply lemma

And obtained the following new goals

Rule argument

spec: DMSQ, name:listp-in-postfix1, seq:⊦ listp(rl + rl0, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2 + rl0, vs, ps, owns)), subst:[ps, rl, owns, r, rl0, vs] → [ps, orl0 + vtail1 ', owns0, vhead, qrl, vs0]

Additional information

Used terms: ps, orl0 + vtail1 ', owns0, vhead, qrl, vs0

Proof step 134 from proof Crash-Theorem (previous step) (back to tree)

⊦ listp(rl + rl0, vs, ps, owns) ∧ r ∈ rl → (∃ rl1, rl2. rl = rl1 + r ' + rl2 ∧ listp(r ' + rl2 + rl0, vs, ps, owns))

A spec-lemma

Proof step 135 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, orl0.last, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, orl0 + vtail1 ', owns0, rvowns), orl = orl0 + vtail1 ', vtail0 = vhead, orl0 ≠ [], phead ≠ vtail1, vhead ≠ null, vhead ≠ vtail1, allocrv(pretRefs, prvHp), allocrv(vretRefs0, vrvHp0), listp((orl0 + vtail1 ') + qrl, vs0, ps, owns0), markedp(orl0, vs0), markedp(orl0, ps), nonullRetRefs((orl0 + vtail1 ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(pretRefs, prvHp, rvowns), rvownerok(vretRefs0, vrvHp0, rvowns), some?(vs0[vtail1].deqID), some?(ps[vtail1].deqID), unmarkedp(qrl, ps), unmarkedp(qrl, vs0), phead ∈ orl0, vhead ∈ orl0, phead ∈ vs0, vhead ∈ vs0, phead ∈ ps, vhead ∈ ps, ¬ dups(qrl), ¬ dups(orl0), ¬ dups(orl0 + qrl), ¬ vtail1 ∈ qrl, ¬ vtail1 ∈ orl0, ps[vhead].next ≠ null → ps[vhead].next ∈ ps, vs0[orl0.last].next ≠ null ∧ vs0[vs0[orl0.last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[orl0.last].next].deqID.tid)) = Claimed(vs0[vs0[orl0.last].next].deqID.tid, vs0[vs0[orl0.last].next].val), listp((orl0 + vtail1 ') + qrl, vs0, ps, owns0) ∧ vhead ∈ orl0 + vtail1 ' → (∃ rl1, rl2. orl0 + vtail1 ' = rl1 + vhead ' + rl2 ∧ listp(vhead ' + rl2 + qrl, vs0, ps, owns0)), vtail = (vtail1 ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vtail1 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

Heuristic simplifier applied rule simplifier

And obtained the following new goals

Rule argument

Additional information

Used simplifier rules: a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ (x + a ' + y).last = (a ' + y).last; (s) ⊦ ¬ a ∈ x ∧ ¬ a ∈ y → (dups(x + a ' + y) ↔ dups(x + y)); (s) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ a ≠ b → (a ∈ b ' + x ↔ a ∈ x); (s) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ x + a ' + y ≠ []; (s) ⊦ a ∈ b ' ↔ a = b; (s) ⊦ x = y + a ' + z → a ∈ x; (s) ⊦ # y = # y0 → (x + y = z + y0 ↔ x = z ∧ y = y0); (ws) ⊦ a ∈ y → a ∈ x + y; (s) in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist] inex: ⊦ x + a ' = y + b ' + z ∧ a ≠ b → (∃ z0. z = z0 + a '); (forward) from specification list[reflist] in-16: ⊦ x + a ' = y + b ' + z ∧ a ≠ b → a ∈ z; (forward) from specification list[reflist] ⊦ a ∈ x → a ∈ x + y; (s)

Proof step 136 from proof Crash-Theorem (previous step) (back to tree)

dom vrvHp0 = dom prvHp, gs = mkgs(ps, vs0, phead, (vhead ' + qrl0).last, vtail, vretRefs0, pretRefs, vrvHp0, prvHp, qrl, (rl1 + vhead ' + qrl0) + vtail1 ', owns0, rvowns), orl = (rl1 + vhead ' + qrl0) + vtail1 ', vtail0 = vhead, phead ≠ vtail1, vhead ≠ null, vhead ≠ vtail1, allocrv(vretRefs0, vrvHp0), allocrv(pretRefs, prvHp), listp(((rl1 + vhead ' + qrl0) + vtail1 ') + qrl, vs0, ps, owns0), listp(vhead ' + (qrl0 + vtail1 ') + qrl, vs0, ps, owns0), markedp(qrl0, ps), markedp(rl1, ps), markedp(qrl0, vs0), markedp(rl1, vs0), nonullRetRefs(((rl1 + vhead ' + qrl0) + vtail1 ') + qrl, vs0, vretRefs0, pretRefs), rvownerok(vretRefs0, vrvHp0, rvowns), rvownerok(pretRefs, prvHp, rvowns), some?(ps[vhead].deqID), some?(vs0[vhead].deqID), some?(ps[vtail1].deqID), some?(vs0[vtail1].deqID), unmarkedp(qrl, vs0), unmarkedp(qrl, ps), vhead ∈ ps, phead ∈ ps, vhead ∈ vs0, phead ∈ vs0, ¬ dups(rl1 + qrl0), ¬ dups(qrl0 + qrl), ¬ dups(rl1), ¬ dups(rl1 + qrl0 + qrl), ¬ dups(qrl0), ¬ dups(qrl), ¬ vhead ∈ qrl, ¬ vhead ∈ qrl0, ¬ vhead ∈ rl1, ¬ vtail1 ∈ qrl, ¬ vtail1 ∈ qrl0, ¬ vtail1 ∈ rl1, vs0[(vhead ' + qrl0).last].next ≠ null ∧ vs0[vs0[(vhead ' + qrl0).last].next].deqID ≠ None → rvowns(vretRefs0(vs0[vs0[(vhead ' + qrl0).last].next].deqID.tid)) = Claimed(vs0[vs0[(vhead ' + qrl0).last].next].deqID.tid, vs0[vs0[(vhead ' + qrl0).last].next].val), ps[vhead].next ≠ null → ps[vhead].next ∈ ps, phead ∈ rl1 ∨ phead = vhead ∨ phead ∈ qrl0, vtail = (vtail1 ' + qrl).last ∨ qrl ≠ [] ∧ vtail = (vtail1 ' + qrl.butlast).last ∧ vs0[qrl.last].deqID = None ∧ vs0[qrl.last].next = null ⊦ ⟪while ps[vtail0].next ≠ null do vtail0 := ps[vtail0].next; gs := (((((gs.vs:= ps).vhead:= vhead).vtail:= vtail0).vretRefs:= pretRefs).vrvHp:= prvHp).owns:= (λ r. None)⟫ ( DMSQGInv(gs) ∧ gs.ps = gs.vs ∧ gs.vs = ps ∧ gs.phead = phead ∧ gs.vretRefs = pretRefs ∧ gs.vrvHp = prvHp ∧ gs.qrl = qrl ∧ gs.orl = orl ∧ gs.owns = (λ r. None) ∧ (gs.vs[gs.vhead].next ≠ null → gs.vs[gs.vs[gs.vhead].next].deqID = None) ∧ gs.vhead = gs.orl.last ∧ gs.vtail = (gs.orl + gs.qrl).last), ps[vhead].next ≠ null ∧ ps[ps[vhead].next].deqID ≠ None

User applied rule constructor cut

And closed the goal!

Rule argument

var: qrl0

Additional information

Used simplifier rules: reflist freely generated by [], +;; ⊦ markedp(rl + rl0, vs) ↔ markedp(rl, vs) ∧ markedp(rl0, vs); (s) ⊦ markedp(r ', vs) ↔ some?(vs[r].deqID); (s) ⊦ listp(r ' + r0 ' + rl, vs, ps, owns) ↔ owns(r) = None ∧ r ∈ vs ∧ r ∈ ps ∧ r ≠ null ∧ ¬ weak?(vs[r].deqID) ∧ (ps[r].deqID = vs[r].deqID ∨ ps[r].deqID = None) ∧ ps[r].val = vs[r].val ∧ ps[r].next = vs[r].next ∧ ps[r].next = r0 ∧ listp(r0 ' + rl, vs, ps, owns); (s) ⊦ dups(x + a ' + y) ↔ a ∈ x ∨ a ∈ y ∨ dups(x + y); (ws) ⊦ a ∈ x + y ↔ a ∈ x ∨ a ∈ y; (s) ⊦ dups(a ' + x) ↔ a ∈ x ∨ dups(x); (s) ⊦ a ∈ b ' + x ↔ ¬(a ≠ b ∧ ¬ a ∈ x); (ss) ⊦ a ∈ b ' + x ↔ a = b ∨ a ∈ x; (s) ⊦ a + x = a ' + x; (s) ⊦ (b ' + a ' + x).last = (a ' + x).last; (s) a: ⊦ ¬ dups(x + y) → ¬ dups(x) ∧ ¬ dups(y); (forward) from specification list-dup[reflist] ⊦ some?(opttid) → opttid ≠ None; (as) listp-twoin: ⊦ listp(rl + r ' + r0 ' + rl0, vs, ps, owns) → ps[r].next = r0; (forward) from specification DMSQ ⊦ ¬ listp(null ' + rl0, vs, ps, owns); (s) a: ⊦ (x + y) + z = x + y + z; from specification list[reflist] ⊦ markedp([], vs); (s) ⊦ [] + x = x; (s) ⊦ ¬ dups([]); (s) ⊦ ¬ a ∈ []; (s) ⊦ x + [] = x; (s) ⊦ a '.last = a; (s) in-15: ⊦ x + a ' = y → a ∈ y; (forward) from specification list[reflist] ⊦ a ≠ b → (a ∈ b ' + x ↔ a ∈ x); (s) ⊦ a ∈ b ' ↔ a = b; (s)

Proof Tree for Crash-Theorem (View conclusion)

The following image shows the proof tree of sequent calculus. The proved goal is at the bottom. You can click on the nodes to view additional rules. See here for details on colors in a proof tree.

the proof tree