Proof step 73 from proof inv-gc (previous step)

Heuristic symbolic execution applied rule skip right

And obtained the following new goals

• Proof step 74:

  # ax = # ndx3, # lax = # gndx0, ndx3 + ndx2 = nodes(gstore2 ↓, jro0)[referenced-adrs(lax.to-adr(n1), fi2)], gstore2[n1] = groupblock(gndx0, lax, m0), abs-log(nx0, gstore ↓) = abs-log(nx, gstore2 ↓) + ax, fi = fi2[ndx3.keys, ax], nodes(gstore ↓, jro) = nodes(gstore2 ↓, jro0)[ax, ndx3], remove-refsize0(gstore1) = gstore1, elems nx0 ∩ jro = ∅, (nx0.toSet \ nx.toSet) ∩ dom gstore2 = ∅, elems nx ∩ jro0 = ∅, dom gstore1 ∪ elems nx0 = dom gstore ∪ jro, dom gstore1 ∪ elems nx = dom gstore2 ∪ jro0, jro1 < jmaxino0, fi2[ndx3.keys, ax].inos < jmaxino0, fi1.inos < pmaxino, fi2.inos < jmaxino, p_os < pmaxino, jro1 < jmaxino, inv(gstore \ jro), inv(gstore2 \ jro0), inv(gstore1), inv(groupblock(gndx0, lax, m0)), new(ax, nodes(gstore2 ↓, jro0)), refsize-cons(gstore, range fi2[ndx3.keys, ax]), refsize-cons(gstore1, range fi1), refsize-cons(gstore2, range fi2), ri-cons(fi2, nodes(gstore2 ↓, jro0)), ri-inj(fi2[ndx3.keys, ax]), ri-inj(fi1), ri-inj(fi2), n1 ∈ gstore2, n1 ∈ gstore1, pmaxino ≤ jmaxino0, jmaxino ≤ jmaxino0, pmaxino ≤ jmaxino, fi2[ndx3.keys, ax].lebs ⊆ dom gstore, dom gstore2 ⊆ dom gstore, range fi2 ⊆ dom nodes(gstore2 ↓, jro0), fi1.lebs ⊆ dom gstore1, fi2.lebs ⊆ dom gstore2, fi2.lebs ⊆ dom gstore, ⌊err⌋ , jvalid0 ∧ nx0 ≠ [] → valid-journal-head(nx0, gstore, jro), jvalid4 ∧ nx ≠ [] → valid-journal-head(nx, gstore2, jro0) ⊦ ⟪if err = ESUCCESS then index_deallocate_gnd#(n1, fi, fi1, gstore1, nx0, pmaxino, p_os, jvalid; jro, gstore; err)⟫ ( (jvalid0 ∧ nx0 ≠ [] → valid-journal-head(nx0, gstore, jro)) ∧ fi.lebs ⊆ dom gstore ∧ refsize-cons(gstore, range fi) ∧ inv(gstore \ jro) ∧ elems nx0 ∩ jro = ∅ ∧ dom gstore1 ∪ elems nx0 = dom gstore ∪ jro), jmaxino = 0, jmaxino0 = 0, m0 = 0, pmaxino = 0, dups(nx0), dups(ax), dups(nx), dups(ndx3.keys + ndx2.keys), dups(ndx3.keys), dups(ndx2.keys), n1 ∈ jro0, n1 ∈ nx

Rule argument

Additional information