[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 159 from proof wlar-erase-worker-helper (previous step)

wlar[n] = wl-entry(m0, stat0), apebs0[n] = mkpeb(aehdr, avhdr, buf, m1, false), apebs = apebs0[n, mkpeb(aehdr0, avhdr0, buf0, m2, false)], vtblpnum0 = n, vtblpnum1 = 1, wlar1 = wlar[n, wlar[n].ec:= (wlar[n].ec + 1)], n < # wlar, n < # apebs0, ubi-io-peb-inv(mkpeb(aehdr0, avhdr0, buf0, m2, false)), ⌊err0⌋ ⊦ [while err0 ≠ ESUCCESS ∧ vtblpnum1 ≤ UBI_ERASE_RETRIES do {wlar1[vtblpnum0].ec := wlar1[vtblpnum0].ec + 1; aubi_io_sync_erase#(vtblpnum0; apebs; err0); if err0 = ESUCCESS then aubi_io_write_echdr#(vtblpnum0, aechdr(wlar1[vtblpnum0].ec); apebs; err0) else debug#("ubi: erase of PEB " + vtblpnum0.toString + " failed"); vtblpnum1 := vtblpnum1 + 1}; if err0 = ESUCCESS then wlar1[vtblpnum0] := wl-entry(wlar1[vtblpnum0].ec, free) else {aubi_io_markbad#(vtblpnum0; apebs; err0); if err0 = ESUCCESS then wlar1[vtblpnum0] := wl-entry(wlar1[vtblpnum0].ec, erroneous) else debug#("ubi: marking PEB " + vtblpnum0.toString + " as bad failed")}] ∃ m, stat. m ≠ 0 ∧ m ≤ UBI_ERASE_RETRIES + 1 ∧ wlar1 = wlar[n, wl-entry(m0 + m, stat)], avhdr0.valid?

User applied rule invariant right

And obtained the following new goals

Proof step 160:
apebs0[n] = mkpeb(aehdr, avhdr, buf, m1, false), wlar[n] = wl-entry(m0, stat0), apebs = apebs0[n, mkpeb(aehdr0, avhdr0, buf0, m2, false)], vtblpnum0 = n, vtblpnum1 = 1, wlar1 = wlar[n, wlar[n].ec:= (wlar[n].ec + 1)], n < # apebs0, n < # wlar, ubi-io-peb-inv(mkpeb(aehdr0, avhdr0, buf0, m2, false)), ⌊err0⌋ ⊦ avhdr0.valid?, n < # wlar ∧ id○(wlar, wlar1, n) ∧ vtblpnum1 ≠ 0 ∧ vtblpnum1 ≤ UBI_ERASE_RETRIES + 1 ∧ (∃ stat. wlar1 = wlar[n, wl-entry(wlar[n].ec + vtblpnum1, stat)])
Proof step 162:
apebs0[n] = mkpeb(aehdr, avhdr, buf, m1, false), wlar[n] = wl-entry(m0, stat0), apebs1 = apebs0[n, mkpeb(aehdr0, avhdr0, buf0, m2, false)], vtblpnum = 1, vtblpnum0 = n, wlar0 = wlar[n, wlar[n].ec:= (wlar[n].ec + 1)], n < # apebs0, n < # wlar, id○(wlar, wlar1, n), ubi-io-peb-inv(mkpeb(aehdr0, avhdr0, buf0, m2, false)), vtblpnum1 ≤ UBI_ERASE_RETRIES + 1, vtblpnum1 ≤ UBI_ERASE_RETRIES, ⌊err⌋ , ∃ stat. wlar1 = wlar[n, wl-entry(wlar[n].ec + vtblpnum1, stat)] ⊦ [wlar1[vtblpnum0].ec := wlar1[vtblpnum0].ec + 1; aubi_io_sync_erase#(vtblpnum0; apebs; err0); if err0 = ESUCCESS then aubi_io_write_echdr#(vtblpnum0, aechdr(wlar1[vtblpnum0].ec); apebs; err0) else debug#("ubi: erase of PEB " + vtblpnum0.toString + " failed"); vtblpnum1 := vtblpnum1 + 1] ( n < # wlar ∧ id○(wlar, wlar1, n) ∧ vtblpnum1 ≠ 0 ∧ vtblpnum1 ≤ UBI_ERASE_RETRIES + 1 ∧ (∃ stat. wlar1 = wlar[n, wl-entry(wlar[n].ec + vtblpnum1, stat)])), err0 = ESUCCESS, vtblpnum1 = 0, avhdr0.valid?
Proof step 221:
apebs0[n] = mkpeb(aehdr, avhdr, buf, m1, false), wlar[n] = wl-entry(m0, stat0), apebs1 = apebs0[n, mkpeb(aehdr0, avhdr0, buf0, m2, false)], vtblpnum = 1, vtblpnum0 = n, wlar0 = wlar[n, wlar[n].ec:= (wlar[n].ec + 1)], n < # apebs0, n < # wlar, id○(wlar, wlar1, n), ubi-io-peb-inv(mkpeb(aehdr0, avhdr0, buf0, m2, false)), vtblpnum1 ≤ UBI_ERASE_RETRIES + 1, ⌊err⌋ , ∃ stat. wlar1 = wlar[n, wl-entry(wlar[n].ec + vtblpnum1, stat)] ⊦ [if err0 = ESUCCESS then wlar1[vtblpnum0] := wl-entry(wlar1[vtblpnum0].ec, free) else {aubi_io_markbad#(vtblpnum0; apebs; err0); if err0 = ESUCCESS then wlar1[vtblpnum0] := wl-entry(wlar1[vtblpnum0].ec, erroneous) else debug#("ubi: marking PEB " + vtblpnum0.toString + " as bad failed")}] ∃ m, stat. m ≠ 0 ∧ m ≤ UBI_ERASE_RETRIES + 1 ∧ wlar1 = wlar[n, wl-entry(m0 + m, stat)], vtblpnum1 = 0, avhdr0.valid?, err0 ≠ ESUCCESS ∧ vtblpnum1 ≤ UBI_ERASE_RETRIES

Rule argument

fma: n < # wlar ∧ id○(wlar, wlar1, n) ∧ vtblpnum1 ≠ 0 ∧ vtblpnum1 ≤ UBI_ERASE_RETRIES + 1 ∧ (∃ stat. wlar1 = wlar[n, wl-entry(wlar[n].ec + vtblpnum1, stat)]) pos: right 1

Additional information

Used formulas: n < # wlar ∧ id○(wlar, wlar1, n) ∧ vtblpnum1 ≠ 0 ∧ vtblpnum1 ≤ UBI_ERASE_RETRIES + 1 ∧ (∃ stat. wlar1 = wlar[n, wl-entry(wlar[n].ec + vtblpnum1, stat)])