LLARRAYELT: \CHECKARRAYBLOCK says checking trailer consistency but checks header instead

[nbriggs]

In \CHECKARRAYBLOCK, one of the COND clauses reads

  ((NEQ (fetch (ARRAYBLOCK INUSE) of BASE)
        (NOT FREE))
    (SETQ ERROR "ARRAYBLOCK Trailer INUSE bit set wrong"))

but it checked the INUSE bit of BASE (the header) earlier, and here it should be checking of TRAILER.
It doesn't affect the normal operation, but it would mean that a mis-set INUSE bit in the trailer would not be detected. This error has been there since at least 13-Feb-1985, the last time this code was edited.

[rmkaplan]

We might get a hint if we keep track of the block lengths of the failures. E.g. the THISLINE block is about 770 cells, all of the font blocks are about either 128 or 256.

…

On Jun 13, 2025, at 5:13 PM, Nick Briggs ***@***.***> wrote: nbriggs created an issue (Interlisp/medley#2189) <#2189> In \CHECKARRAYBLOCK, one of the COND clauses reads ((NEQ (fetch (ARRAYBLOCK INUSE) of BASE) (NOT FREE)) (SETQ ERROR "ARRAYBLOCK Trailer INUSE bit set wrong")) but it checked the INUSE bit of BASE (the header) earlier, and here it should be checking of TRAILER. It doesn't affect the normal operation, but it would mean that a mis-set INUSE bit in the trailer would not be detected. This error has been there since at least 13-Feb-1985, the last time this code was edited. — Reply to this email directly, view it on GitHub <#2189>, or unsubscribe <https://github.com/notifications/unsubscribe-auth/AQSTUJPAN7X4TXYFW6RA5FT3DNSMNAVCNFSM6AAAAAB7JKBGACVHI2DSMVQWIX3LMV43ASLTON2WKOZTGE2DKMJUGEZDCOI>. You are receiving this because you are subscribed to this thread.