On Sun Sep 13, 2026 at 12:35 PM CEST, Mario Domenech Goulart via Chicken-users wrote: > Hi Daniel, > > On Sat, 12 Sep 2026 23:21:27 -0400 Daniel Colascione via Chicken-users > <[email protected]> wrote: > >> The following asserts with DEBUGBUILD=1, 6.0.0 >> >> (import srfi-69) >> (eq?-hash (lambda (x) x)) >> >> [panic] Low-level type assertion (C_truep(C_bytevectorp(C_VAL1(str)))) >> failed > > Thanks for reporting the issue. I've created ticket 709d35 for it: > http://bugs.call-cc.org/ticket/t70/709d35e1e095dc761505457151d2adc19bbbef47/ >
The assertion failure is caused by a failed access to the data of a lambda-info object, which is technically a bytevector, but for which C_bytevectorp fails. I suggest adding a special case in the recursive hash procedure that dispatches on object types (or leave out the lambda-info slot when hashing procedures, altogether). (I will add this info to the ticket) cheers, felix
