On Thu, Jun 29, 2023 at 09:30:19AM +0200, Theo Buehler wrote: > I wrote versions of this diff several times in the past but never sent > it out. A question by claudio encouraged me... > > cryptowarnx() and cryptoerrx() fail at showing openssl error stacks > in a pleasant way as no amount of lipstick can make this pig pretty. > I don't think these stacks should be shown to the user and they are > not a real debugging aid either as I don't recall that this ever made > things any easier for me. > > This mechanically replaces cryptowarnx() with warnx() and cryptoerrx() > with either errx(1, ...) or err(1, ...) depending on whether the only > possible error is allocation failure as that might give a useful hint. > > Regress will need a trivial diff and I'll send a PR to portable once > this is in.
OK job@
