Re: backtrace_on_internal_error

Hi,

On 2023-12-20 10:08:42 +0100, Jelte Fennema-Nio wrote:
> On Sun, 10 Dec 2023 at 00:14, Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:
> > I'm not actually sure that the fe-secure.c part of v3-0002 is
> > necessary, because it's guarding plain recv(2) which really shouldn't
> > return -1 without setting errno. Still, it's a pretty harmless
> > addition.
>
> v3-0002 seems have a very similar goal to v23-0002 in my non-blocking
> and encrypted cancel request patchset here:
> https://www.postgresql.org/message-id/flat/CAGECzQQirExbHe6uLa4C-sP%3DwTR1jazR_wgCWd4177QE-%3DVFDw%40mail.gmail.com#0b6cc1897c6d507cef49a3f3797181aa
>
> Would it be possible to merge that on instead or at least use the same
> approach as that one (i.e. return -2 on EOF). Otherwise I have to
> update that patchset to match the new style of communicating that
> there is an EOF. Also I personally think a separate return value for
> EOF clearer when reading the code than checking for errno being 0.

Tom's patch imo doesn't really introduce anything really new - we already deal
with EOF that way in other places. And it's how the standard APIs deal with
the issue. I'd not design it this way on a green field, but given the current
state Tom's approach seems more sensible...

Greetings,

Andres Freund