On Sun, May 28, 2023 at 8:37 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
> I gather it'd catch things like NodeTag enum assignments changing,
> which is something we really need to have a check for.
Right. Any ABI break that involves machine-generated translation units
seems particularly prone to being overlooked. Just eyeballing code
(and perhaps double-checking struct layout using pahole) seems
inadequate.
I'll try to come up with a standard abi-compliance-checker Postgres
workflow once I'm back from pgCon. It already looks like
abi-compliance-checker is capable of taking most of the guesswork out
of ABI compatibility in stable releases, without any real downside,
which is encouraging. I have spent very little time on this, so it's
quite possible that some detail or other was overlooked.
--
Peter Geoghegan