Re: Assertion failure in hash_kill_items()

Hi,

On 2026-03-17 13:40:10 -0400, Andres Freund wrote:
> On 2026-03-17 19:15:10 +0200, Heikki Linnakangas wrote:
> > I bumped into an assertion failure, while playing with variants of the test
> > case that Alexander Kuzmenkov wrote to exercise hash index page cleanup [1].
> > This is master-only, related to the recent changes in how buffers are marked
> > dirty.
>
> Sorry, I had hoped to push a fix for that already, after it was reported in
> https://postgr.es/m/vjtmvwvbxt7w5uyacxpzibpj65ewcb7uqaqbhd4arvnjbp5jqz%405ksdh6fsyqve
> but real life intervened.
>
> I was planning to commit it together with an addition to
> src/test/modules/index/specs/killtuples.spec
>
> Unfortunately that made the change a good bit more verbose, as a naive
> addition would report a number of buffer accesses that seemed not necessarily
> reliable to me. So I updated the 'result' step to just return true/false
> depending on whether there were any accesses.
>
> I'll go and work on pushing that.

Done, as of f5eb854ab6d.

Greetings,

Andres