Re: BUG #16036: Segmentation fault while doing an update

Hi,

On 2019-10-03 21:30:21 -0700, Andres Freund wrote:
> Here's 1, 2 & 4, working on fixing 3). While we now have OK-ish
> coverage for nodeModifyTable.c and nodeLockRows.c triggered EPQ, there's
> not much for the trigger case.

Here's both the fix and tests.

I think the test adds a good bit coverage. What's lacking is primarily
testing foreign table related paths, but that seems separate enough, and
would need to be done in a separate contrib/ test anyway.

I think there's a good case to be made to backpatch the tests further
than 12, but I'm not sure about it? They do pass (with one error message
about a failure to delete changed to a failure to update, we didn't use
to be able to discern) back to 9.6, requiring

commit 38f8bdcac4982215beb9f65a19debecaf22fd470
Author: Robert Haas <rhaas(at)postgresql(dot)org>
Date: 2016-02-11 08:36:30 -0500

Modify the isolation tester so that multiple sessions can wait.

Greetings,

Andres Freund