Re: Weird test mixup

Michael Paquier <michael@paquier.xyz>

View thread View on postgr.es

From: Michael Paquier <michael@paquier.xyz>

To: Heikki Linnakangas <hlinnaka@iki.fi>

Cc: Tom Lane <tgl@sss.pgh.pa.us>, pgsql-hackers <pgsql-hackers@postgresql.org>

Date: 2024-03-18T01:04:45Z

Lists: pgsql-hackers

Attachments

v2-0001-Introduce-runtime-conditions-in-module-injection_.patch (text/x-diff) patch v2-0001
v2-0002-Switch-GIN-tests-with-injection-points-to-be-conc.patch (text/x-diff) patch v2-0002

On Sat, Mar 16, 2024 at 08:40:21AM +0900, Michael Paquier wrote:
> Linking all the points to a PID with a injection_points_attach_local()
> that switches a static flag while registering a before_shmem_exit() to
> do an automated cleanup sounds like the simplest approach to me based
> on what I'm reading on this thread.

Please find a patch to do exactly that, without touching the backend
APIs.  0001 adds a new function call injection_points_local() that can
be added on top of a SQL test to make it concurrent-safe.  0002 is the
fix for the GIN tests.

I am going to add an open item to not forget about all that.

Comments are welcome.
--
Michael

Commits

injection_points: Store runtime conditions in private area
- 267d41dc4f41 17.0 landed
Introduce private data area for injection points
- 33181b48fd0e 17.0 landed
injection_points: Fix incorrect spinlock acquisition
- 597f66942d0b 17.0 landed
Make GIN tests using injection points concurrent-safe
- 5105c9079681 17.0 landed
injection_points: Fix race condition with local injection point tests
- f4083c497510 17.0 landed
injection_points: Introduce runtime conditions
- f587338dec87 17.0 landed
Make GIN test using injection points repeatable
- 997db123c054 17.0 landed
Fix backstop in gin test if injection point is not reached
- d802ff06d021 17.0 landed
Try to unbreak injection-fault tests in the buildfarm
- 85f65d7a26fc 17.0 landed
Disable tests using injection points in installcheck
- e2e3b8ae9ed7 17.0 landed