Re: Injection points: preloading and runtime arguments

On Sat, Jun 08, 2024 at 04:52:25PM +0500, Andrey M. Borodin wrote:
> Alvaro, here’s the test for multixact CV sleep that I was talking
> about on PGConf.
> It is needed to test [0]. It is based on loaded injection
> points.

> This technique is not committed yet, but the patch looks good.

OK, cool.  I'll try to get that into the tree once v18 opens up.  I
can see that GetNewMultiXactId() opens a critical section.  I am
slightly surprised that you need both the SQL function
injection_points_load() and the macro INJECTION_POINT_LOAD().
Wouldn't one or the other be sufficient?

The test takes 20ms to run here, which is good enough.

+   INJECTION_POINT_LOAD("GetNewMultiXactId-done");
[...]
+   INJECTION_POINT("GetNewMultiXactId-done");
[...]
+   INJECTION_POINT("GetMultiXactIdMembers-CV-sleep");

Be careful about the naming here.  All the points use lower case
characters currently.

+# and another multixact have no offest yet, we must wait until this offset

s/offest/offset/.

> When all prerequisites are ready I will post it to corresponding
> thread and create CF item.

OK, let's do that.
--
Michael