On Tue, 28 Nov 2023 at 11:21, Andres Freund wrote:
> Hm, does this ever matter from a performance POV? The current code does look
> simpler to read to me. If the overhead is relevant, I'd instead just move the
> code into a static inline?
I didn't particularly find the code in examine_variable()
Hi,
On 2023-11-24 17:06:25 +1300, David Rowley wrote:
> While working on the patch in [1], I noticed that ever since
> 00b41463c, it's now suboptimal to do the following:
>
> switch (bms_membership(relids))
> {
> case BMS_EMPTY_SET:
>/* handle empty set */
>break;
> case B
On Fri, 24 Nov 2023 at 19:54, Richard Guo wrote:
> +1 to the idea.
>
> I think you have a typo in distribute_restrictinfo_to_rels. We should
> remove the call of bms_singleton_member and use relid instead.
Thanks for reviewing. I've now pushed this.
David
On Fri, Nov 24, 2023 at 12:06 PM David Rowley wrote:
> In the attached, I've adjusted the code to use the latter of the two
> above methods in 3 places. In examine_variable() this reduces the
> complexity of the logic quite a bit and saves calling bms_is_member()
> in addition to bms_singleton_m
While working on the patch in [1], I noticed that ever since
00b41463c, it's now suboptimal to do the following:
switch (bms_membership(relids))
{
case BMS_EMPTY_SET:
/* handle empty set */
break;
case BMS_SINGLETON:
/* call bms_singleton_member() and handle singleton
