> On 18 Apr 2024, at 22:28, Nathan Bossart <nathandboss...@gmail.com> wrote: > > On Thu, Apr 18, 2024 at 10:23:01AM -0500, Nathan Bossart wrote: >> On Thu, Apr 18, 2024 at 09:24:53AM +0200, Daniel Gustafsson wrote: >>> That does indeed seem like a saner approach. Since we look up the relkind >>> we >>> can also remove the is_index parameter to binary_upgrade_set_pg_class_oids >>> since we already know that without the caller telling us? >> >> Yeah. It looks like that's been possible since commit 9a974cb, so I can >> write a prerequisite patch for this. > > Here's a new patch set with this change.
From a read-through they look good, a very nice performance improvement in an important path. I think it would be nice with some comments on the BinaryUpgradeClassOids struct (since the code using it is thousands of lines away), and a comment on the if (oids == NULL) block explaining the caching. -- Daniel Gustafsson