Horton

Re: Recovering from detoast-related catcache invalidations

Tom Lane <tgl@sss.pgh.pa.us>

View thread
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Xiaoran Wang <fanfuxiaoran@gmail.com>
Cc: pgsql-hackers@lists.postgresql.org
Date: 2024-01-11T22:21:03Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Fix catcache invalidation of a list entry that's being built

  2. Cope with inplace update making catcache stale during TOAST fetch.

  3. Add previous commit to .git-blame-ignore-revs.

  4. Re-pgindent catcache.c after previous commit.

  5. Cope with catcache entries becoming stale during detoasting.

Xiaoran Wang <fanfuxiaoran@gmail.com> writes:
>>> The detection of "get an invalidation" could be refined: what I did
>>> here is to check for any advance of SharedInvalidMessageCounter,
>>> which clearly will have a significant number of false positives.

> I have reviewed your patch, and it looks good.  But instead of checking for
> any advance of SharedInvalidMessageCounter ( if the invalidate message is
> not related to the current tuple, it is a little expensive)  I have another
> idea:  we can recheck the visibility of the tuple with CatalogSnapshot(the
> CatalogSnapthot must be refreshed if there is any SharedInvalidMessages) if
> it is not visible, we re-fetch the tuple, otherwise, we can continue to use
> it as it is not outdated.

Maybe, but that undocumented hack in SetHintBits seems completely
unacceptable.  Isn't there a cleaner way to make this check?

Also, I'm pretty dubious that GetNonHistoricCatalogSnapshot rather
than GetCatalogSnapshot is the right thing, because the catcaches
use the latter.

			regards, tom lane