Re: missing LockBuffer(buffer, BUFFER_LOCK_SHARE) in trigger.c GetTupleForTrigger?

> heap_fetch() though looks very similar has an important difference. It
> might be reading the tuple without lock on it and we need the buffer lock
> to protect against concurrent update/delete to the tuple. AFAIK once the
> tuple is passed through qualification checks etc, even callers of
> heap_fetch() can read the tuple data without any lock on the buffer itself.

Sure, but the page header isn't accessed there, just the tuple data
itself which always stays at the same place on the page.
(A normal heap_fetch shouldn't be worried about updates/deletions to its
tuple, MVCC to the rescue.)