Peter Geoghegan <pg@bowt.ie> writes:
> On Sun, Jun 23, 2019 at 3:36 AM Thomas Munro <thomas.munro@gmail.com> wrote:
>> Pushed. Thanks!
> I wonder what the comment is supposed to mean.
> I think that it's addressing the situation prior to commit 70508ba7aed
> in 2003, which was the point when the "fast" root concept was
> introduced.
Yeah. I did some research into the provenance of that comment when
Thomas pushed the change. It's *old*. The whole para exists verbatim
in Postgres v4r2, src/backend/access/nbtree/nbtpage.c dated 1993-12-10
(in my copy of that tarball). The only change since then has been to
change the whitespace for 4-space tabs.
Even more interesting, the same para also exists verbatim in
v4r2's src/backend/access/nobtree/nobtpage.c, which is dated 1991-10-29
in the same tarball. (If you're wondering, "nobtree" seems to stand
for "no-overwrite btree"; so I suppose it went the way of all flesh
when Stonebraker lost interest in write-once mass storage.) So presumably
this comment dates back to some common ancestor of the mainline btree code
and the no-overwrite code, which must have been even older than the 1991
date.
This is only marginally relevant to what we should do about it today,
but I think it's reasonable to conclude that the current locking
considerations are nearly unrelated to what they were when the comment
was written.
> I think that the whole sentence about "the standard class of race
> conditions" should go. There is no more dance. Nothing in
> _bt_getroot() is surprising to me. The other comments explain things
> comprehensively.
+1
regards, tom lane