RE: btree split logic is fragile in the presence of lar ge index items - Mailing list pgsql-hackers

From Mikheev, Vadim
Subject RE: btree split logic is fragile in the presence of lar ge index items
Date
Msg-id 8F4C99C66D04D4118F580090272A7A23018C67@SECTORBASE1
Whole thread Raw
Responses Re: btree split logic is fragile in the presence of lar ge index items  (Tom Lane <tgl@sss.pgh.pa.us>)
List pgsql-hackers
> > > What about unique key insertions ?
> > 
> > We'll have to find leftmost key in this case and do what we do now.
> >
> 
> Currently the page contains the leftmost key is the target page of
> insertion and is locked exclusively but it may be different in extra
> TID implementation. There may be a very rare deadlock possibility.

First, Tom is not going to do TID implementation now...
But anyway while we hold lock on a page we are able to go right
and lock pages (and we do this now). There is no possibility for
deadlock here: backward scans always unlock page before reading/locking
left page.

Vadim


pgsql-hackers by date:

Previous
From: "Stephan Szabo"
Date:
Subject: Re: Re: [GENERAL] PRIMARY KEY & INHERITANCE (fwd)
Next
From: Larry Rosenman
Date:
Subject: mac.c