| From: | Claudio Freire <klaussfreire(at)gmail(dot)com> |
|---|---|
| To: | Peter Geoghegan <pg(at)bowt(dot)ie> |
| Cc: | Pavan Deolasee <pavan(dot)deolasee(at)gmail(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>, Simon Riggs <simon(dot)riggs(at)2ndquadrant(dot)com> |
| Subject: | Re: Faster inserts with mostly-monotonically increasing values |
| Date: | 2018-03-06 00:48:50 |
| Message-ID: | CAGTBQpaYXRVmp+x+FZ+hABWe6AKCq7X06zj-ikktk1_jtY70nw@mail.gmail.com |
| Views: | Raw Message | Whole Thread | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-hackers |
On Mon, Mar 5, 2018 at 9:37 PM, Claudio Freire <klaussfreire(at)gmail(dot)com> wrote:
> Assuming the rightmost page is the first page the value could be on,
> it already does get an exclusive buffer lock.
That made me check, and:
+ /*
+ * Acquire exclusive lock on the buffer before doing any checks. This
+ * ensures that the index state cannot change, as far as the rightmost
+ * part of the index is concerned.
+ */
+ LockBuffer(buf, BUFFER_LOCK_EXCLUSIVE);
BTree code uses BT_READ and BT_WRITE instead, so that should be:
LockBuffer(buf, BT_WRITE)
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2018-03-06 00:49:06 | Re: [COMMITTERS] pgsql: Fix inadequate locking during get_rel_oids(). |
| Previous Message | Michael Paquier | 2018-03-06 00:47:47 | Re: Cache lookup errors with functions manipulation object addresses |