Horton

Re: buffer locking fix for lazy_scan_heap

Robert Haas <robertmhaas@gmail.com>

View thread
From: Robert Haas <robertmhaas@gmail.com>
To: 高增琦 <pgf00a@gmail.com>
Cc: pgsql-hackers <pgsql-hackers@postgresql.org>
Date: 2015-08-04T01:19:34Z
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. Rearrange lazy_scan_heap to avoid visibility map race conditions.

On Mon, Aug 3, 2015 at 12:03 AM, 高增琦 <pgf00a@gmail.com> wrote:
> sorry for asking this really old commit.
>
> http://git.postgresql.org/gitweb/?p=postgresql.git;a=commitdiff;h=7ab9b2f3b79177e501a1ef90ed004cc68788abaf
>
> I could not understand why "releases the lock on the buffer" is
> a problem since vacuum will lock and check the page bit again before
> set the vm. Did I missed something?

Hmm, you might be right.  It certainly seems safer to do all the work
without releasing the buffer lock temporarily in the middle, but maybe
the old way wasn't actually broken.

-- 
Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company