Skip site navigation (1) Skip section navigation (2)

Re: stuck spin lock with many concurrent users

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Tatsuo Ishii <t-ishii(at)sra(dot)co(dot)jp>
Cc: Inoue(at)tpf(dot)co(dot)jp, pgsql-hackers(at)postgresql(dot)org
Subject: Re: stuck spin lock with many concurrent users
Date: 2001-07-05 13:47:03
Message-ID: 2728.994340823@sss.pgh.pa.us (view raw or flat)
Thread:
Lists: pgsql-hackers
Tatsuo Ishii <t-ishii(at)sra(dot)co(dot)jp> writes:
> If so, what about increase the dead lock timer proportional to the
> length of the waiting holder queue?

I don't think that's a good idea; it's not solving the problem, only
reducing performance, and in a fairly arbitrary way at that.  (The
length of the particular wait queue you happen to be on is no measure
of the total number of processes waiting for locks.)

The real problem is in the spinlock implementation --- deadlock checking
is only one place where lots of processes might gang up on the same
spinlock.  The bufmgr lock is another one.

			regards, tom lane

In response to

pgsql-hackers by date

Next:From: Tom LaneDate: 2001-07-05 13:54:43
Subject: Re: Re: Buffer access rules, and a probable bug
Previous:From: Zeugswetter Andreas SBDate: 2001-07-05 12:27:01
Subject: AW: Re: Backup and Recovery

Privacy Policy | About PostgreSQL
Copyright © 1996-2014 The PostgreSQL Global Development Group