Re: Do we need so many hint bits? - Mailing list pgsql-hackers

From Jeff Davis
Subject Re: Do we need so many hint bits?
Date
Msg-id 1353350330.10198.133.camel@jdavis-laptop
Whole thread Raw
In response to Re: Do we need so many hint bits?  (Atri Sharma <atri.jiit@gmail.com>)
Responses Re: Do we need so many hint bits?  (Atri Sharma <atri.jiit@gmail.com>)
List pgsql-hackers
On Mon, 2012-11-19 at 23:50 +0530, Atri Sharma wrote:

> Sorry If I am being a bit naive, but shouldnt a simple mutex work in
> the case when a process wants to change the VM bit in cache?
> 
> Mutex would be cheaper than locks.
> 
I thought mutexes are locks?

The point is to avoid taking new locks (or mutexes) during a read of the
VM bit, because there is concern that it could cause contention. If we
lock the entire VM page, that represents many, many data pages, so it's
worrisome.

Regards,Jeff Davis





pgsql-hackers by date:

Previous
From: Tomas Vondra
Date:
Subject: Re: too much pgbench init output
Next
From: Atri Sharma
Date:
Subject: Re: Do we need so many hint bits?