Re: [HACKERS] new heap manager mmalloc

From: Bruce Momjian <maillist(at)candle(dot)pha(dot)pa(dot)us>
To: scrappy(at)hub(dot)org (The Hermit Hacker)
Cc: voss(at)vocalweb(dot)de, pgsql-hackers(at)postgreSQL(dot)org
Subject: Re: [HACKERS] new heap manager mmalloc
Date: 1999-01-28 17:46:52
Message-ID: 199901281746.MAA01525@candle.pha.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

>
> Two things against it...
>
> First, its a Linux-ism...he's got it ported to Win and Linux, that's it...

Actually, our problem is not malloc itself. Most Unix OS's have pretty
good malloc's, tuned to their OS. The problem is the number of times we
call it.

Massimo's idea of having several alloc contexts, some of which supply
memory from a backend-managed pool is a good idea. When I was working
on a SQL backend design, I thought this would be a very good way to go.
SQL databases have a nice end-of-transaction free-it-all point that can
make use of such a give-me-the-memory and don't worry about freeing it
until I am done with the transaction model.

--
Bruce Momjian | http://www.op.net/~candle
maillist(at)candle(dot)pha(dot)pa(dot)us | (610) 853-3000
+ If your life is a hard drive, | 830 Blythe Avenue
+ Christ can be your backup. | Drexel Hill, Pennsylvania 19026

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Tom Lane 1999-01-28 18:43:05 Theory and practice of free software
Previous Message D'Arcy J.M. Cain 1999-01-28 17:23:21 Re: [HACKERS] Problem with multiple SUMs