From: | "Luke Lonergan" <llonergan(at)greenplum(dot)com> |
---|---|
To: | "Csaba Nagy" <nagy(at)ecircle-ag(dot)com>, "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | "Qingqing Zhou" <zhouqq(at)cs(dot)toronto(dot)edu>, "postgres hackers" <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: Automatically setting work_mem |
Date: | 2006-03-18 08:05:19 |
Message-ID: | C040FF3F.1F72F%llonergan@greenplum.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers pgsql-patches |
Csaba,
On 3/17/06 7:07 AM, "Csaba Nagy" <nagy(at)ecircle-ag(dot)com> wrote:
> It worths a look at how apache Derby does with query planning, where a
> planned query is actually a compiled Java class, i.e. the executable
> byte code which will run to fetch the results, created and compiled by
> the planner... interesting approach, allows for lots of flexibility at
> run-time, but probably won't work with C :-)
We've looked at using the open source llvm compiler to create an
intermediate representation of the plan, then generate machine code and
dispatch for execution.
This would have the advantage of being able to place runtime constants into
the intermediate representation as constants (like the address of a
comparator function or operator), then let the compiler optimize them out,
hoist, etc. You can't do this at compile time, and there would be no change
of the nice abstract code in the executor.
It's on our list - anyone else interested?
- Luke
From | Date | Subject | |
---|---|---|---|
Next Message | Dave Page | 2006-03-18 09:13:58 | Re: [HACKERS] PostgreSQL Anniversary Proposals -- Important Update |
Previous Message | Luke Lonergan | 2006-03-18 07:59:51 | Re: PostgreSQL Anniversary Proposals -- Important |
From | Date | Subject | |
---|---|---|---|
Next Message | Luke Lonergan | 2006-03-18 21:21:30 | Re: Automatically setting work_mem |
Previous Message | Tom Lane | 2006-03-18 05:59:16 | Re: Automatically setting work_mem |