Re: Variable-length FunctionCallInfoData - Mailing list pgsql-hackers

From Andres Freund
Subject Re: Variable-length FunctionCallInfoData
Date
Msg-id 20180605174612.fzcw6rzjsrilf4sw@alap3.anarazel.de
Whole thread Raw
In response to RE: Variable-length FunctionCallInfoData  (serge@rielau.com)
Responses Re: Variable-length FunctionCallInfoData
List pgsql-hackers
Hi,

On 2018-06-05 10:40:22 -0700, serge@rielau.com wrote:
> Big +1 on this one.

Cool.


> Here is what we did. It's very crude, but minimized the amount of pain:

I think I'd rather go for my approach in core though. Needlessly
carrying around a bunch of pointers, and adding the necessary
indirections on accesses, and more allocations don't seem to buy us that
much. Nor do I really like the hackyness...

Greetings,

Andres Freund


pgsql-hackers by date:

Previous
From: serge@rielau.com
Date:
Subject: RE: Variable-length FunctionCallInfoData
Next
From: Jeff Davis
Date:
Subject: Re: Spilling hashed SetOps and aggregates to disk