From: Tom Lane <tgl@sss.pgh.pa.us>
To: Simon Riggs <simon@2ndquadrant.com>
cc: Andres Freund <andres@anarazel.de>, Tomas Vondra <tv@fuzzy.cz>,
        pgsql-committers <pgsql-committers@postgresql.org>
Subject: Re: pgsql: Generational memory allocator
In-reply-to: 
 <CANP8+jKCX0sTqR8ZexTMxr03Eu3wsuwV2qit0EB0TVorMeSXBg@mail.gmail.com>
References: <E1eHa4J-0006hI-Q8@gemulon.postgresql.org>
 <20171123001649.y2nf7422iay6cw3v@alap3.anarazel.de>
 <CANP8+jKoT=Pg5W_3FcKiDf+FMrJhFYacxt_g5E0=tkmb5s8M6g@mail.gmail.com>
 <c903f275-2150-fa52-64bf-dca7b53ebf8d@fuzzy.cz>
 <16449.1511548313@sss.pgh.pa.us> <4618.1511573138@sss.pgh.pa.us>
 <bf84d940-90d4-de91-19dd-612e011007f4@fuzzy.cz>
 <6189.1511587105@sss.pgh.pa.us> <8991.1511639441@sss.pgh.pa.us>
 <20171125200014.qbewtip5oydqsklt@alap3.anarazel.de>
 <21466.1511644760@sss.pgh.pa.us> <2875.1511646362@sss.pgh.pa.us>
 <CANP8+jKRt0Yenw8Ln4J4u=UsDVkCBgPmjDE2buYou-H9VTMZjw@mail.gmail.com>
 <16797.1511718406@sss.pgh.pa.us>
 <CANP8+jKCX0sTqR8ZexTMxr03Eu3wsuwV2qit0EB0TVorMeSXBg@mail.gmail.com>
Comments: In-reply-to Simon Riggs <simon@2ndquadrant.com>
	message dated "Mon, 27 Nov 2017 05:45:33 +1100"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <28143.1511722417.1@sss.pgh.pa.us>
Date: Sun, 26 Nov 2017 13:53:37 -0500
Message-ID: <28144.1511722417@sss.pgh.pa.us>
Precedence: bulk

Simon Riggs <simon@2ndquadrant.com> writes:
> On 27 November 2017 at 04:46, Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> Well, I'm concerned about the possibility of a lot of palloc thrashing
>> if the first bunch of records it reads happen to have steadily increasing
>> sizes.  However, rather than doubling, it might be sufficient to set a
>> robust minimum on the first allocation, ie use something along the lines
>> of Max(1024, MAXALIGN(state->main_data_len)).

> Agreed.

> I was just researching what that number should be... and I was
> thinking that we should use the maximum normal tuple size, which I
> think is

> TOAST_TUPLE_THRESHOLD +
> SizeOfXLogRecord +
> SizeOfXLogRecordDataHeaderLong

Well, let's not overthink this, because anything under 8K is going to
be rounded up to the next power of 2 anyway by aset.c.  Based on this
point I'd say that BLCKSZ/2 or BLCKSZ/4 would be reasonable candidates
for the minimum.

			regards, tom lane