To: "Alvaro Herrera" <alvherre@commandprompt.com>
Cc: "Simon Riggs" <simon@2ndquadrant.com>,  "Tom Lane" <tgl@sss.pgh.pa.us>,
	"pgsql-hackers" <pgsql-hackers@postgresql.org>,
	"List pgsql-patches" <pgsql-patches@postgresql.org>
Subject: Re: [HACKERS] Hint Bits and Write I/O
In-Reply-To: <20080618143212.GE5077@alvh.no-ip.org> (Alvaro Herrera's message
	of "Wed\, 18 Jun 2008 10\:32\:12 -0400")
User-Agent: Gnus/5.11 (Gnus v5.11) Emacs/22.1 (gnu/linux)
References: <1211916951.4489.232.camel@ebony.site>
	<15901.1211931165@sss.pgh.pa.us>
	<1213797209.9468.126.camel@ebony.site>
	<20080618143212.GE5077@alvh.no-ip.org>
From: Gregory Stark <stark@enterprisedb.com>
Organization: EnterpriseDB
Date: Fri, 27 Jun 2008 15:25:47 +0100
Message-ID: <87lk0rq7is.fsf@oxford.xeocode.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii

"Alvaro Herrera" <alvherre@commandprompt.com> writes:

> If only VACUUM is going to set "flexible" to off, maybe it's better to
> leave the APIs as they are and have a global that's set by VACUUM only
> (and reset in a PG_CATCH block).

Ugh. Perhaps it would be simpler to have a wrapper function HTSV() macro which
passes flexible=true to HTSV_internal(). Then vacuum can call HTSV_internal().

I'm not sure what the performance tradeoff is between having an extra argument
to HTSV and having HTSV check a global which messes with optimizations.


-- 
  Gregory Stark
  EnterpriseDB          http://www.enterprisedb.com
  Ask me about EnterpriseDB's RemoteDBA services!