Re: GiST penalty functions [PoC] - Mailing list pgsql-hackers

From Tom Lane
Subject Re: GiST penalty functions [PoC]
Date
Msg-id 20302.1472734029@sss.pgh.pa.us
Whole thread Raw
In response to Re: GiST penalty functions [PoC]  (Andrew Borodin <borodin@octonica.com>)
Responses Re: GiST penalty functions [PoC]  (Andrew Borodin <borodin@octonica.com>)
List pgsql-hackers
Andrew Borodin <borodin@octonica.com> writes:
> Does every supported Postgres platform conforms to IEEE 754 floating
> point specification?

Possibly, but it is against project policy to allow code to assume so.
That pack_float function is absolutely not acceptable IMV, and would
still be if it were adequately documented, which it's far away from
being.

On general grounds, I would forget all the "inline"s.  Modern compilers
are generally able to make their own decisions about it, and trying to put
your thumb on the scales this heavily is not likely to improve the code.

Haven't really read the patch, just responding to a couple points you
mentioned in the intro.
        regards, tom lane



pgsql-hackers by date:

Previous
From: Michael Paquier
Date:
Subject: Re: pg_basebackup stream xlog to tar
Next
From: Andrew Borodin
Date:
Subject: Re: GiST penalty functions [PoC]