Subject: Re: Greatest Common Divisor
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Fabien COELHO <coelho@cri.ensmp.fr>, Stephen Frost <sfrost@snowman.net>,
 Dean Rasheed <dean.a.rasheed@gmail.com>,
 Chapman Flack <chap@anastigmatix.net>,
 PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
References: <adbd3e0b-e3f1-5bbc-21db-03caf1cef0f7@2ndquadrant.com>
 <alpine.DEB.2.21.1912281848540.24861@pseudo>
 <dbf86808-91fa-3262-f939-2ef3d38ec242@2ndquadrant.com>
 <alpine.DEB.2.21.1912290815520.889@pseudo>
 <5E08D947.9000902@anastigmatix.net>
 <alpine.DEB.2.21.1912291755510.14206@pseudo>
 <CAEZATCX9YgkWupVFcJ5ixkEJMyf2X5ZecN8K5r12oOB8mDHzWQ@mail.gmail.com>
 <20200102145712.GN3195@tamriel.snowman.net> <16242.1577977961@sss.pgh.pa.us>
 <6855727c-3836-21cc-18f9-e79227b99079@2ndquadrant.com>
 <alpine.DEB.2.21.2001032010470.12030@pseudo>
 <5bf72c6b-1290-2ba0-770b-8a13e899fd98@2ndquadrant.com>
 <13445.1578095358@sss.pgh.pa.us>
From: Vik Fearing <vik.fearing@2ndquadrant.com>
Message-ID: <c2fa7d6e-3733-51da-2967-58d2ae72267f@2ndquadrant.com>
Date: Sat, 4 Jan 2020 01:21:32 +0100
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:68.0) Gecko/20100101
 Thunderbird/68.2.2
MIME-Version: 1.0
In-Reply-To: <13445.1578095358@sss.pgh.pa.us>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: 7bit
Content-Language: en-US
Precedence: bulk

On 04/01/2020 00:49, Tom Lane wrote:
> Vik Fearing <vik.fearing@2ndquadrant.com> writes:
>> On 03/01/2020 20:14, Fabien COELHO wrote:
>>> The point of swapping is to a void possibly expensive modulo, but this
>>> should be done on absolute values, otherwise it may not achieve its
>>> purpose as stated by the comment?
>> Ah, true. How widespread are these architectures that need this special
>> treatment? Is it really worth handling?
> On some older RISC architectures, integer division is really slow, like
> slower than floating-point.  I'm not sure if that's true on any platform
> people still care about though.  In recent years, CPU architects have been
> able to throw all the transistors they needed at such problems.  On a
> machine with single-cycle divide, it's likely that the extra
> compare-and-branch is a net loss.


OK.


> Might be worth checking it on ARM in particular, as being a RISC
> architecture that's still popular.


I don't know how I would check this.


> Also, if we end up having a "numeric" implementation, it absolutely is
> worth it for that, because there is nothing cheap about numeric_div.


The patch includes a numeric version, and I take care to short-circuit
everything I can.


> I'd be sort of inclined to have the swap in the other implementations
> just to keep the algorithms as much alike as possible.


They can't quite be the same behavior because numeric doesn't have the
unrepresentable -INT_MIN problem, and integers don't have NaN.

-- 

Vik Fearing