From: Jose Luis Tallon <jltallon@adv-solutions.net>
To: David Rowley <david.rowley@2ndquadrant.com>
To: PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>
Cc: Thomas Munro <thomas.munro@enterprisedb.com>
Subject: Re: Using POPCNT and other advanced bit manipulation instructions
Date: Thu, 20 Dec 2018 11:58:43 +0100
Message-ID: <df95d1e8-8e63-256a-2d3e-ed2f8cff2f37@adv-solutions.net> (raw)
In-Reply-To: <CAKJS1f9WTAGG1tPeJnD18hiQW5gAk59fQ6WK-vfdAKEHyRg2RA@mail.gmail.com>
References: <CAKJS1f9WTAGG1tPeJnD18hiQW5gAk59fQ6WK-vfdAKEHyRg2RA@mail.gmail.com>
On 20/12/18 6:53, David Rowley wrote:
> Back in 2016 [1] there was some discussion about using the POPCNT
> instruction to improve the performance of counting the number of bits
> set in a word. Improving this helps various cases, such as
> bms_num_members and also things like counting the allvisible and
> frozen pages in the visibility map.
>
> [snip]
>
> I've put together a very rough patch to implement using POPCNT and the
> leading and trailing 0-bit instructions to improve the performance of
> bms_next_member() and bms_prev_member(). The correct function should
> be determined on the first call to each function by way of setting a
> function pointer variable to the most suitable supported
> implementation. I've not yet gone through and removed all the
> number_of_ones[] arrays to replace with a pg_popcount*() call.
IMVHO: Please do not disregard potential optimization by the compiler
around those calls.. o_0 That might explain the reduced performance
improvement observed.
Not that I can see any obvious alternative to your implementation right
away ...
> That
> seems to have mostly been done in Thomas' patch [3], part of which
> I've used for the visibilitymap.c code changes. If this patch proves
> to be possible, then I'll look at including the other changes Thomas
> made in his patch too.
>
> What I'm really looking for by posting now are reasons why we can't do
> this. I'm also interested in getting some testing done on older
> machines, particularly machines with processors that are from before
> 2007, both AMD and Intel.
I can offer a 2005-vintage Opteron 2216 rev3 (bought late 2007) to test
on. Feel free to toss me some test code.
cpuinfo flags: fpu de tsc msr pae mce cx8 apic mca cmov pat clflush
mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt lm 3dnowext 3dnow
rep_good nopl extd_apicid eagerfpu pni cx16 hypervisor lahf_lm
cmp_legacy 3dnowprefetch vmmcall
> 2007-2008 seems to be around the time both
> AMD and Intel added support for POPCNT and LZCNT, going by [4].
Thanks
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-hackers@postgresql.org
Cc: jltallon@adv-solutions.net, david.rowley@2ndquadrant.com, pgsql-hackers@lists.postgresql.org, thomas.munro@enterprisedb.com
Subject: Re: Using POPCNT and other advanced bit manipulation instructions
In-Reply-To: <df95d1e8-8e63-256a-2d3e-ed2f8cff2f37@adv-solutions.net>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox