From: | John Naylor <john(dot)naylor(at)enterprisedb(dot)com> |
---|---|
To: | Amit Khandekar <amitdkhan(dot)pg(at)gmail(dot)com> |
Cc: | Heikki Linnakangas <hlinnaka(at)iki(dot)fi>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: [POC] verifying UTF-8 using SIMD instructions |
Date: | 2021-03-09 11:43:52 |
Message-ID: | CAFBsxsGXNQ3d=WZLdBdeVBzE9vesRJbK+a4aBRoNreUfLrnxOw@mail.gmail.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On Tue, Mar 9, 2021 at 5:00 AM Amit Khandekar <amitdkhan(dot)pg(at)gmail(dot)com>
wrote:
>
> Hi,
>
> Just a quick question before I move on to review the patch ... The
> improvement looks like it is only meant for x86 platforms.
Actually it's meant to be faster for all platforms, since the C fallback is
quite a bit different from HEAD. I've found it to be faster on ppc64le. An
earlier version of the patch was a loser on 32-bit Arm because of alignment
issues, but if you could run the test script attached to [1] on 64-bit Arm,
I'd be curious to see how it does on 0002, and whether 0003 and 0004 make
things better or worse. If there is trouble building on non-x86 platforms,
I'd want to fix that also.
(Note: 0001 is not my patch, and I just include it for the tests)
> Can this be
> done in a portable way by arranging for auto-vectorization ? Something
> like commit 88709176236caf. This way it would benefit other platforms
> as well.
I'm fairly certain that the author of a compiler capable of doing that in
this case would be eligible for some kind of AI prize. :-)
[1]
https://www.postgresql.org/message-id/06d45421-61b8-86dd-e765-f1ce527a5a2f@iki.fi
--
John Naylor
EDB: http://www.enterprisedb.com
From | Date | Subject | |
---|---|---|---|
Next Message | houzj.fnst@fujitsu.com | 2021-03-09 11:54:04 | Questions about CommandIsReadOnly |
Previous Message | Arseny Sher | 2021-03-09 10:56:18 | Enlarge IOS vm cache |