Horton

Re: Detection of hadware feature => please do not use signal

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Thomas Munro <thomas.munro@gmail.com>
Cc: Heikki Linnakangas <hlinnaka@iki.fi>, Bastien Roucariès <rouca@debian.org>, pgsql-bugs@lists.postgresql.org
Date: 2024-11-01T04:44:35Z
Lists: pgsql-bugs 
Thomas Munro <thomas.munro@gmail.com> writes:
> On Fri, Nov 1, 2024 at 7:25 AM Tom Lane <tgl@sss.pgh.pa.us> wrote:
>> It occurs to me to wonder whether the existing code works on Windows.
>> Windows-on-ARM wasn't a thing we thought about in 2018, but it's
>> a reasonable target now.

> I looked into that[1] and decided that I was going to ignore it
> completely, because:
> [ oodles o' details ]

Hmm.  So it seems like we could do this:

* Linux, FreeBSD, OpenBSD: do run-time test as recommended
* Windows, macOS: assume that supported ARM hardware can do this
* anything else: assume no hardware CRC

The only thing I'd be even a bit sad about there is not having
NetBSD support.  Maybe we just need to research that a bit more,
or maybe we have to wait for/lobby for them to do what the other
BSDen have done.  But in any case I'm not seeing NetBSD-on-ARM
as a platform that's important enough to block movement on this,
let alone anything else in the "anything else" category.

An alternative to "assume no hardware CRC" could be

* anything else: use the existing SIGILL test

But I do sympathize with Bastien's concerns about that.

			regards, tom lane

Commits

  1. Update configure probes for CFLAGS needed for ARM CRC instructions.

  2. Support runtime CRC feature probing on NetBSD/ARM using sysctl().

  3. Use auxv to check for CRC32 instructions on ARM.

  4. Improve our method for probing the availability of ARM CRC instructions.