From: tornaria <tornaria@users.noreply.github.com>
To: ml@inbox.vuxu.org
Subject: Re: New package: gf2x-1.3.0
Date: Fri, 02 Apr 2021 21:42:57 +0200 [thread overview]
Message-ID: <20210402194257.ahctoYw_7MELHIMs9VbOrDNJcoBxSoVLcxVbYlfvRkA@z> (raw)
In-Reply-To: <gh-mailinglist-notifications-41a7ca26-5023-4802-975b-f1789d68868e-void-packages-29783@inbox.vuxu.org>
[-- Attachment #1: Type: text/plain, Size: 870 bytes --]
New comment by tornaria on void-packages repository
https://github.com/void-linux/void-packages/pull/29783#issuecomment-812682362
Comment:
I think instead of `--disable-pclmul` we really need `--disable-sse3`?
In fact, baseline `x86_64` has only up to sse2 (see e.g. https://en.wikipedia.org/wiki/X86-64#Microarchitecture_Levels).
Eventually, it would be nice to discuss a way to be able to offer binaries compiled with more features, as there are huge differences to be made for some math software. E.g. gf2x testsuite runs 2-3x faster for me with pclmul enabled than without it.
For the moment what I'm doing is adding an option "native" (disabled by default) which allows one to compile for the build machine.
To catch some of these I'm compiling on a cascade-lake cpu, then runing make check on a nehalem cpu (however, the nehalem already has sse4.2).
next prev parent reply other threads:[~2021-04-02 19:42 UTC|newest]
Thread overview: 26+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-03-27 17:05 [PR PATCH] " dkwo
2021-03-27 17:47 ` [PR PATCH] [Updated] " dkwo
2021-03-28 6:00 ` ericonr
2021-03-28 10:20 ` [PR PATCH] [Updated] " dkwo
2021-03-31 17:20 ` [PR REVIEW] " abenson
2021-03-31 17:23 ` abenson
2021-03-31 17:23 ` abenson
2021-04-01 9:20 ` dkwo
2021-04-01 12:43 ` ericonr
2021-04-01 15:06 ` sgn
2021-04-01 15:06 ` sgn
2021-04-01 19:06 ` dkwo
2021-04-01 19:21 ` ericonr
2021-04-01 19:35 ` gt7-void
2021-04-02 8:20 ` [PR PATCH] [Updated] " dkwo
2021-04-02 8:24 ` dkwo
2021-04-02 12:08 ` dkwo
2021-04-02 12:09 ` dkwo
2021-04-02 19:42 ` tornaria [this message]
2021-04-03 17:40 ` [PR PATCH] [Updated] " dkwo
2021-04-04 15:00 ` [PR REVIEW] " tornaria
2021-04-04 15:00 ` tornaria
2021-04-04 17:39 ` [PR PATCH] [Updated] " dkwo
2021-05-06 18:17 ` [PR PATCH] [Closed]: " dkwo
2021-08-14 18:07 ` dkwo
2021-08-14 18:07 ` [PR PATCH] [Closed]: " dkwo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210402194257.ahctoYw_7MELHIMs9VbOrDNJcoBxSoVLcxVbYlfvRkA@z \
--to=tornaria@users.noreply.github.com \
--cc=ml@inbox.vuxu.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).