public inbox for libc-alpha@sourceware.org
 help / color / mirror / Atom feed
From: "Mallappa, Premachandra" <Premachandra.Mallappa@amd.com>
To: Florian Weimer <fweimer@redhat.com>
Cc: "libc-alpha@sourceware.org" <libc-alpha@sourceware.org>,
	"gcc@gcc.gnu.org" <gcc@gcc.gnu.org>,
	"llvm-dev@lists.llvm.org" <llvm-dev@lists.llvm.org>,
	"x86-64-abi@googlegroups.com" <x86-64-abi@googlegroups.com>,
	"H.J. Lu" <hjl.tools@gmail.com>, "matz@suse.de" <matz@suse.de>,
	Tom Stellard <tstellar@redhat.com>, Jeff Law <law@redhat.com>
Subject: RE: New x86-64 micro-architecture levels
Date: Wed, 22 Jul 2020 16:45:08 +0000	[thread overview]
Message-ID: <DM5PR12MB1787C9B6F13978965B62CCBDE7790@DM5PR12MB1787.namprd12.prod.outlook.com> (raw)
In-Reply-To: <87imegn3s0.fsf@oldenburg2.str.redhat.com>

[AMD Public Use]


> That's deliberate, so that we can use the same x86-* names for 32-bit library selection (once we define matching micro-architecture levels there).

Understood.

> If numbers are out, what should we use instead?
> x86-sse4, x86-avx2, x86-avx512?  Would that work?

Yes please, I think we have to choose somewhere, above would be more descriptive

> Let's merge Level B into level C then?

I would vote for this.

>> Also we would also like to have dynamic loader support for "zen" / 
>> "zen2" as a version of "Level D" and takes preference over Level D, 
>> which may have super-optimized libraries from AMD or other vendors.

> *That* shouldn't be too hard to implement if we can nail down the selection criteria.  Let's call this Zen-specific Level C x86-zen-avx2 for the sake of exposition.

Some way of specifying a superset of "level C" , that "C" will capture fully.

Zen/zen2 takes precedence over Level C, but not Level D, but falls back to "Level C" or "x86-avx2" but not "x86-avx".

I think it is better to run a x86-zen on a x86-avx2 or x86-avx compared to running on a base x86_64 config.

> With the levels I proposed, these aspects are covered.  But if we start to create vendor-specific forks in the feature progression, things get complicated.
I am not strictly proposing OS vendors should create/maintain this (it would be nice if they did), but a support to cached load via system-wide-config. This directory may/will contain a subset of system libs.

> Do you think we need to figure this out in this iteration?  If yes, then I really need a semi-formal description of the selection criteria for this x86-zen-avx2 directory, so that I can passed it along with my psABI proposal.

Preference level (decreasing order) (I can only speak for AMD, others please pitch in)
- system wide config to override (in this case x86-zen)
- x86-avx2
- x86-sse4 (or avx, based on how we name and merge Level B)
- default x86_64

  parent reply	other threads:[~2020-07-22 16:45 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-07-10 17:30 Florian Weimer
2020-07-10 19:14 ` Joseph Myers
2020-07-13  7:55   ` Florian Weimer
2020-07-10 21:42 ` H.J. Lu
2020-07-13  6:23   ` Richard Biener
2020-07-13  7:40     ` Florian Weimer
2020-07-13  7:47       ` Jan Beulich
2020-07-13 13:31         ` H.J. Lu
2020-07-13 13:53           ` Jakub Jelinek
2020-07-13  8:57       ` Richard Biener
2020-07-13  6:49   ` Florian Weimer
2020-07-13 13:30     ` H.J. Lu
2020-07-11  7:40 ` Allan Sandfeld Jensen
2020-07-13  6:58   ` Florian Weimer
2020-07-15 14:38 ` Mark Wielaard
2020-07-15 14:45   ` H.J. Lu
2020-07-15 14:56   ` Florian Weimer
2020-07-21 16:05 ` Mallappa, Premachandra
2020-07-21 18:04   ` Florian Weimer
2020-07-22  1:31     ` Dongsheng Song
2020-07-22  8:44       ` Florian Weimer
2020-07-22  9:26         ` Richard Biener
2020-07-22 10:16           ` Florian Weimer
2020-07-22 13:50             ` Richard Biener
2020-07-22 14:21               ` H.J. Lu
2020-07-31 13:06           ` Carlos O'Donell
2020-07-22  7:48     ` Jan Beulich
2020-07-22 10:34       ` Florian Weimer
2020-07-22 11:41         ` Jan Beulich
2020-07-31 13:20         ` Carlos O'Donell
2020-07-22 16:45     ` Mallappa, Premachandra [this message]
2020-07-23 12:44       ` Michael Matz
2020-07-23 13:21         ` H.J. Lu
2020-07-28 15:54       ` Florian Weimer

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=DM5PR12MB1787C9B6F13978965B62CCBDE7790@DM5PR12MB1787.namprd12.prod.outlook.com \
    --to=premachandra.mallappa@amd.com \
    --cc=fweimer@redhat.com \
    --cc=gcc@gcc.gnu.org \
    --cc=hjl.tools@gmail.com \
    --cc=law@redhat.com \
    --cc=libc-alpha@sourceware.org \
    --cc=llvm-dev@lists.llvm.org \
    --cc=matz@suse.de \
    --cc=tstellar@redhat.com \
    --cc=x86-64-abi@googlegroups.com \
    /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 read-only IMAP folder(s) and NNTP newsgroup(s).