From mboxrd@z Thu Jan 1 00:00:00 1970 X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on inbox.vuxu.org X-Spam-Level: X-Spam-Status: No, score=-1.0 required=5.0 tests=DKIM_SIGNED,DKIM_VALID, HTML_MESSAGE,MAILING_LIST_MULTI,RCVD_IN_DNSWL_NONE,RCVD_IN_MSPIKE_H2 autolearn=ham autolearn_force=no version=3.4.4 Received: (qmail 23447 invoked from network); 23 Aug 2021 04:14:12 -0000 Received: from tb-ob21.topicbox.com (173.228.157.67) by inbox.vuxu.org with ESMTPUTF8; 23 Aug 2021 04:14:12 -0000 Received: from tb-mx1.topicbox.com (tb-mx1.nyi.icgroup.com [10.90.30.61]) by tb-ob21.topicbox.com (Postfix) with ESMTP id 3F87B39F79 for ; Mon, 23 Aug 2021 00:14:09 -0400 (EDT) (envelope-from bounce.mMe78513510ae4df2da186c73a.r522be890-2105-11eb-b15e-8d699134e1fa@9fans.bounce.topicbox.com) Received: by tb-mx1.topicbox.com (Postfix, from userid 1132) id F2EE133B7064; Mon, 23 Aug 2021 00:14:08 -0400 (EDT) ARC-Authentication-Results: i=2; topicbox.com; arc=pass; dkim=pass (2048-bit rsa key sha256) header.d=iitbombay-org.20150623.gappssmtp.com header.i=@iitbombay-org.20150623.gappssmtp.com header.b=OSRmOXc/ header.a=rsa-sha256 header.s=20150623 x-bits=2048; dmarc=none policy.published-domain-policy=none policy.applied-disposition=none policy.evaluated-disposition=none (p=none,d=none,d.eval=none) policy.policy-from=p header.from=iitbombay.org; spf=pass smtp.mailfrom=bakul@iitbombay.org smtp.helo=mail-oo1-f52.google.com; x-internal-arc=fail (as.1.topicbox.com=pass, ams.1.topicbox.com=fail (body has been altered)) (Message modified while forwarding at Topicbox) ARC-Message-Signature: i=2; a=rsa-sha256; c=relaxed/relaxed; d= topicbox.com; h=from:content-type:mime-version:subject:date :references:to:in-reply-to:message-id:list-help:list-id :list-post:list-subscribe:reply-to:content-transfer-encoding :list-unsubscribe; s=sysmsg-1; t=1629692048; bh=2r2bxpfQdVlgLNQe KnCvIVv+x8HvXMfypWR8E3Ufqbc=; b=PkStBRbRenaFYN4f9RchfOgN0LiDIcei brx4SRybacFgyS88B2QcRJg7LeYVfjWJyoLEEs79/OLA2p2gSLmgGOe9t5Pf0pJh VicmYVxEVz8eeUU5ijO2JUkU8culzr8q5q1gohuCLMPYMLCi7FTUn7FZ2I9vN1Jh 92FRWPVgdnc= ARC-Seal: i=2; a=rsa-sha256; cv=pass; d=topicbox.com; s=sysmsg-1; t= 1629692048; b=me/ISjYaUc2AS+bTZgT1k+twzq6IshdSX16TQ4pWefWS9SGusG X5OBhSnEXsA622MB+SHmjANwogoRdCSt6CP86Vxg0UF4RNPWAZXNzPdn1RyuIdhA cEaMrnbIGaLewl/ZDbc2SC8HKhaDeAUJ3Qv0rAwFlBc78eEArBg2FeXdo= Authentication-Results: topicbox.com; arc=pass; dkim=pass (2048-bit rsa key sha256) header.d=iitbombay-org.20150623.gappssmtp.com header.i=@iitbombay-org.20150623.gappssmtp.com header.b=OSRmOXc/ header.a=rsa-sha256 header.s=20150623 x-bits=2048; dmarc=none policy.published-domain-policy=none policy.applied-disposition=none policy.evaluated-disposition=none (p=none,d=none,d.eval=none) policy.policy-from=p header.from=iitbombay.org; spf=pass smtp.mailfrom=bakul@iitbombay.org smtp.helo=mail-oo1-f52.google.com; x-internal-arc=fail (as.1.topicbox.com=pass, ams.1.topicbox.com=fail (body has been altered)) (Message modified while forwarding at Topicbox) X-Received-Authentication-Results: tb-mx1.topicbox.com; arc=none (no signatures found); bimi=skipped (DMARC did not pass); dkim=pass (2048-bit rsa key sha256) header.d=iitbombay-org.20150623.gappssmtp.com header.i=@iitbombay-org.20150623.gappssmtp.com header.b=OSRmOXc/ header.a=rsa-sha256 header.s=20150623 x-bits=2048; dmarc=none policy.published-domain-policy=none policy.applied-disposition=none policy.evaluated-disposition=none (p=none,d=none,d.eval=none) policy.policy-from=p header.from=iitbombay.org; iprev=pass smtp.remote-ip=209.85.161.52 (mail-oo1-f52.google.com); spf=pass smtp.mailfrom=bakul@iitbombay.org smtp.helo=mail-oo1-f52.google.com; x-aligned-from=pass (Address match); x-google-dkim=pass (2048-bit rsa key) header.d=1e100.net header.i=@1e100.net header.b=YK6Pch+C; x-me-sender=none; x-ptr=pass smtp.helo=mail-oo1-f52.google.com policy.ptr=mail-oo1-f52.google.com; x-return-mx=pass header.domain=iitbombay.org policy.is_org=yes (MX Records found: alt3.aspmx.l.google.com,alt1.aspmx.l.google.com,aspmx.l.google.com,alt4.aspmx.l.google.com,alt2.aspmx.l.google.com); x-return-mx=pass smtp.domain=iitbombay.org policy.is_org=yes (MX Records found: alt3.aspmx.l.google.com,alt1.aspmx.l.google.com,aspmx.l.google.com,alt4.aspmx.l.google.com,alt2.aspmx.l.google.com); x-tls=pass smtp.version=TLSv1.2 smtp.cipher=ECDHE-RSA-AES256-GCM-SHA384 smtp.bits=256/256; x-vs=clean score=0 state=0 DKIM-Signature: v=1; a=rsa-sha256; c=relaxed; d=9fans.net; h=from :content-type:mime-version:subject:date:references:to :in-reply-to:message-id:list-help:list-id:list-post :list-subscribe:reply-to:content-transfer-encoding :list-unsubscribe; s=dkim-1; bh=n3JY0EGTCbc7jkimy1e92pJjIwcJsfKg Bk4yZ14bfQY=; b=kqoUZ8xTjbE2v6RTqNc2e/6N81JwtiwlGNwsXrP3Gku8tKNM CIrMc+bZMJ21DvnaQihjRhZRgg+GT0eFDFN6FF2dbEDV6RkFP1aJ8fnDHe4MWL6V kwNQ9YpGDDDvjLvZmZw1fXkavDo1Bu3rkZDIqRBoJi27ej3KWoUP1CDh95g= Received: from tb-mx1.topicbox.com (localhost.local [127.0.0.1]) by tb-mx1.topicbox.com (Postfix) with ESMTP id E53A433B6C5A for <9fans@9fans.net>; Mon, 23 Aug 2021 00:13:56 -0400 (EDT) (envelope-from bakul@iitbombay.org) Received: from tb-mx1.topicbox.com (localhost [127.0.0.1]) by tb-mx1.topicbox.com (Authentication Milter) with ESMTP id CB06D86A270; Mon, 23 Aug 2021 00:13:56 -0400 ARC-Seal: i=1; a=rsa-sha256; cv=none; d=topicbox.com; s=arcseal; t= 1629692036; b=CeV2n0iiurn6OMrBGS1TzHfeKFqbXwCswnT19keyySSj8iMSWM 0lKWUHnGTLgKO8nOFo9LnCxUGxlEcjJGEx14OEZu/iZ7E80dZrEjMfkFB7Xh34J7 LdK65hw19fMtcV7LTcS8ANJ92J/Wt0Rvxp0xpxTcZkO7gGrpZZme8A780hJXFayo mR8ZBGPn6hb6Wf3Ze/qKNMVyt4bZrilOLhJP5Am8wMeZoKKgrXzX+yeGA8Ns6QMk xReEfwfmsxbJ6bKhEGEedurEasqjEvx9RnIsM2c6RrHnFzYLN/J2s6YGaMLr6qlR HMaCZ8VHKN5zLVyhQEJ8ldbRv0+8A+471srQ== ARC-Message-Signature: i=1; a=rsa-sha256; c=relaxed/relaxed; d= topicbox.com; h=from:content-type:mime-version:subject:date :references:to:in-reply-to:message-id; s=arcseal; t=1629692036; bh=fI3CZA9hYlCC9eHRbCqcJdBMV5iqoUmh+CYnn4HQBag=; b=PGLSDL1sWNaL cFK5ZrLGh7TaIajGUKel4ZuEBFuel2meipdwkrrsrFeDq2aY/RFq7f4ULlj2Kzd5 uC5VaNkdiwi12YvSH4i1LlI6s35U20ZfExQ6ZCblQjzkwQ8HE+ItT6rA6ClIushb iva5HcvZ7ECd1qVQtJ582PZO/Y0BfPNa3zHdu3CfwT8nW8T/n+rQ5m7iCQTFB20g suxjXZjpbvagQUagTTvzznX7MjDINNslLQepsKRGHDZqIt+KZ843WsvhE7LdPUyT +27BukvEezkB2XVbdbOLJa4kLJe6k8KCPWOEw73F6+QQoTLLM496pF+Dmxrodu3e PErklv4lRQ== ARC-Authentication-Results: i=1; tb-mx1.topicbox.com; arc=none (no signatures found); bimi=skipped (DMARC did not pass); dkim=pass (2048-bit rsa key sha256) header.d=iitbombay-org.20150623.gappssmtp.com header.i=@iitbombay-org.20150623.gappssmtp.com header.b=OSRmOXc/ header.a=rsa-sha256 header.s=20150623 x-bits=2048; dmarc=none policy.published-domain-policy=none policy.applied-disposition=none policy.evaluated-disposition=none (p=none,d=none,d.eval=none) policy.policy-from=p header.from=iitbombay.org; iprev=pass smtp.remote-ip=209.85.161.52 (mail-oo1-f52.google.com); spf=pass smtp.mailfrom=bakul@iitbombay.org smtp.helo=mail-oo1-f52.google.com; x-aligned-from=pass (Address match); x-google-dkim=pass (2048-bit rsa key) header.d=1e100.net header.i=@1e100.net header.b=YK6Pch+C; x-me-sender=none; x-ptr=pass smtp.helo=mail-oo1-f52.google.com policy.ptr=mail-oo1-f52.google.com; x-return-mx=pass header.domain=iitbombay.org policy.is_org=yes (MX Records found: alt3.aspmx.l.google.com,alt1.aspmx.l.google.com,aspmx.l.google.com,alt4.aspmx.l.google.com,alt2.aspmx.l.google.com); x-return-mx=pass smtp.domain=iitbombay.org policy.is_org=yes (MX Records found: alt3.aspmx.l.google.com,alt1.aspmx.l.google.com,aspmx.l.google.com,alt4.aspmx.l.google.com,alt2.aspmx.l.google.com); x-tls=pass smtp.version=TLSv1.2 smtp.cipher=ECDHE-RSA-AES256-GCM-SHA384 smtp.bits=256/256; x-vs=clean score=0 state=0 X-ME-VSCause: gggruggvucftvghtrhhoucdtuddrgedvtddruddtgedgjeelucetufdoteggodetrfdotf fvucfrrhhofhhilhgvmecuhfgrshhtofgrihhlpdggtfgfnhhsuhgsshgtrhhisggvpdfu rfetoffkrfgpnffqhgenuceurghilhhouhhtmecufedttdenucenucfjughrpefhtggguf fffhfvjgfkofesrgdtmherhhdtvdenucfhrhhomhepuegrkhhulhcuufhhrghhuceosggr khhulhesihhithgsohhmsggrhidrohhrgheqnecuggftrfgrthhtvghrnhepuedufeetue evgfdugfekgfehfefgfeelieetueeludeufedvgfduieetteehgeeunecuffhomhgrihhn pegrrhhiiihonhgrrdgvughupdgrmhgurdgtohhmpdhtohhpihgtsghogidrtghomhenuc fkphepvddtledrkeehrdduiedurdehvddpuddtjedrvdduhedrvddvfedrvddvleenucev lhhushhtvghrufhiiigvpedtnecurfgrrhgrmhepihhnvghtpedvtdelrdekhedrudeiud drhedvpdhhvghlohepmhgrihhlqdhoohduqdhfhedvrdhgohhoghhlvgdrtghomhdpmhgr ihhlfhhrohhmpeeosggrkhhulhesihhithgsohhmsggrhidrohhrgheq X-ME-VSScore: 0 X-ME-VSCategory: clean Received-SPF: pass (iitbombay.org: Sender is authorized to use 'bakul@iitbombay.org' in 'mfrom' identity (mechanism 'include:_spf.google.com' matched)) receiver=tb-mx1.topicbox.com; identity=mailfrom; envelope-from="bakul@iitbombay.org"; helo=mail-oo1-f52.google.com; client-ip=209.85.161.52 Received: from mail-oo1-f52.google.com (mail-oo1-f52.google.com [209.85.161.52]) (using TLSv1.2 with cipher ECDHE-RSA-AES256-GCM-SHA384 (256/256 bits)) (No client certificate requested) by tb-mx1.topicbox.com (Postfix) with ESMTPS for <9fans@9fans.net>; Mon, 23 Aug 2021 00:13:56 -0400 (EDT) (envelope-from bakul@iitbombay.org) Received: by mail-oo1-f52.google.com with SMTP id t1-20020a4a54010000b02902638ef0f883so5001723ooa.11 for <9fans@9fans.net>; Sun, 22 Aug 2021 21:13:56 -0700 (PDT) X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20161025; h=x-gm-message-state:from:mime-version:subject:date:references:to :in-reply-to:message-id; bh=fI3CZA9hYlCC9eHRbCqcJdBMV5iqoUmh+CYnn4HQBag=; b=YK6Pch+CKLJ9B/kQRjfEeiIRE0rc8Zdf1LPXnwd36vZbG4hOZJGDGE1SwfXoyu+/R+ gK6ngEm8zgUkd0+J/XxpFJ4mHUfLcGSJ16zJLOBialYT1gZ9WyUl6Y8ZCap+pGIhSoVT GiYl9thXGfOwHWbvU2JV0PbbU5qKAWTqC48UoexnTje4n+Xb2nL1VKP8UzooRJOf3QTA TfP3q2TagADGJGlJ+2W7UGb7wyN8SdHmYRLbjcICkOKdnyCSNcLNvdkG2Tz4fmcAA2P6 gF2Juhsrdf1YLdFBCIILQKy2K6fyhs4GcFfTo67SWLbAtfz+LvvZ5ZJYQ+dejd6Lz7lv xTFA== X-Gm-Message-State: AOAM532w2U7vqzyKJ0swS0Gno1wgEAywVHBmxrsQ3BOaVl/Q+6FlP8Nt eLSt4hb8GPG5td1Vn3tCwHf3UrYo2qjohU63 X-Google-Smtp-Source: ABdhPJwORiO6oL1vz3inIePFS2uWtd3qCzxNKnGLXsYGQsZr9CNs70xNHQMJF1nMoMaZACgBXwtpyg== X-Received: by 2002:a4a:ba0f:: with SMTP id b15mr24234163oop.31.1629692035472; Sun, 22 Aug 2021 21:13:55 -0700 (PDT) Received: from smtpclient.apple (107-215-223-229.lightspeed.sntcca.sbcglobal.net. [107.215.223.229]) by smtp.gmail.com with ESMTPSA id u194sm2637360oie.37.2021.08.22.21.13.54 for <9fans@9fans.net> (version=TLS1_2 cipher=ECDHE-ECDSA-AES128-GCM-SHA256 bits=128/128); Sun, 22 Aug 2021 21:13:55 -0700 (PDT) From: Bakul Shah Content-Type: multipart/alternative; boundary="Apple-Mail=_DB8EC89F-9353-4544-9AA6-EF3045E8F931" Mime-Version: 1.0 (Mac OS X Mail 14.0 \(3654.120.0.1.13\)) Subject: Re: [9fans] Drawterm GPU (was: Software philosophy) Date: Sun, 22 Aug 2021 21:13:53 -0700 References: <92764e35-f5cf-460a-91df-050ba471e6dd@sirjofri.de> <9352EE7C-AE94-4C1C-8738-5A1DA8ECE5A7@iitbombay.org> To: 9fans <9fans@9fans.net> In-Reply-To: Message-Id: <1007A30B-A109-41BA-BD4A-FF06AED8DFEC@iitbombay.org> X-Mailer: Apple Mail (2.3654.120.0.1.13) Topicbox-Policy-Reasoning: allow: sender is a member Topicbox-Message-UUID: 8c4ca5c2-03c8-11ec-b24e-eda59957c845 Archived-At: =?UTF-8?B?PGh0dHBzOi8vOWZhbnMudG9waWNib3guY29tL2dyb3Vwcy85?= =?UTF-8?B?ZmFucy9UYWQyOWJmYzIyM2RjNGZiZS1NZTc4NTEzNTEwYWU0ZGYyZGExODZj?= =?UTF-8?B?NzNhPg==?= List-Help: List-Id: "9fans" <9fans.9fans.net> List-Post: List-Software: Topicbox v0 List-Subscribe: Precedence: list Reply-To: 9fans <9fans@9fans.net> Content-Transfer-Encoding: 7bit List-Unsubscribe: , Topicbox-Delivery-ID: 2:9fans:437d30aa-c441-11e9-8a57-d036212d11b0:522be890-2105-11eb-b15e-8d699134e1fa:Me78513510ae4df2da186c73a:1:svHpCGMEsCsb4CRumhTKQLnr6NlhoJKv83yShKc6CNU --Apple-Mail=_DB8EC89F-9353-4544-9AA6-EF3045E8F931 Content-Transfer-Encoding: quoted-printable Content-Type: text/plain; charset="UTF-8" Don't high end GPUs have thousands of "cores"? Even high end CPUs don't hav= e more than a few dozen cores to 128 or so. While each kind's cores are ver= y different, seems to me GPU/CPU paths have diverged for good. Or we need s= ome massive shift in programming languages + compilers. I lack imagination = how. Still, the thought of the CPUs gaining the complexity of the graphics = engine scares me! -- Bakul > On Aug 22, 2021, at 12:09 PM, Paul Lalonde wro= te: >=20 > I'm pretty sure we're still re-inventing, though it's the CPU's turn to g= ain some of the complexity of the graphics engine. >=20 > Paul >=20 > On Sun, Aug 22, 2021, 12:05 PM Bakul Shah > wrote: > Thanks. Looks like Sutherland's "Wheel of Reincarnation " has not only stopped but exploded :-) Or stopped being applicable. >=20 > -- Bakul >=20 >> On Aug 22, 2021, at 9:23 AM, Paul Lalonde > wrote: >>=20 >> It got complicated because there's no stable interface or ISA. The hard= ware evolved from fixed-function to programmable in a commercial environmen= t where the only meaningful measure was raw performance per dollar at many = price points. Every year the hardware spins and becomes more performant, u= sually faster than Moore's law. With 3D APIs hiding the hardware details t= here is no pressure to make the hardware interface uniform, pretty, or neat= . And with the need for performance there are dozens of fixed function uni= ts that effectively need their own sub-drivers while coordinating at high p= erformance with the other units.=20 >> The system diagrams for GPUs look complex, but they are radical simplifi= cations of what's really on the inside. >>=20 >> Intel really pioneered the open driver stacks, but performance generally= wasn't there. That might be changing now, but I don't know if their recen= tly announced discrete product line will be driver-compatible. >>=20 >> Paul >>=20 >>=20 >> On Sun, Aug 22, 2021 at 8:48 AM Bakul Shah > wrote: >> The FreeBSD amdgpu.ko is over 3Mbytes of compiled code. Not counting the= "firmware" that gets loaded on the GPU board. drm/amd/amdgpu has 200K+ lin= es of source code. drm/amd over 2M lines of code. Intel's i915 seems to be = about 1/10th the amd size. AIUI, this is linux GPU driver code, more or les= s unchanged (FreeBSD has shim code to use it). How did the interface to an = SIMD processor get so complicated? >>=20 >>> On Aug 22, 2021, at 6:44 AM, Paul Lalonde > wrote: >>>=20 >>> I'd love to see GPU support for Plan9. This discussion falls right in= to my professional capacity. I'll say that people generally *grossly* unde= restimate the complexity of a modern GPU and of its supporting software sta= ck. The GPU driver is effectively a second operating system with shared me= mory and DMA interfaces to the host. Even bringing up a modern GPU for jus= t compute tasks is a very large endeavour. >>>=20 >>> That being said, if you want real hardware support, the best place to s= tart is currently AMD's open-source stack. Ignoring the Vulkan bit, unders= tanding their platform abstraction layer (PAL) and shader ISA (https://deve= loper.amd.com/wp-content/resources/Vega_Shader_ISA_28July2017.pdf ) is = the base. The lower hardware levels are reasonably well-described in linux= 's libdrm and its AMD support in amdgpu. >>>=20 >>> Opinions on how to bring this to Plan9? I don't really have any - it's= a huge pile of work with minimal benefit. If you're looking for lightweig= ht graphics, WebGL is a doable path, and almost certainly the right way to = experiment with Plan9-like interfaces to graphics hardware. >>>=20 >>> Paul >>>=20 >>>=20 >>>=20 >>> On Sun, Aug 22, 2021 at 5:30 AM sirjofri > wrote: >>>=20 >>> 22.08.2021 14:10:20 Stuart Morrow >: >>> > Also: >>> >> people have discussed that for years >>> > >>> > They have? I mean I might have seen occasionally someone vaguely >>> > going "some sort of GPU support would be cool to have". That isn't >>> > discussion. >>>=20 >>> I've even heard of someone actually making GPU stuff work on plan 9. I'= ve=20 >>> only heard from their partner, who made a cute glenda thing on a piece = of=20 >>> cloth. I chatted with her a little and told her she should encourage he= r=20 >>> partner for some discussion about this in our channels. It looked like= =20 >>> it's some academic work, but I don't know any details about it. >>>=20 >>> Worst case, someone already has a proper and good GPU implementation fo= r=20 >>> Plan 9 and nobody knows about it. >>>=20 >>> sirjofri >>>=20 >>> Btw if the said person reads this: it would be nice to learn some=20 >>> details. >>>=20 >>> ------------------------------------------ >>> 9fans: 9fans >>> Permalink: https://9fans.topicbox.com/groups/9fans/Tad29bfc223dc4fbe-Md= 3d5cd693c12f948ad4720bc >>> Delivery options: https://9fans.topicbox.com/groups/9fans/subscription = >>=20 >>=20 >>=20 >> -- Bakul >>=20 >=20 > 9fans / 9fans / see discussions + participants + delivery=C2=A0options Permalink ------------------------------------------ 9fans: 9fans Permalink: https://9fans.topicbox.com/groups/9fans/Tad29bfc223dc4fbe-Me7851= 3510ae4df2da186c73a Delivery options: https://9fans.topicbox.com/groups/9fans/subscription --Apple-Mail=_DB8EC89F-9353-4544-9AA6-EF3045E8F931 Content-Transfer-Encoding: quoted-printable Content-Type: text/html; charset="UTF-8" Don't= high end GPUs have thousands of "cores"? Even high end CPUs don&= #39;t have more than a few dozen cores to 128 or so. While each kind's = cores are very different, seems to me GPU/CPU paths have diverged for good.= Or we need some massive shift in programming languages + compilers. I lack= imagination how. Still, the thought of the CPUs gaining the complexity of = the graphics engine scares me!

-- Bakul

On Aug 22, 2021, at 12:09 PM,= Paul Lalonde <pa= ul.a.lalonde@gmail.com> wrote:

I'm pretty sure = we're still re-inventing, though it's the CPU's turn to gain so= me of the complexity of the graphics engine.
Paul

On= Sun, Aug 22, 2021, 12:05 PM Bakul Shah <bakul@iitbombay.org> wrote:
Thanks. Looks like Sutherland's "= Wheel of Reincarnation" has not only stopped but exploded :-) = Or stopped being applicable.

-- Bakul

On Aug 22, 2021, at 9:23 A= M, Paul Lalonde <paul.a.lalonde@gmail.com> wrote= :

It got = complicated because there's no stable interface or ISA.  The hardw= are evolved from fixed-function to programmable in a commercial environment= where the only meaningful measure was raw performance per dollar at many p= rice points.  Every year the hardware spins and becomes more performan= t, usually faster than Moore's law.  With 3D APIs hiding the hardw= are details there is no pressure to make the hardware interface uniform, pr= etty, or neat.  And with the need for performance there are dozens of = fixed function units that effectively need their own sub-drivers while coor= dinating at high performance with the other units. 
Th= e system diagrams for GPUs look complex, but they are radical simplificatio= ns of what's really on the inside.

Intel really pioneered the open driver stacks, but perfor= mance generally wasn't there.  That might be changing now, but I d= on't know if their recently announced discrete product line will be dri= ver-compatible.

Paul

On Sun, Aug 22, 2021 at 8:48 AM Bakul Shah <bakul@iitbo= mbay.org> wrote:
The FreeBSD amdgpu.ko = is over 3Mbytes of compiled code. Not counting the "firmware" tha= t gets loaded on the GPU board. drm/amd/amdgpu has 200K+ lines of source co= de. drm/amd over 2M lines of code. Intel's i915 seems to be about 1/10t= h the amd size. AIUI, this is linux GPU driver code, more or less unchanged= (FreeBSD has shim code to use it). How did the interface to an SIMD proces= sor get so complicated?

On Aug 22, 2021, at 6:44 = AM, Paul Lalonde <paul.a.lalonde@gmail.com> wrot= e:

I'= d love to see  GPU support for Plan9.  This discussion falls righ= t into my professional capacity.  I'll say that people generally *= grossly* underestimate the complexity of a modern GPU and of its supporting= software stack.  The GPU driver is effectively a second operating sys= tem with shared memory and DMA interfaces to the host.  Even bringing = up a modern GPU for just compute tasks is a very large endeavour.

That being said, if you w= ant real hardware support, the best place to start is currently AMD's o= pen-source stack.  Ignoring the Vulkan bit, understanding their platfo= rm abstraction layer (PAL) and shader ISA (https://developer.amd.com/wp-content/resourc= es/Vega_Shader_ISA_28July2017.pdf) is the base.  The lower hardwar= e levels are reasonably well-described in linux's libdrm and its A= MD support in amdgpu.

Opinions on how to bring this to Plan9?  I don't really hav= e any - it's a huge pile of work with minimal benefit.  If you'= ;re looking for lightweight graphics, WebGL is a doable path, and almost ce= rtainly the right way to experiment with Plan9-like interfaces to graphics = hardware.

Paul<= /div>



On Sun, Aug 22, 2021 at 5:30 AM sirjofri <sirjofri+ml-9fans@sirjofri.de> wrote:

22.08.2021 14:10:20 Stuart Morrow <morrow.stuart@gmail.c= om>:
> Also:
>> people have discussed that for years
>
> They have?  I mean I might have seen occasionally someone vaguely=
> going "some sort of GPU support would be cool to have". = ; That isn't
> discussion.

I've even heard of someone actually making GPU stuff work on plan 9. I&= #39;ve
only heard from their partner, who made a cute glenda thing on a piece of <= br class=3D"" /> cloth. I chatted with her a little and told her she should encourage her partner for some discussion about this in our channels. It looked like
it's some academic work, but I don't know any details about it.

Worst case, someone already has a proper and good GPU implementation for Plan 9 and nobody knows about it.

sirjofri

Btw if the said person reads this: it would be nice to learn some
details.

------------------------------------------
9fans: 9fans
Permalink: https://9fans.topicbox.com/groups/9fans/Tad29bfc223dc4fbe-M= d3d5cd693c12f948ad4720bc
Delivery options: https://= 9fans.topicbox.com/groups/9fans/subscription



-- Bakul
<= br class=3D"" />


= --Apple-Mail=_DB8EC89F-9353-4544-9AA6-EF3045E8F931--