From nobody Tue Apr 4 05:57:45 2023 X-Original-To: multimedia@mlmmj.nyi.freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2610:1c1:1:606c::19:1]) by mlmmj.nyi.freebsd.org (Postfix) with ESMTP id 4PrH9l4HSYz431Ch for ; Tue, 4 Apr 2023 05:58:03 +0000 (UTC) (envelope-from kob6558@gmail.com) Received: from mail-yb1-xb34.google.com (mail-yb1-xb34.google.com [IPv6:2607:f8b0:4864:20::b34]) (using TLSv1.3 with cipher TLS_AES_128_GCM_SHA256 (128/128 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256 client-signature RSA-PSS (2048 bits) client-digest SHA256) (Client CN "smtp.gmail.com", Issuer "GTS CA 1D4" (verified OK)) by mx1.freebsd.org (Postfix) with ESMTPS id 4PrH9k5skxz3F0Y; Tue, 4 Apr 2023 05:58:02 +0000 (UTC) (envelope-from kob6558@gmail.com) Authentication-Results: mx1.freebsd.org; dkim=pass header.d=gmail.com header.s=20210112 header.b=ZNsBvy1T; spf=pass (mx1.freebsd.org: domain of kob6558@gmail.com designates 2607:f8b0:4864:20::b34 as permitted sender) smtp.mailfrom=kob6558@gmail.com; dmarc=pass (policy=none) header.from=gmail.com Received: by mail-yb1-xb34.google.com with SMTP id m16so17071499ybk.0; Mon, 03 Apr 2023 22:58:02 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20210112; t=1680587881; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:from:to:cc:subject:date:message-id:reply-to; bh=rNEDdDOQ1+Ic8e00plp6jhgtq463AKROJNUha999MaM=; b=ZNsBvy1TpHj0lrxVeUk8qAhvlA7V+DjnGkJcQ2AC70n3W7Eny1LGmuUzNF3c/I3mrU BGy/smUE2Br1cKHIVrgg8GeZb87X2cDwQLJRwLX0PUOQzSqCtZ4Bg/SgZFfd+n3jWBKc /roisJE+AasLI1eWEavIhTCsrBlcn2x4bdUtKxSjDMgVPVWGPy/ksSTwOSysfjjL/Nml hc1kClKrSadzPDnNM5YL6kEDHRieuCupTIfOq4sweKbfyq0HwsqV+K42g+Z3w1onpT/s dOoMSYf2eTdoFo3fYfA72BxHsOldAXUwyAaBo6aNmdGE1kDypwCPnLUzWCXJHGvQn445 02nA== X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20210112; t=1680587881; h=cc:to:subject:message-id:date:from:in-reply-to:references :mime-version:x-gm-message-state:from:to:cc:subject:date:message-id :reply-to; bh=rNEDdDOQ1+Ic8e00plp6jhgtq463AKROJNUha999MaM=; b=uTdsVcZOB5qxrCyY4+8vhIj/Qh2fijViDFEZyMJqo+zCLaqj1oUtVPX0Ynmd/cJiWw iwIb//80MhD0C4qvkbnwm3vd6Q3aork1hmxxP8Clmh94/74Bmb0Avrbren4u7MldIcd2 Sle4hJF8g6XkEleMIxtTK63Cl7ZhNtlGGlu7zYxrcSUUL3xu3AQziHtsEKsEooacmxlC 6XSOj5bdpPj4127rVYCItoQxCn2ml1H6evfvpdBxgtgp2NQ75afB6klhaRQQRFY742Un ngDNTiqc396DipoMxTzr7PjocKIDCWsGW8+JBEpTobP1r0j1TgzKGV0g9dIx5hVloZUc 1jmw== X-Gm-Message-State: AAQBX9eVdXn17os5tUepx1L9aFEaYoMDcZArEma4egmqPL0yIaeQ2tr/ /u8/OffpwSThBvQV5vYG/Y5ydXu/Jl1lQZDs0UZM1nwvxTc= X-Google-Smtp-Source: AKy350ZdN+ECudYLLYJel5iCGwx7ifSOzo75rN+CMFWYn+Z7dlFbk00hPu/RmbKflMTKimSBNoqJvuWhAUyIs+fHEYQ= X-Received: by 2002:a25:7491:0:b0:b48:5eaa:a804 with SMTP id p139-20020a257491000000b00b485eaaa804mr991740ybc.0.1680587881376; Mon, 03 Apr 2023 22:58:01 -0700 (PDT) List-Id: Multimedia discussions List-Archive: https://lists.freebsd.org/archives/freebsd-multimedia List-Help: List-Post: List-Subscribe: List-Unsubscribe: Sender: owner-freebsd-multimedia@freebsd.org MIME-Version: 1.0 References: In-Reply-To: From: Kevin Oberman Date: Mon, 3 Apr 2023 22:57:45 -0700 Message-ID: Subject: Re: Poor performance with Alder Lake graphics (ThinkPad T16) To: Jan Beich Cc: multimedia@freebsd.org Content-Type: multipart/alternative; boundary="000000000000aef08105f87c5b2a" X-Spamd-Result: default: False [-3.70 / 15.00]; NEURAL_HAM_MEDIUM(-1.00)[-1.000]; NEURAL_HAM_LONG(-1.00)[-1.000]; NEURAL_HAM_SHORT(-1.00)[-1.000]; DMARC_POLICY_ALLOW(-0.50)[gmail.com,none]; FORGED_SENDER(0.30)[rkoberman@gmail.com,kob6558@gmail.com]; R_DKIM_ALLOW(-0.20)[gmail.com:s=20210112]; R_SPF_ALLOW(-0.20)[+ip6:2607:f8b0:4000::/36]; MIME_GOOD(-0.10)[multipart/alternative,text/plain]; RCVD_TLS_LAST(0.00)[]; MLMMJ_DEST(0.00)[multimedia@freebsd.org]; MIME_TRACE(0.00)[0:+,1:+,2:~]; RCVD_IN_DNSWL_NONE(0.00)[2607:f8b0:4864:20::b34:from]; ASN(0.00)[asn:15169, ipnet:2607:f8b0::/32, country:US]; FREEMAIL_ENVFROM(0.00)[gmail.com]; RCPT_COUNT_TWO(0.00)[2]; MID_RHS_MATCH_FROMTLD(0.00)[]; FROM_NEQ_ENVFROM(0.00)[rkoberman@gmail.com,kob6558@gmail.com]; FROM_HAS_DN(0.00)[]; ARC_NA(0.00)[]; DKIM_TRACE(0.00)[gmail.com:+]; TO_DN_SOME(0.00)[]; TO_MATCH_ENVRCPT_ALL(0.00)[]; FREEMAIL_FROM(0.00)[gmail.com]; DWL_DNSWL_NONE(0.00)[gmail.com:dkim]; RCVD_COUNT_TWO(0.00)[2] X-Rspamd-Queue-Id: 4PrH9k5skxz3F0Y X-Spamd-Bar: --- X-ThisMailContainsUnwantedMimeParts: N --000000000000aef08105f87c5b2a Content-Type: text/plain; charset="UTF-8" Content-Transfer-Encoding: quoted-printable On Sun, Apr 2, 2023 at 9:51=E2=80=AFAM Jan Beich wrote= : > Kevin Oberman writes: > > > Since update to drm-550-kmod I noticed n seeming improvement over scfb.= I > > ran glxgears and saw agour 900 fps. > > glxgears should never show more FPS than monitor refresh rate unless it > uses software rendering or VSYNC is disabled via vblank_mode=3D0 in > environ(7) or ~/.drirc. If you need to benchmark GPU use at least > glmark2 + vkmark.h Yes, I meant 515 and I was running novsync. glmark2 reports: ** GLX does not support GLX_EXT_swap_control or GLX_MESA_swap_control! ** Failed to set swap interval. Results may be bounded above by refresh rate. =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D glmark2 2023.01 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D OpenGL Information GL_VENDOR: Mesa/X.org GL_RENDERER: llvmpipe (LLVM 15.0.7, 256 bits) GL_VERSION: 4.5 (Compatibility Profile) Mesa 22.3.7 Surface Config: buf=3D32 r=3D8 g=3D8 b=3D8 a=3D8 depth=3D32 stencil=3D0= samples=3D0 Surface Size: 800x600 windowed =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D [build] use-vbo=3Dfalse: FPS: 188 FrameTime: 5.330 ms [build] use-vbo=3Dtrue: FPS: 175 FrameTime: 5.718 ms [texture] texture-filter=3Dnearest: FPS: 492 FrameTime: 2.033 ms [texture] texture-filter=3Dmipmap: FPS: 377 FrameTime: 2.655 ms [shading] shading=3Dgouraud: FPS: 127 FrameTime: 7.888 ms [shading] shading=3Dblinn-phong-inf: FPS: 128 FrameTime: 7.819 ms [shading] shading=3Dphong: FPS: 123 FrameTime: 8.162 ms [shading] shading=3Dcel: FPS: 104 FrameTime: 9.682 ms [bump] bump-render=3Dhigh-poly: FPS: 72 FrameTime: 13.924 ms [bump] bump-render=3Dnormals: FPS: 314 FrameTime: 3.191 ms [bump] bump-render=3Dheight: FPS: 351 FrameTime: 2.852 ms [effect2d] kernel=3D0,1,0;1,-4,1;0,1,0;: FPS: 383 FrameTime: 2.612 ms [effect2d] kernel=3D1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 404 FrameTime: 2.4= 76 ms [pulsar] light=3Dfalse:quads=3D5:texture=3Dfalse: FPS: 433 FrameTime: 2.312= ms [desktop] blur-radius=3D5:effect=3Dblur:passes=3D1:separable=3Dtrue:windows= =3D4: FPS: 177 [desktop] effect=3Dshadow:windows=3D4: FPS: 214 FrameTime: 4.677 ms [buffer] columns=3D200:interleave=3Dfalse:update-dispersion=3D0.9:update-fraction=3D= 0.5:update-method=3Dmap: FPS: 94 FrameTime: 10.693 ms [buffer] columns=3D200:interleave=3Dfalse:update-dispersion=3D0.9:update-fraction=3D= 0.5:update-method=3Dsubdata: FPS: 92 FrameTime: 10.904 ms [buffer] columns=3D200:interleave=3Dtrue:update-dispersion=3D0.9:update-fraction=3D0= .5:update-method=3Dmap: FPS: 88 FrameTime: 11.398 ms [ideas] speed=3Dduration: FPS: 132 FrameTime: 7.579 ms [jellyfish] : FPS: 95 FrameTime: 10.626 ms [terrain] : FPS: 13 FrameTime: 77.267 ms [shadow] : FPS: 92 FrameTime: 10.976 ms [refract] : FPS: 23 FrameTime: 43.605 ms [conditionals] fragment-steps=3D0:vertex-steps=3D0: FPS: 178 FrameTime: 5.6= 40 ms [conditionals] fragment-steps=3D5:vertex-steps=3D0: FPS: 145 FrameTime: 6.9= 31 ms [conditionals] fragment-steps=3D0:vertex-steps=3D5: FPS: 241 FrameTime: 4.1= 61 ms [function] fragment-complexity=3Dlow:fragment-steps=3D5: FPS: 167 FrameTime= : 6.002 ms [function] fragment-complexity=3Dmedium:fragment-steps=3D5: FPS: 151 FrameT= ime: 6.626 ms [loop] fragment-loop=3Dfalse:fragment-steps=3D5:vertex-steps=3D5: FPS: 146 FrameTime: 6.892 ms [loop] fragment-steps=3D5:fragment-uniform=3Dfalse:vertex-steps=3D5: FPS: 1= 52 FrameTime: 6.600 ms [loop] fragment-steps=3D5:fragment-uniform=3Dtrue:vertex-steps=3D5: FPS: 15= 0 FrameTime: 6.709 ms =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D glmark2 Score: 192 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D vkmark was less successful: =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D vkmark 2017.08-29-gd872846 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D Vendor ID: 0x10005 Device ID: 0x0 Device Name: llvmpipe (LLVM 15.0.7, 256 bits) Driver Version: 1 Device UUID: 76616c2d750000000000000000000000 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D [vertex] device-local=3Dtrue: FPS: 143 FrameTime: 6.993 ms [vertex] device-local=3Dfalse: FPS: 117 FrameTime: 8.547 ms [texture] anisotropy=3D0: FPS: 318 FrameTime: 3.145 ms [texture] anisotropy=3D16: At that point, the benchmark hungup up and, after waiting a few minutes, I killed it. While I have little understanding of these benchmarks, the numbers look pretty bad as does the warning at the start of the run. (That note was repeated at the start of each test.) Assuming drm-550-kmod is typo for drm-515-kmod. > > > [vo/gpu] VT_GETMODE failed: Inappropriate ioctl for device > > [vo/gpu/opengl] Failed to set up VT switcher. Terminal switching will b= e > > unavailable. > > VT_GETMODE isn't supposed to be used outside of --gpu-context=3Ddrm and > maybe --gpu-context=3Ddisplayvk. Both are intended for playing videos > with the best performance directly on KMS console like /dev/ttyv0. > > In short, ignore this error due to auto-detection scrambling to find any > accelerated GPU context. > > > [vo/gpu/opengl] Suspected software renderer or indirect context. > [...] > > error: Kernel is too old (4.16+ required) or unusable for Iris. > > Check your dmesg logs for loading failures. > [...] > > libEGL warning: DRI2: failed to authenticate > > Looks like OpenGL acceleration failed under Xorg for some reason. > Maybe caused by https://github.com/freebsd/edrm-kmod/issues/175 > Nope. glxinfo looks fine. Can't believe that this graphics is MUCH worse than my 12 year old Ivy Lake system. > Am I doing something wrong or is there a lack of support for Alder Lake > > graphics? N.B. Alder Lake graphics is Iris and I think Gallium might be > > needed. > > If /dev/dri exists and kmscube runs fine then your GPU is properly > supported. > It exists and kmscube runs. > > VA-API is separate from OpenGL. VA-API on Alder Lake needs > libva-intel-media-driver > but the actual support depends on PCI ID e.g., may require newer version > than then one packaged. > OK. I am now running libva-intel-media-driver. Still not seeing the video acceleration that I would expect. 12 threads running and hte total CPU usage is 9& to play a 1392x1072 video. > > > Should I use xf86-intel-drive? > > Did you miss Intel copyright in libwayland and Wayland code in Mesa?\ Never looked, I'll admit. > > Intel stopped contributing to modesetting years ago, sometime after > Ice Lake launch. xf86-video-intel was deprecated by Intel before that. > modesetting is still maintained by others and doesn't need GPU-specific > support but has many bugs compared to Wayland rendering e.g., atomic > mode setting is broken. This is likely a limitation of DDX layer in Xorg > compared to other X11 servers. > I was unaware that Intel dropped out of the modesetting work. I was aware that xf86-video-intel was really out of date which is why I moved to modesetting. xf86-video-intel may work if you add Alder Lake PCI IDs similar to Tiger > Lake. > However, performance on Cannon Lake and later is likely very poor due to > > https://gitlab.freedesktop.org/xorg/driver/xf86-video-intel/-/commit/7181= c5a41c3f Yes, that looks like a problem. Alder Lake is 12th Gen. and I don't supose that a new backend is on the horizon, but my 10th Gen Comet Lake system had much better performance and it should have had that issue, too. Color me very confused. --=20 Kevin Oberman, Part time kid herder and retired Network Engineer E-mail: rkoberman@gmail.com PGP Fingerprint: D03FB98AFA78E3B78C1694B318AB39EF1B055683 --000000000000aef08105f87c5b2a Content-Type: text/html; charset="UTF-8" Content-Transfer-Encoding: quoted-printable
On Sun, Apr 2, 2023 at 9:51=E2= =80=AFAM Jan Beich <jbeich@freebsd= .org> wrote:
Kevin Oberman <rkoberman@gmail.com> writes:

> Since update to drm-550-kmod I noticed n seeming improvement over scfb= . I
> ran glxgears and saw agour 900 fps.

glxgears should never show more FPS than monitor refresh rate unless it
uses software rendering or VSYNC is disabled via vblank_mode=3D0 in
environ(7) or ~/.drirc. If you need to benchmark GPU use at least
glmark2 + vkmark.h
=C2=A0
Yes, I meant 515 and I was running novsync.
glmark2 reports:
** GLX does not support GLX_EXT_swa= p_control or GLX_MESA_swap_control!
** Failed to set swap interval. Resu= lts may be bounded above by refresh rate.
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
=C2=A0 =C2= =A0 glmark2 2023.01
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
=C2=A0 =C2=A0 OpenGL Information=
=C2=A0 =C2=A0 GL_VENDOR: =C2=A0 =C2=A0 =C2=A0Mesa/X.org
=C2=A0 =C2= =A0 GL_RENDERER: =C2=A0 =C2=A0llvmpipe (LLVM 15.0.7, 256 bits)
=C2=A0 = =C2=A0 GL_VERSION: =C2=A0 =C2=A0 4.5 (Compatibility Profile) Mesa 22.3.7=C2=A0 =C2=A0 Surface Config: buf=3D32 r=3D8 g=3D8 b=3D8 a=3D8 depth=3D32 = stencil=3D0 samples=3D0
=C2=A0 =C2=A0 Surface Size: =C2=A0 800x600 windo= wed
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D
[build] use-vbo=3Dfalse: FPS: 188 F= rameTime: 5.330 ms
[build] use-vbo=3Dtrue: FPS: 175 FrameTim= e: 5.718 ms
[texture] texture-filter=3Dnearest: FPS: 492 Fra= meTime: 2.033 ms
[texture] texture-filter=3Dmipmap: FPS: 377= FrameTime: 2.655 ms
[shading] shading=3Dgouraud: FPS: 127 F= rameTime: 7.888 ms
[shading] shading=3Dblinn-phong-inf: FPS:= 128 FrameTime: 7.819 ms
[shading] shading=3Dphong: FPS: 123= FrameTime: 8.162 ms
[shading] shading=3Dcel: FPS: 104 FrameTime: 9.68= 2 ms
[bump] bump-render=3Dhigh-poly: FPS: 72 FrameTime: 13.924 ms=
[bump] bump-render=3Dnormals: FPS: 314 FrameTime: 3.19= 1 ms
[bump] bump-render=3Dheight: FPS: 351 FrameTime: 2.852 ms
[effect2d] kernel=3D0,1,0;1,-4,1;0,1,0;: FPS: 383 FrameTime: 2.612 m= s
[effect2d] kernel=3D1,1,1,1,1;1,1,1,1,1;1,1,1,1,1;: FPS: 404 Fr= ameTime: 2.476 ms
[pulsar] light=3Dfalse:quads=3D5:texture=3Dfals= e: FPS: 433 FrameTime: 2.312 ms
[desktop] blur-radius=3D5:effect= =3Dblur:passes=3D1:separable=3Dtrue:windows=3D4: FPS: 177
[d= esktop] effect=3Dshadow:windows=3D4: FPS: 214 FrameTime: 4.677 ms
[buffer] columns=3D200:interleave=3Dfalse:update-dispersion=3D0.9:update-f= raction=3D0.5:update-method=3Dmap: FPS: 94 FrameTime: 10.693 ms
[= buffer] columns=3D200:interleave=3Dfalse:update-dispersion=3D0.9:update-fra= ction=3D0.5:update-method=3Dsubdata: FPS: 92 FrameTime: 10.904 ms
[buffer] columns=3D200:interleave=3Dtrue:update-dispersion=3D0.9:update-fr= action=3D0.5:update-method=3Dmap: FPS: 88 FrameTime: 11.398 ms
[i= deas] speed=3Dduration: FPS: 132 FrameTime: 7.579 ms
[jellyfish] = <default>: FPS: 95 FrameTime: 10.626 ms
[terrain] <defau= lt>: FPS: 13 FrameTime: 77.267 ms
[shadow] <defau= lt>: FPS: 92 FrameTime: 10.976 ms
[refract] <default&g= t;: FPS: 23 FrameTime: 43.605 ms
[conditionals] fragment-ste= ps=3D0:vertex-steps=3D0: FPS: 178 FrameTime: 5.640 ms
[condi= tionals] fragment-steps=3D5:vertex-steps=3D0: FPS: 145 FrameTime: 6.931 ms<= /div>
[conditionals] fragment-steps=3D0:vertex-steps=3D5: FPS: 241= FrameTime: 4.161 ms
[function] fragment-complexity=3Dlow:fr= agment-steps=3D5: FPS: 167 FrameTime: 6.002 ms
[function] fr= agment-complexity=3Dmedium:fragment-steps=3D5: FPS: 151 FrameTime: 6.626 ms=
[loop] fragment-loop=3Dfalse:fragment-steps=3D5:vertex-step= s=3D5: FPS: 146 FrameTime: 6.892 ms
[loop] fragment-steps=3D= 5:fragment-uniform=3Dfalse:vertex-steps=3D5: FPS: 152 FrameTime: 6.600 ms
[loop] fragment-steps=3D5:fragment-uniform=3Dtrue:vertex-step= s=3D5: FPS: 150 FrameTime: 6.709 ms

=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D
=C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2= =A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 =C2=A0 glmark2 Score: 192 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D

vkmark was less succe= ssful:
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
=C2=A0 =C2=A0 vkmark 2017.08-29-= gd872846
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D
=C2=A0 =C2=A0 Vendor ID: =C2=A0 =C2=A0 =C2= =A00x10005
=C2=A0 =C2=A0 Device ID: =C2=A0 =C2=A0 =C2=A00x0
=C2=A0 = =C2=A0 Device Name: =C2=A0 =C2=A0llvmpipe (LLVM 15.0.7, 256 bits)
=C2=A0= =C2=A0 Driver Version: 1
=C2=A0 =C2=A0 Device UUID: =C2=A0 =C2=A076616c= 2d750000000000000000000000
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D= =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
[vertex] device-local=3Dtr= ue: FPS: 143 FrameTime: 6.993 ms
[vertex] device-local=3Dfalse: FPS: 117= FrameTime: 8.547 ms
[texture] anisotropy=3D0: FPS: 318 FrameTime: 3.145= ms
[texture] anisotropy=3D16:
At that point, the benchma= rk hungup up and, after waiting a few minutes, I killed it.
=
While I have little understanding of these benchmarks, = the numbers look pretty bad as does the warning at the start of the run. (T= hat note was repeated at the start of each test.)

=
Assuming drm-550-kmod is typo for drm-515-kmod.

> [vo/gpu] VT_GETMODE failed: Inappropriate ioctl for device
> [vo/gpu/opengl] Failed to set up VT switcher. Terminal switching will = be
> unavailable.

VT_GETMODE isn't supposed to be used outside of --gpu-context=3Ddrm and=
maybe --gpu-context=3Ddisplayvk. Both are intended for playing videos
with the best performance directly on KMS console like /dev/ttyv0.

In short, ignore this error due to auto-detection scrambling to find any accelerated GPU context.

> [vo/gpu/opengl] Suspected software renderer or indirect context.
[...]
> error: Kernel is too old (4.16+ required) or unusable for Iris.
> Check your dmesg logs for loading failures.
[...]
> libEGL warning: DRI2: failed to authenticate

Looks like OpenGL acceleration failed under Xorg for some reason.
Maybe caused by https://github.com/freebsd/edrm-kmod/issues/175
=C2=A0
<= div>Nope. glxinfo looks fine.

Can't believe that this graphics is MUCH wor= se than my 12 year old Ivy Lake system.
> Am I doing something wrong or is there a lack of support for Alder Lak= e
> graphics? N.B. Alder Lake graphics is Iris and I think Gallium might b= e
> needed.

If /dev/dri exists and kmscube runs fine then your GPU is properly supporte= d.

It exists and kmscube runs.

VA-API is separate from OpenGL. VA-API on Alder Lake needs libva-intel-medi= a-driver
but the actual support depends on PCI ID e.g., may require newer version than then one packaged.

OK. I am n= ow running libva-intel-media-driver. Still not seeing the video acceleratio= n that I would expect. 12 threads running and hte total CPU usage is 9&= to play a 1392x1072 video.

>=C2=A0 Should I use xf86-intel-drive?

Did you miss Intel copyright in libwayland and Wayland code in Mesa?\

Never looked, I'll ad= mit.

Intel stopped contributing to modesetting years ago, sometime after
Ice Lake launch. xf86-video-intel was deprecated by Intel before that.
modesetting is still maintained by others and doesn't need GPU-specific=
support but has many bugs compared to Wayland rendering e.g., atomic
mode setting is broken. This is likely a limitation of DDX layer in Xorg compared to other X11 servers.

I w= as unaware that Intel dropped out of the modesetting work. I was aware that= xf86-video-intel was really out of date which is why I moved to modesettin= g.

xf86-video-intel may work if you add Alder Lake PCI IDs similar to Tiger La= ke.
However, performance on Cannon Lake and later is likely very poor due to https://gitlab.free= desktop.org/xorg/driver/xf86-video-intel/-/commit/7181c5a41c3f

Yes, that looks like a problem. Alder Lake is = 12th Gen. and I don't supose that a new backend is on the horizon, but = my 10th Gen Comet Lake system had much better performance and it should hav= e had that issue, too.

Color me very con= fused.
--
=
Kevin Oberman, Part ti= me kid herder and retired Network Engineer
E-mail: rkoberman@gmail.com
P= GP Fingerprint: D03FB98AFA78E3B78C1694B318AB39EF1B055683
<= /div>
--000000000000aef08105f87c5b2a--