Package Details: libva-intel-driver-g45-h264 1:2.4.1-2

Git Clone URL: https://aur.archlinux.org/libva-intel-driver-g45-h264.git (read-only, click to copy)
Package Base: libva-intel-driver-g45-h264
Description: Video Acceleration (VA) API for Linux
Upstream URL: https://01.org/linuxmedia/vaapi
Licenses: MIT
Conflicts: libva-intel-driver
Provides: libva-intel-driver
Replaces: libva-intel-driver
Submitter: pedromj
Maintainer: alium
Last Packager: alium
Votes: 65
Popularity: 0.066158
First Submitted: 2014-08-04 02:05 (UTC)
Last Updated: 2021-04-17 16:03 (UTC)

Required by (32)

Sources (1)

Latest Comments

boarim commented on 2022-07-11 07:43 (UTC)

@brikler

Thank you for your reply. I tried with gpu-context=wayland but it yields the same error as with the defaults (gpu-context=auto).

brikler commented on 2022-07-10 17:06 (UTC) (edited on 2022-07-10 18:44 (UTC) by brikler)

@boarim

--gpu-context=<sys>
    The value auto (the default) selects the GPU context. You can also pass help to get a complete list of compiled in backends (sorted by autoprobe order).

auto
    auto-select (default)
…
…
wayland
    Wayland/EGL

read more: https://man.archlinux.org/man/mpv.1 it helps …perhaps ;)

boarim commented on 2022-07-10 15:31 (UTC)

Running under Wayland, I cannot manage to make video hardware acceleration work. I have the same error than brikler here, with the same configuration when using mpv. In addition it makes my Wayland session crash.

If I try to set vo=vaapi, I get the same error than fine2006 under XWayland.

Tested on a H264 encoded video.

Here is my setup:

$ vainfo
vainfo: VA-API version: 1.15 (libva 2.15.0)
vainfo: Driver version: Intel i965 driver for Intel(R) GM45 Express Chipset - 2.4.1
vainfo: Supported profile and entrypoints
      VAProfileMPEG2Simple            : VAEntrypointVLD
      VAProfileMPEG2Main              : VAEntrypointVLD
      VAProfileH264ConstrainedBaseline: VAEntrypointVLD
      VAProfileH264Main               : VAEntrypointVLD
      VAProfileH264High               : VAEntrypointVLD

ViNi-Arco commented on 2022-02-27 22:38 (UTC)

Thank you very much, it works wonders

brikler commented on 2021-11-28 12:14 (UTC) (edited on 2022-03-01 07:20 (UTC) by brikler)

hi,

i have a problem with MPV/vlc and vaapi:

Audio/Video desynchronisation detected! Possible reasons include too slow
hardware, temporary CPU spikes, broken drivers, and broken files. Audio
position will not match to the video (see A-V status field).
i965: Failed to submit batchbuffer: Input/output errorCache: 273s/26MB
Abgebrochen (Speicherabzug geschrieben)
#mpv.conf
--hwdec=vaapi
--vo=gpu
--hwdec-codecs=auto
intel-gpu-top: Intel Cantiga (Gen4)

can anybody help me to troubleshot and fix this problem?

edit: it seems vo=vaapi solves my problem, because it avoid buggy opengel.

Un1Gfn commented on 2021-10-10 02:21 (UTC)

please add python2 to makedepends=()

alium commented on 2021-06-07 18:57 (UTC)

vo=gpu is the preferred choice in any case and includes VA-API support via hwdec=vaapi or vaapi-copy.

fine2006 commented on 2021-06-07 18:55 (UTC)

Running mpv --vo=vaapi

[vaapi] libva: vaGetDriverNameByIndex() failed with unknown libva error, driver_name = (null)
[vaapi] Failed to initialize VAAPI: unknown libva error
Error opening/initializing the selected video_out (--vo) device.
Video: no video

alium commented on 2021-04-17 16:05 (UTC) (edited on 2021-04-17 16:06 (UTC) by alium)

@solarkraft: thanks for comment. I checked the conflict between intel-gpu-tools and libva-intel-driver-g45-h264 (which in the past really existed) and conflict between files exist no more, so i removed intel-gpu-tools from conflicts=() array.

solarkraft commented on 2021-04-17 15:05 (UTC) (edited on 2021-04-17 15:08 (UTC) by solarkraft)

intel-gpu-tools and libva-intel-driver-g45-h264 are in conflict :/ How do they conflict? No way around this?

VxlerieUwU commented on 2020-10-09 11:40 (UTC)

@aviallon GM965 =/= gm45

alium commented on 2020-10-03 07:12 (UTC)

@aviallon rebuild this package with newest libva, maybe it help

aviallon commented on 2020-10-03 00:04 (UTC)

Hello, has anyone had success with this iGPU and va-api ?

00:02.0 VGA compatible controller: Intel Corporation Mobile GM965/GL960 Integrated Graphics Controller (primary) (rev 0c) (prog-if 00 [VGA controller])
        Subsystem: Hewlett-Packard Company Compaq 6710b
        Control: I/O+ Mem+ BusMaster+ SpecCycle- MemWINV- VGASnoop- ParErr- Stepping- SERR- FastB2B- DisINTx-
        Status: Cap+ 66MHz- UDF- FastB2B+ ParErr- DEVSEL=fast >TAbort- <TAbort- <MAbort- >SERR- <PERR- INTx-
        Latency: 0
        Interrupt: pin A routed to IRQ 16
        Region 0: Memory at e4300000 (64-bit, non-prefetchable) [size=1M]
        Region 2: Memory at d0000000 (64-bit, prefetchable) [size=256M]
        Region 4: I/O ports at 4000 [size=8]
        Expansion ROM at 000c0000 [virtual] [disabled] [size=128K]
        Capabilities: [90] MSI: Enable- Count=1/1 Maskable- 64bit-
                Address: 00000000  Data: 0000
        Capabilities: [d0] Power Management version 3
                Flags: PMEClk- DSI+ D1- D2- AuxCurrent=0mA PME(D0-,D1-,D2-,D3hot-,D3cold-)
                Status: D0 NoSoftRst- PME-Enable- DSel=0 DScale=0 PME-
                Bridge: PM- B3+
        Kernel driver in use: i915
        Kernel modules: i915, intelfb

Sadly, vainfo is not happy, and I can't find why on the internet.

$ vainfo
libva error: /usr/lib/dri/i965_drv_video.so init failed
vaInitialize failed with error code -1 (unknown libva error),exit

Any help (including saying it just doesn't work with this particular GPU or go see in this other place) is appreciated.

usuariopolivalen commented on 2020-08-24 20:20 (UTC)

for me smplayer with mpv is not working in either normal or experimental version. But at least experimental ones don't hang my gpu.. the only thing is that doesn't do the job but not break all the system as the normal ones.

[ffmpeg/video] h264: Hardware does not offer any usable surface formats. it's my problem that has the output of mpv. Vo vaapi and vaapi with and without copy. It doesn't mind.

usuariopolivalen commented on 2020-08-21 10:24 (UTC)

@MaheshWiz you have my same old laptop, but mines has t6700 I think.

My advice is to use smplayer with mpv that is a lot better than VLC and has less overhead using vaapi. VLC is a lot worse than mpv for vaapi accelerated video. On the other hand, thanks for the update too. It was needed for so many old computers!

VxlerieUwU commented on 2020-08-15 09:55 (UTC)

MPV works using --vo=vaapi

MaheshWiz commented on 2020-08-13 18:50 (UTC) (edited on 2020-08-13 18:56 (UTC) by MaheshWiz)

If using mpv or video clip is the only way to test your driver. its working 100%.!! I was only putting my own findings to help others use VLC player & Chromium to extract maximum benefits of HW Accel using g45 driver & never meant to demean you in any way. I can only thank you for great help in updating the driver for us. BTW I liked your experimental driver more since its lot smoother compare to the regular.

My vainfo: VAProfileMPEG2Simple : VAEntrypointVLD VAProfileMPEG2Main : VAEntrypointVLD VAProfileH264ConstrainedBaseline: VAEntrypointVLD VAProfileH264Main : VAEntrypointVLD VAProfileH264High : VAEntrypointVLD

alium commented on 2020-08-13 15:24 (UTC) (edited on 2020-08-13 15:25 (UTC) by alium)

@MaheshWiz and other guys: DON'T use chromium and youtube as standart for h264 testing on g45, please. Use mpv and some movie or clip in h264 instead.

Chrome/chromium have vaapi experimental only and not everytime stable, youtube offer in many cases AV1 codec instead h264 (i tested it with h264ify and derivates and still was AV1)

MaheshWiz commented on 2020-08-13 14:53 (UTC) (edited on 2020-08-13 15:02 (UTC) by MaheshWiz)

I was giving feedback for the experimental driver. While the earlier comment was related to the normal g45 driver via Vlc. I was using chromium to view 1080p video using the experimental driver it ran smoothly compare to regular version.. But only drawback was high CPU usage. 6-15% CPU is only achievable VIA VLC player using my settings. I will post screenshot soon.

alium commented on 2020-08-13 14:04 (UTC)

@MaheshWiz: "cpu keeps hitting 40%-80% usage" <-- that means rather, that GPU is not used for HW decoding. Because before you wrote cpu usage: 6-15% and that should be if GPU HW decoding in use.

can you please put here 'vainfo' output?

MaheshWiz commented on 2020-08-13 13:50 (UTC) (edited on 2020-08-13 13:52 (UTC) by MaheshWiz)

@alium: I am currently using your experimental version, works similar but cpu keeps hitting 40%-80% usage. Both version shows intel_gpu_top -> Render/3D busy 60-80%. Does it mean gpu rendering ? colors in vidoes look much better in experimental version, don't know could be my opinion.

alium commented on 2020-08-13 10:17 (UTC) (edited on 2020-08-13 10:17 (UTC) by alium)

@MaheshWiz it should be a different implementation of the g45 driver.

@undevdecatos are you really use experimental driver?? can you please share info from 'vaapi'?

VxlerieUwU commented on 2020-08-13 09:42 (UTC)

Can confirm VLC works with @MahesWizs settings on the 2.4.1-experimental release

MaheshWiz commented on 2020-08-13 07:35 (UTC)

Using the 2.4.1. whats the difference in experimental driver ?

alium commented on 2020-08-13 03:52 (UTC)

@MahesWiz: thanks for report! Are you using a 'normal' driver or have you used an this experimental one?

MaheshWiz commented on 2020-08-12 23:02 (UTC) (edited on 2020-08-13 07:36 (UTC) by MaheshWiz)

Thanks for update !! Able to get HW accel in my Dell-1545/Cpu-T9800/gma4500MHD,

Using modesetting intel driver with AccelMethod= "glamor".

VLC can play 1920x1080 youtube videos (cpu usage: 6-15%) - cpu temp (45°-55°C). Chromium can play 720p youtube videos using h264ify with additional flags.

my environment vars :- LIBVA_DRIVER_NAME=i965 VAAPI_MPEG4_ENABLED=true VDPAU_DRIVER=va_gl MESA_LOADER_DRIVER_OVERRIDE=i965

My settings for VLC Select - Tools - Preferences -> Video = OpenGL video output

Select - Tools - Preferences ->-> Input/Codecs -> HW Accel decoding = VA-API decoder via DRM

VxlerieUwU commented on 2020-08-12 21:28 (UTC) (edited on 2020-08-12 21:35 (UTC) by VxlerieUwU)

Stil experiencing a GPU hang using VLC/FFMPEG using 2.4.1

[ 706.460691] i915 0000:00:02.0: GPU HANG: ecode 4:1:0xf3ffffff, in vlc [20518], hang on rcs0

[ 706.460849] i915 0000:00:02.0: Resetting chip for hang on rcs0

[ 714.466656] i915 0000:00:02.0: Resetting chip for hang on rcs0

Getting an error from VLC using the experimental code:

[00007fc8b00ca950] vaapi generic error: config doesn't support forced fourcc

alium commented on 2020-08-12 14:13 (UTC)

those who still have the g45 ask about testing a new version of the experimental driver. I wonder if it works and if the VAAPI acceleration really works with the G45

https://bitbucket.org/alium/g45-h264/downloads/intel-driver-g45-h264-2.4.1-experimental.tar.gz

Sha256sums: 9bd6b973c8e9b15718da6165eef3e377687d0d8dd6bcd31cc1c3c3718b041115

alium commented on 2020-08-12 10:50 (UTC) (edited on 2020-08-12 10:51 (UTC) by alium)

updated to 2.4.1 (nothing new for g45). Please test it.

Note: End of August will release Firefox 80 with VAAPI support on X11! Maybe will works HW acceleration for G45.

cockroach commented on 2020-07-25 01:18 (UTC) (edited on 2020-07-25 01:21 (UTC) by cockroach)

I'm also getting GPU hangs on a G45 chip. This is what dmesg says: [ 6799.043367] i915 0000:00:02.0: GPU HANG: ecode 4:1:0xf3ffffff, in vlc [8637], hang on rcs0 [ 6799.043418] i915 0000:00:02.0: Resetting chip for hang on rcs0 [ 6807.042017] i915 0000:00:02.0: Resetting chip for hang on rcs0 [ 6815.042121] i915 0000:00:02.0: Resetting chip for hang on rcs0 [ 6823.042023] i915 0000:00:02.0: Resetting chip for hang on rcs0 [ 6831.042028] i915 0000:00:02.0: Resetting chip for hang on rcs0 [ 6839.042036] i915 0000:00:02.0: Resetting chip for hang on rcs0 [ 6839.260347] traps: vlc[8559] general protection fault ip:7ff569443c26 sp:7ff55f581a70 error:0 in libc-2.31.so[7ff56942a000+14d000] [ 6847.042014] i915 0000:00:02.0: Resetting chip for hang on rcs0 This happens with version 2.4.0-2.

VxlerieUwU commented on 2020-07-17 23:39 (UTC) (edited on 2020-07-17 23:40 (UTC) by VxlerieUwU)

yes, Mesa DRI Mobile Intel® GM45 Express Chipset (CTG).

alium commented on 2020-07-16 13:35 (UTC)

@agm28011997: amdgpu can't use Intel's video driver (use mesa driver) for decoding...

@undevdecatos: can be produced by kernel/drm.. have you G45 graphics card?

VxlerieUwU commented on 2020-07-15 13:42 (UTC) (edited on 2020-07-15 13:42 (UTC) by VxlerieUwU)

Attempting to decode a h264 video on vlc/ffmpeg results in a gpu crash: Here are my dmesg logs:

[11979.738905] i915 0000:00:02.0: GPU HANG: ecode 4:1:f3ffffff, in systemd-logind [389]

[11979.739083] i915 0000:00:02.0: Resetting chip for stopped heartbeat on rcs0

[11979.844470] i915 0000:00:02.0: systemd-logind[389] context reset due to GPU hang

[11982.748329] i915 0000:00:02.0: GPU HANG: ecode 4:1:f3ffffff, in systemd-logind [389] [11982.752958] i915 0000:00:02.0: Resetting chip for stopped heartbeat on rcs0

[11982.856564] i915 0000:00:02.0: systemd-logind[389] context reset due to GPU hang

usuariopolivalen commented on 2020-07-12 02:55 (UTC)

thanks @alium I had tested on amdgpu already and it works well, but not as efficient as chromium I think. I have not tested deeply. In the older one I have not tested yet

alium commented on 2020-07-07 20:37 (UTC)

thanks for report @Martine! I expected it not to work. The good news is that firefox 80 will be released in late August, with VAAPI support for X11

Martine commented on 2020-07-06 09:48 (UTC)

Chromium spews a little bit more info: MESA-LOADER: failed to retrieve device information MESA-LOADER: failed to open i915 (search paths /usr/lib/dri) failed to load driver: i915 MESA-LOADER: failed to open kms_swrast (search paths /usr/lib/dri) failed to load driver: kms_swrast MESA-LOADER: failed to open swrast (search paths /usr/lib/dri) failed to load swrast driver

The non-experimemtal driver does not give this error

Martine commented on 2020-07-06 09:38 (UTC)

@Alium, you were correct.EXPERIMENTAL builds fine with the patches, my bad.

However after installing vainfo says: Segmentation fault: core dumped

I had hoped with running mpv from cmdline I got more info but the moment I switch to VAAPI mpv crashes with the same error.

alium commented on 2020-07-05 11:21 (UTC) (edited on 2020-07-05 11:21 (UTC) by alium)

@Martine did you use the gcc-10 patches that are in my PKGBUILD? it does not look like that... ;-)

Martine commented on 2020-07-05 11:01 (UTC)

Tried to build the 2.4.0-EXPERIMENTAL build but fails with the following error:


/usr/bin/ld: ./.libs/libi965_drv_video.a(libi965_drv_video_la-gen9_hevc_encoder.o):(.rodata+0x2a20): multiple definition of `gen9_hevc_encoder_kernels'; ./.libs/libi965_drv_video.a(libi965_drv_video_la-gen9_hevc_enc_kernels_binary.o):(.rodata+0x0): first defined here
collect2: error: ld returned 1 exit status
There seems to be a fix for it, not sure if you imported it already https://github.com/intel/intel-vaapi-driver/issues/503

Thanks for looking into it!

alium commented on 2020-06-28 07:54 (UTC) (edited on 2020-06-28 08:24 (UTC) by alium)

exist not so much solution for old G45:

can you try install libva1 https://www.archlinux.org/packages/community/x86_64/libva1/ and older driver: https://bitbucket.org/alium/g45-h264/downloads/intel-driver-g45-h264-1.8.2.tar.gz

Maybe will Chromium works.

Or you can try use newest firefox under wayland (have VAAPI support, but no X11 support).

Or you can try https://bitbucket.org/alium/g45-h264/downloads/intel-driver-g45-h264-2.4.0-EXPERIMENTAL.tar.gz with enabled VPP for G45 (should add VAProfileNone). I am not sure if that will works (probably on 99% not), but volunteers can test it :-)

you just need modify in PKGBUILD source=() and generate new sums

Nact commented on 2020-06-23 14:23 (UTC)

Package builds fine with intel-gpu-tools installed, at least for me.

usuariopolivalen commented on 2020-06-20 01:25 (UTC)

apart from this, thanks for all.

usuariopolivalen commented on 2020-06-20 01:20 (UTC)

@Martine @alium I am using vaapi and chromium since too much time ago. But.. there is a problem since a few months, it's the exact thing that martine said. Do you know something about this?

hans0 commented on 2020-05-20 19:02 (UTC)

@alium works : ) Thank you very much!

Hekuran commented on 2020-05-19 19:27 (UTC)

@alium aye, it works now! Thanks!

alium commented on 2020-05-19 14:47 (UTC)

Thank you for report, Hekuran!

I prepared new tarball and added gcc10 fixes. Compilation works again (tested), can somebody please test the new tarball?

Hekuran commented on 2020-05-19 11:34 (UTC) (edited on 2020-05-19 11:43 (UTC) by Hekuran)

I' not able to build. i even tried 2018's v2.3.0, couldn't that either. (I have a ThinkPad X200, core 2 duo P8600). Here are my makepkg logs:

latest v2.4.0: https://pastebin.com/raw/TWusMkb9

old v2.3.0: https://pastebin.com/raw/14WZxvfk

Martine commented on 2020-05-17 19:55 (UTC)

For hardware decoding in chromium there is a feature missing. According to comment #324 in https://bbs.archlinux.org/viewtopic.php?id=244031&p=13 chromium-vaapi requires 'VAProfileNone` which is not supported?

My Vainfo reports

vainfo: VA-API version: 1.7 (libva 2.7.1)
vainfo: Driver version: Intel i965 driver for Intel(R) GM45 Express Chipset - 2.4.0
vainfo: Supported profile and entrypoints
      VAProfileMPEG2Simple            : VAEntrypointVLD
      VAProfileMPEG2Main              : VAEntrypointVLD
      VAProfileH264ConstrainedBaseline: VAEntrypointVLD
      VAProfileH264Main               : VAEntrypointVLD
      VAProfileH264High               : VAEntrypointVLD

Is it possible to add this into the driver?

alium commented on 2020-05-14 15:58 (UTC)

@Martine: thanks for report!

Martine commented on 2020-05-14 15:33 (UTC)

Driver version 2.3.0 and 2.4.0 both work with Manjaro kernel 4.19.122 LTS. They don't work with 5.4.40 LTS and 5.6.10 as of writing. I keep the 4.19 for now.

Martine commented on 2020-05-14 14:39 (UTC) (edited on 2020-05-14 14:54 (UTC) by Martine)

Thanks. Downgrading succeeded but the problem persists. I downgraded mesa too but it didn'n matter either. Not sure how to proceed.

-edit downgrading from 5.6 to 4.19 solved the problem. Will test other kernels.

alium commented on 2020-05-14 12:10 (UTC) (edited on 2020-05-14 12:11 (UTC) by alium)

just build manually, download PKGBUILD, change that in PKGBUILD and build new package:

pkgver = 2.3.0
source = https://bitbucket.org/alium/g45-h264/downloads/intel-driver-g45-h264-2.3.0.tar.gz
sha256sums = 59754576de8bc69412ca0554b3a28406b0562cbf2616eca0d3c261528e81c641

older sources are still available

Martine commented on 2020-05-14 11:53 (UTC) (edited on 2020-05-14 12:03 (UTC) by Martine)

@alium thanks for you reply; however I can't downgrade.

pamac downgrade libva-intel-driver-g45-h264 says Unable to downgrade. Is that correct?

alium commented on 2020-05-14 11:43 (UTC) (edited on 2020-05-14 11:44 (UTC) by alium)

@Martine, thank you for report. I am very sorry, i have no HW anymore for test i. It may also be related to the Intel driver and drm, there have been some major changes (normally GPU crash not because broken VAAPI). Intel's developers no longer give any support for this piece of code. After all, the entire driver is only in 'maintenance' mode. I can only advise, install the previous version and put in libva-intel-driver-g45-h264 IgnorePkg (in pacman.conf).

Martine commented on 2020-05-14 11:18 (UTC) (edited on 2020-05-14 11:20 (UTC) by Martine)

And the GPU crash dump

GPU HANG: ecode 4:1:f3ffffff, in Xorg [899]
Kernel: 5.6.12-1-MANJARO x86_64
Driver: 20200114
Time: 1589447133 s 764013 us
Boottime: 43944 s 368998 us
Uptime: 43925 s 832130 us
Capture: 4308060608 jiffies; 18244 ms ago
Active process (on ring rcs0): Xorg [899]
Reset count: 0
Suspend count: 0
Platform: GM45
Subplatform: 0x0
PCI ID: 0x2a42
PCI Revision: 0x07
PCI Subsystem: 1043:1862
IOMMU enabled?: 0
RPM wakelock: yes
PM suspended: no
GT awake: yes
EIR: 0x00000000
IER: 0x02028053
PGTBL_ER: 0x00000000
FORCEWAKE: 0x00000000
DERRMR: 0x00000000
  fence[0] = 40be0000408309d
  fence[1] = 3d1900003cda01d
  fence[2] = 4ec7000048c80bd
  fence[3] = 2e9a00002a9b0fd
  fence[4] = 9f5900009ca809d
  fence[5] = 78da0000765109d
  fence[6] = 40820000404301d
  fence[7] = 3fbd00003f7e01d
  fence[8] = 94c700008ec80bd
  fence[9] = 3c9900003c5a01d
  fence[10] = adb50000a9b607d
  fence[11] = 3b1400003ad503d
  fence[12] = 3c5900003c1a01d
  fence[13] = 5a13000054140bd
  fence[14] = 3edb00003e940bd
  fence[15] = 00000000
rcs0 command stream:
  CCID:  0x0309410d
  START: 0x00305000
  HEAD:  0x268027ec [0x00002788]
  TAIL:  0x00002c48 [0x000027f8, 0x00002810]
  CTL:   0x00003001
  MODE:  0x00000040
  HWS:   0x00303000
  ACTHD: 0x00000000 268027ec
  IPEIR: 0x00000000
  IPEHR: 0x0c000000
  INSTDONE: 0xffffffff
  SC_INSTDONE: 0xbfffffff
  batch: [0x00000000_039a2000, 0x00000000_039a3000]
  BBADDR: 0x00000000_07129068
  BB_STATE: 0x00000080
  INSTPS: 0x26116c38
  INSTPM: 0x00000000
  FADDR: 0x00000000 003077e8
  engine reset count: 0
  Active context: Xorg[899] prio 0, guilty 0 active 0
rcs0 --- NULL context = 0xffffffff ffffffff
:f'SuPC"#b$+7KLR*.nU-#_N21+ikcS/V4s%/EdR'![+.n%($Ia@[&lD6A#V*6I7tdc:&j-==I<$.cjp>.chj&DH6EZY#TKjr[)f3G'6Td-LCag:#t1f]UR7(D^XC[J-"g'jPTRD'8IBkR,Ps)n.$E>k;rCmk(!.,bhRFJ=hEU4=dIH,Sm(374:g$h!'#7SbQ5<E!5.jJDs'ZAFqIahUcm6>*!=A.Act4]$6VB\&-N6rZ&&qPWqE55]$=;JiT3;$p7!qg#Wr.X1!j;'k[d_e`lR_GoAknJpkU$m^cX7:k\e=&+EGQ.p&Gk!aIb:6FgLpO^\e0'p/?7/L52BuRoeOWLiI)LASflf'BF:c&51#Teasd/q"GEolsqBn(P!LhIrP#LYtFBj_k@QWI.d(`(i)`<q.%dMNI2)=77o*SV97)r(hi`-_LB@Y*R46+cpo530_6KS_%Pt$$(e]77ZB)Wp;C23qEU!hShV7:AVm9V36>h$6gs,rW<Mar6<n06reUF6\ln5^`qO)77H2#+e3(ON)XL"HR`D]r]h1+,LtlUJb)\aqM)Ff3PW+F7Au+b/+4`9]1``eUId5,g+pJO'a<2X.J@fcH6Zp\[(_G\rkk*@=Y[?98i<su)g]Y\F_6K:LUb/Z6$A#0Ac9k:ME[O0iW>Kuj,+%sA5K`_%efTOTc8pj<&FIj5[MU(41EAAgcbkUH>6K:HJEkd76Eu#Rcmo5^]g5afD)EM%1.L2F*jXADn51$<*>a.B7C3TFWSL-Y6LW6Y&m!l=9(/4MPU:d(?/\RM<[&j<e2[*FH3nb1L2!TmgTHXr`S;3,6$/Q*XJKpdK@^8qgMUie+hGo<adN!,41<rT8[Ykmet1cF[;T]D@sMopCT$Ogeqj:g>K620fAQ.uU9sR?o-E94?B*@\Jst&^+%a"6T-]ogJnc(.:ttD#+1o1SHq:Z'.fSNZP]5Nm9I0%PPB2uJS+"QNAP?d.2!6AZ::XA;GhlM=hYi(s:<aIS3\HKb.0X(A,t$H)XZ3a[<RoWT5UUe_I]M5!RkH)=L4*F&ni`I)9q[?6%$4At(o+N'G;pp%mH)@c\cQjtaaJ2h5up@"E^7Hf]<-8Hr@l+(`NG1@XJFUHPYXIn3mG.o&S2`IY\Jq_R^,^mN)h>OaRUPkc6+\kNkfo8\YQ.n7]/+:@OL)+d-RhQ3s&XT@p\$i-^NcS#!\`>2lI\a"WMilL/Q\<:mg"6N$Q-KD8^,^3L'/u+tAW8""K*&4]n'hHtQi'<5rafl*3S#7t.VFMOL'Ip<A_!"B02I8f76#i:Q>7`cf`d+3IU"4"CS89QC#B9TAbZ4PDQ3P$)=0So>j&>Lak[^`K*+-fd0FiL>RnPUgMoou`Y>+)U6MC)&B-@QepFYm@HB#IL>O:D9olp^]nB)gIgbe2n&_Qc^#I^0M>W$\_Wo,T)M19:*e'H,_"V4R'NNAmMOK+6)uMf6Pq'[s!qTk?G"!XBWMhW:h.Z8Y@C9oWcjZ/]$ht"Y*\ZU2GF#HW1_58]U&d8IKBE91EhO\iqm)l'Y90VhC+,)@_nPj%a\QK^b,L'Q#B"'1Vk!7U6DTYc;3i.hOh3H?G-f1;;<3F'H,s"F'QI0ppPtR+9G'37Lrfi5E-:jGjp?EG>Z)]2b*H1;JdS=?M%2g"J*F2\Qh4n34Tr%Mo'4Y%WD0h"mo87/2/@HbX(o>uZc!`n;Q1[^R'bc)qdMYfM`Gh[IZ-Iti/^o2lI+XlE@N%@i7PW,&ac6:*u:Hg:1)7_n!<WmW"9"X3HA*qXA25d"t<^it0j6'$ul,f-c2H]R+[<L^CcE\<"im(@G:4\X\Z$2%/>`40X()7k5Mk5Rc\T#!tOgIVHQ&FFLY<EC-],2OVD,#E5Z^mjJJSuq[!\AbJudH`S1r@,'45q&C;n'd5dP[&Ym9#_kuddN3Dlu8S`ojXYa7Y6EG:7W8"Sah?]/B%Llc<1Cpc,gfEY5@G$6:\rT9^,n_pfYJT/,>*^F[g/neLXXg$969F=Z@mRCogD/6"+LcMp?Q'0")):2fq,'jWh;>^L^R8&DS#QmTMpT2cAA26otG?21%#oe?Ss\0'eZram/B&R(%,,0V%.OVuJfl:X8"VLd_!:AuC'iX):P-DJO[Ie7"*.![XK/<LGkM.SR6`T(:h->]`p.i$AQa4foF%V(VH8^62^+9=H56MkOQDjAf<C':]PY7-7Kf*1<C0AV;Y8\=Y$rBZC1$$7'R/2^]$l`BdeC,kk(T&dtp^g2sJK8\kgMhm]NCT4'K(f'.+aoqYM"qMQ/JN]Hu+9+YX/p9/l\6G$Z).H4'%aZCrds#E3P^(Tp51Yc.1UI@M<1M&3f]cM8`f:N!p$R.C`fW@XG6Tf+C&pI2M&La/2r1bWA;LCA:F=\Ka_%#`l:YVpM+TEBB%WEfQ.-`JC2h`=1%ih/&j5(kVCqY*Ffo&G='%50dO#q"@6qJc_X9G$1I]hsqnO]kK>?Y+Tg)SZ8c5rt,A-`so0ANG\G9Uf_Nb1"a\QO9l)+[$T]=qbaVFXIf/.(pBaG:R"]8NSaia[_.mQ-@QY^m:siFL(Nlq/:;`,0nhnZk1P.Hm=n:IW*Ea\QhKRq\-u3%C*Y(cA'q^JW+&mjpo.05aP94Bq<uDYWnA((U$^FZEOXM=T!-EL)e9Z2Feq"naU*P<>XX2AWLTd)lde/7Y3dq/6"R19;O&';l%Xr-8p89!]"k<t2X4gEoTX6t*e6K8pdY5N<>3:$l(ghI1muM4LkF=7%66nHloJ`ibo;#mtP3"u/)`(YSDM=6!nn"E7^YNJ2R4J3Wb&-jjm:DVZ8SV-Q-;Iq_%@@L$aY,!]V&b7BZg"f?\ZF.5M-T)=oN.Z?u@B6)d4*$=L:a5OU(SuqbU.LY`7QH*ff*lt#M2g0r0imLs:ikmIF&G-sUBnIK=`R$K[43.(OapCgT4P\JCD-e^K?*jMmeS_Z':e:)2jPFr<TC\=/5Qq\p9[D)lNaMT?rB&au>Qt:ApORJeYI"7Q]*J)DN*1/oI(2\i>\_H@+gU(XI(;t<\Q?t:=(T]ofl;?k[G8rtfDbnB\"ip7OafIhHD(]D!!&ts
rcs0 --- HW Status = 0x00000000 00303000
:cL%-H!!!TsIkUj*TZOP=_p/sM^Cak>!HmR*ec5[M!Jg=W!!!#C
rcs0 --- batch = 0x00000000 039a2000
:Q^sXk5ndp-c+h74ZJ28EKI^k2kh>5u6C!2F8@U'6+JTYZ$kP]oQ\s=iKb5`+4X-[+Dd4'(nL(@Y#eV5ILP+:>6^%"bXAG#k/_MIjo4IT&m$e)4qsU#u4$,C7gW.tj9OHEuXZm/#3k$4LC^'_>]C#>\0D-+K5YJ=(_t\:iGceIPlVr7:LY8haF'`2S-tA2?!b=08atNs+<Q"X'Kt,Ie2AQDJ/b'k$dg@`)/#o*d3WFW8o3'fe`KDmaLbSV*B:s.TW52t+\NjmVL:NIYdsL)4Ft3*>'NY)$fK,";0;?$+9**3,-KTR+n4T5AN"G[(O>497P0;,t'=pU)L5mf'Y\VKmFc%(sdD^`9)H3_6@YTmoP-A[QZDSk7_tNGI0?SjZZtRk!)%RJC5C.g>G7=#-,r7Phqi'DeGLQP3->mi;^PM>97l-:'):,EP6Uqn$Y\DCfc6sToQHlEE(cflqqIcAr5A:'#(ZV?AFDb*LXR`o&L3\319GaeMMSWFr(HYE6mNpFoM--`<K,JnsL5mcK1fkLI`=3ehik_Y+ZIUF;TqdMM//kQkX^[$F)Nq==\7bL`Di]7g$>TMpj5*):\Q2DmpRLYIH"=e%4lTKlT>?N$m6aAY//l-Yh-u,!1iZpP#(c(S=J;LJUoYL#"on)s^>4SEP?O"XgE8BI^pOq8Gfj^6iSu/pqi3TKpZ,%a.!`:q/MqqS@2$`0aqbd2X*u19^QRSlnWQqW?JjDHbJT?uZe*#6n^\0[IHs6]Wpf?6p'PU6FJo/1#LL'^s&jgY>*_+ChYQ,Z5LTA+emHJTP\R2dZQ(@=)K$-SqUT)g\*'T([&iC&L&0m.ch[BNSCOk2`)g*q9&J1r+/rUO9!^jlE^Po=KT0ktlqUp>C]1&VqaPKRE\<Xg!!!"%
rcs0 --- ring = 0x00000000 00305000
:Qd)%F@0B?5?2oTm5Sh-dBLk]`!1j4aTI59H"s)O;h8T,4h;-oC#H+DpL3?*'kHR;!G&GG"aip/#pGq\d8"U?Q/F$PO-!HAD&tpg;;T8Y[qk[ZoY.4M;`pj^c67Hj!q1$nDrnkBdoB5u9,?iYU,tSX'=/u\]!fjf9NVMc@K8Lg>-u`4#L+%*EJq\jM6^S0IjF`==##O6NP"nj9%U^"^;#)B%9i0sU,W-*qf*/8Tcf.SbHA=18J2Vlipl[HoU!*hL=s##X8Qm+LbRL9:+!dDN3ZPpr$_d,_,b(LO*5HkBU%1d$hkA7)H<_GO[-Se),QeM`ot^hnlcC=f&Vi&K"Wk=>nKSllNeoa7-9Z;S'^+UM8Mn>Pk`F"QA(LmOR@e#M+BW[i@#LYF^*H>]Z-an#.`\[n#.$M8dS@%/3[>XTf_*'=[V+8;89Puh8M92gk`P:?`V0ViEB_JF6u_8o8WLP@V$-AcS!dEs354=Kk9$*:L)1>-Y&<1YBO%e)M/h,s@q&<`7<-p+&o5tUe<&OZ:Sa<<gcA:#*HF#"*ZZ['4M$DF3t-5(4E\`6._'T;o79#Gr:X:k7IT%3%H0MAZ:!j<3J.,mcg6eh!=iN:+Cb3n8J"7VplbhE+2igg#eo5j3(NK`M$-25*5Ng`)>m%"6P/Ah@'L6*q'9qiOQI&bM%>b7,Y8JRjp[@I&!B8@.RTR9JYYJ$jHHB"6^S'FA3NR3P@Q\)QjuLE5"R@oa$:O'*bDN:!!!#I
rcs0 --- HW context = 0x00000000 03a93000
:QL12eg!`cA["$+3>p@Dk!h)Q/QNU1o\cA*A"B>8k[ZD()TcgS$!`6]]&.oJJ&h+AKKr&CWKqX$I6:=WcF=l`I/L&UKG'@4p]^3jicHlAH\_jm/h#5N2F1KVD3S_pBrf?OKZ+.mp/8V"^@_a.QO6rP?SBp@Y>8*J+f#/1L?BNRZqduj(0FaUqFoV^Q+o[!>O8p;iEMRZJ^Ar7VhF$\t)LuE)"Q0\2FXE3[L4#,g'lo2>5DGX#<)_gg!'#3B*RplC73Ful,bXiP4`<Nk5QDYX4Rk)GR6[=a:V[UiOa$j-cD6N$fXfW3Eo5.Cd!1a\%cX5KpC3\i\O>+.Iks!_oLI9kFMM85>jq0'F.78%Wn^-F,Hq?,9BmRJYUj+R=gTHuX-gSs!Y/m9>qEPLd]>]1OAkcEW3ao*<[nS)0TWCc>A\LC]GAi1a?N<Hd"FG*JY]G#;8WfbI?CD`?;NuWX/oM:Z#!j=:(g.i5]l7%,sQ7dQugd?+8ud%[u4Mt>$e=oo0&F#R=gsk"!*42',btKkL"U/TrYQ_[-5NGE.^pdEF!7ll;C*aKA-/T3Wk[/a3O=MJmdN7WeqE,a`=bB[F&5p_gi"@ILcn'=HgDgC`bO8M8>("%X%CtF$a%UYUJeu>h!;kF$+7[[$HWNF=qoH^#ZY6eUHR8S0P924uSri.m^nOK+V$Oo_^Zfl+`.@W&DoToJ9#6:NVD@qldm;`N#.m*)AmZA_^PANe/VgT[7i<Y['Nm4Ijt=<#5-'_)YN[Ya4g\TgB"-+l4Ks`FLAc\ODEI25:n&JYJ`g7\)C@/e/H:ph-F%Ru<VMBO,Qk`KnNpP,n66big-;@k0.HdZ)M,/]k+R.q[Ac%^5&\)1R%(%W%'@Y.q%P&V';+5FdqoUs;3<)nfcqbppEGbKs^"6_8=E:-B%[/$/=\:#XueI>p'W4T+IqKQg5%\3&tr#5^.%2qPQ.o%;pUERna1NI6W7rj%^&[.,X`R@H5cJ"@\IRKBB8Nf1,'>I&,Y.B0<TMjj6%poS&YA@.Od!WI@td,nm8M]`5uEBcGr!F')^HSb^hXbH'$V?0%Pd:I]Z>!cMCGXl(g1UT^2cVXPk)NSWK@3Wb*XD.B&qIQQMdkKk)$CDm-7Q^F@N7eJH)FpUS7/EV6g$p0;2]mP#f]oVdB]:1S'q(V'Dj*oXbH'6adbHDGLXZ6U>5k*)@9<$uGtl"l!JQDUb[]\De4Q'04I2sYJQB_4gn\.Me;]3LO)=M]SOf$K0JAs?DUd)S;H/osfXK*:#\qFa6'c@:+!%ER:0Jj/HAX]:`Sg>tNSht#'Yd^!R)b/(#qlb+cMPn.]0pKJHrHQY?@t"HBt29:FLmVFA>-T^f%.;PmNUT'"35!h2_s=RI8iVPp*00[I^'a%YN+fL,2q4l*.a#b-/n0X3#/Je:9aA?&ug0#.)_2*m?auBMq#0?rLZP2q`NSJcJ3*Uo4;Q8P@Q&2*5tm>2#>,K(>E?gfGNE4OXDKOj0Qb!lJP7I:#C_A8t2jmeYf^<,@eeOl/_7?6r9,^Vb%T.39*G&SgdY,Ks>jKDd@%X6`+iG&QF2)A0XftX`VUoDc6/(RofFUpZ<$nLrUG;Q.6CBoG.%T:oJXVXr(L.bTd&9UEQH*R0@SVnb>mNIh$H`6upIt>4o,L/cD=AfaiTanO_P=i8K-pY:'t0BS9o;aXjMf[i<#t?eJ#^OqclG(6MX4O)!@\Zfu7M`22V]2%KUl<G5%c#(fZKf4%tfK#tQH`Bacj!i7+UC0rc;=OhoWIf(aO%gg^WSQ'p+++aHC2l\f\*=7HMBX)TRBfj4hb5BOg#T:)2WLffWcZ7)rF,-)@8n,2PXA@0mj1:J4"=jg/%rqfB4];(]T$:qYeS6n*UD[]H)\\JhoD9V=3(]?EL]%_M6"7t3pnuc'EN*db8kG&tT^oOc[5nTN+DPfMO0Up[.t16\.LMW\44#a*n8tqkF4+Gm":9(-2L:tt=`tZqP%/+-)Q^HBqFmbTN8L\VcU=^rcha]g+keZ^,,6IHgJANl's?u!rZ)Bt`Ga^X/LjOA1%(cZ1'WsF*nFe$]HmK6XpmYYE@E<+a=]H0nYkFSAi\XqRrXnrHeui)n?QYBB9,B6+Je\\KDmX-4(Za"VR81S>]&.#Z>Ci[^MDkar5JRNm[g?,;e?_4pta!G$qIR'ZQ[.Yrmr^,o$)bb'=>!FH.@N\Uj%deif:LC[f$YKm[Pmt,J/]Kna3!ZAl\?'`T`>!6'Y)dn_gB7FL'N#ld#D&"N7^:KnA\H5*=_"#mfpagr.R`f:RMp'iA6ddVI'R>;g^uZ3\(j&q'/(>q1O#:=%6eTAU"3]h,<C%<f3hcm$M%.O/!;oG#YJ4#Ya9SYAYq'*:p8%:LD7oH$3Teu7MiJY&_&iT$9>j6qmf%;M0;:UBB)%k.[@h%m<`M83hS;N+]aB<5U,/`.)m15Q>Qd2cB\?.SgV/Y_TsWSk@s=e3&`_4?fN6+G,"dWf!16MBt8k@lKc^=@U_egb-ZNHo/IjM=V'$F1K[<t2l?'f0J-/-jh^#-j\08//:ZD/eEG*3IG*=e"JpmEPWnmR0Y9GAL>`Kgam$U?.;AV+2dTc@6qh8Y`muG5D(YpHpl?n^9O(ibVB2hSd@W]t;DWd^^I_q)>:PBY*XdH8-M*hPnup:ZF`5ljYdZS"eUcL"E3bH[Q6P0]<@0mQ8[UFpOicmJ1rnn"mZbQ8i&@*ABQe`P_C)itUpHTMBPaVQ)[Xk\aIdI)XSRHSI:U@cSJ\caW41R&urgGa;ZmmA0m(it'c.gusO,dZp!RZO5*lGkFJ:gs#WE_C3V8*k(lrH7'CJ"Ee]X0(rm8'kKOS3'!8hf@s#2G8r^$ldLdBs&1LRkfa`R?Mu1Y"fsSqYSXD$VEr5#ft0ZX*ha`jX1D4jmO,@!+&0/'<""Nb/CHUcp-5NYWY\[An:S\l2P25p_M)rAL]($Fm#Ae;r#<\OHWTp3N<T?[q;>SN(6%SG=[,5t1c?`K[J$p*r].Gu!(!rYJ5DY(li7q%m_Z:,Sk!tFHXB_;iU5%hH"oJ;#Js%3o3VP!'A=HSQ%u[i*lbGXL;Viu!!%XY55P5J"LnFD?%/i!Gb4T+[d6/tnFsVOs$PMs3(,\G:-gH4kCd(_I*DVSn/T)kTDrW7qENNE)R%Qg!"fX;
Num Pipes: 2
Pipe [0]:
  Power: on
  SRC: 059f0383
  STAT: 18400206
Plane [0]:
  CNTR: d8004400
  STRIDE: 00001800
  ADDR: 00000000
  SURF: 08ec8000
  TILEOFF: 00000000
Cursor [0]:
  CNTR: 04000027
  POS: 03110363
  BASE: 00024000
Pipe [1]:
  Power: on
  SRC: 03ff02ff
  STAT: 10400206
Plane [1]:
  CNTR: 01000000
  STRIDE: 00001000
  ADDR: 00000000
  SURF: 00000000
  TILEOFF: 00000000
Cursor [1]:
  CNTR: 00000000
  POS: 00000000
  BASE: 00000000
CPU transcoder: A
  Power: on
  CONF: c0000050
  HTOTAL: 06df059f
  HBLANK: 06df059f
  HSYNC: 05ff05df
  VTOTAL: 038f0383
  VBLANK: 038f0383
  VSYNC: 03870384
CPU transcoder: B
  Power: on
  CONF: 00000000
  HTOTAL: 06df059f
  HBLANK: 06df059f
  HSYNC: 05ff05df
  VTOTAL: 038f0383
  VBLANK: 038f0383
  VSYNC: 03870384
engines: 5
gen: 4
gt: 0
iommu: disabled
memory-regions: 5
page-sizes: 1000
platform: GM45
ppgtt-size: 0
ppgtt-type: 0
is_mobile: yes
is_lp: no
require_force_probe: no
is_dgfx: no
has_64bit_reloc: no
gpu_reset_clobbers_display: no
has_reset_engine: no
has_fpga_dbg: no
has_global_mocs: no
has_gt_uc: no
has_l3_dpf: no
has_llc: no
has_logical_ring_contexts: no
has_logical_ring_elsq: no
has_logical_ring_preemption: no
has_pooled_eu: no
has_rc6: no
has_rc6p: no
has_rps: no
has_runtime_pm: no
has_snoop: yes
has_coherent_ggtt: yes
unfenced_needs_alignment: no
hws_needs_physical: no
cursor_needs_physical: no
has_csr: no
has_ddi: no
has_dp_mst: no
has_dsb: no
has_dsc: no
has_fbc: yes
has_gmch: yes
has_hdcp: no
has_hotplug: yes
has_ipc: no
has_modular_fia: no
has_overlay: no
has_psr: no
overlay_needs_physical: no
supports_tv: yes
slice total: 0, mask=0000
subslice total: 0
EU total: 0
EU per subslice: 0
has slice power gating: no
has subslice power gating: no
has EU power gating: no
CS timestamp frequency: 0 kHz
Unavailable
Has logical contexts? yes
scheduler: 0
i915.vbt_firmware=(null)
i915.modeset=-1
i915.lvds_channel_mode=0
i915.panel_use_ssc=-1
i915.vbt_sdvo_panel_type=-1
i915.enable_dc=-1
i915.enable_fbc=0
i915.enable_psr=-1
i915.disable_power_well=1
i915.enable_ips=1
i915.invert_brightness=0
i915.enable_guc=0
i915.guc_log_level=-1
i915.guc_firmware_path=(null)
i915.huc_firmware_path=(null)
i915.dmc_firmware_path=(null)
i915.mmio_debug=0
i915.edp_vswing=0
i915.reset=3
i915.inject_probe_failure=0
i915.fastboot=-1
i915.enable_dpcd_backlight=0
i915.force_probe=*
i915.fake_lmem_start=0
i915.alpha_support=yes
i915.enable_hangcheck=yes
i915.prefault_disable=no
i915.load_detect_test=no
i915.force_reset_modeset_test=no
i915.error_capture=yes
i915.disable_display=no
i915.verbose_state_checks=yes
i915.nuclear_pageflip=no
i915.enable_dp_mst=yes
i915.enable_gvt=no

vainfo says:

vainfo: VA-API version: 1.7 (libva 2.7.1)
vainfo: Driver version: Intel i965 driver for Intel(R) GM45 Express Chipset - 2.4.0
vainfo: Supported profile and entrypoints
      VAProfileMPEG2Simple            : VAEntrypointVLD
      VAProfileMPEG2Main              : VAEntrypointVLD
      VAProfileH264ConstrainedBaseline: VAEntrypointVLD
      VAProfileH264Main               : VAEntrypointVLD
      VAProfileH264High               : VAEntrypointVLD

Martine commented on 2020-05-14 11:08 (UTC) (edited on 2020-05-14 11:17 (UTC) by Martine)

Some more info: mpv output

   0.312][v][vd] Codec list:
[   0.312][v][vd]     h264 - H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10
[   0.312][v][vd]     h264_v4l2m2m (h264) - V4L2 mem2mem H.264 decoder wrapper
[   0.312][v][vd]     h264_qsv (h264) - H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10 (Intel Quick Sync Video acceleration)
[   0.312][v][vd]     h264_cuvid (h264) - Nvidia CUVID H264 decoder
[   0.312][v][vd] Opening decoder h264
[   0.313][v][vd] Looking at hwdec h264-vaapi...
[   0.313][v][vo/gpu] Loading hwdec driver 'vaapi-egl'
[   0.313][v][vo/gpu/vaapi-egl] using VAAPI EGL interop
[   0.313][v][vo/gpu/vaapi-egl] Trying to open a x11 VA display...
[   0.313][d][vo/gpu/vaapi-egl/vaapi] libva: VA-API version 1.7.0
[   0.314][d][vo/gpu/vaapi-egl/vaapi] libva: Trying to open /usr/lib/dri/iHD_drv_video.so
[   0.314][d][vo/gpu/vaapi-egl/vaapi] libva: va_openDriver() returns -1
[   0.314][d][vo/gpu/vaapi-egl/vaapi] libva: Trying to open /usr/lib/dri/i965_drv_video.so
[   0.346][d][vo/gpu/vaapi-egl/vaapi] libva: Found init function __vaDriverInit_1_7
[   0.347][d][vo/gpu/vaapi-egl/vaapi] libva: va_openDriver() returns 0
[   0.347][v][vo/gpu/vaapi-egl/vaapi] Initialized VAAPI: version 1.7
[   0.347][d][ffmpeg] AVHWDeviceContext: VAAPI driver: Intel i965 driver for Intel(R) GM45 Express Chipset - 2.4.0.
[   0.347][d][ffmpeg] AVHWDeviceContext: Driver not found in known nonstandard list, using standard behaviour.
[   0.347][v][vo/gpu/vaapi-egl] Going to probe surface formats (may log bogus errors)...
[   0.349][d][vo/gpu/vaapi-egl] Supported formats:
[   0.349][d][vo/gpu/vaapi-egl]  yuv420p
[   0.349][d][vo/gpu/vaapi-egl]  nv12
[   0.349][v][vo/gpu/vaapi-egl] Done probing surface formats.
[   0.349][v][vo/gpu] Loading hwdec driver 'cuda-nvdec'
[   0.350][v][vo/gpu/cuda-nvdec] Failed to load CUDA symbols
[   0.350][v][vo/gpu] Loading failed.
[   0.350][v][vo/gpu] Loading hwdec driver 'vdpau-gl'
[   0.350][v][vo/gpu] Loading failed.
[   0.350][v][vo/gpu] Loading hwdec driver 'drmprime-drm'
[   0.350][v][vo/gpu/drmprime-drm] Failed to retrieve DRM fd from native display.
[   0.350][v][vo/gpu] Loading failed.
[   0.350][v][vd] Trying hardware decoding via h264-vaapi.
[   0.351][v][vd] Pixel formats supported by decoder: vdpau cuda vaapi_vld yuv420p
[   0.351][v][vd] Codec profile: High (0x64)
[   0.371][v][vd] Requesting pixfmt 'vaapi_vld' from decoder.
[   0.391][d][ffmpeg/video] h264: Reinit context to 1920x800, pix_fmt: vaapi_vld
[   0.433][v][vd] Selected codec: h264 (H.264 / AVC / MPEG-4 AVC / MPEG-4 part 10)
[   0.434][v][vf] User filter list:
[   0.434][v][vf]   (empty)
[   0.435][v][ad] Codec list:
[   0.435][v][ad]     ac3 - ATSC A/52A (AC-3)
[   0.435][v][ad]     ac3_fixed (ac3) - ATSC A/52A (AC-3)
[   0.435][v][ad] Opening decoder ac3
[   0.436][v][ad] Requesting 1 threads for decoding.
[   0.436][v][ad] Selected codec: ac3 (ATSC A/52A (AC-3))
[   0.437][v][af] User filter list:
[   0.437][v][af]   (empty)
[   0.440][v][sub/ass] Shaper: FriBidi 0.19.7 (SIMPLE) HarfBuzz-ng 2.6.5 (COMPLEX)
[   0.440][v][sub/ass] Setting up fonts...
[   0.455][v][sub/ass] Using font provider fontconfig
[   0.456][v][sub/ass] Done.
[   0.456][v][cplayer] Starting playback...
[   0.470][v][af] [in] 48000Hz 5.1(side) 6ch floatp
[   0.470][v][af] [userspeed] 48000Hz 5.1(side) 6ch floatp
[   0.471][v][af] [userspeed] (disabled)
[   0.471][v][af] [convert] 48000Hz 5.1(side) 6ch floatp
[   0.471][v][vd] Pixel formats supported by decoder: vdpau cuda vaapi_vld yuv420p
[   0.471][v][vd] Codec profile: High (0x64)
[   0.472][v][vd] Requesting pixfmt 'vaapi_vld' from decoder.
[   0.472][d][ffmpeg/video] h264: Reinit context to 1920x800, pix_fmt: vaapi_vld
[   0.483][i][vd] Using hardware decoding (vaapi).
[   0.483][v][vd] Decoder format: 1920x798 vaapi[nv12] bt.709/bt.709/bt.1886/limited/auto CL=unknown
[   0.483][v][vd] Using container aspect ratio.
[   0.483][v][vf] [in] 1920x798 vaapi[nv12] bt.709/bt.709/bt.1886/limited/display SP=1.000000 CL=mpeg2/4/h264
[   0.484][v][vf] [userdeint] 1920x798 vaapi[nv12] bt.709/bt.709/bt.1886/limited/display SP=1.000000 CL=mpeg2/4/h264
[   0.484][v][vf] [userdeint] (disabled)
[   0.484][v][vf] [autorotate] 1920x798 vaapi[nv12] bt.709/bt.709/bt.1886/limited/display SP=1.000000 CL=mpeg2/4/h264
[   0.484][v][vf] [autorotate] (disabled)
[   0.484][v][vf] [convert] 1920x798 vaapi[nv12] bt.709/bt.709/bt.1886/limited/display SP=1.000000 CL=mpeg2/4/h264
[   0.484][v][vf] [convert] (disabled)
[   0.484][v][vf] [out] 1920x798 vaapi[nv12] bt.709/bt.709/bt.1886/limited/display SP=1.000000 CL=mpeg2/4/h264
[   0.485][d][osc] osc_init 
[   0.485][v][ao] Trying audio driver 'pulse'
[   0.485][v][ao/pulse] requested format: 48000 Hz, 5.1(side) channels, floatp
[   0.486][v][ao/pulse] Library version: 13.0.0
[   0.486][v][ao/pulse] Proto: 33
[   0.486][v][ao/pulse] Server proto: 4294967295
[   0.489][v][ao/pulse] Channel layouts:
[   0.489][v][ao/pulse]  - #fl
[   0.489][v][ao/pulse]  - #fr
[   0.489][v][ao/pulse]  - #fc
[   0.489][v][ao/pulse]  - #lfe
[   0.489][v][ao/pulse]  - #bl
[   0.489][v][ao/pulse]  - #br
[   0.489][v][ao/pulse]  - #flc
[   0.489][v][ao/pulse]  - #frc
[   0.490][v][ao/pulse]  - #bc
[   0.490][v][ao/pulse]  - #sl
[   0.490][v][ao/pulse]  - #sr
[   0.490][v][ao/pulse]  - #tc
[   0.490][v][ao/pulse]  - #tfl
[   0.490][v][ao/pulse]  - #tfc
[   0.490][v][ao/pulse]  - #tfr
[   0.490][v][ao/pulse]  - #tbl
[   0.490][v][ao/pulse]  - #tbc
[   0.490][v][ao/pulse]  - #tbr
[   0.490][v][ao/pulse] result: 5.1(side)
[   0.494][v][ao/pulse] device buffer: 4800 samples.
[   0.495][v][ao/pulse] using soft-buffer of 9600 samples.
[   0.495][i][cplayer] AO: [pulse] 48000Hz 5.1(side) 6ch float
[   0.495][v][cplayer] AO: Description: PulseAudio audio output
[   0.496][v][autoconvert] inserting resampler
[   0.496][v][swresample] format change, reinitializing resampler
[   0.496][v][swresample] 48000Hz 5.1(side) floatp -> 48000Hz 5.1(side) float
[   0.501][v][af] [out] 48000Hz 5.1(side) 6ch float
[   0.502][w][ffmpeg/video] h264: Increasing reorder buffer to 1
[   0.571][i][cplayer] VO: [gpu] 1920x798 vaapi[nv12]
[   0.571][v][cplayer] VO: Description: Shader-based GPU Renderer
[   0.571][v][vo/gpu] reconfig to 1920x798 vaapi[nv12] bt.709/bt.709/bt.1886/limited/display SP=1.000000 CL=mpeg2/4/h264
[   6.673][d][vo/gpu] screen size: 1440x900
[   9.732][v][vo/gpu] Resize: 1441x599
[   9.733][v][vo/gpu] Window size: 1441x599 (Borders: l=0 t=0 r=0 b=0)
[   9.733][v][vo/gpu] Video source: 1920x798 (1:1)
[   9.733][v][vo/gpu] Video display: (0, 0) 1920x798 -> (0, 0) 1441x599
[   9.733][v][vo/gpu] Video scale: 0.750521/0.750627
[   9.733][v][vo/gpu] OSD borders: l=0 t=0 r=0 b=0
[   9.733][v][vo/gpu] Video borders: l=0 t=0 r=0 b=0
[   9.733][v][vo/gpu] Reported display depth: 8
[   9.733][v][vo/gpu] Testing FBO format rgba16f
[   9.733][d][vo/gpu] Resizing texture: 16x16
[   9.733][v][vo/gpu] Using FBO format rgba16f.
[   9.733][v][vo/gpu] No advanced processing required. Enabling dumb mode.
[   9.733][v][vo/gpu] Resize: 1441x599
[   9.733][v][vo/gpu] Window size: 1441x599 (Borders: l=0 t=0 r=0 b=0)
[   9.733][v][vo/gpu] Video source: 1920x798 (1:1)
[   9.733][v][vo/gpu] Video display: (0, 0) 1920x798 -> (0, 0) 1441x599
[   9.733][v][vo/gpu] Video scale: 0.750521/0.750627
[   9.733][v][vo/gpu] OSD borders: l=0 t=0 r=0 b=0
[   9.733][v][vo/gpu] Video borders: l=0 t=0 r=0 b=0
[   9.733][v][vo/gpu] Reported display depth: 8
[   9.751][d][vo/gpu/opengl] vertex shader source:
[   9.751][d][vo/gpu/opengl] [  1] #version 120
[   9.751][d][vo/gpu/opengl] [  2] #define tex1D texture1D
[   9.751][d][vo/gpu/opengl] [  3] #define tex3D texture3D
[   9.751][d][vo/gpu/opengl] [  4] #define texture texture2D
[   9.751][d][vo/gpu/opengl] [  5] #define LUT_POS(x, lut_size) mix(0.5 / (lut_size), 1.0 - 0.5 / (lut_size), (x))
[   9.751][d][vo/gpu/opengl] [  6] attribute vec2 vertex_position;
[   9.751][d][vo/gpu/opengl] [  7] attribute vec2 vertex_texcoord0;
[   9.751][d][vo/gpu/opengl] [  8] varying vec2 texcoord0;
[   9.751][d][vo/gpu/opengl] [  9] attribute vec2 vertex_texcoord1;
[   9.751][d][vo/gpu/opengl] [ 10] varying vec2 texcoord1;
[   9.751][d][vo/gpu/opengl] [ 11] void main() {
[   9.751][d][vo/gpu/opengl] [ 12] gl_Position = vec4(vertex_position, 1.0, 1.0);
[   9.751][d][vo/gpu/opengl] [ 13] texcoord0 = vertex_texcoord0;
[   9.751][d][vo/gpu/opengl] [ 14] texcoord1 = vertex_texcoord1;
[   9.751][d][vo/gpu/opengl] [ 15] }
[   9.752][d][vo/gpu/opengl] fragment shader source:
[   9.752][d][vo/gpu/opengl] [  1] #version 120
[   9.752][d][vo/gpu/opengl] [  2] #define tex1D texture1D
[   9.752][d][vo/gpu/opengl] [  3] #define tex3D texture3D
[   9.752][d][vo/gpu/opengl] [  4] #define texture texture2D
[   9.752][d][vo/gpu/opengl] [  5] #define LUT_POS(x, lut_size) mix(0.5 / (lut_size), 1.0 - 0.5 / (lut_size), (x))
[   9.752][d][vo/gpu/opengl] [  6] varying vec2 texcoord0;
[   9.752][d][vo/gpu/opengl] [  7] varying vec2 texcoord1;
[   9.752][d][vo/gpu/opengl] [  8] uniform mat3 colormatrix;
[   9.752][d][vo/gpu/opengl] [  9] uniform vec3 colormatrix_c;
[   9.752][d][vo/gpu/opengl] [ 10] uniform vec3 src_luma;
[   9.752][d][vo/gpu/opengl] [ 11] uniform vec3 dst_luma;
[   9.752][d][vo/gpu/opengl] [ 12] uniform sampler2D texture0;
[   9.752][d][vo/gpu/opengl] [ 13] uniform vec2 texture_size0;
[   9.752][d][vo/gpu/opengl] [ 14] uniform mat2 texture_rot0;
[   9.752][d][vo/gpu/opengl] [ 15] uniform vec2 texture_off0;
[   9.752][d][vo/gpu/opengl] [ 16] uniform vec2 pixel_size0;
[   9.752][d][vo/gpu/opengl] [ 17] uniform sampler2D texture1;
[   9.752][d][vo/gpu/opengl] [ 18] uniform vec2 texture_size1;
[   9.752][d][vo/gpu/opengl] [ 19] uniform mat2 texture_rot1;
[   9.752][d][vo/gpu/opengl] [ 20] uniform vec2 texture_off1;
[   9.752][d][vo/gpu/opengl] [ 21] uniform vec2 pixel_size1;
[   9.752][d][vo/gpu/opengl] [ 22] void main() {
[   9.752][d][vo/gpu/opengl] [ 23] vec4 color = vec4(0.0, 0.0, 0.0, 1.0);
[   9.752][d][vo/gpu/opengl] [ 24] color.r = 1.000000 * vec4(texture(texture0, texcoord0)).r;
[   9.752][d][vo/gpu/opengl] [ 25] color.gb = 1.000000 * vec4(texture(texture1, texcoord1)).rg;
[   9.752][d][vo/gpu/opengl] [ 26] color = color.rgbr;
[   9.752][d][vo/gpu/opengl] [ 27] color.rgb = mat3(colormatrix) * color.rgb + colormatrix_c;
[   9.752][d][vo/gpu/opengl] [ 28] color.a = 1.0;
[   9.752][d][vo/gpu/opengl] [ 29] // color mapping
[   9.752][d][vo/gpu/opengl] [ 30] color.rgb = vec3(1.000000);
[   9.752][d][vo/gpu/opengl] [ 31] color.rgb = vec3(1.000000);
[   9.752][d][vo/gpu/opengl] [ 32] gl_FragColor = color;
[   9.752][d][vo/gpu/opengl] [ 33] }
[   9.783][d][vo/gpu/opengl] shader link log (status=1): 
[   9.828][v][cplayer] first video frame after restart shown
[   9.836][v][cplayer] audio sync: sync_to_video=1, offset=0.000000
[   9.836][d][cplayer] Run command: change-list, flags=64, args=["shared-script-properties", "append", "osc-margins=0.000000,0.000000,0.000000,0.000000"]
[   9.837][v][cplayer] Set property: shared-script-properties -> 1
[   9.838][v][cplayer] starting audio playback
[   9.839][v][cplayer] playback restart complete @ 0.000000
[   9.839][d][osc] osc_init 
[   9.841][d][cplayer] Run command: change-list, flags=64, args=["shared-script-properties", "append", "osc-margins=0.000000,0.000000,0.000000,0.000000"]
[   9.845][v][cplayer] Set property: shared-script-properties -> 1
[   9.879][v][vo/gpu/x11] Disabling screensaver.
[  15.844][w][cplayer] 
[  15.844][w][cplayer] Audio/Video desynchronisation detected! Possible reasons include too slow
[  15.844][w][cplayer] hardware, temporary CPU spikes, broken drivers, and broken files. Audio
[  15.844][w][cplayer] position will not match to the video (see A-V status field).
[  15.844][w][cplayer] 
[  39.166][d][cplayer] Run command: quit, flags=73, args=["4"]
[  39.166][v][cplayer] EOF code: 5
[ 39.171][d][ad] Uninit decoder. [ 39.172][d][vd] Uninit decoder. [ 39.178][d][cplayer] Terminating demuxers... [ 39.179][d][cplayer] Done terminating demuxers. [ 39.179][v][cplayer] finished playback, success (reason 3) [ 39.179][i][cplayer] [ 39.179][i][cplayer] [ 39.179][i][cplayer] Exiting... (Quit) [ 39.179][d][stats] Exiting... [ 39.180][d][cplayer] Run command: change-list, flags=64, args=["shared-script-properties", "remove", "osc-margins"] [ 39.181][v][cplayer] Set property: shared-script-properties -> 1 [ 39.182][d][osc] Exiting... [ 39.187][d][console] Exiting... [ 39.193][d][ytdl_hook] Exiting... [ 39.202][v][ao/pulse] draining... [ 39.210][v][vo/gpu/x11] Enabling screensaver. [ 39.276][d][vo/gpu] flushing shader cache [ 39.287][d][vo/gpu/x11] uninit ...

Martine commented on 2020-05-14 08:57 (UTC) (edited on 2020-05-14 09:00 (UTC) by Martine)

There is indeed a complete system hangup since april on my system when using hardware acceleration. When playing a H264 video my system crashes with the following output

mei 13 22:50:12 PinkiePie kernel: i915 0000:00:02.0: GPU HANG: ecode 4:1:f3ffffff, in kwin_x11 [1763]
mei 13 22:50:12 PinkiePie kernel: GPU hangs can indicate a bug anywhere in the entire gfx stack, including userspace.
mei 13 22:50:12 PinkiePie kernel: Please file a new bug report at https://gitlab.freedesktop.org/drm/intel/issues/new.
mei 13 22:50:12 PinkiePie kernel: Please see https://gitlab.freedesktop.org/drm/intel/-/wikis/How-to-file-i915-bugs for details.
mei 13 22:50:12 PinkiePie kernel: drm/i915 developers can then reassign to the right component if it's not a kernel issue.
mei 13 22:50:12 PinkiePie kernel: The GPU crash dump is required to analyze GPU hangs, so please always attach it.
mei 13 22:50:12 PinkiePie kernel: GPU crash dump saved to /sys/class/drm/card0/error
mei 13 22:50:12 PinkiePie kernel: i915 0000:00:02.0: Resetting chip for stopped heartbeat on rcs0
mei 13 22:50:12 PinkiePie kernel: i915 0000:00:02.0: kwin_x11[1763] context reset due to GPU hang
mei 13 22:50:15 PinkiePie kernel: i915 0000:00:02.0: GPU HANG: ecode 4:1:f3ffffff
mei 13 22:50:15 PinkiePie kernel: i915 0000:00:02.0: Resetting chip for stopped heartbeat on rcs0
mei 13 22:50:21 PinkiePie kernel: i915 0000:00:02.0: GPU HANG: ecode 4:1:f3ffffff, in plasmashell [1771]
mei 13 22:50:21 PinkiePie kernel: i915 0000:00:02.0: Resetting chip for stopped heartbeat on rcs0
mei 13 22:50:21 PinkiePie kernel: i915 0000:00:02.0: plasmashell[1771] context reset due to GPU hang

damianduzy commented on 2020-04-20 09:58 (UTC) (edited on 2020-04-20 09:59 (UTC) by damianduzy)

@alium Sorry about the delay, I wasn't checking the AUR for the last couple of days. Here you go: https://bin.disroot.org/?57ebe721b6564307#9kLhJrGwreDynxqh4vcYgNb2nuLUHzxuvmLvfZEf4AU3

alium commented on 2020-04-14 11:13 (UTC)

@damianduzy, thanks for report! can you please provide more info from mpv??

mpv -v file.src

damianduzy commented on 2020-04-14 09:53 (UTC)

For me this doesn't work, the whole session freezes and the video doesn't play, there's audio for only a second and then nothing, I have to kill mpv to unfreeze everything.

alium commented on 2020-04-03 08:17 (UTC)

Hi, i prepared update for latest version of libva-intel-driver (2.4.0). I hope, maybe still use this package. If yes, please test! :-)

alium commented on 2019-06-09 18:03 (UTC)

@Ulin: hi, why you need 32-bit version of this package? have you 32-bit media player?

Ulin commented on 2019-06-09 15:53 (UTC) (edited on 2019-06-09 15:53 (UTC) by Ulin)

Does this make sense in a 32-bit version under Multilib ?? There is no package for 32 bits in AUR ...

vainfo: VA-API version: 1.4 (libva 2.4.1)
vainfo: Driver version: Intel i965 driver for Intel(R) GM45 Express Chipset - 2.3.0
vainfo: Supported profile and entrypoints
      VAProfileMPEG2Simple            : VAEntrypointVLD
      VAProfileMPEG2Main              : VAEntrypointVLD
      VAProfileH264ConstrainedBaseline: VAEntrypointVLD
      VAProfileH264Main               : VAEntrypointVLD
      VAProfileH264High               : VAEntrypointVLD

nanaoa commented on 2019-04-13 16:22 (UTC) (edited on 2019-04-13 16:32 (UTC) by nanaoa)

Rigth now libva-intel-driver-g45-h264 is installed, but I got same output with libva-intel-driver. Should I have more supported codecs with this package?

alium commented on 2019-04-13 12:11 (UTC)

@nanaoa: is this vainfo from libva-intel-driver-g45-h264 or libva-intel-driver? If libva-intel-driver, keep it, seems, that you have some better graphics card, which is suported by main driver.

nanaoa commented on 2019-04-13 10:49 (UTC) (edited on 2019-04-13 11:13 (UTC) by nanaoa)

@alium sure:

vainfo: VA-API version: 1.4 (libva 2.4.1)
vainfo: Driver version: Intel i965 driver for Intel(R) Haswell Mobile - 2.3.0
vainfo: Supported profile and entrypoints
      VAProfileMPEG2Simple            : VAEntrypointVLD
      VAProfileMPEG2Simple            : VAEntrypointEncSlice
      VAProfileMPEG2Main              : VAEntrypointVLD
      VAProfileMPEG2Main              : VAEntrypointEncSlice
      VAProfileH264ConstrainedBaseline: VAEntrypointVLD
      VAProfileH264ConstrainedBaseline: VAEntrypointEncSlice
      VAProfileH264Main               : VAEntrypointVLD
      VAProfileH264Main               : VAEntrypointEncSlice
      VAProfileH264High               : VAEntrypointVLD
      VAProfileH264High               : VAEntrypointEncSlice
      VAProfileH264MultiviewHigh      : VAEntrypointVLD
      VAProfileH264MultiviewHigh      : VAEntrypointEncSlice
      VAProfileH264StereoHigh         : VAEntrypointVLD
      VAProfileH264StereoHigh         : VAEntrypointEncSlice
      VAProfileVC1Simple              : VAEntrypointVLD
      VAProfileVC1Main                : VAEntrypointVLD
      VAProfileVC1Advanced            : VAEntrypointVLD
      VAProfileNone                   : VAEntrypointVideoProc
      VAProfileJPEGBaseline           : VAEntrypointVLD

alium commented on 2019-04-12 18:50 (UTC)

@nanaoa: can you please share your vainfo output?

nanaoa commented on 2019-04-12 13:49 (UTC)

I replaced libva-intel-driver by libva-intel-driver-g45-h264 but I get same output from vainfo with both packages. Is this correct?

I have a 4th gen i7 laptop (GMA4500MHD?).

Ulin commented on 2018-12-29 06:22 (UTC)

Yes, it seems like it works. H264 accelerates.

vainfo

vainfo: VA-API version: 1.3 (libva 2.3.0) vainfo: Driver version: Intel i965 driver for Intel(R) GM45 Express Chipset - 2.3.0 vainfo: Supported profile and entrypoints VAProfileMPEG2Simple : VAEntrypointVLD VAProfileMPEG2Main : VAEntrypointVLD VAProfileH264ConstrainedBaseline: VAEntrypointVLD VAProfileH264Main : VAEntrypointVLD VAProfileH264High : VAEntrypointVLD

alium commented on 2018-12-28 10:50 (UTC) (edited on 2018-12-28 10:50 (UTC) by alium)

should works, but please test, i have G45 no more

j1simon commented on 2018-04-07 14:39 (UTC)

ok, that was the problem. This package should be added to "conflicts", don't you think?

alium commented on 2018-04-07 14:37 (UTC)

@j1simon: uninstall intel-gpu-tools, this package broke compilation

j1simon commented on 2018-04-07 14:15 (UTC) (edited on 2018-04-07 14:16 (UTC) by j1simon)

libdrm 2.4.91-3
libglvnd 1.0.0-1
libva 2.1.0-1
libva-intel-driver 2.1.0-1
mesa 18.0.0-2
libva-utils 2.1.0-1

These are my installed libraries. All from official repositories.

alium commented on 2018-04-07 14:11 (UTC)

@j1simon: something is wrong or broken on your OS... tested, can be compiled without problems https://pastebin.com/U4Nvxb25

j1simon commented on 2018-04-07 14:05 (UTC)

It doesn't compile: https://pastebin.com/6mLApuch

hans0 commented on 2018-02-19 18:47 (UTC)

Thank you @alium, the new version works :)

alium commented on 2018-02-13 16:46 (UTC)

should work, but please test!!

usuariopolivalen commented on 2017-12-04 21:35 (UTC)

have you tried the last mesa? it functions with this aur package or not?

usuariopolivalen commented on 2017-11-18 12:51 (UTC)

much thanks @alium and much thnaks @eduardoeae i have compiled the mesa 17.1 and it is working perfectly again, the pc relifes with that libva and mesa package because of the videos, it is a pity that mesa developpers make something not to work and not fix it

chaseadam commented on 2017-11-16 19:24 (UTC)

@alium this looks like great work. Happy to have someone keeping hardware functional. I see there is a set of source downloads, but how do you create those? Is there a git or similar version control tool which you are using to create the updated drivers?

usuariopolivalen commented on 2017-11-15 19:18 (UTC)

@eduardoeae I sent you an email to the mail you have in aur

usuariopolivalen commented on 2017-11-14 19:30 (UTC)

much thanks @eduardoeae, I am trying after I've finished the studies

eduardoeae commented on 2017-11-14 12:37 (UTC) (edited on 2017-11-14 13:53 (UTC) by eduardoeae)

#DELETED LINK This PKGBUILDs replace mesa and lib32-mesa, so you are gonna want to blacklist those in pacman.conf. I'm going to clean it up and submit it to AUR when I have time. # Already in AUR: https://aur.archlinux.org/packages/mesa17.1/ https://aur.archlinux.org/packages/lib32-mesa17.1/

usuariopolivalen commented on 2017-11-14 10:00 (UTC)

@eduardoeae can you upload the mesa package for me pls? I haven't got the packages in my pc

usuariopolivalen commented on 2017-11-12 19:57 (UTC) (edited on 2017-11-12 19:57 (UTC) by usuariopolivalen)

my pc is freezing again, no errors either, all it is like before, very annoying problem with mesa.. when do they fix it?? is there a bug filled?? PD: thanks you for all on the other hand!:)

eduardoeae commented on 2017-11-12 17:51 (UTC)

YES! IT WORKS! You are awesome. PD: mesa 17.2 does freezes with vaapi for me. 17.1 works alright.

usuariopolivalen commented on 2017-11-11 21:02 (UTC)

@eduardoeae I hadn't downgrade yet, not really needed for me, i don't use this pc usually, at least not for videos, have you fixed the mpv error?

usuariopolivalen commented on 2017-11-11 20:58 (UTC)

AO: [pulse] 44100Hz stereo 2ch float mpv: i965_avc_ildb.c:599: i965_avc_ildb_decode_init: Assertion `avc_ildb_context->urb.cs_start + avc_ildb_context->urb.num_cs_entries * avc_ildb_context->urb.size_cs_entry <= i965->intel.device_info->urb_size' failed. i have this error

eduardoeae commented on 2017-11-11 14:33 (UTC)

Mesa is 17.1.10. Because 17.2.x freezes mpv even with --hwdec=no --vo=opengl. All the rest is up-to-date. # Just tried current mesa and I got the same error. ## Mesa 17.2.4 doesn't freeze anymore for me.

eduardoeae commented on 2017-11-11 13:58 (UTC)

mpv aborts: mpv: i965_avc_ildb.c:599: i965_avc_ildb_decode_init: Assertion `avc_ildb_context->urb.cs_start + avc_ildb_context->urb.num_cs_entries * avc_ildb_context->urb.size_cs_entry <= i965->intel.device_info->urb_size' failed.

eduardoeae commented on 2017-11-11 01:08 (UTC) (edited on 2017-11-11 01:19 (UTC) by eduardoeae)

@agm28011997: I have been using previous mesa, because of freezes when playing videos, even with opengl. @alium: ffmpeg now requieres a library from official libva package. It doesn't work otherwise.

usuariopolivalen commented on 2017-11-10 11:59 (UTC)

I can't reproduce any video with vaapi .. in mpv and chromium for example my pc slow down and lag, and sometimes freezes, are there people like me with problems since the update of mesa but not fixed in some weeks?

jfrankenau commented on 2017-10-21 08:29 (UTC)

My thanks as well, @alium!

Raziel23 commented on 2017-10-17 09:10 (UTC)

I think it's time to move to a newer hardware. Nevertheless thank you @alium for maintaing this package for so long.

hans0 commented on 2017-09-29 17:59 (UTC)

What a pity :(

hans0 commented on 2017-06-02 18:39 (UTC)

@alium: Works for me, too! Great work, keep it up! Thanks :)

Raziel23 commented on 2017-05-31 19:09 (UTC)

@alium: For me it works. Thanks for updating the package.

hans0 commented on 2017-04-25 21:01 (UTC)

@Raziel23 works like charm :) @alium Thank you very much : )

hans0 commented on 2017-04-25 19:49 (UTC)

@Raziel23 I didn't know that, so I will try mpv-git plus patch.

Raziel23 commented on 2017-04-25 19:17 (UTC)

@hans0: I can't reproduce your issue when playing 1080p h264 video on mpv-git. bomi-git base on mpv player, but it contains mpv source code in its tree that wasn't update for nearly 2 years, see here: https://github.com/xylosper/bomi/tree/master/src So basically you use 2 years old mpv player that may contains some old bugs. So I can't speak about such old version. Furthermore bomi project is not develop anymore: https://github.com/xylosper/bomi/issues/490 For me it works without stutters when playing 1080p h264 video on mpv-git plus this driver with patch from @tegularius.

hans0 commented on 2017-04-25 15:14 (UTC) (edited on 2017-04-25 15:25 (UTC) by hans0)

I tried this driver with bomi-git: 720p acceleration works, 1080p stutters (without the patch from @tegularius).

Raziel23 commented on 2017-04-22 15:45 (UTC)

@tegularius: Yes your patch works without patching mpv or ffmpeg. @alium: Please apply the patch from @tegularius.

tegularius commented on 2017-04-22 13:44 (UTC) (edited on 2017-04-22 14:04 (UTC) by tegularius)

Parts of the driver had not been patched for support h264. It seems that the driver support for ironlake is the same as it should be for g4x-devices, so I patched it to return the same surface-attribute-information. Works as far as I can tell. https://gist.github.com/anonymous/e5383bd3c78bf389d124fcecde5a1556

Raziel23 commented on 2017-04-21 17:27 (UTC) (edited on 2017-04-21 19:37 (UTC) by Raziel23)

@tegularius: I dug a little bit about this issue. When mpv is compiled against ffmpeg 3.2.4 it uses an old vaapi infrastructure, which means that vaapi is handled by mpv player. On the other hand when mpv is compiled against ffmpeg 3.3 it uses a new API, which means that vaapi is handled by ffmpeg. They differently handle vaapi. FFmpeg rely on codec/format configuration that is returned by libva-intel-driver-g45-h264. I don't know if this can be fixed in libva-intel-driver-g45-h264 (I looked into a source code but don't have enough knowledge), but there are two other solution that works for me: 1. Use old vaapi infrastructure in mpv when compiled against ffmpeg 3.3. Disclaimer: old vaapi infrastructure maybe removed from mpv in the future. Compile mpv with the following flags: https://gist.github.com/anonymous/3e0daaedc6fcfa1e41365696c79883ba 2. Fix ffmpeg 3.3 with patch which do: if hardware configuration returned by libva-intel-driver-g45-h264 driver doesn't have any relevant format then returns a full set of image formats and check one by one if any format is relevant to decode h264: https://gist.github.com/anonymous/8670b535237f0e650947092586e9446d Use either one solution. I don't know if this can be fixed in libva-intel-driver-g45-h264 package. Edit: Upload patches to gist.

tegularius commented on 2017-04-20 19:45 (UTC) (edited on 2017-04-20 19:46 (UTC) by tegularius)

Sadly doesn't seem to work, mpv says: [vd] Pixel formats supported by decoder: vdpau vaapi_vld yuv420p [vd] Codec profile: Main (0x4d) [ffmpeg/video] h264: Hardware does not offer any usable surface formats. [vd] Pixel formats supported by decoder: vdpau yuv420p [vd] Codec profile: Main (0x4d) [ffmpeg/video] h264: Reinit context to 1280x720, pix_fmt: yuv420p [vd] Falling back to software decoding. vainfo reports: libva info: VA-API version 0.40.0 libva info: va_getDriverName() returns 0 libva info: Trying to open /usr/lib/dri/i965_drv_video.so libva info: Found init function __vaDriverInit_0_40 libva info: va_openDriver() returns 0 vainfo: VA-API version: 0.40 (libva ) vainfo: Driver version: Intel i965 driver for Intel(R) GM45 Express Chipset - 1.8.1 vainfo: Supported profile and entrypoints VAProfileMPEG2Simple : VAEntrypointVLD VAProfileMPEG2Main : VAEntrypointVLD VAProfileH264ConstrainedBaseline: VAEntrypointVLD VAProfileH264Main : VAEntrypointVLD VAProfileH264High : VAEntrypointVLD

pedromj commented on 2016-09-18 04:50 (UTC) (edited on 2016-09-18 04:50 (UTC) by pedromj)

As we stated in previous comments, neither me nor @alium (if nothing changed) have access to G45 chipset. If any of you, current users of G45 chipsets, can maintain this package, please contribute a fixed version and become the new maintainer. Thank you.

kwsk commented on 2016-09-18 03:00 (UTC)

libva 1.7.2 was released resolving dependencies... looking for conflicting packages... error: failed to prepare transaction (could not satisfy dependencies) :: libva-intel-driver-g45-h264: installing libva (1.7.2-1) breaks dependency 'libva=1.7.1'

pedromj commented on 2016-01-27 15:35 (UTC) (edited on 2016-01-27 15:36 (UTC) by pedromj)

Actually @alium has been the main contributor to this package. I have added him to the list of co-maintainers so he can update it when a new release is out. Moreover, I no longer have access to a machine with a G45 chipset, so I cannot test if this work before updating it.

Raziel23 commented on 2016-01-27 15:15 (UTC)

@alium: I repeat here on AUR what I sent to you in an email, thus all interested users will know. I found a cause of my issue [1]. After applying a patch [2] from there on intel-driver-g45-h264-1.6.2.tar.gz source green artifacts disappeared. This patch is already in mainline [3]. Please apply this patch in PKGBUILD or intel-driver-g45-h264-1.6.2.tar.gz source. I want to thank you for maintaining this source. BTW. You should be a co-maintainer of libva-intel-driver-g45-h264 package on AUR. [1] https://bugs.freedesktop.org/show_bug.cgi?id=93499 [2] https://bugs.freedesktop.org/attachment.cgi?id=120803 [3] http://cgit.freedesktop.org/vaapi/intel-driver/commit/?id=10c4cdd103f730a9e07ffcf0f940e8f63ee5738d

Raziel23 commented on 2016-01-25 08:47 (UTC)

@alium: Thank you for your response. I will send you an email with details.

Raziel23 commented on 2016-01-24 14:14 (UTC)

@alium: With your new version I get green artifacts at the bottom of a video. With previous version this issue doesn't exist. I don't know if this can be fixed, I just want let you know about this.

pedromj commented on 2015-11-11 19:21 (UTC)

What has been abandoned is the development of tne code. It is updated to maintain compatibility with new libva versions, but not really developed and therefore it may not work in the future.

SBeaver commented on 2015-11-11 18:22 (UTC)

The arch wiki claims that this package is abandoned. If not please update that entry.

pedromj commented on 2015-09-22 07:57 (UTC)

Updated, please test.

pedromj commented on 2015-04-16 13:46 (UTC)

Updated, please test.

cb474 commented on 2015-01-01 04:33 (UTC)

Thanks alium.

alium commented on 2014-12-31 10:21 (UTC)

@cb474: you need update to libva-1.5.0 in [extra], after update you can see: [alois@toshiba ~]$ vainfo libva info: VA-API version 0.37.0 libva info: va_getDriverName() returns 0 libva info: Trying to open /usr/lib/dri/i965_drv_video.so libva info: Found init function __vaDriverInit_0_37 libva info: va_openDriver() returns 0 vainfo: VA-API version: 0.37 (libva 1.5.0)

cb474 commented on 2014-12-31 10:04 (UTC)

I get this error when trying to build: configure: error: Package requirements (libva >= 0.37) were not met: Requested 'libva >= 0.37' but version of libva is 0.36.0 Consider adjusting the PKG_CONFIG_PATH environment variable if you installed software in a non-standard prefix. Alternatively, you may set the environment variables LIBVA_DEPS_CFLAGS and LIBVA_DEPS_LIBS to avoid the need to call pkg-config. See the pkg-config man page for more details.

pedromj commented on 2014-12-30 20:00 (UTC)

Updated, please test.

pedromj commented on 2014-10-09 09:56 (UTC)

Tested and uploaded. Please, check it works right. Thank you.

City-busz commented on 2014-08-04 04:24 (UTC)

libva-driver-intel-g45-h264 merged into libva-intel-driver-g45-h264.

pedromj commented on 2014-08-04 02:07 (UTC)

Hi all, The requested name change is totally reasonable, so I've done it. Please, find the new package in: https://aur.archlinux.org/packages/libva-intel-driver-g45-h264/ [Do not forget to vote, so we may be able to get it included in the official repositories sometime in the future :-D]. Regards, Pedro

City-busz commented on 2014-08-01 14:07 (UTC)

Please rename this package to libva-intel-driver-g45-h264 to make it consistent with the package in [extra]. After you uploaded it in the new name, I'll merge this one.

pedromj commented on 2014-05-18 08:42 (UTC)

Updated, please check. Regards, Pedro

alium commented on 2014-05-10 21:26 (UTC)

libva 1.3.1 is out. I merged the master tree, and I added/enabled VPP (video postprocesing) now, please test before the deployment. For me works 720p x264 videos in vlc and mpv. vainfo: VA-API version: 0.35 (libva 1.3.1) vainfo: Driver version: Intel i965 driver - 1.3.2.pre1 vainfo: Supported profile and entrypoints VAProfileMPEG2Simple : VAEntrypointVLD VAProfileMPEG2Main : VAEntrypointVLD VAProfileH264ConstrainedBaseline: VAEntrypointVLD VAProfileH264Main : VAEntrypointVLD VAProfileH264High : VAEntrypointVLD VAProfileNone : VAEntrypointVideoProc Link for test: https://downloads.sourceforge.net/project/g45h264/testing/intel-driver-g45-h264-20140510.tar.gz - sha256sums=('13bcea803592d2175ac18a0825fb6096e95ded181cbf3eea02e12567d04343c0') I waiting for clarification from Intel's developer that the VPP on the G45 is really supported/implemented. If no problems, I move it from testing.

pedromj commented on 2014-03-26 15:31 (UTC)

Updated version. Please test.

commented on 2013-09-24 15:28 (UTC)

i installed the package ,vainfo is proper. And everything else seems to work fine.

pedromj commented on 2013-09-23 18:49 (UTC)

Hi, I've updated the package. Please test it and let me know any problem. Thank you. Regards, Pedro

pedromj commented on 2013-09-23 18:49 (UTC)

Hi, I've updated the package. Please test it and let me know any problem. Thank you. Regards, Pedro

pedromj commented on 2013-07-02 12:28 (UTC)

Updated package. Please test it and report any problem. Thank you.

stativ commented on 2013-07-02 10:26 (UTC)

trapanator: I just did a simple test, where I first checked the output of "vainfo" and then I tried to play a file that is encoded in h264 using "mplayer -vo vaapi file.mp4"

andrea76 commented on 2013-07-02 10:17 (UTC)

@stativ: how do you test your card with these libs?

stativ commented on 2013-07-02 09:24 (UTC)

alium: I did a quick test of commit 4568e24f0382d2d03215bcaf78eb0eae68f88ae5 and it seems to work fine on my GMA 4500MHD.

pedromj commented on 2013-07-01 12:00 (UTC)

Uploaded the updated package. Please, test it and give us feedback as @alium mentioned. Thank you.

pedromj commented on 2013-07-01 10:43 (UTC)

It still does not work and I tried what you said. The SHA256 sums differ from the ones you posted. Please, point me to the source package you used to get the hash: sha256sums=('f391d8fa2e109295d722ae7ba3f069dc6dba0c8d7c23549d4b7e35542f1f8bed') or send it to me (pedromj@gmail.com) because the file I get from the following URL: https://downloads.sourceforge.net/project/g45h264/intel-driver-g45-h264-20130701.tar.gz has the following hash: sha256sums=('a8dbe9fb92b3ccc6f1718121a6f168614cdbe7814cea8a65fe19283b82b8d78d') so I wonder where is the file you have used. Thank you.

pedromj commented on 2013-07-01 10:06 (UTC)

The new file does not match the SHA256sum you posted. However, even after updating the PKGBUILD to match the SHA256sum of the new file, it does not work properly, does not show H264 profile.

pedromj commented on 2013-07-01 09:55 (UTC)

This version does not work properly: $ vainfo libva info: VA-API version 0.34.0 libva info: va_getDriverName() returns 0 libva info: Trying to open /usr/lib/dri/i965_drv_video.so libva info: Found init function __vaDriverInit_0_34 libva info: va_openDriver() returns 0 vainfo: VA-API version: 0.34 (libva 1.2.1) vainfo: Driver version: Intel i965 driver - 1.2.1.pre1 vainfo: Supported profile and entrypoints VAProfileMPEG2Simple : VAEntrypointVLD VAProfileMPEG2Main : VAEntrypointVLD With previous versions it did offer support H264 profile for my hardware (X4500MHD). Alium, please, test it and let me know how to fix it so I can upgrade the package. Thank you.

pedromj commented on 2013-04-13 15:48 (UTC)

I've revised the PKGBUILD, fixed some little problems, tested it builds and works properly, and uploaded the new package version. Thanks alium.

andrea76 commented on 2013-04-13 12:26 (UTC)

thank you, alium. Now I have h264 accelerarion, too, on my 4500. Remember to uninstall libva-intel-driver before.

pedromj commented on 2013-04-07 20:28 (UTC)

Do not hesitate to get the PKGBUILD and adapt it to work with current version. I can review and upload it (do not forget to add your contribution line). Thank you.

andrea76 commented on 2013-04-07 20:03 (UTC)

the branch 'g45-h264' is removed from GIT, because it isn't compatible with libva-1.1.0 anymore :-( Anyone fixed this problem?

pedromj commented on 2013-02-04 13:32 (UTC)

I've tried but does not easily work. If you can provide a PKGBUILD, I'll give it a try and update this. Thank you.

commented on 2013-02-03 20:10 (UTC)

Suggestion - change pkgbuild to use "git clone --depth 1" to reduce network traffic and download times. We don't need the entire history, just the current snapshot.

pedromj commented on 2013-01-24 18:15 (UTC)

I have tested and uploaded the version provided by "vbmithr".

arkusuma commented on 2013-01-22 09:39 (UTC)

I'm orphaning this package.

vbmithr commented on 2013-01-22 09:21 (UTC)

To the maintainer: update it or orphan it. Here is a PKGBUILD that works and take account on input of previous comments: http://pastebin.com/DQRM8MJB

chx commented on 2013-01-12 00:01 (UTC)

I needed to install libtool to get this package installing.

wapmorgan commented on 2013-01-11 14:46 (UTC)

If during compiling you received this message - http://pastebin.com/VHVBt6qC - then do (replace "wapmorgan" in dirs to your name) : 0. Delete libva-intel-driver packages (if you installed it) $ sudo pacman -R libva-intel-driver 1. Change dir to the source dir (e.g. /tmp/yaourt-tmp-wapmorgan/aur-libva-driver-intel-g45-h264/src/intel-driver) $ cd /tmp/yaourt-tmp-wapmorgan/aur-libva-driver-intel-g45-h264/src/intel-driver 2. Edit "configure.ac": replace AM_CONFIG_HEADER([src/config.h]) to AC_CONFIG_HEADERS([src/config.h]) (25 line); $ sed -i 's/AM_CONFIG_HEADER/AC_CONFIG_HEADERS/' configure.ac 3. Go to dir where PKGBUILD stays $ cd /tmp/yaourt-tmp-wapmorgan/aur-libva-driver-intel-g45-h264 4. Comment git clone command $ sed -i 's/git clone/#git clone/' PKGBUILD 5. Compile/install packages $ makepkg -i

commented on 2012-11-27 17:01 (UTC)

instead of conflicts=('libva-driver-intel') Should be conflicts=('libva-intel-driver') provides=('libva-intel-driver=$pkgver')

commented on 2012-09-21 20:46 (UTC)

Built from modifying PKGBUILD worked fine. Thanks for this heads up.

commented on 2012-09-21 20:31 (UTC)

Vote here if you want this fixed. https://bugs.archlinux.org/task/31491

commented on 2012-09-21 20:27 (UTC)

How do you downgrade to 1.0.16? I only fount 1.0.15.

alium commented on 2012-09-11 10:26 (UTC)

https://bugs.archlinux.org/task/31491

alium commented on 2012-09-11 07:09 (UTC)

@protomog: it is because libva 1.1.0 have API changes, please switch your version libva to 1.0.16. 1.1.0 is reverted and exist no more.

commented on 2012-09-10 22:05 (UTC)

Fails to compile: In file included from intel_batchbuffer.h:9:0, from gen6_mfc.c:35: intel_driver.h:112:1: error: unknown type name 'Bool' intel_driver.h:113:1: error: unknown type name 'Bool' In file included from i965_render.h:36:0, from i965_drv_video.h:206, from gen6_mfc.c:38: i965_post_processing.h:370:1: error: unknown type name 'Bool' i965_post_processing.h:372:1: error: unknown type name 'Bool' In file included from i965_drv_video.h:206:0, from gen6_mfc.c:38: i965_render.h:82:1: error: unknown type name 'Bool' i965_render.h:83:1: error: unknown type name 'Bool' In file included from i965_encoder.h:39:0, from gen6_mfc.c:39: gen6_vme.h:86:1: error: unknown type name 'Bool' gen6_vme.h:87:1: error: unknown type name 'Bool' In file included from i965_encoder.h:40:0, from gen6_mfc.c:39: gen6_mfc.h:103:1: error: unknown type name 'Bool' gen6_mfc.h:104:1: error: unknown type name 'Bool' gen6_mfc.c:955:1: error: unknown type name 'Bool' gen6_mfc.c: In function 'gen6_mfc_context_init': gen6_mfc.c:957:12: error: 'True' undeclared (first use in this function) gen6_mfc.c:957:12: note: each undeclared identifier is reported only once for each function it appears in gen6_mfc.c: At top level: gen6_mfc.c:960:1: error: unknown type name 'Bool' gen6_mfc.c: In function 'gen6_mfc_context_destroy': gen6_mfc.c:990:12: error: 'True' undeclared (first use in this function) gen6_mfc.c: At top level: gen6_mfc.c:388:13: warning: 'gen6_mfc_avc_directmode_state' defined but not used [-Wunused-function] gen6_mfc.c:599:1: warning: 'gen6_mfc_avc_insert_object' defined but not used [-Wunused-function] gen6_mfc.c: In function 'gen6_mfc_context_destroy': gen6_mfc.c:991:1: warning: control reaches end of non-void function [-Wreturn-type] gen6_mfc.c: In function 'gen6_mfc_context_init': gen6_mfc.c:958:1: warning: control reaches end of non-void function [-Wreturn-type] make[3]: *** [i965_drv_video_la-gen6_mfc.lo] Error 1 make[3]: Leaving directory `/tmp/yaourt-tmp-name/aur-libva-driver-intel-g45-h264/src/intel-driver/src' make[2]: *** [all-recursive] Error 1 make[2]: Leaving directory `/tmp/yaourt-tmp-name/aur-libva-driver-intel-g45-h264/src/intel-driver/src' make[1]: *** [all] Error 2 make[1]: Leaving directory `/tmp/yaourt-tmp-name/aur-libva-driver-intel-g45-h264/src/intel-driver/src'

commented on 2011-11-27 10:03 (UTC)

Does this replace xf86-video-intel?

arkusuma commented on 2011-11-27 09:53 (UTC)

No it doesn't.

commented on 2011-11-27 09:32 (UTC)

Does this replace xf86-video-intel?

arkusuma commented on 2011-11-04 11:26 (UTC)

I don't think so. This is the output of libva-driver-intel-1.0.15-1 from extra on my laptop (4500MHD graphic): $ vainfo libva: VA-API version 0.32.0 libva: va_getDriverName() returns 0 libva: Trying to open /usr/lib/dri/i965_drv_video.so libva: va_openDriver() returns 0 vainfo: VA-API version: 0.32 (libva 1.0.15) vainfo: Driver version: Intel i965 driver - 1.0.15 vainfo: Supported profile and entrypoints VAProfileMPEG2Simple : VAEntrypointVLD VAProfileMPEG2Main : VAEntrypointVLD

wonder commented on 2011-11-04 11:10 (UTC)

hmm i was thinking g45-h264 was merged in master.