You are not logged in.

#2601 2013-09-29 13:20:03

Anon11
Member
Registered: 2012-05-20
Posts: 36

Re: Discussion: the ati/radeon open source drivers & radeon repository

archie0 wrote:

Yes it does. I've tested it with openarena full screen, went to settings -> system, and it was using the dedicated card; but to me it seems it performs worse than with the IGP. Glxgears gives me around 1000 FPS with the dedicated and between 750-950 with the IGP, the fps with the IGP are fluctuating though.

Couple of things for performance

1. I think the 7000 series (southern islands) and beyond support is not yet upto par with earlier ones ( like 6000).
    Also, I think dpm for for SI went only in kernel 3.12 , so you can try mainline kernels - you may get improvement if you default clock is low

2. You can try sb backend ( this is enabled by default in mesa 9.3)
       set -    R600_DEBUG=sb  or   R600_DEBUG=sb,nollvm
 
    This should improve performance in 6000 too.

Also for the full screen thing - do you setup the env variables and xrandr stuff before starting xorg ?
Thats one thing i didnt try yet - other wise its a bug for my card sad

Offline

#2602 2013-09-29 14:54:12

archie0
Member
Registered: 2012-02-26
Posts: 144

Re: Discussion: the ati/radeon open source drivers & radeon repository

Anon11 wrote:

Honestly, it makes no difference if I run the xrandr command on boot or not.
Here's how I tested: ran the xrandr command, exported the variable, ran the game.
That's it. What DE are you using? Could be DE related

Offline

#2603 2013-09-30 07:51:35

Anon11
Member
Registered: 2012-05-20
Posts: 36

Re: Discussion: the ati/radeon open source drivers & radeon repository

archie0 wrote:

Honestly, it makes no difference if I run the xrandr command on boot or not.
Here's how I tested: ran the xrandr command, exported the variable, ran the game.
That's it. What DE are you using? Could be DE related

Didnt work , seems to be driver bug - gpu is working ( in radeontop ) and i think everything is ok execpt that the screen is black.
Hopefully will get solved in future kernel releases

Offline

#2604 2013-09-30 08:03:36

agapito
Member
From: Who cares.
Registered: 2008-11-13
Posts: 641

Re: Discussion: the ati/radeon open source drivers & radeon repository

@lordheavy: Could you add glamor-egl-git to the repo? It brings xv support for all radeonsi cards.

http://phoronix.com/forums/showthread.p … post360423

Thank you.


Excuse my poor English.

Offline

#2605 2013-10-02 09:40:26

lordheavy
Developer
From: France
Registered: 2010-04-01
Posts: 292

Re: Discussion: the ati/radeon open source drivers & radeon repository

agapito wrote:

@lordheavy: Could you add glamor-egl-git to the repo? It brings xv support for all radeonsi cards.

http://phoronix.com/forums/showthread.p … post360423

Thank you.

Pushed a glamor-egl-git (with xv support enabled) package, on which xf86-video-ati-git package niow depends.

[lordh@archMain ~]$ xvinfo
X-Video Extension version 2.2
screen #0
  Adaptor #0: "GLAMOR Textured Video"
    number of ports: 16
    port base: 63
    operations supported: PutImage 
    supported visuals:
      depth 24, visualID 0x21
    number of attributes: 5
      "XV_BRIGHTNESS" (range -1000 to 1000)
              client settable attribute
              client gettable attribute (current value is 0)
      "XV_CONTRAST" (range -1000 to 1000)
              client settable attribute
              client gettable attribute (current value is 0)
      "XV_SATURATION" (range -1000 to 1000)
              client settable attribute
              client gettable attribute (current value is 0)
      "XV_HUE" (range -1000 to 1000)
              client settable attribute
              client gettable attribute (current value is 0)
      "XV_COLORSPACE" (range 0 to 1)
              client settable attribute
              client gettable attribute (current value is 0)
    maximum XvImage size: 8192 x 8192
    Number of image formats: 2
      id: 0x32315659 (YV12)
        guid: 59563132-0000-0010-8000-00aa00389b71
        bits per pixel: 12
        number of planes: 3
        type: YUV (planar)
      id: 0x30323449 (I420)
        guid: 49343230-0000-0010-8000-00aa00389b71
        bits per pixel: 12
        number of planes: 3
        type: YUV (planar)

Please also note that i've disabled PrettyStackTrace support in lib32-llvm package, so now mono games are working in Steam https://bugs.freedesktop.org/show_bug.cgi?id=60929

Last edited by lordheavy (2013-10-02 09:45:28)

Offline

#2606 2013-10-02 10:14:33

agapito
Member
From: Who cares.
Registered: 2008-11-13
Posts: 641

Re: Discussion: the ati/radeon open source drivers & radeon repository

Thank you!


Excuse my poor English.

Offline

#2607 2013-10-02 13:41:00

IncredibleLaser
Member
From: Germany, NRW
Registered: 2008-07-16
Posts: 158

Re: Discussion: the ati/radeon open source drivers & radeon repository

Thanks for your repo. I use it and the performance is really good. Also props to AMD for their docs and all the people writing those drivers, you make my next graphics card an easier choice.

Offline

#2608 2013-10-02 15:15:20

archie0
Member
Registered: 2012-02-26
Posts: 144

Re: Discussion: the ati/radeon open source drivers & radeon repository

Glamor is still not working for me. I've installed the git version of the ati driver, added AccelMethod "glamor" to my Device section, but X freezes when loading the OpenGL extensions. What am I doing wrong?

Offline

#2609 2013-10-07 08:45:33

IncredibleLaser
Member
From: Germany, NRW
Registered: 2008-07-16
Posts: 158

Re: Discussion: the ati/radeon open source drivers & radeon repository

Some more information would be nice. Which card do you use? It works here with "Mobility Radeon HD 4650/5165".  My /etc/X11/xorg.conf.d/20-radeon.conf looks like this:

Section "Module"
        Load "dri2"
        Load "glamoregl" 
EndSection

Section "Device"
    Identifier "Radeon"
    Driver "radeon"
    Option "AccelMethod" "glamor"
EndSection

and it works.

Offline

#2610 2013-10-13 15:22:53

AntAntAnt
Member
Registered: 2012-12-28
Posts: 11

Re: Discussion: the ati/radeon open source drivers & radeon repository

Hi all,

I got X randomly crashing at boot. Restarting the computer multiple times finally get me to KDM.
Here is Xorg log when it crashes:

[     2.507] (++) using VT number 7

[     2.512] (II) [KMS] drm report modesetting isn't supported.
[     2.512] (EE) Screen 0 deleted because of no matching config section.
[     2.512] (II) UnloadModule: "radeon"
[     2.512] (EE) Device(s) detected, but none match those in the config file.
[     2.512] (EE) 
Fatal server error:
[     2.512] (EE) no screens found(EE) 
[     2.512] (EE) 

And the same part when it works:

[     2.404] (++) using VT number 7

[     2.412] (II) [KMS] Kernel modesetting enabled.
[     2.412] (II) RADEON(0): Creating default Display subsection in Screen section
        "Default Screen Section" for depth/fbbpp 24/32
[     2.412] (==) RADEON(0): Depth 24, (--) framebuffer bpp 32
[     2.412] (II) RADEON(0): Pixel depth = 24 bits stored in 4 bytes (32 bpp pixmaps)
[     2.412] (==) RADEON(0): Default visual is TrueColor
[     2.412] (==) RADEON(0): RGB weight 888
[     2.412] (II) RADEON(0): Using 8 bits per RGB (8 bit DAC)
[     2.412] (--) RADEON(0): Chipset: "ATI Mobility Radeon HD 4670" (ChipID = 0x9488)
[     3.255] (II) Loading sub module "dri2"
[     3.255] (II) LoadModule: "dri2"
[     3.255] (II) Module "dri2" already built-in
[     3.255] (II) Loading sub module "exa"
[     3.255] (II) LoadModule: "exa"
[     3.255] (II) Loading /usr/lib/xorg/modules/libexa.so
[     3.257] (II) Module exa: vendor="X.Org Foundation"
[     3.257]    compiled for 1.14.3, module version = 2.6.0
[     3.257]    ABI class: X.Org Video Driver, version 14.1

Any idea what is causing that? Could that be hardware issue?

Offline

#2611 2013-10-13 15:28:42

snack
Member
From: Italy
Registered: 2009-01-13
Posts: 861

Re: Discussion: the ati/radeon open source drivers & radeon repository

@AntAntAnt: it's a known bug related to R600+DPM, I think. Check this: https://bugs.freedesktop.org/show_bug.cgi?id=66963

Offline

#2612 2013-10-13 15:42:46

AntAntAnt
Member
Registered: 2012-12-28
Posts: 11

Re: Discussion: the ati/radeon open source drivers & radeon repository

Actually I enabled dpm a few days ago but I was already having this issue. I will remove it just in case though.
Also, the effects described in the bug seems different than mine as I am just reaching tty1 when X crashes.
Thanks for the pointer though.

Offline

#2613 2013-10-13 15:47:34

snack
Member
From: Italy
Registered: 2009-01-13
Posts: 861

Re: Discussion: the ati/radeon open source drivers & radeon repository

AntAntAnt wrote:

Actually I enabled dpm a few days ago but I was already having this issue. I will remove it just in case though.
Also, the effects described in the bug seems different than mine as I am just reaching tty1 when X crashes.
Thanks for the pointer though.

Sorry for the misleading answer. I thought I recognize the same symptoms of my Mobile HD3470 in your post but I was wrong.

Offline

#2614 2013-10-13 18:59:37

KlavKalashj
Member
Registered: 2008-10-09
Posts: 376

Re: Discussion: the ati/radeon open source drivers & radeon repository

This bug (which I am experiencing) sounds relevant, could this be your problem too? https://bugs.freedesktop.org/show_bug.cgi?id=69723

Offline

#2615 2013-10-13 19:07:21

AntAntAnt
Member
Registered: 2012-12-28
Posts: 11

Re: Discussion: the ati/radeon open source drivers & radeon repository

I don't think so as I was experimenting the problem even with dpm not set.
I turned it off anyway, so I am going to see if it gets better during the next boots.

Offline

#2616 2013-10-16 23:18:02

skottish
Forum Fellow
From: Here
Registered: 2006-06-16
Posts: 7,942

Re: Discussion: the ati/radeon open source drivers & radeon repository

Thanks for the [mesa-git] repo lordheavy. I appreciate it.

Is there any compelling reason to use xf86-video-ati-git for a 6850? Other than the vdpau dl-close bug that's already been reported upstream, everything seems to be working well with the non-git driver.

Offline

#2617 2013-10-19 16:32:57

ddeeds
Member
Registered: 2013-10-02
Posts: 8

Re: Discussion: the ati/radeon open source drivers & radeon repository

gee wrote:

Hey guys,

is there any prebuild mesa, or at least pkgfile, with CKoenig changes to vdpau titled "NV_vdpau_interop v3", and kernel with Anssi's "hda - hdmi: Add ATI/AMD multi-channel audio support"?

This looks super interesting to me!

Radeon, VDPAU, HD Audio (including DTS-HD, TrueHD, LPCM 7.1), XBMC plus a 50% decrease in CPU usage over AMD Catalyst Drivers

What you get:

- Bitstream Audio (including DTS-HD, TrueHD, LPCM 7.1)
- mpeg2, mpeg4, h264, vc1 accleration
- Bob deinterlacing via vdpau
- 24 fps Video Playback (AMD and Intel Haswell CPUs)
- XBMC runs with 50% less CPU usage over AMD Catalyst Drivers

Last edited by ddeeds (2013-10-20 01:14:04)

Offline

#2618 2013-10-20 07:16:07

gee
Member
Registered: 2006-11-29
Posts: 313

Re: Discussion: the ati/radeon open source drivers & radeon repository

ddeeds wrote:
gee wrote:

Hey guys,

is there any prebuild mesa, or at least pkgfile, with CKoenig changes to vdpau titled "NV_vdpau_interop v3", and kernel with Anssi's "hda - hdmi: Add ATI/AMD multi-channel audio support"?

This looks super interesting to me!

Radeon, VDPAU, HD Audio (including DTS-HD, TrueHD, LPCM 7.1), XBMC plus a 50% decrease in CPU usage over AMD Catalyst Drivers

What you get:

- Bitstream Audio (including DTS-HD, TrueHD, LPCM 7.1)
- mpeg2, mpeg4, h264, vc1 accleration
- Bob deinterlacing via vdpau
- 24 fps Video Playback (AMD and Intel Haswell CPUs)
- XBMC runs with 50% less CPU usage over AMD Catalyst Drivers

Awesome post thank you!

Offline

#2619 2013-10-21 07:09:48

kaszak696
Member
Registered: 2009-05-26
Posts: 543

Re: Discussion: the ati/radeon open source drivers & radeon repository

Hey guys, anyone tried running Brütal Legend on radeon R600g driver? I don't know if i should download it on linux, 8GB of transfer doesn't grow on trees wink


'What can be asserted without evidence can also be dismissed without evidence.' - Christopher Hitchens
'There's no such thing as addiction, there's only things that you enjoy doing more than life.' - Doug Stanhope
GitHub Junkyard

Offline

#2620 2013-10-27 14:42:26

haagch
Member
Registered: 2013-08-17
Posts: 209

Re: Discussion: the ati/radeon open source drivers & radeon repository

There is intel-dri-git in the repository, could you also add xf86-video-intel-git?

Since very recently I got "AIGLX error: i965 exports no extensions (/usr/lib/xorg/modules/dri/i965_dri.so: undefined symbol: __driDriverExtensions)" and no 3d acceleration anymore. Does anyone know what's up with that? Is that related to the new glx implementation and do I need a newer X server?

[ 76109.446] (II) LoadModule: "intel"
[ 76109.446] (II) Loading /usr/lib/xorg/modules/drivers/intel_drv.so
[ 76109.446] (II) Module intel: vendor="X.Org Foundation"
[ 76109.446]    compiled for 1.14.2, module version = 2.21.15
[ 76109.446]    Module class: X.Org Video Driver
[ 76109.446]    ABI class: X.Org Video Driver, version 14.1
[ 76109.446] (II) LoadModule: "vesa"
[ 76109.446] (WW) Warning, couldn't open module vesa
[ 76109.446] (II) UnloadModule: "vesa"
[ 76109.446] (II) Unloading vesa
[ 76109.446] (EE) Failed to load module "vesa" (module does not exist, 0)
[ 76109.446] (II) LoadModule: "modesetting"
[ 76109.447] (WW) Warning, couldn't open module modesetting
[ 76109.447] (II) UnloadModule: "modesetting"
[ 76109.447] (II) Unloading modesetting
[ 76109.447] (EE) Failed to load module "modesetting" (module does not exist, 0)
[ 76109.447] (II) LoadModule: "fbdev"
[ 76109.447] (WW) Warning, couldn't open module fbdev
[ 76109.447] (II) UnloadModule: "fbdev"
[ 76109.447] (II) Unloading fbdev
[ 76109.447] (EE) Failed to load module "fbdev" (module does not exist, 0)
[ 76109.447] (II) intel: Driver for Intel(R) Integrated Graphics Chipsets:
        i810, i810-dc100, i810e, i815, i830M, 845G, 854, 852GM/855GM, 865G,
        915G, E7221 (i915), 915GM, 945G, 945GM, 945GME, Pineview GM,
        Pineview G, 965G, G35, 965Q, 946GZ, 965GM, 965GME/GLE, G33, Q35, Q33,
        GM45, 4 Series, G45/G43, Q45/Q43, G41, B43, HD Graphics,
        HD Graphics 2000, HD Graphics 3000, HD Graphics 2500,
        HD Graphics 4000, HD Graphics P4000, HD Graphics 4600,
        HD Graphics 5000, HD Graphics P4600/P4700, Iris(TM) Graphics 5100,
        HD Graphics 4400, HD Graphics 4200, Iris(TM) Pro Graphics 5200
[ 76109.447] (++) using VT number 1

[ 76109.448] (II) intel(0): Creating default Display subsection in Screen section
        "Default Screen Section" for depth/fbbpp 24/32
[ 76109.448] (==) intel(0): Depth 24, (--) framebuffer bpp 32
[ 76109.448] (==) intel(0): RGB weight 888
[ 76109.448] (==) intel(0): Default visual is TrueColor
[ 76109.448] (--) intel(0): Integrated Graphics Chipset: Intel(R) HD Graphics 4000
[ 76109.448] (--) intel(0): CPU: x86-64, sse2, sse3, ssse3, sse4.1, sse4.2, avx
[ 76109.448] (**) intel(0): Framebuffer tiled
[ 76109.448] (**) intel(0): Pixmaps tiled
[ 76109.448] (**) intel(0): "Tear free" disabled
[ 76109.448] (**) intel(0): Forcing per-crtc-pixmaps? no
[ 76109.449] (II) intel(0): Output LVDS1 has no monitor section
[ 76109.450] (--) intel(0): found backlight control interface acpi_video0 (type 'firmware')
[ 76109.451] (II) intel(0): Output VGA1 has no monitor section
[ 76109.451] (II) intel(0): Output HDMI1 has no monitor section
[ 76109.451] (II) intel(0): Output DP1 has no monitor section
[ 76109.451] (II) intel(0): Output HDMI2 has no monitor section
[ 76109.451] (II) intel(0): Output HDMI3 has no monitor section
[ 76109.451] (II) intel(0): Output DP2 has no monitor section
[ 76109.451] (II) intel(0): Output DP3 has no monitor section
[ 76109.451] (--) intel(0): Output LVDS1 using initial mode 1920x1080 on pipe 0
[ 76109.451] (--) intel(0): Output HDMI1 using initial mode 1920x1080 on pipe 1
[ 76109.451] (==) intel(0): DPI set to (96, 96)
[ 76109.451] (II) Loading sub module "dri2"
[ 76109.451] (II) LoadModule: "dri2"
[ 76109.451] (II) Module "dri2" already built-in
[ 76109.451] (==) Depth 24 pixmap format is 32 bpp
[ 76109.451] (II) intel(0): SNA initialized with Ivybridge (gen7, gt2) backend
[ 76109.451] (==) intel(0): Backing store disabled
[ 76109.451] (==) intel(0): Silken mouse enabled
[ 76109.451] (II) intel(0): HW Cursor enabled
[ 76109.451] (II) intel(0): RandR 1.2 enabled, ignore the following RandR disabled message.
[ 76109.453] (==) intel(0): DPMS enabled
[ 76109.453] (II) intel(0): [DRI2] Setup complete
[ 76109.453] (II) intel(0): [DRI2]   DRI driver: i965
[ 76109.453] (II) intel(0): direct rendering: DRI2 Enabled
[ 76109.453] (==) intel(0): hotplug detection: "enabled"
[ 76109.453] (--) RandR disabled
[ 76109.459] (EE) AIGLX error: i965 exports no extensions (/usr/lib/xorg/modules/dri/i965_dri.so: undefined symbol: __driDriverExtensions)
[ 76109.460] (EE) AIGLX: reverting to software rendering
[ 76109.460] (II) AIGLX: Screen 0 is not DRI capable
[ 76109.469] (II) AIGLX: Loaded and initialized swrast
[ 76109.469] (II) GLX: Initialized DRISWRAST GL provider for screen 0

Offline

#2621 2013-10-27 17:01:24

lordheavy
Developer
From: France
Registered: 2010-04-01
Posts: 292

Re: Discussion: the ati/radeon open source drivers & radeon repository

haagch wrote:

There is intel-dri-git in the repository, could you also add xf86-video-intel-git?

Since very recently I got "AIGLX error: i965 exports no extensions (/usr/lib/xorg/modules/dri/i965_dri.so: undefined symbol: __driDriverExtensions)" and no 3d acceleration anymore. Does anyone know what's up with that? Is that related to the new glx implementation and do I need a newer X server?

It's related to the "megadriver" feature, xserver needs a new symbol; you'll need to rebuild xorg-server with this patch here:
http://cgit.freedesktop.org/xorg/xserve … 8eceaeb893

It's also possible to have the xf86-video-intel-git package, i will add it in the repo.

Last edited by lordheavy (2013-10-27 17:19:57)

Offline

#2622 2013-10-27 19:33:30

haagch
Member
Registered: 2013-08-17
Posts: 209

Re: Discussion: the ati/radeon open source drivers & radeon repository

Thanks, much appreciated. smile

I can confirm that with only this patch everything (read: glxinfo) works.
It can surprise people because the mesa version in your repository already has it but the xserver from core doesn't.

Offline

#2623 2013-10-28 06:22:28

quelcom
Member
From: Helsinki
Registered: 2009-04-12
Posts: 4

Re: Discussion: the ati/radeon open source drivers & radeon repository

Any tips for getting stable and reliable HD video playback with Evergreen based graphic cards? My HTPC has an E2 APU that uses a Radeon HD 7340 (Palm/Wrestler) and although I have followed the wiki instructions the HD video playback is driving me crazy.

The computer is only used to play videos. When I run XBMC through the xinit command (no window manager) there is about 80% chances that it will get stuck right after trying to play an HD video. When this happens I need to go back to the TTY that had the xinit command and stop it. Then I repeat the process until it works. When it does work, I could play all HD videos I want just fine.. until I close XBMC.

Do you have better experiences with mpv or vlc? I tried them both but the CPU usage was really high so I do not think it was using the GPU for the decoding.

Offline

#2624 2013-10-29 04:22:51

snowblind
Member
Registered: 2013-07-18
Posts: 39

Re: Discussion: the ati/radeon open source drivers & radeon repository

lordheavy wrote:
haagch wrote:

There is intel-dri-git in the repository, could you also add xf86-video-intel-git?

Since very recently I got "AIGLX error: i965 exports no extensions (/usr/lib/xorg/modules/dri/i965_dri.so: undefined symbol: __driDriverExtensions)" and no 3d acceleration anymore. Does anyone know what's up with that? Is that related to the new glx implementation and do I need a newer X server?

It's related to the "megadriver" feature, xserver needs a new symbol; you'll need to rebuild xorg-server with this patch here:
http://cgit.freedesktop.org/xorg/xserve … 8eceaeb893

It's also possible to have the xf86-video-intel-git package, i will add it in the repo.

Is this at all related to this?

LIBGL_DEBUG=verbose glxinfo
name of display: :0.0
libGL: OpenDriver: trying /tls/i965_dri.so
libGL: OpenDriver: trying /i965_dri.so
libGL error: dlopen /i965_dri.so failed (/i965_dri.so: cannot open shared object file: No such file or directory)
libGL: OpenDriver: trying /usr/lib32/xorg/modules/dri/tls/i965_dri.so
libGL: OpenDriver: trying /usr/lib32/xorg/modules/dri/i965_dri.so
libGL error: dlopen /usr/lib32/xorg/modules/dri/i965_dri.so failed (/usr/lib32/xorg/modules/dri/i965_dri.so: wrong ELF class: ELFCLASS32)
libGL error: unable to load driver: i965_dri.so
libGL error: driver pointer missing
libGL error: failed to load driver: i965

This also happens when I use DRI_PRIME=1 to offload it to radeonsi. However, I can symlink both the i965 and radeonsi driver modules to the root directory (/) and it works. Was the prefix for the 64bit driver messed up?

Last edited by snowblind (2013-10-29 04:24:52)

Offline

#2625 2013-10-29 17:15:16

haagch
Member
Registered: 2013-08-17
Posts: 209

Re: Discussion: the ati/radeon open source drivers & radeon repository

No.

/usr/lib/xorg/modules/dri/i965_dri.so is owned by intel-dri-git 59335.8bd4476-1

Check if you have that file, if not (re)install intel-dri-git

If it is there, check whether you have $LIBGL_DRIVERS_PATH set to something that doesn't include /usr/lib/xorg/modules/dri.

Offline

Board footer

Powered by FluxBB