Fedora 10, fglrx 8.543 and AVIVO?

[asun]

How do I enable AVIVO with the current setup of:
Fedora 10
fglrx 8.543 (Catalyst 8.11)
Radeon HD 2400
mplayer 1.0

I remember clearly that AVIVO works really well in Fedora 9. I could play HD clips, even 1080p without any hitches. With my current F10 setup, the CPU is pegged frequently when playing HD video and the picture is jerky, which means the decoding is not offloaded to the GPU. The message from mplayer also does indicate that AVIVO is enabled (it used to be in F9).

Here is my relevant xorg.conf:
Section "Device"
Identifier "aticonfig-Device[0]-0"
Driver "fglrx"
Option "OpenGLOverlay" "off"
Option "VideoOverlay" "off"
Option "TexturedVideo" "on"
Option "ShadowFB" "on"
Option "UseFastTLS" "1"
Option "Textured2D" "on"
Option "TexturedXrender" "off"
Option "RenderAccel" "true"
Option "TexturedVideoSync" "on"
BusID "PCI:2:0:0"
EndSection

Oh, by the way if I turn on TexturedXrender, X will crash over and over again. Can anyone help?

[Kano]

You can buy a cheap Nvidia 8400 GS and begin to play with vdpau enabled mplayer, xine, mythtv or whatever. There is still some tearing, but the cpu has not much to do. No way to use AVIVO today, ask ATI when this will happen or if it will be better than VDPAU.

[asun]

Alright, I'm now officially asking ATI as to why AVIVO is not working anymore :) I clearly remember it worked a few versions of Catalyst prior to this one.

[bridgman]

What do you mean by AVIVO ? It means different things to different people, but it started as "Advanced Video In / Video Out", ie a combination of the new 10-bit display controller pipe on 5xx and higher plus a bunch of video processing algorithms for improved noise reduction, de-interlacing etc...

From your earlier email I think you're talking about accelerated Xv ?

[asun]

What do you mean by AVIVO ? It means different things to different people, but it started as "Advanced Video In / Video Out", ie a combination of the new 10-bit display controller pipe on 5xx and higher plus a bunch of video processing algorithms for improved noise reduction, de-interlacing etc...

From your earlier email I think you're talking about accelerated Xv ?

Ah, thanks for the response. The AVIVO I mentioned means offloading some video processing to the GPU. I noticed going from Fedora 9 to Fedora 10, some hi-def videos start to become jerky when playing on the HDTV. My investigation brings it down to the video processing is pegging out the CPU (using top) which never happened in Fedora 9. Not sure what you mean by accelerated xv, but xvinfo shows a bunch of info and mplayer can use xv output.

Since I'm not sure whether it's a problem with kernel, X, mplayer or Catalyst driver, I'm asking the knowledgeable people here first. :)

[bridgman]

(scratches head)

I didn't think fglrx even ran on F10 right now, at least that's what I had been told. Are you 100% sure you are actually running fglrx ?

Can you pastebin your xorg log ? That will give us a good idea what is going on.

[asun]

Ah yes, I did install the fglrx from F9's livna/rpmfusion repository. The instruction was posted in fedoraforum. The whole Xorg.0.log is too long to post here. Here is the snippet. If you need any other part, please let me know.


(II) Loading /usr/lib64/xorg/modules//libfb.so
(II) Module fb: vendor="X.Org Foundation"
compiled for 1.5.3, module version = 1.0.0
ABI class: X.Org ANSI C Emulation, version 0.4
(II) Loading sub module "ramdac"
(II) LoadModule: "ramdac"
(II) Module "ramdac" already built-in
(==) fglrx(0): NoAccel = NO
(II) Loading sub module "xaa"
(II) LoadModule: "xaa"

(II) Loading /usr/lib64/xorg/modules//libxaa.so
(II) Module xaa: vendor="X.Org Foundation"
compiled for 1.5.3, module version = 1.2.0
ABI class: X.Org Video Driver, version 4.1
(==) fglrx(0): NoDRI = NO
(II) Loading sub module "fglrxdrm"
(II) LoadModule: "fglrxdrm"

(II) Reloading /usr/lib64/xorg/modules/linux//libfglrxdrm.so
(==) fglrx(0): Capabilities: 0x00000000
(==) fglrx(0): CapabilitiesEx: 0x00000000
(==) fglrx(0): cpuFlags: 0x4000001f
(==) fglrx(0): OpenGL ClientDriverName: "fglrx_dri.so"
(**) fglrx(0): ATI GART size: 255 MB
(II) fglrx(0): [pcie] 261120 kB allocated
(II) fglrx(0): [drm] DRM buffer queue setup: nbufs = 100 bufsize = 65536
(**) fglrx(0): UseFastTLS=1
(==) fglrx(0): BlockSignalsOnLock=1
(II) fglrx(0): Direct rendering enabled
(II) fglrx(0): Interrupt handler installed at IRQ 16.
(II) fglrx(0): Exposed events to the /proc interface
(II) fglrx(0): Finished Initialize PPLIB!
(--) Depth 24 pixmap format is 32 bpp
(II) do I need RAC? No, I don't.
(II) resource ranges after preInit:
[0] -1 0 0xffffffff - 0xffffffff (0x1) MX[B]
[1] -1 0 0x000f0000 - 0x000fffff (0x10000) MX[B]
<snip>
<snip>
[27] 0 0 0x000003b0 - 0x000003bb (0xc) IS[B](OprU)
[28] 0 0 0x000003c0 - 0x000003df (0x20) IS[B](OprU)
(II) fglrx(0): driver needs X.org 1.4.x.y with x.y >= 99.906
(II) fglrx(0): detected X.org 7.4.3.0
(II) Loading extension ATIFGLRXDRI
(II) fglrx(0): doing DRIScreenInit
(II) fglrx(0): DRIScreenInit for fglrx driver
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 12, (OK)
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 12, (OK)
drmOpenByBusid: Searching for BusID PCI:2:0:0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 12, (OK)
drmOpenByBusid: drmOpenMinor returns 12
drmOpenByBusid: drmGetBusid reports PCI:2:0:0
(II) fglrx(0): [drm] DRM interface version 1.0
(II) fglrx(0): [drm] created "fglrx" driver at busid "PCI:2:0:0"
(II) fglrx(0): [drm] added 8192 byte SAREA at 0x2000
(II) fglrx(0): [drm] mapped SAREA 0x2000 to 0x7f9b3b24a000
(II) fglrx(0): [drm] framebuffer handle = 0x3000
(II) fglrx(0): [drm] added 1 reserved context for kernel
(II) fglrx(0): DRIScreenInit done
(II) fglrx(0): Kernel Module Version Information:
(II) fglrx(0): Name: fglrx
(II) fglrx(0): Version: 8.54.3
(II) fglrx(0): Date: Oct 10 2008
(II) fglrx(0): Desc: ATI FireGL DRM kernel module
(II) fglrx(0): Kernel Module version matches driver.
(II) fglrx(0): Kernel Module Build Time Information:
(II) fglrx(0): Build-Kernel UTS_RELEASE: 2.6.27.9-159.fc10.x86_64
(II) fglrx(0): Build-Kernel MODVERSIONS: no
(II) fglrx(0): Build-Kernel __SMP__: no
(II) fglrx(0): Build-Kernel PAGE_SIZE: 0x1000
(II) fglrx(0): [drm] register handle = 0x00004000
(II) fglrx(0): DRI initialization successfull!
(II) fglrx(0): FBADPhys: 0xc0000000 FBMappedSize: 0x01068000
(II) fglrx(0): FBMM initialized for area (0,0)-(1920,2240)
(II) fglrx(0): FBMM auto alloc for area (0,0)-(1920,1080) (front color buffer - assumption)
(II) fglrx(0): Largest offscreen area available: 1920 x 1160
(==) fglrx(0): Backing store disabled
(II) Loading extension FGLRXEXTENSION
(II) Loading extension ATITVOUT
(**) fglrx(0): DPMS enabled
(**) fglrx(0): Textured Video is enabled.
(II) LoadModule: "glesx"

(II) Loading /usr/lib64/xorg/modules//glesx.so
(II) Module glesx: vendor="X.Org Foundation"
compiled for 1.4.99.906, module version = 1.0.0
ABI class: X.Org Server Extension, version 1.1
(II) Loading extension GLESX
(II) fglrx(0): GLESX enableFlags = 26
(**) fglrx(0): Option "XaaNoOffscreenPixmaps" "on"
(II) fglrx(0): Using XFree86 Acceleration Architecture (XAA)
Screen to screen bit blits
Solid filled rectangles
Solid Horizontal and Vertical Lines
Driver provided ScreenToScreenBitBlt replacement
Driver provided FillSolidRects replacement
(II) fglrx(0): GLESX is enabled
(II) LoadModule: "amdxmm"

(II) Loading /usr/lib64/xorg/modules//amdxmm.so
(II) Module amdxmm: vendor="X.Org Foundation"
compiled for 1.4.99.906, module version = 1.0.0
ABI class: X.Org Server Extension, version 1.1
(WW) fglrx(0): Option "ShadowFB" is not used
(WW) fglrx(0): Option "EXAVSync" is not used
(WW) fglrx(0): Option "RenderAccel" is not used
(WW) fglrx(0): Option "VendorName" is not used
(WW) fglrx(0): Option "ModelName" is not used
(II) fglrx(0): X context handle = 0x1
(II) fglrx(0): [DRI] installation complete
(==) fglrx(0): Silken mouse enabled
(==) fglrx(0): Using hardware cursor
(==) RandR enabled
(II) Initializing built-in extension MIT-SHM
(II) Initializing built-in extension XInputExtension
(II) Initializing built-in extension XTEST
(II) Initializing built-in extension XKEYBOARD
(II) Initializing built-in extension XINERAMA
(II) Initializing built-in extension XFIXES
(II) Initializing built-in extension RENDER
(II) Initializing built-in extension RANDR
(II) Initializing built-in extension COMPOSITE
(II) Initializing built-in extension DAMAGE
(II) Initializing built-in extension XEVIE
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 13, (OK)
drmOpenByBusid: Searching for BusID PCI:2:0:0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 13, (OK)
drmOpenByBusid: drmOpenMinor returns 13
drmOpenByBusid: drmGetBusid reports PCI:2:0:0
(WW) AIGLX: 3D driver claims to not support visual 0x23

[bridgman]

Ah yes, I did install the fglrx from F9's livna/rpmfusion repository. The instruction was posted in fedoraforum. The whole Xorg.0.log is too long to post here. Here is the snippet. If you need any other part, please let me know.

Can you put the log up on pastebin http://www.pastebin.ca and post a link ?

[asun]

Thanks. The link is http://www.pastebin.ca/1294063

By the way, I updated the fglrx to 8.561 and there is still no offloading the video processing to GPU. The log above is from the new fglrx.

[bridgman]

There's an error message in the log I haven't seen before :

(EE) AIGLX error: dlopen of /usr/lib/dri/fglrx_dri.so failed (/usr/lib/dri/fglrx_dri.so: wrong ELF class: ELFCLASS32)
(EE) AIGLX: reverting to software rendering

Not sure if that will affect your video playback. I assume you are using Xv for video output ?

[Kizo]

Hello,
I have been reading these forums for period of time, and just decided to register.
I'm rather new at linux, but I have a radeon graphics card, and I am aware of a problem with this file /usr/lib/dri/fglrx_dri.so on 64bit systems. You can try removing the file, and make a link to its 64bit version, like this:
rm /usr/lib/dri/fglrx_dri.so
ln -s /usr/lib64/dri/fglrx_dri.so /usr/lib/dri/fglrx_dri.so

This problem is also mentioned here: http://www.phoronix.com/forums/showthread.php?t=14027 , in the second post.
As I said, I'm new to linux, but I think the problem may be that AIGLX, or fglrx (or something :) ) may be accessing the wrong file.
If I'm wrong, I appologise for the intrusion

[asun]

Ah.. how can I forget. :o Thanks to Kizo, I linked to the 64-bit version. That removes the error but there is still no offloading the video processing to GPU yet. mplayer still uses 100% CPU when playing hi-def video.

The new Xorg.0.log is here: http://www.pastebin.ca/1294233

Again, thanks to all the suggestions. Oh, yes, I'm using the xv video output for mplayer. The xv itself is fine, just not offloading video decoding to the GPU. After digging around for a while, it seems that what I'm looking for is XvMC. Bridgman, do you know what's the current status of it?

[bridgman]

This is where I get confused. We never supported XvMC (or anything else that accelerated video decoding) in older drivers either, so I don't see how there could have been a regression. There has to be something else going on here, like not getting Xv acceleration or something...

[asun]

This is where I get confused. We never supported XvMC in older drivers either, so I don't see how there could have been a regression...

That's odd. I clearly remember a while back (during F9), mplayer was spewing something about AVIVO supported and the playback was really smoother that it was now. Oh well, I guess it'll remain a mystery for now. Is it possible because of the TexturedXrender that is malfunctioning now?

[bridgman]

I'm not sure if TexturedXRender is supported yet or not; definitely try turning it off, but it shouldn't really affect playing back through Xv.

I'll take a look through the mplayer source and see if I can find what *they* mean by AVIVO ;)

I looked through your latest X log and didn't see anything out of the ordinary at all. Starting to have the first doubts about whether this is a driver issue at all...

[asun]

I just checked with fgrlx 8.561, "TexturedXrender" is not causing X crash anymore. It could very well be problem with different mplayer versions. I hope the AVIVO still leaves some traces in the mplayer source. Please keep me posted and thanks bridgman.