ATI radeon Sticky

[Wedge_]

madmax2001: it'd help if you could post exactly what problems you're having. Does anything in this thread help?

wallace1819: the "Permission denied" error looks like it's caused by portage rather than the drivers themselves. Try searching the forums and see if you can find anything on it.

[madmax2001]

Fair enough, I'm at work. When I get home I'll post something more detailed.

[Wedge_]

I am getting really tired of these multi-posts I clicked the Submit button once, and I get a bunch of the damn things

[wallace1819]

ok i've seen a half dozen versions of the "correct" kernel config for the following situation...

2.6.0-test5 kernel
x86 ati-drivers (using internal NOT kernel AGP) support
radeon 9000 (r250)

what is the right way to do this. is it...

agp
----
1. agpgart support compiled directly into kernel with NO drivers chosen
2. agpgart support compiled directly into kernel with motherboard driver compiled as module
3. agpgart support compiled directly into kernel with motherboard driver compiled directly into kernel
4. agpgart support NOT compiled into kernel

DRM
------
A. drm support compiled directly into kernel with NO drivers chosen
B. drm support compiled directly into kernel with video driver compiled as module
C. drm support compiled directly into kernel with video driver compiled directly into kernel
D. drm support NOT compiled into kernel

what is the correct combination of agp and dri?


thx for the help,
wallace

[Wedge_]

The fglrx driver won't work if other DRM modules are loaded. I think it should work if they're compiled as modules and aren't loaded, but I'm not 100% sure. Anyway, the best option is probably to compile both agpgart and the appropriate motherboard driver as modules (make sure you load both). Then compile in DRM support + ATI Radeon support as a module. That should give you the greatest flexibility - if you compile in AGP support, it can stop the internal AGP in the fglrx driver from working.

However, if the fglrx driver works well enough for you, then you can remove DRM from the kernel altogether, it isn't needed.

[wallace1819]

thx

i completly removed drm from the kernel and compiled agpgart in directly and my MB driver as a module and loaded it. and have stoped using the internel agp.

glxgears gives ~1480
fgl_glxgears gives ~108

this is within KDE running at 1600X1200

do these number sound right for a radeon 9000 mobility? I see lots of number like

glxgears ~4000
fgl_glxgears ~400

do people usuall text at lower resolutions to produce number like this or do think there is some more tweaking i could do some where?

thx
wallace

[wallace1819]

thx

i completly removed drm from the kernel and compiled agpgart in directly and my MB driver as a module and loaded it. and have stoped using the internel agp.

glxgears gives ~1480
fgl_glxgears gives ~108

this is within KDE running at 1600X1200

do these number sound right for a radeon 9000 mobility? I see lots of number like

glxgears ~4000
fgl_glxgears ~400

do people usuall text at lower resolutions to produce number like this or do think there is some more tweaking i could do some where?

thx
wallace

[Wedge_]

I think lowering your resolution will usually increase a glxgears score, but it isn't a proper benchmark anyway. Were the other scores you saw running at the same color depth? Remember that the fglrx driver only supports 24 bit at the moment - 16 bit would be faster. The 2.9.x drivers are also faster than the current 3.2.x drivers - I lose 1000fps in glxgears between 2.9.x and 3.2.x. You certainly seem to have 3D acceleration working though, which is good

[zeroK]

1. I want to thank everyone for their great tips on getting some speed out of my graphic card
2. Could it be that people with glxgears framerates from > 4000 have something over their glxgears window? *g*

[QuantumZ]

Yes, I know there are many different threads concerning the Radeon cards, including a FAQ and a 'Radeon Sticky' thread. I've done my best to read through all available documentation, but if I have missed something, and the answer to my problem is glaring me in the face, please feel free to slap me about the head and point me in the proper direction. That being said....

1.) I emerged XFree86.
2.) Then I configured my kernel. I have done this for many years, so I promise my kernel options are tight and correct. MTRR enabled, AGP enabled as module, my correct chipset chosen, and DRM disabled in kernel.

Code: Select all

#CONFIG_DRM is not set
...
CONFIG_AGP=m
CONFIG_AGP_INTEL=y
...
CONFIG_MTRR=y

3.) I then emerged Xfree86 DRM (Versions Listed Below.)

4.) I configured my XF86Config file. (By hand.) The values are correct, and the important sections are included here.

Code: Select all

Section "DRI"
    Mode 0666
EndSection
...
Load     "glx"
Load     "dri"

5.) Now, here in begins the problems. When I run glxinfo it reports that I have Direction Rendering running. Relevant sections listed.

Code: Select all

name of display: :0.0
display: :0 screen: 0
direct rendering: Yes
...
OpenGL vendor string: Tungsten Graphics, Inc.
OpenGL renderer: Mesa DRI Radeon 20020611 AGP 4x x86/MMX/SSE TCL
OpenGL version: 1.2 Mesa 4.04  

6.) This should bode well, but it does not. When I start up glxgears, I get no errors, but I also get strangely awful performance. And what I found suspicious is that this performance is absolutely constant, never fluctuating.

Code: Select all

932 frames in 5.0 seconds = 184.400 FPS
1000 frames in 5.0 seconds = 200.000 FPS
1000 frames in 5.0 seconds = 200.000 FPS
1000 frames in 5.0 seconds = 200.000 FPS

It goes on like this, everytime I run glxgears. It will start with something between 175 and 200 and then it will stay at 200, for however long I run it, no matter what else I am doing. I had my machine recompiling the kernel, and I still got precisely 200.000 FPS.

What is more strange, is that in order to test any mistakes I may have made, I disabled DRI and and accel, using the following Options in the XF86Config file.

Code: Select all

Option "no_accel"  "true"
Option "no_dri"  "true"

When I disabled Accel and DRI, I ran glxinfo and got expected information. Direct Rendering was off, and my render string was Mesa GLX Indirect. What troubled me was that when I ran glxgears I received higher scores, than I did when Direct Rendering was enabled!

Code: Select all

1754 frames in 5.0 seconds = 350.800 FPS
1850 frames in 5.0 seconds = 370.000 FPS
1872 frames in 5.0 seconds = 374.400 FPS
1926 frames in 5.0 seconds = 385.200 FPS

The only EE (Errors) or WW (Warnings) in my XFree86.0.log file are as follows:

Code: Select all

(WW) Open APM Failed (/dev/apm_bios) (No such file or directory)
(WW) Radeon(0): Bad V_BIOS checksum

This confuses me to no end, I have tried everything that I have found to try. (I may have used a linux system for a long time, but my graphics were always strictly 2D.) If there is a bug in glxgears or a bug in the xfree-drm package that I have missed please let me know. If I have been remiss or incompetent in something I apologize, please also let me know that so I can fix this problem and move on with my life. Lastly, if this post is redundant you'll have to forgive me, it's just imperative that I find an answer.

I am running a Dell Inspiron 5100.
2.4 Ghz. 512Mb RAM.
Chipset: i845G. Bios: A06
Radeon Mobility M7 LW [Radeon Mobility 7500] (rev 0)
Kernel - 'Gaming-Sources' (2.4.20 w/ gaming-r3 patches)
XFree86 Version 4.3.0
XFree-DRM Version 4.3.0-r6

[xlyz]

good news:

http://forums.gentoo.org/viewtopic.php?t=89975

[vincer]

Parak: I tried your instructions but still no go. I even did a completely fresh install.

I'm throwing in the towel and going back to Mandrake. I've just put way too much time into this and haven't gotten ANYWHERE. Still the same blank screen as two weeks ago. It's too bad I was really excited about Gentoo. Maybe later I guess.

Thanks for your help, guys.

v

[antipop]

I tried reading the whole thread through but it's too long.

Code: Select all

$ time glxinfo
name of display: :1.0
display: :1  screen: 0
direct rendering: Yes
server glx vendor string: SGI
server glx version string: 1.2
server glx extensions:
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_EXT_import_context
client glx vendor string: SGI
client glx version string: 1.2
client glx extensions:
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_EXT_import_context
GLX extensions:
    GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_EXT_import_context
OpenGL vendor string: ATI Technologies Inc.
OpenGL renderer string: Radeon 8500 DDR Pentium 4 (SSE2)
OpenGL version string: 1.3 (X4.3.0-3.2.5)
OpenGL extensions:
    GL_ARB_multitexture, GL_EXT_texture_env_add, GL_EXT_compiled_vertex_array,
    GL_S3_s3tc, GL_ARB_point_parameters, GL_ARB_texture_border_clamp,
    GL_ARB_texture_compression, GL_ARB_texture_cube_map,
    GL_ARB_texture_env_add, GL_ARB_texture_env_combine,
    GL_ARB_texture_env_crossbar, GL_ARB_texture_env_dot3,
    GL_ARB_texture_mirrored_repeat, GL_ARB_transpose_matrix,
    GL_ARB_vertex_blend, GL_ARB_vertex_buffer_object, GL_ARB_vertex_program,
    GL_ARB_window_pos, GL_ATI_element_array, GL_ATI_envmap_bumpmap,
    GL_ATI_fragment_shader, GL_ATI_map_object_buffer,
    GL_ATI_texture_env_combine3, GL_ATI_texture_mirror_once,
    GL_ATI_vertex_array_object, GL_ATI_vertex_attrib_array_object,
    GL_ATI_vertex_streams, GL_ATIX_texture_env_combine3,
    GL_ATIX_texture_env_route, GL_ATIX_vertex_shader_output_point_size,
    GL_EXT_abgr, GL_EXT_bgra, GL_EXT_blend_color, GL_EXT_blend_func_separate,
    GL_EXT_blend_minmax, GL_EXT_blend_subtract, GL_EXT_clip_volume_hint,
    GL_EXT_draw_range_elements, GL_EXT_fog_coord, GL_EXT_multi_draw_arrays,
    GL_EXT_packed_pixels, GL_EXT_point_parameters, GL_EXT_rescale_normal,
    GL_EXT_polygon_offset, GL_EXT_secondary_color,
    GL_EXT_separate_specular_color, GL_EXT_stencil_wrap,
    GL_EXT_texgen_reflection, GL_EXT_texture3D,
    GL_EXT_texture_compression_s3tc, GL_EXT_texture_cube_map,
    GL_EXT_texture_edge_clamp, GL_EXT_texture_env_combine,
    GL_EXT_texture_env_dot3, GL_EXT_texture_filter_anisotropic,
    GL_EXT_texture_lod_bias, GL_EXT_texture_object, GL_EXT_texture_rectangle,
    GL_EXT_vertex_array, GL_EXT_vertex_shader, GL_HP_occlusion_test,
    GL_NV_texgen_reflection, GL_NV_blend_square, GL_NV_occlusion_query,
    GL_SGI_color_matrix, GL_SGIS_texture_edge_clamp,
    GL_SGIS_texture_border_clamp, GL_SGIS_texture_lod,
    GL_SGIS_generate_mipmap, GL_SGIS_multitexture, GL_SUN_multi_draw_arrays
glu version: 1.3
glu extensions:
    GLU_EXT_nurbs_tessellator, GLU_EXT_object_space_tess
 
   visual  x  bf lv rg d st colorbuffer ax dp st accumbuffer  ms  cav
 id dep cl sp sz l  ci b ro  r  g  b  a bf th cl  r  g  b  a ns b eat
----------------------------------------------------------------------
0x23 24 tc  0 24  0 r  y  .  8  8  8  8  0 24  8 16 16 16 16  0 0 Slow
0x24 24 tc  0 24  0 r  .  .  8  8  8  8  0 24  8 16 16 16 16  0 0 Slow
0x25 24 tc  0 24  0 r  y  .  8  8  8  8  0 24  0 16 16 16 16  0 0 Slow
0x26 24 tc  0 24  0 r  .  .  8  8  8  8  0 24  0 16 16 16 16  0 0 Slow
0x27 24 tc  0 24  0 r  y  .  8  8  8  8  0 24  8  0  0  0  0  0 0 None
0x28 24 tc  0 24  0 r  .  .  8  8  8  8  0 24  8  0  0  0  0  0 0 None
0x29 24 tc  0 24  0 r  y  .  8  8  8  8  0 24  0  0  0  0  0  0 0 None
0x2a 24 tc  0 24  0 r  .  .  8  8  8  8  0 24  0  0  0  0  0  0 0 None
0x2b 24 dc  0 24  0 r  y  .  8  8  8  8  0 24  8 16 16 16 16  0 0 Slow
0x2c 24 dc  0 24  0 r  .  .  8  8  8  8  0 24  8 16 16 16 16  0 0 Slow
0x2d 24 dc  0 24  0 r  y  .  8  8  8  8  0 24  0 16 16 16 16  0 0 Slow
0x2e 24 dc  0 24  0 r  .  .  8  8  8  8  0 24  0 16 16 16 16  0 0 Slow
0x2f 24 dc  0 24  0 r  y  .  8  8  8  8  0 24  8  0  0  0  0  0 0 None
0x30 24 dc  0 24  0 r  .  .  8  8  8  8  0 24  8  0  0  0  0  0 0 None
0x31 24 dc  0 24  0 r  y  .  8  8  8  8  0 24  0  0  0  0  0  0 0 None
0x32 24 dc  0 24  0 r  .  .  8  8  8  8  0 24  0  0  0  0  0  0 0 None
 
real    0m6.483s
user    0m0.045s
sys     0m0.019s

Has anyone ran into problem like mine?

Loading of Enemy Territory takes ~10 seconds before I see anything.
I didn't have this problem when I was using 2.4.x kernels.

[yardbird]

@QuantumZ: Do you have some form of vertical refresh synchronization with monitor's refresh rate? If so disable it, it will keep your fps down as close to monitor refresh rate as possible.

[schoett]

Well, I managed to compile the 3.2.5 driver for kernel 2.6.0-test5, but I still get a completely black screen immediately when starting X.

The problem disappears with Option "UseInternalAGPGART" "yes"

This only works if there is no kernel agpgart module available (which otherwise gets loaded automatically and prevents the internal AGPgart from working).

glxgears with ATI Radeon 9600 TX, 1280x1024x(24 in 32), AGP 8x: 2393 FPS.

This does not work any more with 2.6.0-test6

Code: Select all

(II) fglrx(0): UMM area:     0xd8700000 (size=0x07900000)
(II) fglrx(0): driver needs XFree86 version: 4.3.x
(II) fglrx(0): detected XFree86 version: 4.3.0
(II) Loading extension ATIFGLRXDRI
(II) fglrx(0): doing DRIScreenInit
drmOpenDevice: minor is 0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is -1, (No such device)
drmOpenDevice: open result is -1, (No such device)
drmOpenDevice: Open failed
drmOpenDevice: minor is 0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is -1, (No such device)
drmOpenDevice: open result is -1, (No such device)
drmOpenDevice: Open failed
[drm] failed to load kernel module "agpgart"
drmOpenDevice: minor is 0
drmOpenDevice: node name is /dev/dri/card0
drmOpenDevice: open result is 6, (OK)
drmGetBusid returned ''
(II) fglrx(0): [drm] loaded kernel module for "fglrx" driver
(II) fglrx(0): [drm] created "fglrx" driver at busid "PCI:1:0:0"
(II) fglrx(0): [drm] added 8192 byte SAREA at 0xe0a6b000
(II) fglrx(0): [drm] mapped SAREA 0xe0a6b000 to 0x401e5000
(II) fglrx(0): [drm] framebuffer handle = 0xd8000000
(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: 3.2.5
(II) fglrx(0):     Date: Aug  6 2003
(II) fglrx(0):     Desc: ATI Fire GL 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:        2.6.0-test6
(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 = 0xe8420000
(II) fglrx(0): [agp] Mode=0x1f004e0b bridge: 0x1039/0x0648
(II) fglrx(0): [agp] AGP disable mask 0x00000000
(II) fglrx(0): [agp] enabling AGP with mode=0x1f004f0a
(EE) fglrx(0): [agp] Failed to set AGP mode!
(EE) fglrx(0): cannot init AGP
(II) fglrx(0): [drm] removed 1 reserved context for kernel
(II) fglrx(0): [drm] unmapping 8192 bytes of SAREA 0xe0a6b000 at 0x401e5000
(WW) fglrx(0): ***********************************************
(WW) fglrx(0): * DRI initialization failed!                  *
(WW) fglrx(0): * (maybe driver kernel module missing or bad) *
(WW) fglrx(0): * 2D acceleraton available (MMIO)             *
(WW) fglrx(0): * no 3D acceleration available                *
(WW) fglrx(0): ********************************************* *

There is some interesting "dmesg" output:

Code: Select all

fglrx: module license 'Proprietary. (C) 2002 - ATI Technologies, Starnberg, GERMANY' taints kernel.
[fglrx] Maximum main memory to use for locked dma buffers: 432 MBytes.
[fglrx] module loaded - fglrx 3.2.5 [Aug  6 2003] on minor 0
Fire GL built-in AGP-support
Based on agpgart interface v0.99 (c) Jeff Hartmann
agpgart: Maximum main memory to use for agp memory: 439M
agpgart: Detected SiS 645dx chipset
agpgart: AGP aperture is 128M @ 0xd0000000
Power management callback for AGP chipset installed
[fglrx] AGP detected, AgpState   = 0x1f004e0b (hardware caps of chipset)
AGP: Found 2 AGPv3 devices
AGP: Doing enable for AGPv3
agpgart: Found an AGP 3.0 compliant device. 
[fglrx:_r6x_CheckAGPCommand] *ERROR* query for AGP device capabilities failed
[fglrx:firegl_unlock] *ERROR* Process 3897 using kernel context 0

With "UseInternalAGPGART" "no", I get a complete system hang, as before. Guess I'll have to wait for October 8

[dpowers]

What a thread. I have a problem. I am using a Radeon 7000 with Xfree-4.3 and xfree-drm. I also have an nvidia card thrown in the mix for good measure. Thus, I have a 3 monitor display. Anyway, after upgrading to X4.3, I was getting garbled video in the radeon. After some tweaking in XF86Config (namely removing the line that related Screen 3 to the 2nd port of the radeon card in the ServerLayout section), I now have the nvidia monitor working and both radeon monitors working, EXCEPT the radeon monitors are mirrors of each other. So instead of 3 monitors, I only have 2. Thoughts?

[bushwakko]

I'm having problems compiling the ati-drivers under 2.6.0-test6-mm1 , anyone else have this, or a fix perhaps.

[Wedge_]

bushwakko: I think 3.2.5-r1 should fix the problems with -mm1, but I haven't tried it yet.

[Packetlos]

I also have the problem:

fglrx: module license 'Proprietary. (C) 2002 - ATI Technologies, Starnberg, GERMANY' taints kernel.
fglrx: Unknown symbol tsk_euid

I am using ati-drivers-3.2.5-r1 on Linux version 2.6.0-test6

[alshain]

I also have the problem:

fglrx: module license 'Proprietary. (C) 2002 - ATI Technologies, Starnberg, GERMANY' taints kernel.
fglrx: Unknown symbol tsk_euid

I am using ati-drivers-3.2.5-r1 on Linux version 2.6.0-test6

Me too - it was fine with mm1 but I'm using mm2 and the dreaded symbol error has returned.

Andrew (looking forward to the 8th)

[Wedge_]

If you're running 2.6.0-test6 (ie not test6-mm#), try commenting out this line in the ebuild:

Code: Select all

patch < ${FILESDIR}/3.2.5-linux-2.6.0-test6-mm.patch

It doesn't seem to be required if you aren't using an -mm kernel.

[Malcolm]

If you're running 2.6.0-test6 (ie not test6-mm#), try commenting out this line in the ebuild:

Code: Select all

patch < ${FILESDIR}/3.2.5-linux-2.6.0-test6-mm.patch

It doesn't seem to be required if you aren't using an -mm kernel.

This worked for me.

I'm also getting a QT error message when I tryed to compile the first time:

Code: Select all

make: *** [fireglcontrol.qt3.gcc3.2.3] Error 1
rm moc_Page.cpp moc_TVoutAdjustPage.cpp moc_AdjustPage.cpp moc_InfoPage.cpp moc_DualScreenPage. cpp moc_Sheet.cpp
 
!!! ERROR: media-video/ati-drivers-3.2.5-r1 failed.
!!! Function src_compile, Line 90, Exitcode 2
!!! (no error message)

After doing a USE="-qt" it emerged properly but anyone have an idea what this isn't compiling?

[Malcolm]

If you're running 2.6.0-test6 (ie not test6-mm#), try commenting out this line in the ebuild:

Code: Select all

patch < ${FILESDIR}/3.2.5-linux-2.6.0-test6-mm.patch

It doesn't seem to be required if you aren't using an -mm kernel.

This worked for me.

I'm also getting a QT error message when I tryed to compile the first time:

Code: Select all

make: *** [fireglcontrol.qt3.gcc3.2.3] Error 1
rm moc_Page.cpp moc_TVoutAdjustPage.cpp moc_AdjustPage.cpp moc_InfoPage.cpp moc_DualScreenPage. cpp moc_Sheet.cpp
 
!!! ERROR: media-video/ati-drivers-3.2.5-r1 failed.
!!! Function src_compile, Line 90, Exitcode 2
!!! (no error message)

After doing a USE="-qt" it emerged properly but anyone have an idea what this isn't compiling?

[Wedge_]

Post some more of the output, I think the actual error is missing.

[Malcolm]

Weird, when I tried to emerge it again it actually worked (:)).

Only problem I'm having now is under glxinfo and fglrxinfo I'm getting

Code: Select all

OpenGL vendor string: Mesa project: www.mesa3d.org
OpenGL renderer string: Mesa GLX Indirect
OpenGL version string: 1.3 Mesa 4.0.4

Kinda anoying. I get ~250FPS in glxgears and when I try to do a fgl_glxgears I get

Code: Select all

Error: couldn't get fbconfig

I have a Rad 9500 128mb, ran opengl-update ati, donno why its not workin :S