quick_and_dirty_2d pushed to master

newer
xf86-video-radeonhd:CS: 3 commit(s)

older
Re: [radeonhd] quick_and_dirty_2d...

Alex Deucher

25 Jul 2008 25 Jul '08

22:56

Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you. Enjoy! Alex -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org

Show replies by date

Vladimir Grebenschikov

28 Jul 28 Jul

09:43

On Fri, 2008-07-25 at 18:56 -0400, Alex Deucher wrote:

...

Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you.

I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image. Like attached picture. Is there any cure for that problem ? (Gnome developers said that problem is due to bad X11 drivers) 3D and 2D acceleration works well for me. (stellarium, torcs, mplayer -vo xv) But google-earth (on FreeBSD under Linux emulation) starts in software GL emulation mode, still no cure ? % googleearth unknown chip id 0x7145, can't guess. libGL warning: 3D driver returned no fbconfigs. libGL error: InitDriver failed libGL error: reverting to (slow) indirect rendering ... % grep RADEONHD /var/log/Xorg.0.log (II) RADEONHD: X driver for the following AMD GPG (ATI) graphics devices: (II) RADEONHD: version 1.2.1, built from git branch quick_and_dirty_2d, commit cdff6eb3 (**) RADEONHD(0): Depth 24, (--) framebuffer bpp 32 (**) RADEONHD(0): Option "SWcursor" (**) RADEONHD(0): Option "RROutputOrder" "PANEL" (**) RADEONHD(0): Option "DRI" (**) RADEONHD(0): Selected XAA 2D acceleration. (--) RADEONHD(0): Detected an M54 on a Lenovo Thinkpad T60 (2007) (==) RADEONHD(0): Write-combining range (0xee100000,0x10000) was already clear (II) RADEONHD(0): Mapped IO at 0x28653000 (size 0x00010000) (II) RADEONHD(0): Getting BIOS copy from legacy VBIOS location (II) RADEONHD(0): ATOM BIOS Rom: (II) RADEONHD(0): Analog TV Default Mode: 135043410 (--) RADEONHD(0): VideoRAM: 131072 kByte (II) RADEONHD(0): Framebuffer space used by Firmware (kb): 20 (II) RADEONHD(0): Start of VRAM area used by Firmware: 0x7ffb000 (II) RADEONHD(0): AtomBIOS requests 20kB of VRAM scratch space (II) RADEONHD(0): AtomBIOS VRAM scratch base: 0x7ffb000 (II) RADEONHD(0): Default Engine Clock: 452000 (II) RADEONHD(0): Default Memory Clock: 352000 (II) RADEONHD(0): Maximum Pixel ClockPLL Frequency Output: 1100000 (II) RADEONHD(0): Minimum Pixel ClockPLL Frequency Output: 0 (II) RADEONHD(0): Maximum Pixel ClockPLL Frequency Input: 13500 (II) RADEONHD(0): Minimum Pixel ClockPLL Frequency Input: 1000 (II) RADEONHD(0): Maximum Pixel Clock: 400000 (II) RADEONHD(0): Reference Clock: 27000 (II) RADEONHD(0): Found libdri 5.4.0. (II) RADEONHD(0): Found libdrm 1.3.0. (II) RADEONHD(0): Found radeon drm 1.29.0. (II) RADEONHD(0): Default Engine Clock: 452000 (II) RADEONHD(0): I2C bus "RHD I2C line 0" initialized. (II) RADEONHD(0): I2C bus "RHD I2C line 1" initialized. (II) RADEONHD(0): I2C bus "RHD I2C line 2" initialized. (II) RADEONHD(0): Detected VGA mode. (II) RADEONHD(0): Minimum Pixel ClockPLL Frequency Output: 0 (II) RADEONHD(0): Maximum Pixel ClockPLL Frequency Output: 1100000 (II) RADEONHD(0): Maximum Pixel Clock: 400000 (II) RADEONHD(0): Reference Clock: 27000 (II) RADEONHD(0): FB: Allocated Cursor Image at offset 0x00000000 (size = 0x00004000) (II) RADEONHD(0): FB: Allocated Cursor Image at offset 0x00004000 (size = 0x00004000) (II) RADEONHD(0): Connector[0] {RHD_CONNECTOR_VGA, "VGA CRT1", RHD_DDC_0, RHD_HPD_NONE, { RHD_OUTPUT_DACA, RHD_OUTPUT_NONE } } (II) RADEONHD(0): Connector[1] {RHD_CONNECTOR_PANEL, "PANEL LCD1", RHD_DDC_2, RHD_HPD_NONE, { RHD_OUTPUT_LVTMA, RHD_OUTPUT_NONE } } (II) RADEONHD(0): Connector[2] {RHD_CONNECTOR_DVI, "DVI-I DFP1", RHD_DDC_1, RHD_HPD_0, { RHD_OUTPUT_TMDSA, RHD_OUTPUT_NONE } } (--) RADEONHD(0): Attaching Output DAC A to Connector VGA 1 (II) RADEONHD(0): LVDS SEQ Dig onto DE: 40 (II) RADEONHD(0): LVDS SEQ DE to BL: 250 (II) RADEONHD(0): LVDS Off Delay: 170 (II) RADEONHD(0): LVDS Duallink: 0x1 (II) RADEONHD(0): LVDS 24Bit: 0x0 (II) RADEONHD(0): LVDS FPDI: 0x0 (II) RADEONHD(0): LVDS Temporal Dither : 0x40 (II) RADEONHD(0): LVDS Spatial Dither : 0x0 (II) RADEONHD(0): LVDS Grey Level: 0x2 (II) RADEONHD(0): AtomBIOS returned 2 Grey Levels (--) RADEONHD(0): Detected a 18bit dual link panel. (--) RADEONHD(0): Attaching Output LVDS to Connector PANEL (--) RADEONHD(0): Attaching Output TMDS A to Connector DVI-D 1 (II) RADEONHD(0): RandR: Adding RRoutput VGA_1 for Output DAC A (II) RADEONHD(0): RandR: Adding RRoutput PANEL for Output LVDS (II) RADEONHD(0): RandR: Adding RRoutput DVI-D_1 for Output TMDS A (II) RADEONHD(0): RandR: Reordering output PANEL .... P.S.: I have recent FreeBSD 8-CURRENT with drm and mesa build from git.

...

Enjoy!

Alex

Alex Deucher

13:59

On Mon, Jul 28, 2008 at 5:43 AM, Vladimir Grebenschikov wrote:

...

On Fri, 2008-07-25 at 18:56 -0400, Alex Deucher wrote:

...
Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you.

I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image.

please try with master. the quick_and_dirty_2d should be considered dead at this point.

...

Like attached picture.

Are you using XAA or EXA?

...

Is there any cure for that problem ? (Gnome developers said that problem is due to bad X11 drivers)

3D and 2D acceleration works well for me. (stellarium, torcs, mplayer -vo xv)

But google-earth (on FreeBSD under Linux emulation) starts in software GL emulation mode, still no cure ?

% googleearth unknown chip id 0x7145, can't guess. libGL warning: 3D driver returned no fbconfigs. libGL error: InitDriver failed libGL error: reverting to (slow) indirect rendering ...

It looks like google earth is probably picking up an old mesa driver are you sure you have mesa installed properly. What does glxinfo show?

...

% grep RADEONHD /var/log/Xorg.0.log (II) RADEONHD: X driver for the following AMD GPG (ATI) graphics devices: (II) RADEONHD: version 1.2.1, built from git branch quick_and_dirty_2d, commit cdff6eb3 (**) RADEONHD(0): Depth 24, (--) framebuffer bpp 32 (**) RADEONHD(0): Option "SWcursor" (**) RADEONHD(0): Option "RROutputOrder" "PANEL" (**) RADEONHD(0): Option "DRI" (**) RADEONHD(0): Selected XAA 2D acceleration. (--) RADEONHD(0): Detected an M54 on a Lenovo Thinkpad T60 (2007) (==) RADEONHD(0): Write-combining range (0xee100000,0x10000) was already clear (II) RADEONHD(0): Mapped IO at 0x28653000 (size 0x00010000) (II) RADEONHD(0): Getting BIOS copy from legacy VBIOS location (II) RADEONHD(0): ATOM BIOS Rom: (II) RADEONHD(0): Analog TV Default Mode: 135043410 (--) RADEONHD(0): VideoRAM: 131072 kByte (II) RADEONHD(0): Framebuffer space used by Firmware (kb): 20 (II) RADEONHD(0): Start of VRAM area used by Firmware: 0x7ffb000 (II) RADEONHD(0): AtomBIOS requests 20kB of VRAM scratch space (II) RADEONHD(0): AtomBIOS VRAM scratch base: 0x7ffb000 (II) RADEONHD(0): Default Engine Clock: 452000 (II) RADEONHD(0): Default Memory Clock: 352000 (II) RADEONHD(0): Maximum Pixel ClockPLL Frequency Output: 1100000 (II) RADEONHD(0): Minimum Pixel ClockPLL Frequency Output: 0 (II) RADEONHD(0): Maximum Pixel ClockPLL Frequency Input: 13500 (II) RADEONHD(0): Minimum Pixel ClockPLL Frequency Input: 1000 (II) RADEONHD(0): Maximum Pixel Clock: 400000 (II) RADEONHD(0): Reference Clock: 27000 (II) RADEONHD(0): Found libdri 5.4.0. (II) RADEONHD(0): Found libdrm 1.3.0. (II) RADEONHD(0): Found radeon drm 1.29.0. (II) RADEONHD(0): Default Engine Clock: 452000 (II) RADEONHD(0): I2C bus "RHD I2C line 0" initialized. (II) RADEONHD(0): I2C bus "RHD I2C line 1" initialized. (II) RADEONHD(0): I2C bus "RHD I2C line 2" initialized. (II) RADEONHD(0): Detected VGA mode. (II) RADEONHD(0): Minimum Pixel ClockPLL Frequency Output: 0 (II) RADEONHD(0): Maximum Pixel ClockPLL Frequency Output: 1100000 (II) RADEONHD(0): Maximum Pixel Clock: 400000 (II) RADEONHD(0): Reference Clock: 27000 (II) RADEONHD(0): FB: Allocated Cursor Image at offset 0x00000000 (size = 0x00004000) (II) RADEONHD(0): FB: Allocated Cursor Image at offset 0x00004000 (size = 0x00004000) (II) RADEONHD(0): Connector[0] {RHD_CONNECTOR_VGA, "VGA CRT1", RHD_DDC_0, RHD_HPD_NONE, { RHD_OUTPUT_DACA, RHD_OUTPUT_NONE } } (II) RADEONHD(0): Connector[1] {RHD_CONNECTOR_PANEL, "PANEL LCD1", RHD_DDC_2, RHD_HPD_NONE, { RHD_OUTPUT_LVTMA, RHD_OUTPUT_NONE } } (II) RADEONHD(0): Connector[2] {RHD_CONNECTOR_DVI, "DVI-I DFP1", RHD_DDC_1, RHD_HPD_0, { RHD_OUTPUT_TMDSA, RHD_OUTPUT_NONE } } (--) RADEONHD(0): Attaching Output DAC A to Connector VGA 1 (II) RADEONHD(0): LVDS SEQ Dig onto DE: 40 (II) RADEONHD(0): LVDS SEQ DE to BL: 250 (II) RADEONHD(0): LVDS Off Delay: 170 (II) RADEONHD(0): LVDS Duallink: 0x1 (II) RADEONHD(0): LVDS 24Bit: 0x0 (II) RADEONHD(0): LVDS FPDI: 0x0 (II) RADEONHD(0): LVDS Temporal Dither : 0x40 (II) RADEONHD(0): LVDS Spatial Dither : 0x0 (II) RADEONHD(0): LVDS Grey Level: 0x2 (II) RADEONHD(0): AtomBIOS returned 2 Grey Levels (--) RADEONHD(0): Detected a 18bit dual link panel. (--) RADEONHD(0): Attaching Output LVDS to Connector PANEL (--) RADEONHD(0): Attaching Output TMDS A to Connector DVI-D 1 (II) RADEONHD(0): RandR: Adding RRoutput VGA_1 for Output DAC A (II) RADEONHD(0): RandR: Adding RRoutput PANEL for Output LVDS (II) RADEONHD(0): RandR: Adding RRoutput DVI-D_1 for Output TMDS A (II) RADEONHD(0): RandR: Reordering output PANEL ....

Can you attach the full log? Alex -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org

Vladimir Grebenschikov

15:58

On Mon, 2008-07-28 at 09:59 -0400, Alex Deucher wrote:

...

On Mon, Jul 28, 2008 at 5:43 AM, Vladimir Grebenschikov wrote:

...
On Fri, 2008-07-25 at 18:56 -0400, Alex Deucher wrote:

...
Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you.

I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image.

please try with master. the quick_and_dirty_2d should be considered dead at this point.

Upgraded, the same.

...

...
Like attached picture.

Are you using XAA or EXA?

No special options in xorg.conf, % egrep XAA\|EXA /var/log/Xorg.0.log (**) RADEONHD(0): Selected XAA 2D acceleration. (II) RADEONHD(0): Using accelerated EXA DownloadFromScreen hook; GART location = 0xe0000000 (II) RADEONHD(0): Using XFree86 Acceleration Architecture (XAA) % Looks like XAA.

...

...
Is there any cure for that problem ? (Gnome developers said that problem is due to bad X11 drivers)

3D and 2D acceleration works well for me. (stellarium, torcs, mplayer -vo xv)

But google-earth (on FreeBSD under Linux emulation) starts in software GL emulation mode, still no cure ?

% googleearth unknown chip id 0x7145, can't guess. libGL warning: 3D driver returned no fbconfigs. libGL error: InitDriver failed libGL error: reverting to (slow) indirect rendering ...

It looks like google earth is probably picking up an old mesa driver are you sure you have mesa installed properly. What does glxinfo show?

% src/mesa/progs/demos/glinfo Warning, RV515 detected, 3D HAHAHAHAHA!!. GL_VERSION: 1.3 Mesa 7.1 GL_EXTENSIONS: GL_ARB_depth_texture GL_ARB_fragment_program GL_ARB_imaging GL_ARB_multisample GL_ARB_multitexture 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_MESAX_texture_float GL_ARB_texture_mirrored_repeat GL_ARB_texture_rectangle GL_ARB_transpose_matrix GL_ARB_vertex_buffer_object GL_ARB_vertex_program GL_ARB_window_pos GL_EXT_abgr GL_EXT_bgra GL_EXT_blend_color GL_EXT_blend_equation_separate GL_EXT_blend_func_separate GL_EXT_blend_logic_op GL_EXT_blend_minmax GL_EXT_blend_subtract GL_EXT_clip_volume_hint GL_EXT_compiled_vertex_array GL_EXT_convolution GL_EXT_copy_texture GL_EXT_draw_range_elements GL_EXT_gpu_program_parameters GL_EXT_histogram GL_EXT_multi_draw_arrays GL_EXT_packed_pixels GL_EXT_polygon_offset GL_EXT_rescale_normal GL_EXT_secondary_color GL_EXT_separate_specular_color GL_EXT_stencil_two_side GL_EXT_stencil_wrap GL_EXT_subtexture GL_EXT_texture GL_EXT_texture3D GL_EXT_texture_edge_clamp GL_EXT_texture_env_add GL_EXT_texture_env_combine GL_EXT_texture_env_dot3 GL_EXT_texture_filter_anisotropic GL_EXT_texture_lod_bias GL_EXT_texture_mirror_clamp GL_EXT_texture_object GL_EXT_texture_rectangle GL_EXT_vertex_array GL_APPLE_packed_pixels GL_ATI_blend_equation_separate GL_ATI_texture_env_combine3 GL_ATI_texture_mirror_once GL_IBM_rasterpos_clip GL_IBM_texture_mirrored_repeat GL_INGR_blend_func_separate GL_MESA_pack_invert GL_MESA_ycbcr_texture GL_MESA_window_pos GL_NV_blend_square GL_NV_light_max_exponent GL_NV_texture_rectangle GL_NV_texgen_reflection GL_NV_vertex_program GL_OES_read_format GL_SGI_color_matrix GL_SGI_color_table GL_SGIS_generate_mipmap GL_SGIS_texture_border_clamp GL_SGIS_texture_edge_clamp GL_SGIS_texture_lod GL_SGIX_depth_texture GL_SUN_multi_draw_arrays GL_RENDERER: Mesa DRI R300 20060815 x86/MMX/SSE2 TCL GL_VENDOR: DRI R300 Project GLU_VERSION: 1.3 GLU_EXTENSIONS: GLU_EXT_nurbs_tessellator GLU_EXT_object_space_tess GLUT_API_VERSION: 5 GLUT_XLIB_IMPLEMENTATION: 15 % Most probably Linux's (I am on FreeBSD) mesa libraries are not installed properly Linux's glxinfo said the same as googleearth: % /compat/linux/usr/X11R6/bin/glxinfo unknown chip id 0x7145, can't guess. libGL warning: 3D driver returned no fbconfigs. libGL error: InitDriver failed libGL error: reverting to (slow) indirect rendering display: :0.0 screen:0 direct rendering: No server glx vendor string: SGI server glx version string: 1.2 server glx extensions: GLX_ARB_multisample, GLX_EXT_visual_info, GLX_EXT_visual_rating, GLX_EXT_import_context, GLX_EXT_texture_from_pixmap, GLX_OML_swap_method, GLX_SGI_make_current_read, GLX_SGIS_multisample, GLX_SGIX_hyperpipe, GLX_SGIX_swap_barrier, GLX_SGIX_fbconfig, GLX_MESA_copy_sub_buffer client glx vendor string: SGI client glx version string: 1.4 ... % pkg_info -I linux_dri\* linux_dri-7.0 Binary Linux DRI libraries for 3D hardware acceleration of %

...

Can you attach the full log?

attached One more problem - X cursor When I am using Option "SWcursor" Everything works well on non-GL applications, but for GL applications (like stellarium) there visible glitches around cursor. Without SWcursor, X cursor starts glitching on usual use-cases, like moving cross cursor type boundaries.

...

Alex -- Vladimir B. Grebenschikov vova@fbsd.ru

Alex Deucher

16:21

On Mon, Jul 28, 2008 at 11:58 AM, Vladimir Grebenschikov wrote:

...

On Mon, 2008-07-28 at 09:59 -0400, Alex Deucher wrote:

...
On Mon, Jul 28, 2008 at 5:43 AM, Vladimir Grebenschikov wrote:

...
On Fri, 2008-07-25 at 18:56 -0400, Alex Deucher wrote:

...
Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you.

I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image.

please try with master. the quick_and_dirty_2d should be considered dead at this point.

Upgraded, the same.

...
...
Like attached picture.

Are you using XAA or EXA?

No special options in xorg.conf, % egrep XAA\|EXA /var/log/Xorg.0.log (**) RADEONHD(0): Selected XAA 2D acceleration. (II) RADEONHD(0): Using accelerated EXA DownloadFromScreen hook; GART location = 0xe0000000 (II) RADEONHD(0): Using XFree86 Acceleration Architecture (XAA) %

Looks like XAA.

Does it help if you switch to EXA? Option "AccelMethod" "EXA"

...

...
...
Is there any cure for that problem ? (Gnome developers said that problem is due to bad X11 drivers)

3D and 2D acceleration works well for me. (stellarium, torcs, mplayer -vo xv)

But google-earth (on FreeBSD under Linux emulation) starts in software GL emulation mode, still no cure ?

% googleearth unknown chip id 0x7145, can't guess. libGL warning: 3D driver returned no fbconfigs. libGL error: InitDriver failed libGL error: reverting to (slow) indirect rendering ...

It looks like google earth is probably picking up an old mesa driver are you sure you have mesa installed properly. What does glxinfo show?

% src/mesa/progs/demos/glinfo Warning, RV515 detected, 3D HAHAHAHAHA!!. GL_VERSION: 1.3 Mesa 7.1 GL_EXTENSIONS: GL_ARB_depth_texture GL_ARB_fragment_program GL_ARB_imaging GL_ARB_multisample GL_ARB_multitexture 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_MESAX_texture_float GL_ARB_texture_mirrored_repeat GL_ARB_texture_rectangle GL_ARB_transpose_matrix GL_ARB_vertex_buffer_object GL_ARB_vertex_program GL_ARB_window_pos GL_EXT_abgr GL_EXT_bgra GL_EXT_blend_color GL_EXT_blend_equation_separate GL_EXT_blend_func_separate GL_EXT_blend_logic_op GL_EXT_blend_minmax GL_EXT_blend_subtract GL_EXT_clip_volume_hint GL_EXT_compiled_vertex_array GL_EXT_convolution GL_EXT_copy_texture GL_EXT_draw_range_elements GL_EXT_gpu_program_parameters GL_EXT_histogram GL_EXT_multi_draw_arrays GL_EXT_packed_pixels GL_EXT_polygon_offset GL_EXT_rescale_normal GL_EXT_secondary_color GL_EXT_separate_specular_color GL_EXT_stencil_two_side GL_EXT_stencil_wrap GL_EXT_subtexture GL_EXT_texture GL_EXT_texture3D GL_EXT_texture_edge_clamp GL_EXT_texture_env_add GL_EXT_texture_env_combine GL_EXT_texture_env_dot3 GL_EXT_texture_filter_anisotropic GL_EXT_texture_lod_bias GL_EXT_texture_mirror_clamp GL_EXT_texture_object GL_EXT_texture_rectangle GL_EXT_vertex_array GL_APPLE_packed_pixels GL_ATI_blend_equation_separate GL_ATI_texture_env_combine3 GL_ATI_texture_mirror_once GL_IBM_rasterpos_clip GL_IBM_texture_mirrored_repeat GL_INGR_blend_func_separate GL_MESA_pack_invert GL_MESA_ycbcr_texture GL_MESA_window_pos GL_NV_blend_square GL_NV_light_max_exponent GL_NV_texture_rectangle GL_NV_texgen_reflection GL_NV_vertex_program GL_OES_read_format GL_SGI_color_matrix GL_SGI_color_table GL_SGIS_generate_mipmap GL_SGIS_texture_border_clamp GL_SGIS_texture_edge_clamp GL_SGIS_texture_lod GL_SGIX_depth_texture GL_SUN_multi_draw_arrays GL_RENDERER: Mesa DRI R300 20060815 x86/MMX/SSE2 TCL GL_VENDOR: DRI R300 Project GLU_VERSION: 1.3 GLU_EXTENSIONS: GLU_EXT_nurbs_tessellator GLU_EXT_object_space_tess GLUT_API_VERSION: 5 GLUT_XLIB_IMPLEMENTATION: 15 %

Most probably Linux's (I am on FreeBSD) mesa libraries are not installed properly

Linux's glxinfo said the same as googleearth: % /compat/linux/usr/X11R6/bin/glxinfo unknown chip id 0x7145, can't guess.

It's picking up an old lib.

...

% pkg_info -I linux_dri\* linux_dri-7.0 Binary Linux DRI libraries for 3D hardware acceleration of %

probably that package contains a old lib, but I'm not really familiar with how the linux emulation works.

...

...
Can you attach the full log?

attached

One more problem - X cursor

When I am using Option "SWcursor"

Everything works well on non-GL applications, but for GL applications (like stellarium) there visible glitches around cursor.

that's do to sw updating the cursor mixed with 3D rendering. you really want to use hw cursors when using 3D.

...

Without SWcursor, X cursor starts glitching on usual use-cases, like moving cross cursor type boundaries.

Perhaps there's an issue with disabling/enabling the cursors when crossing the screen boundaries. Do you have similar problems with radeon? radeonhd handles the cursor enable/disable directly while radeon lets xrandr handle it. Alex -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org

Vladimir Grebenschikov

17:19

On Mon, 2008-07-28 at 12:21 -0400, Alex Deucher wrote:

...

On Mon, Jul 28, 2008 at 11:58 AM, Vladimir Grebenschikov wrote:

...
On Mon, 2008-07-28 at 09:59 -0400, Alex Deucher wrote:

...
On Mon, Jul 28, 2008 at 5:43 AM, Vladimir Grebenschikov wrote:

...
On Fri, 2008-07-25 at 18:56 -0400, Alex Deucher wrote:

...
Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you.

I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image.

please try with master. the quick_and_dirty_2d should be considered dead at this point.

Upgraded, the same.

...
...
Like attached picture.

Are you using XAA or EXA?

No special options in xorg.conf, % egrep XAA\|EXA /var/log/Xorg.0.log (**) RADEONHD(0): Selected XAA 2D acceleration. (II) RADEONHD(0): Using accelerated EXA DownloadFromScreen hook; GART location = 0xe0000000 (II) RADEONHD(0): Using XFree86 Acceleration Architecture (XAA) %

Looks like XAA.

Does it help if you switch to EXA? Option "AccelMethod" "EXA"

Yes, it helps, now page shown correctly, but, ff3 rotating icons (while page loading on tab header) are a bit broken in bottom part, including icon background.

...

...
% pkg_info -I linux_dri\* linux_dri-7.0 Binary Linux DRI libraries for 3D hardware acceleration of %

probably that package contains a old lib, but I'm not really familiar with how the linux emulation works.

Looks like I need to build linux mesa from git, looks tricky on FreeBSD.

...

...
When I am using Option "SWcursor"

Everything works well on non-GL applications, but for GL applications (like stellarium) there visible glitches around cursor.

that's do to sw updating the cursor mixed with 3D rendering. you really want to use hw cursors when using 3D.

...
Without SWcursor, X cursor starts glitching on usual use-cases, like moving cross cursor type boundaries.

Perhaps there's an issue with disabling/enabling the cursors when crossing the screen boundaries. Do you have similar problems with radeon? radeonhd handles the cursor enable/disable directly while radeon lets xrandr handle it.

Did not try with radeon, but there is no such problem with vesa.

...

Alex -- Vladimir B. Grebenschikov vova@fbsd.ru -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org

Marc Dietrich

15:44

Hi, Am Montag 28 Juli 2008 11:43:46 schrieb Vladimir Grebenschikov:

...

On Fri, 2008-07-25 at 18:56 -0400, Alex Deucher wrote:

...
Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you.

I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image.

similar here after starting an opengl app and moving the window. See attached screen shot. I'm using EXA. It does not happen with XAA, but with it, opengl apps are messed up (maybe another problem). I'm using Xserver 1.5, latest mesa/drm from git. Greetings Marc

Alex Deucher

16:02

On Mon, Jul 28, 2008 at 11:44 AM, Marc Dietrich wrote:

...

Hi,

Am Montag 28 Juli 2008 11:43:46 schrieb Vladimir Grebenschikov:

...
On Fri, 2008-07-25 at 18:56 -0400, Alex Deucher wrote:

...
Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you.

I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image.

similar here after starting an opengl app and moving the window. See attached screen shot.

I'm using EXA. It does not happen with XAA, but with it, opengl apps are messed up (maybe another problem). I'm using Xserver 1.5, latest mesa/drm from git.

Can you attach your xorg log? Alex -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org

Marc

29 Jul 29 Jul

08:03

Hi, Am Montag 28 Juli 2008 18:02:46 schrieben Sie:

...

On Mon, Jul 28, 2008 at 11:44 AM, Marc Dietrich

wrote:

...
Hi,

Am Montag 28 Juli 2008 11:43:46 schrieb Vladimir Grebenschikov:

...
On Fri, 2008-07-25 at 18:56 -0400, Alex Deucher wrote:

...
Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you.

I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image.

similar here after starting an opengl app and moving the window. See attached screen shot.

I'm using EXA. It does not happen with XAA, but with it, opengl apps are messed up (maybe another problem). I'm using Xserver 1.5, latest mesa/drm from git.

Can you attach your xorg log?

attached. Sorry that I forgot to mention, that I use a RS690 on openSUSE 11.0/x86_64 with xorg from the X11:Xorg repo. Btw. there was a patch to make radeonhd/radeon driver work with xserver 1.4 on 64-bit machines, but I couldn't find it anymore. Marc -- "Artists and intellectuals support our cause." Lord Arthur Ponsonby, "Falsehood in Wartime: Propaganda Lies of the First World War", 1928

Alex Deucher

13:55

On Tue, Jul 29, 2008 at 4:03 AM, Marc wrote:

...

Hi,

Am Montag 28 Juli 2008 18:02:46 schrieben Sie:

...
On Mon, Jul 28, 2008 at 11:44 AM, Marc Dietrich

wrote:

...
Hi,

Am Montag 28 Juli 2008 11:43:46 schrieb Vladimir Grebenschikov:

...
On Fri, 2008-07-25 at 18:56 -0400, Alex Deucher wrote:

...
Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you.

I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image.

similar here after starting an opengl app and moving the window. See attached screen shot.

I'm using EXA. It does not happen with XAA, but with it, opengl apps are messed up (maybe another problem). I'm using Xserver 1.5, latest mesa/drm from git.

Can you attach your xorg log?

attached.

Sorry that I forgot to mention, that I use a RS690 on openSUSE 11.0/x86_64 with xorg from the X11:Xorg repo.

Ah RS690. Does this options help: Option "UnverifiedFeatures" "False" Several others have reported that this fixes corruption problems on their RS690 systems.

...

Btw. there was a patch to make radeonhd/radeon driver work with xserver 1.4 on 64-bit machines, but I couldn't find it anymore.

radeon should work fine on 64-bit with xserver 1.4. I was running that for quite a while. I'm not sure about radeonhd. Alex -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org

Marc Dietrich

14:17

Hi,

...

...
...
...
...
I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image.

similar here after starting an opengl app and moving the window. See attached screen shot.

I'm using EXA. It does not happen with XAA, but with it, opengl apps are messed up (maybe another problem). I'm using Xserver 1.5, latest mesa/drm from git.

Can you attach your xorg log?

attached.

Sorry that I forgot to mention, that I use a RS690 on openSUSE 11.0/x86_64 with xorg from the X11:Xorg repo.

Ah RS690. Does this options help: Option "UnverifiedFeatures" "False"

already have this.

...

Several others have reported that this fixes corruption problems on their RS690 systems.

...
Btw. there was a patch to make radeonhd/radeon driver work with xserver 1.4 on 64-bit machines, but I couldn't find it anymore.

radeon should work fine on 64-bit with xserver 1.4. I was running that for quite a while. I'm not sure about radeonhd.

I meant this one: https://bugs.freedesktop.org/show_bug.cgi?id=13358 Marc -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org

Alex Deucher

28 Jul 28 Jul

16:21

On Mon, Jul 28, 2008 at 11:44 AM, Marc Dietrich wrote:

...

Hi,

Am Montag 28 Juli 2008 11:43:46 schrieb Vladimir Grebenschikov:

...
On Fri, 2008-07-25 at 18:56 -0400, Alex Deucher wrote:

...
Just a heads up everyone, I've gone ahead and pushed the quick_and_dirty_2d branch to master, so please test and let us know how it how well it works for you.

I am using quick_and_dirty_2d branch with firefox3 and get "magic" messed results when pages has some background image.

similar here after starting an opengl app and moving the window. See attached screen shot.

I'm using EXA. It does not happen with XAA, but with it, opengl apps are messed up (maybe another problem). I'm using Xserver 1.5, latest mesa/drm from git.

Are you on Linux or *BSD? Alex -- To unsubscribe, e-mail: radeonhd+unsubscribe@opensuse.org For additional commands, e-mail: radeonhd+help@opensuse.org

5760

Age (days ago)

5764

Last active (days ago)

List overview

Download

11 comments

4 participants

participants (4)

Alex Deucher
Marc
Marc Dietrich
Vladimir Grebenschikov