Re: [R300] Xorg cvs problem
On Mon, 2004-12-27 at 12:09 -0500, Keith Conger wrote: > > I guess > > > > Option "XaaNoScanlineCPUToScreenColorExpandFill" > > > > works around this? Does this work correctly with the DRI disabled? > > Nope that option doesn't fix mozilla or the Xscreensaver dialog. Not even together with "XaaNoScanlineImageWriteRect" ? > But disabling dri fixes everything. So it definitely sounds like the HOST_DATA_SWAP bits don't have any effect via the CP. :( -- Earthling Michel DÃnzer | Debian (powerpc), X and DRI developer Libre software enthusiast| http://svcs.affero.net/rm.php?r=daenzer --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
R300 status
Just an update on what has been happening with R300 CVS: r300_demo: * several new test modes are available: + --vb-triangles - paint triangles using vertex buffer [WORKS] + --immd-triangles - paint triangles using immediate data [BROKEN] (paints but locks up) + --idx-triangles - paint triangles using data in vertex buffer with vertices specified by embedded indices [WORKS] * changed r300_lib.[c,h] files to be more portable and ease inclusion in other code. r300_driver/drm: * two new commands: + end32 - for stabilizing engine after 3d rendering (to prevent lockups) + packet3_raw - for passing several packet3's directly to the engine r300_driver/r300 * wrote glue to include r300_lib.[c,h] into the driver code * added code to paint QUAD primitives with flat colors. Pretty basic, but no lockups :) Note: no transform is done yet so glxgears displays static (non-moving) gear teeth. best Vladimir Dergachev --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Accessing AGP
Hi all, Can someone more knowledgable explain to me how to properly access AGP space from within Mesa driver ? What I want to do is to copy vertex data (i.e. coordinates, colors, etc) into an array in AGP space so that R300 engine can fetch it from there. Ideally, I would be able to tell Mesa to just store them there to begin with. I looked at R200 MAOS code and I do not understand it - it looks like it does not use GART memory for vertex buffers, but instead users indirect buffers. thank you Vladimir Dergachev --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
[Bug 1859] local variable used before set
Please do not reply to this email: if you want to comment on the bug, go to the URL shown below and enter yourcomments there. https://bugs.freedesktop.org/show_bug.cgi?id=1859 [EMAIL PROTECTED] changed: What|Removed |Added Status|NEW |RESOLVED Resolution||FIXED --- Additional Comments From [EMAIL PROTECTED] 2004-12-27 13:53 --- check_color ignores the size parameter anyway, so this is safe to just init to zero to hush the warning. applied to CVS, thanks. -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [R300] Xorg cvs problem
> I guess > > Option "XaaNoScanlineCPUToScreenColorExpandFill" > > works around this? Does this work correctly with the DRI disabled? Nope that option doesn't fix mozilla or the Xscreensaver dialog. But disabling dri fixes everything. So now I have dri disabled and have removed the above option. > > No big surprise at this point, but thanks for testing. > > Option "DMAForXv" "off" > > should work around that. This does fix the discoloration. -- Keith Conger [EMAIL PROTECTED] http://pimpstation.org --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Weekly IRC meeting reminder
This is just a friendly reminder that the weekly dri-devel IRC meeting will be starting in the #dri-devel channel on irc.freenode.net at 2100 UTC (or 4:00PM EST or 1:00PM PST, if you prefer). Time zone conversion available at: http://www.timezoneconverter.com/cgi-bin/tzc.tzc Logs of previous IRC meetings are available at: http://dri.sourceforge.net/IRC-logs/ --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [R300] Xorg cvs problem
1. In RADEONSetupForScanlineImageWrite dp_gui_master_cntl_clip does not depend on endianness unlike other similar functions.. Is it right ? I tested it with all possible value do not see anychange :( 2. It appears that PIX_ORDER field in gui_cntl is *not* used by microcode. In fact, I can't find DP_GUI_MASTER anywhere in R300 docs. It seems that PIX_ORDER & LSB_TO_MSB are duplicated constant. # define RADEON_GMC_BYTE_PIX_ORDER (1<< 14) # define RADEON_GMC_BYTE_MSB_TO_LSB (0<< 14) # define RADEON_GMC_BYTE_LSB_TO_MSB (1<< 14) Is there any register with similarities with DP_GUI_MASTER in your doc ? Or any which deal with the endianness So, perhaps, R300 hardware would use the same endiannes settings for both MMIO and ACCEL modes ? Tested too but no change. I will try to test again because with all change i made in all place i am no more sure of what i do. My code is becoming a real mess :) Thx for your comments Btw i still do not find a way to make r300_demo works but i am working on it. I think that this maybe related to the problem we have here. best, Jerome Glisse --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [R300] Xorg cvs problem
Michel DÃnzer wrote: On Mon, 2004-12-27 at 09:26 -0500, Keith Conger wrote: Ok I changed to 24 can't remember why I bumped it down to 16 in the first place. Also comment out the Option you had me add the artifacts do not show. Hmm, sounds like the host data byte swapping doesn't work via the CP for some reason... I fear this too after lot of test whatever value i wrote to RADEON_RBBM_GUICNTL or RADEON_DP_GUI_MASTER_CNTL I got the screen corrupted in the same way, i think i should have see some change... There must be some magic incantation to do in some strange place to change the situation. Maybe we should ask ati... They should help us to correct only a stupid byte swapping problem :) Anyway i will browse darwin source see if there is radeon code somewhere available with the magic formula :) best, Jerome Glisse --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [R300] Xorg cvs problem
On Mon, 27 Dec 2004, Michel [ISO-8859-1] Dänzer wrote: On Mon, 2004-12-27 at 09:26 -0500, Keith Conger wrote: Ok I changed to 24 can't remember why I bumped it down to 16 in the first place. Also comment out the Option you had me add the artifacts do not show. Hmm, sounds like the host data byte swapping doesn't work via the CP for some reason... Hi Michel, I'll just try to add some noise here, from the point of view of a person who does not have the hardware to test with and examines the code and docs :) 1. In RADEONSetupForScanlineImageWrite dp_gui_master_cntl_clip does not depend on endianness unlike other similar functions.. Is it right ? 2. It appears that PIX_ORDER field in gui_cntl is *not* used by microcode. In fact, I can't find DP_GUI_MASTER anywhere in R300 docs. So, perhaps, R300 hardware would use the same endiannes settings for both MMIO and ACCEL modes ? best Vladimir Dergachev
Re: [R300] Xorg cvs problem
Hi, Ok I changed to 24 can't remember why I bumped it down to 16 in the first place. Also comment out the Option you had me add the artifacts do not show. However after some testing I did notice the fonts being screwy in mozilla and the Xscreensaver lock dialog. Also I tested Xv and the colors are really off. One last thing is glxgears also doesn't really display anything. This is without dri from the r300 project. Thanks, Keith On Thu, 2004-12-23 at 13:34 -0500, Michel Dänzer wrote: > On Thu, 2004-12-23 at 11:46 -0500, Keith Conger wrote: > > > > This does in fact stop the artifacts. > > Does the problem also occur in depth 24 (your chip should be plenty > powerful to handle that :)? Does Xv work correctly? > > > P.S. I assume the problem also occurs if you don't play with 3D > acceleration at all? > -- Keith Conger [EMAIL PROTECTED] http://pimpstation.org --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [R300] Xorg cvs problem
On Mon, 2004-12-27 at 09:26 -0500, Keith Conger wrote: > > Ok I changed to 24 can't remember why I bumped it down to 16 in the > first place. Also comment out the Option you had me add the artifacts do > not show. Hmm, sounds like the host data byte swapping doesn't work via the CP for some reason... > However after some testing I did notice the fonts being screwy > in mozilla and the Xscreensaver lock dialog. I guess Option "XaaNoScanlineCPUToScreenColorExpandFill" works around this? Does this work correctly with the DRI disabled? > Also I tested Xv and the colors are really off. No big surprise at this point, but thanks for testing. Option "DMAForXv" "off" should work around that. -- Earthling Michel DÃnzer | Debian (powerpc), X and DRI developer Libre software enthusiast| http://svcs.affero.net/rm.php?r=daenzer --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
[Bug 1678] Direct Rendering problems on IGP 320M
Please do not reply to this email: if you want to comment on the bug, go to the URL shown below and enter yourcomments there. https://bugs.freedesktop.org/show_bug.cgi?id=1678 [EMAIL PROTECTED] changed: What|Removed |Added Attachment #1585 is|0 |1 obsolete|| --- Additional Comments From [EMAIL PROTECTED] 2004-12-27 03:03 --- Created an attachment (id=1597) --> (https://bugs.freedesktop.org/attachment.cgi?id=1597&action=view) dsdt.dsl that fixes freezings when starting X by changing some IRQs (commented), see http://groups.yahoo.com/group/amilo -- Configure bugmail: https://bugs.freedesktop.org/userprefs.cgi?tab=email --- You are receiving this mail because: --- You are the assignee for the bug, or are watching the assignee. --- SF email is sponsored by - The IT Product Guide Read honest & candid reviews on hundreds of IT Products from real users. Discover which products truly live up to the hype. Start reading now. http://productguide.itmanagersjournal.com/ -- ___ Dri-devel mailing list Dri-devel@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/dri-devel
