Hello.

        Here is update of the status and bugs we have in the 
drm-openchrome and xorg driver.

xf86-video-openchrome  - SVN repo at  
                        http://svn.openchrome.org/svn/branches/kms_branch

Bugs:

        * http://www.openchrome.org/trac/ticket/413
                - First 6 columns of the LVDS display is duplicated on the 
                  right side of the screen. This is due to the pitch not 
                  equaling the horizontal width of the display.
        * http://www.openchrome.org/trac/ticket/414
                - Hardware cursor doesn't work on VX900. My theory is this
                  is due to bugs is the memory allocater from the VRAM in 
                  the current X org driver.
        * http://www.openchrome.org/trac/ticket/415
                - Wrong LVDS index from the scratch pad is used for the 
                  OLPC. Pulled out my OLPC today to fix this.
        * http://www.openchrome.org/trac/ticket/418
                - EXA is broken on the OLPC. To date this is the only 
                  platform with EXA problems. It was traced to check in
                  r840. This bug will take some deep debugging to figure 
                  out.  
        * DisplayPort detect is just broken. Discovered machines that 
          have LVDS with real EDID. The EDID came off the second I2c bus 
          which happens to be the same port we tested for DP being  present.
          The fix is quite complex so it will not be fixed in UMS. KMS 
          side will address this.

Goals:
        * Fix the EXA bug on OLPC. The plan is to drop XAA after this.
        * Fix up the EXA code to use CreatePixmap2, DestoryPixmap so the
          EXA layer can be used in the TTM and non-TTM environment.
        * Work out a buffer manager to handle the TTM and non-TTM case.
        * KMS support.

drm-opencrhome - Tested with xf86-video-modesetting branch 
                 git repo at 
http://cgit.freedesktop.org/~jsimmons/drm-openchrome

Bugs:
        * Bug discovered is the offset from the BO allcated is not 
          offseted int the VRAM. Need to track it down.
        * Fix the kernel cursor api. Currently it oops the machine.

Goals:
        * Need to implement proper fencing.
        * Hardware accelerate the blitting.
        * Support LVDS KMS - I have early code but it needs love :-)



_______________________________________________
Openchrome-devel mailing list
[email protected]
http://wiki.openchrome.org/mailman/listinfo/openchrome-devel


Reply via email to