See attached shot, usually it looks like image just put into buffer with wrong offset.
Sometimes image repeated down several times (out of usual cursor boundaries)
Often after some time of work with such cursor system crashes.
$ egrep -i curs\|mouse /var/log/Xorg.0.log
(**) |-->Input Device "Mouse0"
(II) LoadModule: "mouse"
(II) Loading /usr/local/lib/xorg/modules/input//mouse_drv.so
(II) Module mouse: vendor="X.Org Foundation"
(II) RADEONHD(0): FB: Allocated Cursor Image at offset 0x00000000 (size = 0x00004000)
(II) RADEONHD(0): FB: Allocated Cursor Image at offset 0x00004000 (size = 0x00004000)
(==) RADEONHD(0): Silken mouse enabled
(**) Mouse0: Device: "/dev/sysmouse"
(**) Mouse0: Protocol: "auto"
(**) Mouse0: always reports core events
(**) Option "Device" "/dev/sysmouse"
(==) Mouse0: Emulate3Buttons, Emulate3Timeout: 50
(**) Mouse0: ZAxisMapping: buttons 4, 5, 6 and 7
(==) Mouse0: YAxisMapping: buttons 4 and 5
(**) Mouse0: EmulateWheel, EmulateWheelButton: 2, EmulateWheelInertia: 10, EmulateWheelTimeout: 200
(**) Mouse0: Buttons: 11
(**) Mouse0: Sensitivity: 1
(II) XINPUT: Adding extended input device "Mouse0" (type: MOUSE)
(**) Mouse0: (accel) keeping acceleration scheme 1
(**) Mouse0: (accel) filter chain progression: 2.00
(**) Mouse0: (accel) filter stage 0: 20.00 ms
(**) Mouse0: (accel) set acceleration profile 0
(II) Mouse0: SetupAuto: hw.iftype is 4, hw.model is 0
(II) Mouse0: SetupAuto: protocol is SysMouse
(II) config/hal: Adding input device PS/2 Mouse
Created attachment 26422 [details]
Created attachment 26423 [details]
Created attachment 26426 [details]
Created attachment 26427 [details]
Doesn't quite look like the old known cursor corruption, but please try to reproduce it with git HEAD.
Cursor code has changed quite a bit since 1.2.5 release.
Tried with latest version
(II) RADEONHD: version 1.2.5, built from git branch (no branch), commit c6f607fd
everithing the same (after restart of X cursor still not changed)
Created attachment 26428 [details]
Xorg.0.log with recent radeonhd driver
Hm. This looks like an endian problem. However, we don't support big endian at all AFAIR, and i586 is little endian anyway. Strange.
I used to see this on my rs690 occasionally when I would move the cursor between head before the last round of cursor fixes (i.e., both cursors have to be set to the same color mode). I haven't seen it since. It's probably related to using ARGB on one cursor and mono on the other which was the basis of the cursor issues.
Does this issue occur with the preferred ati driver (xf86-vide-ati)? If so, please move this to the Driver/Radeon component.
Development of radeonhd has pretty much halted and development focus is on the ati driver. Please see http://www.x.org/wiki/radeonhd
If the issue does not exist in the ati driver (or if there is no response to this message), this bug will be closed as WONTFIX unless someone contributes a patch.
Closing due to lack of response. Please reopen and move to the Driver/Radeon
component if this issue persists with xf86-video-ati