Created attachment 43562 [details] logs reveal more info hi all, Recently i have gotten around with the habit of testing release candidate builds out of the linus's tree . Here is something strange which seems to happening with 2.6.37-rc8+ also i am not sure if this existed earlier . Upon hibernating and booting back , X display becomes weird and unusable , but on normal shutdown/reboot things seem to be fine. Here with attaching logs of Xorg.0.log and kern.log a snippet of kern.log as follows Jan 9 11:04:58 x-moto kernel: [ 4554.844761] done. Jan 9 11:04:58 x-moto kernel: [ 4554.844780] PM: Basic memory bitmaps freed Jan 9 11:04:58 x-moto kernel: [ 4554.844785] video LNXVIDEO:00: Restoring backlight state Jan 9 11:04:58 x-moto kernel: [ 4554.850370] [drm:drm_mode_getfb] *ERROR* invalid framebuffer id Jan 9 11:04:58 x-moto kernel: [ 4554.909592] [drm:radeon_ib_schedule] *ERROR* radeon: couldn't schedule IB(12). Jan 9 11:04:58 x-moto kernel: [ 4554.909596] [drm:radeon_cs_ioctl] *ERROR* Failed to schedule IB ! Jan 9 11:04:58 x-moto kernel: [ 4554.909870] [drm:radeon_ib_schedule] *ERROR* radeon: couldn't schedule IB(13). Jan 9 11:04:58 x-moto kernel: [ 4554.909873] [drm:radeon_cs_ioctl] *ERROR* Failed to schedule IB ! Jan 9 11:04:58 x-moto kernel: [ 4555.005941] [drm:radeon_ib_schedule] *ERROR* radeon: couldn't schedule IB(14). Jan 9 11:04:58 x-moto kernel: [ 4555.005945] [drm:radeon_cs_ioctl] *ERROR* Failed to schedule IB ! Jan 9 11:04:58 x-moto kernel: [ 4555.007755] [drm:radeon_ib_schedule] *ERROR* radeon: couldn't schedule IB(15). Jan 9 11:04:58 x-moto kernel: [ 4555.007758] [drm:radeon_cs_ioctl] *ERROR* Failed to schedule IB ! Jan 9 11:04:58 x-moto kernel: [ 4555.010712] [drm:radeon_ib_schedule] *ERROR* radeon: couldn't schedule IB(0). Jan 9 11:04:58 x-moto kernel: [ 4555.010716] [drm:radeon_cs_ioctl] *ERROR* Failed to schedule IB !
I assumed that this regression is also in 2.6.38, and assigned it to DRI.
Did hibernate ever work properly on your system? If so when and can you bisect?
(In reply to comment #2) > Did hibernate ever work properly on your system? If so when and can you > bisect? It used to work fine with Ubuntu Karmic. And i just found out that the issue seems to be present with the distro(Ubuntu Lucid) kernel(2.6.32-21-generic) as well . So that concludes hibernate never has been working on lucid
I think Karmic used UMS rather than KMS, so it's not a regression as KMS and UMS are basically different driver stacks.
[dropped from the regression list]
If this is still seen with modern kernels please update/re-open the bug