On Mit, 2012-03-14 at 11:08 -0400, Alex Deucher wrote:
2012/3/14 Michel Dänzer michel@daenzer.net:
From: Michel Dänzer michel.daenzer@amd.com
The hardware only takes 27 bits for the offset, so larger offsets are truncated, and the hardware cursor shows random bits other than the intended ones.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=46796
Cc: stable@vger.kernel.org Signed-off-by: Michel Dänzer michel.daenzer@amd.com
Reviewed-by: Alex Deucher alexander.deucher@amd.com
Thanks. In the meantime, it occurred to me there might be a problem when there's less than 128M of VRAM in the first place (so bo->placement.lpfn ends up larger than rdev->mman.bdev.man[TTM_PL_VRAM].size). To be safe, I'll send a v2 patch preventing that.
FWIW, I think we may need a similar fix for the crtc base addresses. They are also limited to 27 bit offsets from the mc address specified in DISPLAY[2]_BASE_ADDR.
Hmm, good point.