On Thu, May 22, 2014 at 6:33 PM, Matt Roper <matthew.d.roper@xxxxxxxxx> wrote: > v3: > - Move integer overflow checking from setplane_internal to setplane > ioctl. The upcoming legacy cursor support via universal planes needs > to maintain current cursor ioctl semantics and not return error for > these extreme values (found via intel-gpu-tools kms_cursor_crc test). Imo that's just the test being silly. I think it's valid to reject such nonsense with -EINVAL and limit the random-walk range of our testcase a bit. The igt testcases are just guidelines for what our ABI is, but not the hard rules. And clarifying the ABI in such cases is imo the right approach. That random-walk testcase just was meant to test correctly clamping. E.g. on hardware with just 12bit for the cursor placement we'd wrap around and show a cursor again that should be off. So as long as you still test that you can restrict the range of the test a bit. -Daniel -- Daniel Vetter Software Engineer, Intel Corporation +41 (0) 79 365 57 48 - http://blog.ffwll.ch _______________________________________________ dri-devel mailing list dri-devel@xxxxxxxxxxxxxxxxxxxxx http://lists.freedesktop.org/mailman/listinfo/dri-devel
