On 01.04.2016 22:12, Michael Thayer wrote:
I have just been adjusting the cursor handling code in the VirtualBox kernel driver and realised that the modesetting user space driver is still adjusting the reported cursor position to take the hot-spot into account, even though it also reports the hot-spot using drmModeSetCursor2(). In theory this would be easy to fix, but that would create a new problem: kernel space would have no easy way to tell whether it was dealing with the old or with the new driver. Other drivers do not have this bug though, and I seem to be the first person to want to use both position and hot-spot information in modesetting. Does anyone have a good idea about how to deal with this? My best so far is for my driver to intercept DRM_IOCTL_MODE_CURSOR2 and just -EINVAL it to force fall-back.
Ah, it gets more fun - the kernel code clearly also makes the same assumption: if a driver does not provide cursor_set2(), it falls back to calling cursor_set() with the same parameters, minus the hot-spot location. Reported cursor position does not take this into account. So this is clearly already a part of the kernel API, though probably as yet with no one actually depending on it.
Regards, Michael -- Michael Thayer | VirtualBox engineer ORACLE Deutschland B.V. & Co. KG | Werkstr. 24 | D-71384 Weinstadt ORACLE Deutschland B.V. & Co. KG Hauptverwaltung: Riesstraße 25, D-80992 München Registergericht: Amtsgericht München, HRA 95603 Komplementärin: ORACLE Deutschland Verwaltung B.V. Hertogswetering 163/167, 3543 AS Utrecht, Niederlande Handelsregister der Handelskammer Midden-Nederland, Nr. 30143697 Geschäftsführer: Alexander van der Ven, Jan Schultheiss, Val Maher _______________________________________________ xorg-devel@lists.x.org: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: https://lists.x.org/mailman/listinfo/xorg-devel
