[Bug 57192] New: X11DRV_SetCursorPos breaks when xinput "Coordinate Transformation Matrix" is customized
https://bugs.winehq.org/show_bug.cgi?id=57192 Bug ID: 57192 Summary: X11DRV_SetCursorPos breaks when xinput "Coordinate Transformation Matrix" is customized Product: Wine Version: 9.1 Hardware: x86-64 OS: Linux Status: UNCONFIRMED Severity: normal Priority: P2 Component: winex11.drv Assignee: wine-bugs(a)winehq.org Reporter: mk939(a)ymail.com Distribution: --- Created attachment 77083 --> https://bugs.winehq.org/attachment.cgi?id=77083 WINEDEBUG=+cursor,+win wine minetest.exe | trimmed to one repetition of the bugged SetCursorPos call Discovered by testing Minetest 5.9.0 [1] in Wine. In-game, the cursor is always moved to the center. However, when the xinput "Coordinate Transformation Matrix" parameter, the cursor moves to unpredictable positions (bottom right corner, for example). The game becomes unplayable. I think it's caused by "X11DRV_SetCursorPos" because skipping "NtUserSetCursorPos" entirely does avoid this issue. The cursor position is no longer centred, but the game does not look as glitchy any more. System specifications: * Ubuntu 22.04, X11, xfce4 * amdgpu driver Reproduced in Wine 9.1, Wine 9.17 and Proton 9.0-2. Reproduced with * Wine's virtual desktop on and off * Window manager decorations on and off (winecfg) * Window manager control on and off (winecfg) Steps to reproduce: 1. xinput list 2. Remember the xinput <ID> of the mouse (Virtual core pointer -> HID) 3. xinput --set-prop <ID> 'Coordinate Transformation Matrix' 1.1 0 0 0 1.1 0 0 0 1 4. Join a world in Minetest and try to play This problem does not occur when the Transformation Matrix is set to "1.0 0 0 0 1.0 0 0 0 1". It also does not occur on Windows due to lack of xinput and X11 in the first place. [1] https://github.com/minetest/minetest/releases/download/5.9.0/minetest-5.9.0-... -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
http://bugs.winehq.org/show_bug.cgi?id=57192 mk939(a)ymail.com changed: What |Removed |Added ---------------------------------------------------------------------------- Resolution|--- |FIXED Status|UNCONFIRMED |RESOLVED --- Comment #1 from mk939(a)ymail.com --- Cannot reproduce any more in Wine 10.12. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
http://bugs.winehq.org/show_bug.cgi?id=57192 Alexandre Julliard <julliard(a)winehq.org> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|RESOLVED |CLOSED --- Comment #2 from Alexandre Julliard <julliard(a)winehq.org> --- Closing bugs fixed in 10.15. -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
http://bugs.winehq.org/show_bug.cgi?id=57192 --- Comment #3 from mk939(a)ymail.com --- In case someone else runs into a similar issue - in the Irrlicht (a framework) source, I encountered a remark about this problem. xinput2 should generally work better. Cross-reference to a 9 year old bug: https://gitlab.freedesktop.org/xorg/xserver/-/issues/600 -- Do not reply to this email, post in Bugzilla using the above URL to reply. You are receiving this mail because: You are watching all bug changes.
participants (1)
-
WineHQ Bugzilla