On Sun, Jun 3, 2012 at 6:25 PM, Rafael Antognolli <antogno...@profusion.mobi> wrote: > On Thu, Feb 9, 2012 at 9:28 AM, Carsten Haitzler <ras...@rasterman.com> wrote: >> On Fri, 3 Feb 2012 17:21:26 -0500 mik firestone <mikf...@gmail.com> said: >> >>> For the longest time, I have been having a really odd problem with vmware. >>> Every time I started a guest OS, I would lose portions of my keyboard as >>> soon as vmware tools started running in the guest. When I say portions, I >>> mean just certain keys -- like "a", "s", "d" but not "q". It was >>> particularly confusing because if I enabled the caps lock and held the >>> shift key down, I had a full keyboard. If I restarted X, it fixed the >>> problem for a few seconds. I tried xev to see what it was seeing and got >>> some very odd results. >>> >>> It was when I googled those results that I found an article suggesting it >>> was a bad combination of the window manager's shortcuts and vmware. I have >>> a lot of custom keyboard shortcuts in e17. When I checked the list of >>> shortcuts against the list of keys that stopped working, it was a perfect >>> map. To test it even further, I deleted all of the keyboard shortcuts and >>> tried vmware again. The only key I lost was 'g'. To test it even further, I >>> started up KDE instead of e17 and I am not seeing the same symptoms at all. >>> I am therefore somewhat certain that this is an e17 thing, not an X or >>> vmware thing. >> >> it is most certainly a vmware thing. e sets up bindings the same way wm's >> have >> since dinosaurs roamed the earth. it grabs keys+modifiers. it just so happens >> that these may clash with what vmware wants by default and it has some >> fallbacks that make it grab keys without modifiers and no one ever tests that >> at vmware. as it's closed it's hard to tell, but if you use something lik >> xscope and dump an x session log out of vmware u may likely find this to be >> the case... > > I'm facing exactly the same problem here. > > From xev log, what I have when let vmware grab the mouse/keyboard, and > then ungrab it, is: > > MappingNotify event, serial 34, synthetic NO, window 0x0, > request MappingModifier, first_keycode 0, count 0 > > This MappingNotify event end's up calling e's ecore event callback > _e_bindings_mapping_change_event_cb(). What I could test is that if I > comment the following lines: > e_managers_keys_ungrab() > e_managers_keys_grab() > > then I stop getting this wrong behavior, though I think other things > may be borked. I have no idea what these key grab/ungrab are for, but > from its code I guess that when it checks which modifiers are enabled > during this event, it may be getting that ctrl+alt is still enabled > (when exiting VMware, since this is the default shortcut). > > You guys who understand a bit more of X, have any idea of what could > we do to solve this?
BTW, I guess that this wasn't a problem before http://trac.enlightenment.org/e/changeset/63308, in which this "feature" was finally made enabled. -- Rafael Antognolli ProFUSION embedded systems http://profusion.mobi ------------------------------------------------------------------------------ Live Security Virtual Conference Exclusive live event will cover all the ways today's security and threat landscape has changed and how IT managers can respond. Discussions will include endpoint security, mobile security and the latest in malware threats. http://www.accelacomm.com/jaw/sfrnl04242012/114/50122263/ _______________________________________________ enlightenment-users mailing list enlightenment-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/enlightenment-users
