Googling the net I've found that this problem returns from time to time since 2005 (first time noticed that nvidia driver messes with input devices). I am experiencing problem with alt-tab (pressing it few times causes keyboard input to be dead until X restart) while using nvidia-drivers-190.42-r3 (current stable) and nvidia-drivers-195.36.15 (~x86 keyword). Problem does not occur when I am using xf86-video-nv-2.1.17 (all with xorg-server-1.7.6, xorg-drivers-1.7 and xorg-x11-7.4-r1). Since I don't exactly remember when it started to occur, I can only suspect that it is somehow connected with recent update of xorg-server which was totally regressive for me (I had to remove second graphics card, second monitor and disable Xinerama as working with it became unstable). Since GLX (very important for me) does not want to work with nv driver, I can't work with Gentoo anymore. Reproducible: Always Steps to Reproduce: 1. Install nvidia-drivers 2. Use full blown desktop environment 3. Open few windows and try to switch them about 100 times using alt+tab (that's what I am typically do: working with computer makes me do many things at the same time!) Actual Results: After unpredictible number of window switching keyboard is dead Expected Results: normal work At first I was suspecting KDE keymaps, as some people suggested me so. Turning them off did not help anything.
"Since GLX (very important for me) does not want to work with nv driver, I can't work with Gentoo anymore." - eselect opengl did the trick.
Now it happend to me again, this time while using xf86-video-nv-2.1.17 driver. I don't know if it was coincidence: changing from nvidia to nv solved problem for few days then today I discovered how to turn on GLX for nv driver and after few hours of worknig alt+tab blocked my keyboard again! I also recalled that I had this problem for some longer time with previous IBM PS/2 keyboard (which recently I replaced with new DELL USB keyboard), thing is it was so rare case that I didn't get annoyed by that, just logged out, logged in and continued. After recent X server upgrade problem started to occur more frequently which sometimes ruined my all day work! As this time it happend while not working with nvidia driver, this bug need to be retagged.
Could you please attach your xorg.conf and Xorg.0.log?!
Two days ago I have upgraded nvidia drivers to: [ebuild R ] x11-drivers/nvidia-drivers-195.36.24 USE="acpi gtk -custom-cflags (-multilib)" 0 kB For two days problem did not occur (however I tried to avoid using alt+tab, I am using taskbar instead) until today when something strange happend: I was typing commands in Terminal window while I accidently clicked on KDE toolbar and started texmaker (to which I have shourtcut from toolbar). I didn't notice that at first, so I didn't stop to type commands in Terminal window. By the time texmaker window was shown, kebyoard was dead! I suspect problem is not related to alt+tab per se, rather this is context switching problem which may lead to keyboard malfunction. Also I got the feeling (which I cannot confirm by evidence) that problem occurs when I try to change context (switch to other application, for example using alt+tab) under heavy load. Thing is, on my Pentium IV HT 3.0GHz and KDE4 it is very easy to cause heavy load.
Created attachment 229539 [details] xorg.conf, as requested
Created attachment 229541 [details] Xorg.0.log as requested
Today I've found that while keyboard is locked in X, it is not dead completely: Ctrl+Alt+F2 brought me to the second text console. It didn't help with anything, as returning to X (Ctrl+Alt+F7) does not unlock the keyboard, the only way to do that is to log out X session completely and return to KDM where keyboard works again and it is possible to type username and password. For some long time I did not try to switch between consoles as for some time I was using two graphics cards (whit xinerama) - while working with that it was not possible to switch to text console as it caused black screen until reboot (it was still possible to blindly log in and type reboot command)
Do you still have the nvidia kernel module loading when you use "nv" ? Thanks
Yes, you're right, I didn't notice that: nvidia module was still loaded at boot time when nv driver was set and eselect opengl was pointning to xorg-x11. Now I unmerged nvidia-drivers completely so the module is not loaded at boot time. So far alt+tab didn't block my keyboard using nv driver, I'll let you know as soon as it occure.
Unmerging the nvidia-drivers package will not remove the module, as /lib/modules is in CONFIG_PROTECT. # find /lib/modules -name nvidia.ko
I noticed that and removed nvidia.ko file manually. I'm sorry I didn't mention that.
Alright, it looks like "nv" isn't doing any harm to your system. Reassigning to Doug who maintains the binary nvidia driver. Please paste the output of nvidia-bug-report.sh Thanks
Created attachment 230655 [details] nvidia-bug-report.sh output as requested It is too big to paste here, so I'm attaching original output file.
Sorry for late response, I wanted to test nv without nvidia.ko loaded a while longer, so I've switched back from nv today morning. After nvidia.ko file removal alt+tab problem did not occur (and I was using it extensively).
I am having the exact same symptoms described in this bug. Similarities to report: At some point after logging into X, my keyboard stops working. It feels like it's triggered by window switching, but I'm not positive. Keyboard still works in other virtual consoles. Switching back to GUI from VC does not restore keyboard. Logging out restores keyboard at login screen. We're both using KDE/KDM. It's very frustrating. ;) Difference: I don't have an nvidia card. I'm using the intel driver. BTW, I do read elog messages, and I did rebuild all X drivers, including xf86-video-intel, after my last merge of xorg-server.
Portage 2.1.8.3 (default/linux/x86/10.0/desktop/kde, gcc-4.3.4, glibc-2.10.1-r1, 2.6.32-gentoo-r7 i686) ================================================================= System uname: Linux-2.6.32-gentoo-r7-i686-Intel-R-_Pentium-R-_4_CPU_3.00GHz-with-gentoo-1.12.13 Timestamp of tree: Tue, 11 May 2010 15:45:01 +0000 app-shells/bash: 4.0_p37 dev-java/java-config: 2.1.10 dev-lang/python: 2.6.4-r1 dev-python/pycrypto: 2.1.0_beta1 dev-util/cmake: 2.6.4-r3 sys-apps/baselayout: 1.12.13 sys-apps/sandbox: 1.6-r2 sys-devel/autoconf: 2.13, 2.65 sys-devel/automake: 1.8.5-r4, 1.9.6-r3, 1.10.3, 1.11.1 sys-devel/binutils: 2.18-r3 sys-devel/gcc: 4.3.4 sys-devel/gcc-config: 1.4.1 sys-devel/libtool: 2.2.6b virtual/os-headers: 2.6.30-r1 ACCEPT_KEYWORDS="x86" ACCEPT_LICENSE="* -@EULA" CBUILD="i686-pc-linux-gnu" CFLAGS="-march=prescott -O2 -pipe -fomit-frame-pointer" CHOST="i686-pc-linux-gnu" CONFIG_PROTECT="/etc /usr/share/X11/xkb /usr/share/config" CONFIG_PROTECT_MASK="/etc/ca-certificates.conf /etc/env.d /etc/env.d/java/ /etc/eselect/postgresql /etc/fonts/fonts.conf /etc/gconf /etc/revdep-rebuild /etc/sandbox.d /etc/terminfo" CXXFLAGS="-march=prescott -O2 -pipe -fomit-frame-pointer" DISTDIR="/usr/portage/distfiles" EMERGE_DEFAULT_OPTS=" --with-bdeps y" FEATURES="assume-digests buildpkg collision-protect distlocks fail-clean fixpackages news parallel-fetch protect-owned sandbox sfperms strict unmerge-logs unmerge-orphans userfetch userpriv userpriv_fakeroot usersandbox verify-rdepend" GENTOO_MIRRORS="http://distfiles.gentoo.org" LANG="en_US.UTF-8" LC_ALL="en_US.UTF-8" LDFLAGS="-Wl,-O1" LINGUAS="en en_US" MAKEOPTS="-j3" PKGDIR="/usr/portage/packages" PORTAGE_CONFIGROOT="/" PORTAGE_RSYNC_OPTS="--recursive --links --safe-links --perms --times --compress --force --whole-file --delete --stats --timeout=180 --exclude=/distfiles --exclude=/local --exclude=/packages" PORTAGE_TMPDIR="/var/tmp" PORTDIR="/usr/portage" PORTDIR_OVERLAY="/var/lib/layman/sunrise /var/lib/layman/nerone" SYNC="rsync://portage.nerone.org/gentoo-portage" USE="X a52 aac acl acpi alsa asf avahi bash-completion branding browserplugin bsf bzip2 cairo caps cddb cdparanoia cleartype cli consolekit cracklib crypt cups curl cxx dbus dri dts dvd dvdr dvdread emboss encode exif faac faad fam fame ffmpeg fftw firefox flac flash ftp gd gdbm gif glib glut gmp gnutls gpm gstreamer gtk hal iconv idn imap imlib ithreads jack java javascript jpeg kde kdehiddenvisibility kipi lame lcms libnotify live lzo mad matroska mbrola mcal memlimit mikmod mjpeg mmap mmx mng modules mozilla mozsvg mp3 mp4 mpeg mplayer mudflap musicbrainz mysql ncurses network network-cron nls nntp nptl nptlonly nsplugin nvidia offensive ogg openexr opengl pam pango pch pcre pdf phonon physfs png postgres ppds pppd pulseaudio python qt3support qt4 readline reflection reiserfs rtc sasl schroedinger sdl semantic-desktop session sndfile speex spell spl sql sse sse2 ssl startup-notification subversion svg symlink sysfs syslog taglib theora threads threadsafe tiff tk truetype unicode usb utempter utf8 vcd vim-syntax vorbis webdav win32codecs wps x264 x86 xattr xcb xcomposite xforms xine xml xorg xosd xulrunner xv xvid zlib" ALSA_CARDS="ali5451 als4000 atiixp atiixp-modem bt87x ca0106 cmipci emu10k1 emu10k1x ens1370 ens1371 es1938 es1968 fm801 hda-intel intel8x0 intel8x0m maestro3 trident usb-audio via82xx via82xx-modem ymfpci" ALSA_PCM_PLUGINS="adpcm alaw asym copy dmix dshare dsnoop empty extplug file hooks iec958 ioplug ladspa lfloat linear meter mmap_emul mulaw multi null plug rate route share shm softvol" APACHE2_MODULES=" alias auth_basic authn_alias authn_default authn_file authz_default authz_groupfile authz_host authz_user autoindex dav dav_fs dir env expires headers include info log_config mime mime_magic negotiation proxy proxy_http rewrite setenvif status unique_id " APACHE2_MPMS="event" CAMERAS="canon directory ptp2" ELIBC="glibc" INPUT_DEVICES="evdev" KERNEL="linux" LCD_DEVICES="bayrad cfontz cfontz633 glk hd44780 lb216 lcdm001 mtxorb ncurses text" LINGUAS="en en_US" RUBY_TARGETS="ruby18" USERLAND="GNU" VIDEO_CARDS="intel vesa" XTABLES_ADDONS="quota2 psd pknock lscan length2 ipv4options ipset ipp2p iface geoip fuzzy condition tee tarpit sysrq steal rawnat logmark ipmark dhcpmac delude chaos account" Unset: CPPFLAGS, CTARGET, FFLAGS, INSTALL_MASK, PORTAGE_COMPRESS, PORTAGE_COMPRESS_FLAGS, PORTAGE_RSYNC_EXTRA_OPTS
So doing a little digging on this one turns up literally no code paths that could really result in this issue based on the most up to date issue (e.g. task switching between foreground applications). I can't find anything similar on the Nvidia forums and then lastly there's a "me too" post on this bug with someone using the Intel driver. This really seems like an issue kernel side to me or some kind of disconnect between X processing input events. To be honest I've got several issues at the office with various machines (Gentoo, Ubuntu 10.04 LTS and latest Arch) with input devices and newer X's. For example my machine (Gentoo) if I leave it over night it refuses to accept mouse input into X until I switch to VT1 and then back to VT7. I've confirmed the kernel is still delivering input events for the mouse, however X couldn't really careless. Arch Linux spits some udev message to syslog and then completely ignores all input until it spits the same message again almost 2 minutes later. Ubuntu, its just a terrible mess we've got half a dozen "me toos" on various launchpad bugs (Ubuntu happens to be our most popular distro at the office). Given all the odd behaviors with X, the fact that these problems occur for people not using nvidia-drivers, I'm going to have to kick this back to the X guys. Hopefully some of the upstream issues start to get resolved and land in some of the newer xorg-server updates and fix this. I know binary drivers are the favorite public whipping boy, but its really doubtful this is an nvidia-drivers issue.
Many updates took place since my last visit here. Yesterday I had to switch KDE windows heavily so I used alt-tab more than 100 times and it did not block. I guess we can close it.
