Bug 1269667
| Summary: | mouse event for unknown monitor (0 >= 0) - mouse lost | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | poma <pomidorabelisima> | ||||||
| Component: | xorg-x11-drv-qxl | Assignee: | Alon Levy <alon> | ||||||
| Status: | CLOSED DEFERRED | QA Contact: | Fedora Extras Quality Assurance <extras-qa> | ||||||
| Severity: | unspecified | Docs Contact: | |||||||
| Priority: | unspecified | ||||||||
| Version: | rawhide | CC: | alon, amit.shah, berrange, cfergeau, crobinso, dracut-maint-list, dwmw2, fziglio, gansalmon, harald, hdegoede, itamar, jonathan, kernel-maint, lyude, madhu.chinakonda, marcandre.lureau, mchehab, pbonzini, rjones, sandmann, virt-maint, xgl-maint, zbyszek | ||||||
| Target Milestone: | --- | ||||||||
| Target Release: | --- | ||||||||
| Hardware: | x86_64 | ||||||||
| OS: | Linux | ||||||||
| Whiteboard: | |||||||||
| Fixed In Version: | Doc Type: | Bug Fix | |||||||
| Doc Text: | Story Points: | --- | |||||||
| Clone Of: | Environment: | ||||||||
| Last Closed: | 2016-01-29 16:18:01 UTC | Type: | Bug | ||||||
| Regression: | --- | Mount Type: | --- | ||||||
| Documentation: | --- | CRM: | |||||||
| Verified Versions: | Category: | --- | |||||||
| oVirt Team: | --- | RHEL 7.3 requirements from Atomic Host: | |||||||
| Cloudforms Team: | --- | Target Upstream Version: | |||||||
| Embargoed: | |||||||||
| Attachments: |
|
||||||||
|
Description
poma
2015-10-07 21:45:00 UTC
- 1st session seen via xrandr: Screen 0: minimum 320 x 200, current 1024 x 768, maximum 8192 x 8192 Virtual-0 disconnected Virtual-1 disconnected Virtual-2 disconnected Virtual-3 disconnected In the above example RE-LOGIN and 2nd X session (with a recuperated mouse functionality) -is- possible, but only due to usage of: /etc/lightdm/lightdm.conf display-setup-script=/usr/bin/RandR xrandr --newmode "widthxheight" ... xrandr --addmode Virtual-0 widthxheight ... xrandr --output Virtual-0 --mode widthxheight ... to override relatively low default display resolution. But in fact without it, the following is actually happening: == RE-LOGIN == /var/log/Xorg.0.log ... [ 126.755] (II) qxl(0): Output Virtual-0 has no monitor section [ 126.755] (II) qxl(0): Output Virtual-1 has no monitor section [ 126.755] (II) qxl(0): Output Virtual-2 has no monitor section [ 126.755] (II) qxl(0): Output Virtual-3 has no monitor section [ 126.755] (II) qxl(0): EDID for output Virtual-0 [ 126.755] (II) qxl(0): EDID for output Virtual-1 [ 126.755] (II) qxl(0): EDID for output Virtual-2 [ 126.755] (II) qxl(0): EDID for output Virtual-3 [ 126.755] (II) qxl(0): Output Virtual-0 disconnected [ 126.755] (II) qxl(0): Output Virtual-1 disconnected [ 126.755] (II) qxl(0): Output Virtual-2 disconnected [ 126.755] (II) qxl(0): Output Virtual-3 disconnected [ 126.755] (WW) qxl(0): No outputs definitely connected, trying again... [ 126.755] (II) qxl(0): Output Virtual-0 disconnected [ 126.755] (II) qxl(0): Output Virtual-1 disconnected [ 126.755] (II) qxl(0): Output Virtual-2 disconnected [ 126.755] (II) qxl(0): Output Virtual-3 disconnected [ 126.755] (WW) qxl(0): Unable to find connected outputs - setting 1024x768 initial framebuffer [ 126.755] (II) qxl(0): Using default gamma of (1.0, 1.0, 1.0) unless otherwise stated. [ 126.755] (II) qxl(0): PreInit complete ... Because of this 2nd X session is not possible. This problem is induced by graphics video modes. Spice Server & video modes: - BROKEN vga=[ VESA >= 320x200x8 ] - WORKING vga=[ VESA <= 80x25 ] or [ VGA ] As reference, VNC Server works with all video modes. See: https://www.kernel.org/doc/Documentation/kernel-parameters.txt To be more precise, this problem is induced by graphics video modes used exclusively with the initramfs image generated in default "hostonly" mode. Mister Hoyer, do you have some idea regarding the initramfs part of this problem, i.e. "broken" "hostonly" mode vs "no-hostonly"(rescue) mode? Created attachment 1081232 [details]
vga=ask
SeaBIOS
(In reply to poma from comment #5) > Mister Hoyer, do you have some idea regarding the initramfs part of this > problem, i.e. "broken" "hostonly" mode vs "no-hostonly"(rescue) mode? ? missing driver in the hostonly part? Check the kernel modules loaded in both variants. Tested also with: - spice-vdagent-0.15.0-5.fc23.x86_64 - spice-vdagent-0.16.0-1.fc23.x86_64 - spice-vdagent-0.16.0-2.fc24.x86_64 - xorg-x11-drv-qxl-0.1.4-4.fc24.x86_64 - xorg-x11-drv-qxl-0.1.4-5.fc24.x86_64 - xorg-x11-drv-qxl-0.1.4-6.fc24.x86_64 - QEMU emulator version 2.3.1 (qemu-2.3.1-7.fc24), Copyright (c) 2003-2008 Fabrice Bellard Sameo. (In reply to Harald Hoyer from comment #7) > (In reply to poma from comment #5) > > Mister Hoyer, do you have some idea regarding the initramfs part of this > > problem, i.e. "broken" "hostonly" mode vs "no-hostonly"(rescue) mode? > > ? missing driver in the hostonly part? > > Check the kernel modules loaded in both variants. Nothing special with modules, diff lsmod-hostonly-sort-awk-print-1.txt lsmod-rescue-sort-awk-print-1.txt 0a1 > 8021q 11a13 > garp 28a31 > mrp QEMU emulator version 2.3.1 (qemu-2.3.1-7.fc22), Copyright (c) 2003-2008 Fabrice Bellard journal - spice-vdagentd debug: ... Oct 09 23:50:27 localhost.localdomain systemd[1]: Starting Agent daemon for Spice guests... Oct 09 23:50:27 localhost.localdomain systemd[1]: spice-vdagentd.service: PID file /var/run/spice-vdagentd/spice-vdagentd.pid not readable (yet?) after start: No such file or directory Oct 09 23:50:27 localhost.localdomain systemd[1]: Started Agent daemon for Spice guests. Oct 09 23:50:29 localhost.localdomain spice-vdagentd[600]: Active session: c1 Oct 09 23:50:34 localhost.localdomain spice-vdagentd[600]: Active session: 1 Oct 09 23:50:35 localhost.localdomain spice-vdagentd[600]: new client accepted: 0x561d98ca78b0, pid: 1244 Oct 09 23:50:35 localhost.localdomain spice-vdagentd[600]: Session for pid 1244: 1 Oct 09 23:50:35 localhost.localdomain spice-vdagentd[600]: 0x561d98ca78b0 sent version, arg1: 0, arg2: 0, size 7 Oct 09 23:50:35 localhost.localdomain spice-vdagentd[600]: 0x561d98ca78b0 is now the active session Oct 09 23:50:36 localhost.localdomain spice-vdagentd[600]: 0x561d98ca78b0 received guest xorg resolution, arg1: 1024, arg2: 768, size 16 Oct 09 23:50:36 localhost.localdomain spice-vdagentd[600]: uinput-update-size: 1024x768 Oct 09 23:50:36 localhost.localdomain spice-vdagentd[600]: screen 0: +0+0 Oct 09 23:50:36 localhost.localdomain spice-vdagentd[600]: opening vdagent virtio channel Oct 09 23:50:36 localhost.localdomain spice-vdagentd[600]: 0x561d98ca78b0 received guest xorg resolution, arg1: 1440, arg2: 900, size 16 Oct 09 23:50:36 localhost.localdomain spice-vdagentd[600]: uinput-update-size: 1440x900 Oct 09 23:50:36 localhost.localdomain spice-vdagentd[600]: screen 0: +0+0 Oct 09 23:50:36 localhost.localdomain spice-vdagentd[600]: 0x561d98ca78b0 received guest xorg resolution, arg1: 1440, arg2: 900, size 16 Oct 09 23:50:36 localhost.localdomain spice-vdagentd[600]: uinput-update-size: 1440x900 Oct 09 23:50:36 localhost.localdomain spice-vdagentd[600]: screen 0: +0+0 Oct 09 23:52:16 localhost.localdomain systemd[1]: Stopping Agent daemon for Spice guests... Oct 09 23:53:46 localhost.localdomain systemd[1]: spice-vdagentd.service: State 'stop-sigterm' timed out. Killing. Oct 09 23:53:46 localhost.localdomain systemd[1]: spice-vdagentd.service: Main process exited, code=killed, status=9/KILL Oct 09 23:53:46 localhost.localdomain systemd[1]: Stopped Agent daemon for Spice guests. Oct 09 23:53:46 localhost.localdomain systemd[1]: spice-vdagentd.service: Unit entered failed state. Oct 09 23:53:46 localhost.localdomain systemd[1]: spice-vdagentd.service: Failed with result 'signal'. Why does not stop properly? QEMU emulator version 2.4.0.1 (qemu-2.4.0.1-2.fc24), Copyright (c) 2003-2008 Fabrice Bellard w/ vga=0x315 journal - spice-vdagentd debug: ... Oct 10 00:18:10 localhost.localdomain systemd[1]: Starting Agent daemon for Spice guests... Oct 10 00:18:10 localhost.localdomain systemd[1]: Started Agent daemon for Spice guests. Oct 10 00:18:18 localhost.localdomain spice-vdagentd[597]: Active session: c1 Oct 10 00:18:30 localhost.localdomain spice-vdagentd[597]: Error getting active session: No data available Oct 10 00:18:30 localhost.localdomain spice-vdagentd[597]: Error getting active session: No data available Oct 10 00:18:30 localhost.localdomain spice-vdagentd[597]: Active session: 1 Oct 10 00:18:34 localhost.localdomain spice-vdagentd[597]: new client accepted: 0x55a477ae8a40, pid: 1278 Oct 10 00:18:34 localhost.localdomain spice-vdagentd[597]: Session for pid 1278: 1 Oct 10 00:18:34 localhost.localdomain spice-vdagentd[597]: 0x55a477ae8a40 sent version, arg1: 0, arg2: 0, size 7 Oct 10 00:18:34 localhost.localdomain spice-vdagentd[597]: 0x55a477ae8a40 is now the active session Oct 10 00:18:34 localhost.localdomain spice-vdagentd[597]: 0x55a477ae8a40 received guest xorg resolution, arg1: 1024, arg2: 768, size 16 Oct 10 00:18:34 localhost.localdomain spice-vdagentd[597]: uinput-update-size: 1024x768 Oct 10 00:18:34 localhost.localdomain spice-vdagentd[597]: screen 0: +0+0 Oct 10 00:18:34 localhost.localdomain spice-vdagentd[597]: opening vdagent virtio channel Oct 10 00:18:34 localhost.localdomain spice-vdagentd[597]: New client connected Oct 10 00:18:34 localhost.localdomain spice-vdagentd[597]: Set max clipboard: 104857600 Oct 10 00:18:35 localhost.localdomain spice-vdagentd[597]: 0x55a477ae8a40 received guest xorg resolution, arg1: 1024, arg2: 768, size 0 Oct 10 00:18:35 localhost.localdomain spice-vdagentd[597]: uinput-update-size: 1024x768 Oct 10 00:18:36 localhost.localdomain spice-vdagentd[597]: 0x55a477ae8a40 sent audio volume sync, arg1: 0, arg2: 0, size 7 Oct 10 00:18:52 localhost.localdomain spice-vdagentd[597]: mouse event for unknown monitor (0 >= 0) ... Oct 10 00:20:32 localhost.localdomain spice-vdagentd[597]: mouse event for unknown monitor (0 >= 0) Oct 10 00:20:48 localhost.localdomain systemd[1]: Stopping Agent daemon for Spice guests... Oct 10 00:20:48 localhost.localdomain spice-vdagentd[597]: (nil) is now the active session Oct 10 00:20:48 localhost.localdomain spice-vdagentd[597]: 0x55a477ae8a40 disconnected Oct 10 00:20:48 localhost.localdomain spice-vdagentd[597]: vdagentd quiting, returning status 0 Oct 10 00:20:48 localhost.localdomain systemd[1]: Stopped Agent daemon for Spice guests. QEMU emulator version 2.4.0.1 (qemu-2.4.0.1-1.fc24), Copyright (c) 2003-2008 Fabrice Bellard w/ vga=0x315 journal - spice-vdagentd debug: ... Oct 10 00:51:18 localhost.localdomain systemd[1]: Starting Agent daemon for Spice guests... Oct 10 00:51:19 localhost.localdomain systemd[1]: Started Agent daemon for Spice guests. Oct 10 00:51:27 localhost.localdomain spice-vdagentd[586]: Active session: c1 Oct 10 00:51:38 localhost.localdomain spice-vdagentd[586]: Active session: 1 Oct 10 00:51:42 localhost.localdomain spice-vdagentd[586]: new client accepted: 0x55ef544e68b0, pid: 1262 Oct 10 00:51:42 localhost.localdomain spice-vdagentd[586]: Session for pid 1262: 1 Oct 10 00:51:42 localhost.localdomain spice-vdagentd[586]: 0x55ef544e68b0 sent version, arg1: 0, arg2: 0, size 7 Oct 10 00:51:42 localhost.localdomain spice-vdagentd[586]: 0x55ef544e68b0 is now the active session Oct 10 00:51:42 localhost.localdomain spice-vdagentd[586]: 0x55ef544e68b0 received guest xorg resolution, arg1: 1440, arg2: 900, size 16 Oct 10 00:51:42 localhost.localdomain spice-vdagentd[586]: uinput-update-size: 1440x900 Oct 10 00:51:42 localhost.localdomain spice-vdagentd[586]: screen 0: +0+0 Oct 10 00:51:43 localhost.localdomain spice-vdagentd[586]: opening vdagent virtio channel Oct 10 00:51:43 localhost.localdomain spice-vdagentd[586]: New client connected Oct 10 00:51:43 localhost.localdomain spice-vdagentd[586]: Set max clipboard: 104857600 Oct 10 00:51:43 localhost.localdomain spice-vdagentd[586]: 0x55ef544e68b0 sent audio volume sync, arg1: 0, arg2: 0, size 7 Oct 10 00:51:43 localhost.localdomain spice-vdagentd[586]: 0x55ef544e68b0 received guest xorg resolution, arg1: 1440, arg2: 900, size 0 Oct 10 00:51:43 localhost.localdomain spice-vdagentd[586]: uinput-update-size: 1440x900 Oct 10 00:51:51 localhost.localdomain spice-vdagentd[586]: mouse event for unknown monitor (0 >= 0) ... Oct 10 00:52:36 localhost.localdomain spice-vdagentd[586]: mouse event for unknown monitor (0 >= 0) Oct 10 00:53:03 localhost.localdomain systemd[1]: Stopping Agent daemon for Spice guests... Oct 10 00:53:03 localhost.localdomain spice-vdagentd[586]: (nil) is now the active session Oct 10 00:53:03 localhost.localdomain spice-vdagentd[586]: 0x55ef544e68b0 disconnected Oct 10 00:53:03 localhost.localdomain spice-vdagentd[586]: vdagentd quiting, returning status 0 Oct 10 00:53:03 localhost.localdomain systemd[1]: Stopped Agent daemon for Spice guests. QEMU emulator version 2.4.0.1 (qemu-2.4.0.1-2.fc24), Copyright (c) 2003-2008 Fabrice Bellard w/ vga=[NORMAL_VGA - standard 80x25 mode - 0 0F00] journal - spice-vdagentd debug: ... Oct 10 00:24:49 localhost.localdomain systemd[1]: Starting Agent daemon for Spice guests... Oct 10 00:24:50 localhost.localdomain systemd[1]: spice-vdagentd.service: PID file /var/run/spice-vdagentd/spice-vdagentd.pid not readable (yet?) after start: No such file or directory Oct 10 00:24:50 localhost.localdomain systemd[1]: Started Agent daemon for Spice guests. Oct 10 00:24:58 localhost.localdomain spice-vdagentd[576]: Active session: c1 Oct 10 00:25:05 localhost.localdomain spice-vdagentd[576]: Active session: 1 Oct 10 00:25:09 localhost.localdomain spice-vdagentd[576]: new client accepted: 0x5653007ee8b0, pid: 1248 Oct 10 00:25:09 localhost.localdomain spice-vdagentd[576]: Session for pid 1248: 1 Oct 10 00:25:09 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 sent version, arg1: 0, arg2: 0, size 7 Oct 10 00:25:09 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 is now the active session Oct 10 00:25:09 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 received guest xorg resolution, arg1: 1024, arg2: 768, size 16 Oct 10 00:25:09 localhost.localdomain spice-vdagentd[576]: uinput-update-size: 1024x768 Oct 10 00:25:09 localhost.localdomain spice-vdagentd[576]: screen 0: +0+0 Oct 10 00:25:09 localhost.localdomain spice-vdagentd[576]: opening vdagent virtio channel Oct 10 00:25:09 localhost.localdomain spice-vdagentd[576]: New client connected Oct 10 00:25:09 localhost.localdomain spice-vdagentd[576]: Set max clipboard: 104857600 Oct 10 00:25:10 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 sent audio volume sync, arg1: 0, arg2: 0, size 7 Oct 10 00:25:15 localhost.localdomain spice-vdagentd[576]: mouse-event: mon 0 311x235 Oct 10 00:25:15 localhost.localdomain spice-vdagentd[576]: mouse: abs-x 311 Oct 10 00:25:15 localhost.localdomain spice-vdagentd[576]: mouse: abs-y 235 Oct 10 00:25:15 localhost.localdomain spice-vdagentd[576]: mouse: syn ... Oct 10 00:25:18 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 received guest xorg resolution, arg1: 1440, arg2: 900, size 16 Oct 10 00:25:18 localhost.localdomain spice-vdagentd[576]: uinput-update-size: 1440x900 Oct 10 00:25:18 localhost.localdomain spice-vdagentd[576]: screen 0: +0+0 Oct 10 00:25:18 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 received guest xorg resolution, arg1: 1440, arg2: 900, size 16 Oct 10 00:25:18 localhost.localdomain spice-vdagentd[576]: uinput-update-size: 1440x900 Oct 10 00:25:18 localhost.localdomain spice-vdagentd[576]: screen 0: +0+0 Oct 10 00:25:19 localhost.localdomain spice-vdagentd[576]: mouse-event: mon 0 1033x555 Oct 10 00:25:19 localhost.localdomain spice-vdagentd[576]: mouse: abs-x 1033 Oct 10 00:25:19 localhost.localdomain spice-vdagentd[576]: mouse: abs-y 555 Oct 10 00:25:19 localhost.localdomain spice-vdagentd[576]: mouse: syn ... Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 received clipboard grab, arg1: 1, arg2: 0, size 4 Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: mouse-event: mon 0 169x109 Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: mouse: btn-left down Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: mouse: syn Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: mouse-event: mon 0 169x109 Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: mouse: btn-left up Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: mouse: syn Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 received clipboard release, arg1: 1, arg2: 0, size 0 Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 received clipboard grab, arg1: 1, arg2: 0, size 4 Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 sent clipboard request, arg1: 1, arg2: 1, size 0 Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 received clipboard data, arg1: 1, arg2: 1, size 25 Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: mouse-event: mon 0 169x109 Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: mouse: btn-left down Oct 10 00:25:49 localhost.localdomain spice-vdagentd[576]: mouse: syn ... Oct 10 00:25:56 localhost.localdomain spice-vdagentd[576]: mouse-event: mon 0 511x235 Oct 10 00:25:56 localhost.localdomain spice-vdagentd[576]: mouse: abs-x 511 Oct 10 00:25:56 localhost.localdomain spice-vdagentd[576]: mouse: abs-y 235 Oct 10 00:25:56 localhost.localdomain spice-vdagentd[576]: mouse: syn Oct 10 00:25:59 localhost.localdomain systemd[1]: Stopping Agent daemon for Spice guests... Oct 10 00:25:59 localhost.localdomain spice-vdagentd[576]: (nil) is now the active session Oct 10 00:25:59 localhost.localdomain spice-vdagentd[576]: 0x5653007ee8b0 disconnected Oct 10 00:25:59 localhost.localdomain spice-vdagentd[576]: vdagentd quiting, returning status 0 Oct 10 00:25:59 localhost.localdomain systemd[1]: Stopped Agent daemon for Spice guests. As seen in the demonstrated logs, in contrast to Rawhide, this this particular problem does not occur on Fedora Stable - 22, with the exception of the other problem - broken "Stopping Agent daemon for Spice guests". With regard to this problem: qemu-2.3.1-4.fc22 BROKEN qemu-2.3.1-5.fc22 BROKEN qemu-2.3.1-6.fc22 BROKEN qemu-2.3.1-7.fc22 OK qemu-2.3.1-7.fc24 (qemu-2.3.1-7.fc22 Rawhide rebuild) BROKEN diff qemu-2.3.1-6.fc22 qemu-2.3.1-7.fc22 : # CVE-2015-7295: virtio-net possible remote DoS (bz #1264393) Patch0016: 0016-virtio-introduce-virtqueue_unmap_sg.patch Patch0017: 0017-virtio-introduce-virtqueue_discard.patch Patch0018: 0018-virtio-net-correctly-drop-truncated-packets.patch # drive-mirror: Fix coroutine reentrance (bz #1266936) Patch0019: 0019-mirror-Fix-coroutine-reentrance.patch # Fix udp socket 'localaddr' (bz #1268708) Patch0020: 0020-util-socket-Add-missing-localaddr-and-localport-opti.patch It turns out only qemu-2.3.1-7 on Fedora 22 works, thanks to some of the five patches. Hmm w/ disabled spice-vdagent auto start: .config/autostart/spice-vdagent.desktop [Desktop Entry] Hidden=true - 1st X session # xrandr --display :0 Screen 0: minimum 320 x 200, current 1024 x 768, maximum 8192 x 8192 Virtual-0 connected 1024x768+0+0 0mm x 0mm 1024x768 59.92*+ 60.02 1920x1200 59.88 1920x1080 59.96 1600x1200 59.87 1680x1050 59.95 1400x1050 59.98 1280x1024 59.89 1440x900 59.89 1280x960 59.94 1280x854 59.89 1280x800 59.81 1280x720 59.86 1152x768 59.78 800x600 59.86 848x480 59.66 720x480 59.71 640x480 59.38 Virtual-1 disconnected Virtual-2 disconnected Virtual-3 disconnected $ /usr/bin/spice-vdagent -d -x spice-vdagent[1605]: 0x55e343a6bdc0 connected to /var/run/spice-vdagentd/spice-vdagent-sock spice-vdagent[1605]: Selection window: 52428801 spice-vdagent[1605]: Screen 0 1024x768+0+0 spice-vdagent[1605]: 0x55e343a6bdc0 sent guest xorg resolution, arg1: 1024, arg2: 768, size 16 spice-vdagent[1605]: net_wm_name: "Xfwm4", has icons: 1 spice-vdagent[1605]: 0x55e343a6bdc0 received version, arg1: 0, arg2: 0, size 7 spice-vdagent[1605]: 0x55e343a6bdc0 sent guest xorg resolution, arg1: 1024, arg2: 768, size 0 spice-vdagent[1605]: unhandled x11 event, type 33, window 631 spice-vdagent[1605]: 0x55e343a6bdc0 received audio volume sync, arg1: 0, arg2: 0, size 7 spice-vdagent[1605]: vdagent_audio_playback_sync mute=yes nchannels=2 spice-vdagent[1605]: vdagent-audio: (playback-left) 0 (%0.00) spice-vdagent[1605]: vdagent-audio: (playback-right) 0 (%0.00) spice-vdagent[1605]: 0x55e343a6bdc0 received clipboard release, arg1: 1, arg2: 0, size 0 spice-vdagent[1605]: primary: received release while not owning client clipboard spice-vdagent[1605]: 0x55e343a6bdc0 received clipboard grab, arg1: 1, arg2: 0, size 4 spice-vdagent[1605]: primary: New selection owner: 52428801 spice-vdagent[1605]: 0x55e343a6bdc0 sent clipboard request, arg1: 1, arg2: 1, size 0 spice-vdagent[1605]: 0x55e343a6bdc0 received clipboard data, arg1: 1, arg2: 1, size 183 spice-vdagent[1605]: 0x55e343a6bdc0 received clipboard release, arg1: 0, arg2: 0, size 0 spice-vdagent[1605]: clipboard: received release while not owning client clipboard spice-vdagent[1605]: 0x55e343a6bdc0 received clipboard grab, arg1: 0, arg2: 0, size 4 spice-vdagent[1605]: clipboard: New selection owner: 52428801 spice-vdagent[1605]: clipboard: sent 5 targets: spice-vdagent[1605]: clipboard: TARGETS spice-vdagent[1605]: clipboard: UTF8_STRING spice-vdagent[1605]: clipboard: text/plain;charset=UTF-8 spice-vdagent[1605]: clipboard: text/plain;charset=utf-8 spice-vdagent[1605]: clipboard: STRING spice-vdagent[1605]: 0x55e343a6bdc0 sent clipboard request, arg1: 0, arg2: 1, size 0 spice-vdagent[1605]: 0x55e343a6bdc0 received clipboard data, arg1: 0, arg2: 1, size 183 spice-vdagent[1605]: clipboard: sent 5 targets: spice-vdagent[1605]: clipboard: TARGETS spice-vdagent[1605]: clipboard: UTF8_STRING spice-vdagent[1605]: clipboard: text/plain;charset=UTF-8 spice-vdagent[1605]: clipboard: text/plain;charset=utf-8 spice-vdagent[1605]: clipboard: STRING spice-vdagent[1605]: clipboard: sent 5 targets: spice-vdagent[1605]: clipboard: TARGETS spice-vdagent[1605]: clipboard: UTF8_STRING spice-vdagent[1605]: clipboard: text/plain;charset=UTF-8 spice-vdagent[1605]: clipboard: text/plain;charset=utf-8 spice-vdagent[1605]: clipboard: STRING spice-vdagent[1605]: 0x55e343a6bdc0 sent clipboard request, arg1: 0, arg2: 1, size 0 spice-vdagent[1605]: 0x55e343a6bdc0 received clipboard data, arg1: 0, arg2: 1, size 183 spice-vdagent[1605]: 0x55e343a6bdc0 sent clipboard request, arg1: 0, arg2: 1, size 0 spice-vdagent[1605]: 0x55e343a6bdc0 received clipboard data, arg1: 0, arg2: 1, size 183 spice-vdagent[1605]: 0x55e343a6bdc0 sent clipboard request, arg1: 0, arg2: 1, size 0 spice-vdagent[1605]: 0x55e343a6bdc0 received clipboard data, arg1: 0, arg2: 1, size 183 spice-vdagent[1605]: 0x55e343a6bdc0 sent clipboard request, arg1: 0, arg2: 1, size 0 spice-vdagent[1605]: 0x55e343a6bdc0 received clipboard data, arg1: 0, arg2: 1, size 183 ... # xrandr --display :0 Screen 0: minimum 320 x 200, current 1024 x 768, maximum 8192 x 8192 Virtual-0 disconnected Virtual-1 disconnected Virtual-2 disconnected Virtual-3 disconnected $ killall spice-vdagent ... spice-vdagent[1605]: 0x55e343a6bdc0 disconnected # xrandr --display :0 Screen 0: minimum 320 x 200, current 1024 x 768, maximum 8192 x 8192 Virtual-0 disconnected Virtual-1 disconnected Virtual-2 disconnected Virtual-3 disconnected After the display once lost, 2nd X session is impossible, w/ or w/o spice-vdagent. An exception exists if used xrandr setup via lightdm/display-setup-script, see comment 2. However mouse -is- lost, despite xrandr routine. With regard to this problem, tested kernels - within domains/guests, within qemu-2.3.1-7.fc22 on Fedora 22 host & qemu-2.4.0.1-1.fc24 on Rawhide host, are: BROKEN: - 4.3.0-0.rc4.git3.1.fc24.x86_64 - 4.3.0-0.rc4.git0.1.fc24.x86_64 - 4.3.0-0.rc3.git1.1.fc24.x86_64 - 4.3.0-0.rc2.git0.1.fc24.x86_64 - 4.3.0-0.rc1.git0.1.fc24.x86_64 - 4.3.0-0.rc0.git9.1.fc24.x86_64 - 4.3.0-0.rc0.git8.1.fc24.x86_64 OK: + 4.3.0-0.rc0.git7.1.fc24.x86_64 + 4.3.0-0.rc0.git6.1.fc24.x86_64 + 4.3.0-0.rc0.git5.1.fc24.x86_64 + 4.2.0-0.rc8.git3.1.fc24.x86_64 + 4.2.0-1.fc24.x86_64 + 4.2.3-200.fc22.x86_64 diff
--- linux-4.3.0-0.rc0.git7.1.fc22.x86_64/drivers/input/misc/uinput.c
+++ linux-4.3.0-0.rc0.git8.1.fc22.x86_64/drivers/input/misc/uinput.c
&
--- vanilla-4.2-git7/drivers/input/misc/uinput.c
+++ vanilla-4.2-git8/drivers/input/misc/uinput.c
@@ -320,10 +320,8 @@
* Check if absmin/absmax/absfuzz/absflat are sane.
*/
- for (cnt = 0; cnt < ABS_CNT; cnt++) {
+ for_each_set_bit(cnt, dev->absbit, ABS_CNT) {
int min, max;
- if (!test_bit(cnt, dev->absbit))
- continue;
min = input_abs_get_min(dev, cnt);
max = input_abs_get_max(dev, cnt);
@@ -41[67],7 +41[45],7 @@
dev->id.product = user_dev->id.product;
dev->id.version = user_dev->id.version;
- for (i = 0; i < ABS_CNT; i++) {
+ for_each_set_bit(i, dev->absbit, ABS_CNT) {
input_abs_set_max(dev, i, user_dev->absmax[i]);
input_abs_set_min(dev, i, user_dev->absmin[i]);
input_abs_set_fuzz(dev, i, user_dev->absfuzz[i]);
--- a/drivers/input/misc/uinput.c
+++ b/drivers/input/misc/uinput.c
@@ -320,8 +320,10 @@
* Check if absmin/absmax/absfuzz/absflat are sane.
*/
- for_each_set_bit(cnt, dev->absbit, ABS_CNT) {
+ for (cnt = 0; cnt < ABS_CNT; cnt++) {
int min, max;
+ if (!test_bit(cnt, dev->absbit))
+ continue;
min = input_abs_get_min(dev, cnt);
max = input_abs_get_max(dev, cnt);
Revert to working state as in 4.3.0-0.rc0.git7.1.fc24.x86_64 i.e. commit ba4e9a6
https://git.kernel.org/cgit/linux/kernel/git/next/linux-next.git/commit/drivers/input/misc/uinput.c?id=ba4e9a6
non effetto positivo.
Omitted "i18n" dracut module will do the trick for a Host-Only initramfs image. /etc/dracut.conf.d/custom.conf omit_dracutmodules+=" i18n " ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /usr/lib/dracut/dracut.conf.d/01-dist.conf # dracut config file customized for RedHat/Fedora. # i18n i18n_vars="/etc/sysconfig/keyboard:KEYTABLE-KEYMAP /etc/sysconfig/i18n:SYSFONT-FONT,FONTACM-FONT_MAP,FONT_UNIMAP" i18n_default_font="eurlatgr" i18n_install_all="yes" stdloglvl=3 sysloglvl=5 install_optional_items+=" vi /etc/virc ps grep cat rm " prefix="/" systemdutildir=/usr/lib/systemd systemdsystemunitdir=/usr/lib/systemd/system systemdsystemconfdir=/etc/systemd/system udevdir=/usr/lib/udev hostonly="yes" hostonly_cmdline="no" early_microcode="yes" ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ /usr/lib/dracut/modules.d/10i18n/README ... 1. Hostonly vs Generic If you invoke dracut with '-H' option, i18n module install script will gather variables values from your configuration files using mappings provided in "/etc/dracut.conf.d/<foo>.conf". Those variables will be put in "etc/vconsole.conf" and "etc/locale.conf" files inside initramfs image. Next it will install only declared font, keymaps and so. When building generic image (dracut without '-H' option), install script copies all content of directories: consolefonts, consoletrans, unimaps and keymaps to image. Take into account that's about 4 MiB. ... Created attachment 1082164 [details]
initramfs-4.3.0-0.rc4.git3.1.fc24.x86_64.img.diff
diff
initramfs-4.3.0-0.rc4.git3.1.fc24.x86_64-omit-i18n.img
initramfs-4.3.0-0.rc4.git3.1.fc24.x86_64.img
It turns out that in case of the SPICE remote computing protocol - if "vga=" higher than standard 80x25 mode is used - works all or nothing for "i18n" dracut module, for the "hostonly" mode. Pardon me, works "nothing" regarding "i18n" dracut module, for the "hostonly" mode. Maybe "setfont" breaks it? Maybe replace setfont temporarily with a shell script and retry. (In reply to Harald Hoyer from comment #22) > Maybe "setfont" breaks it? Maybe replace setfont temporarily with a shell > script and retry. or replace setfont with /bin/true. (In reply to Harald Hoyer from comment #23) > (In reply to Harald Hoyer from comment #22) > > Maybe "setfont" breaks it? Maybe replace setfont temporarily with a shell > > script and retry. > > or replace setfont with /bin/true. diff --git a/modules.d/10i18n/console_init.sh b/modules.d/10i18n/console_init.sh index 2536f14..00e9900 100755 --- a/modules.d/10i18n/console_init.sh +++ b/modules.d/10i18n/console_init.sh @@ -51,7 +51,7 @@ set_font() { [ -n "${FONT_MAP}" ] && trans="-m ${FONT_MAP}" [ -n "${FONT_UNIMAP}" ] && uni="-u ${FONT_UNIMAP}" - setfont ${FONT} -C ${dev} ${trans} ${uni} + /bin/true ${FONT} -C ${dev} ${trans} ${uni} } dev_close() { Same as before. # >console_init.sh # >parse-i18n.sh and it is still broken. So, whatever happens, it is in 'module-setup.sh'. (In reply to poma from comment #24) > (In reply to Harald Hoyer from comment #23) > > (In reply to Harald Hoyer from comment #22) > > > Maybe "setfont" breaks it? Maybe replace setfont temporarily with a shell > > > script and retry. > > > > or replace setfont with /bin/true. > > > diff --git a/modules.d/10i18n/console_init.sh > b/modules.d/10i18n/console_init.sh > index 2536f14..00e9900 100755 > --- a/modules.d/10i18n/console_init.sh > +++ b/modules.d/10i18n/console_init.sh > @@ -51,7 +51,7 @@ set_font() { > [ -n "${FONT_MAP}" ] && trans="-m ${FONT_MAP}" > [ -n "${FONT_UNIMAP}" ] && uni="-u ${FONT_UNIMAP}" > > - setfont ${FONT} -C ${dev} ${trans} ${uni} > + /bin/true ${FONT} -C ${dev} ${trans} ${uni} > } > > dev_close() { > > > Same as before. no.. the real binary. It's called from systemd-vconsole-setup nowadays. The shell script is only used, if it is a non-systemd system. # mv /bin/setfont /bin/setfont.bak # cp /bin/true /bin/setfont # dracut --force # mv /bin/setfont.bak /bin/setfont (In reply to Harald Hoyer from comment #26) > (In reply to poma from comment #24) > > (In reply to Harald Hoyer from comment #23) > > > (In reply to Harald Hoyer from comment #22) > > > > Maybe "setfont" breaks it? Maybe replace setfont temporarily with a shell > > > > script and retry. > > > > > > or replace setfont with /bin/true. > > > > > > diff --git a/modules.d/10i18n/console_init.sh > > b/modules.d/10i18n/console_init.sh > > index 2536f14..00e9900 100755 > > --- a/modules.d/10i18n/console_init.sh > > +++ b/modules.d/10i18n/console_init.sh > > @@ -51,7 +51,7 @@ set_font() { > > [ -n "${FONT_MAP}" ] && trans="-m ${FONT_MAP}" > > [ -n "${FONT_UNIMAP}" ] && uni="-u ${FONT_UNIMAP}" > > > > - setfont ${FONT} -C ${dev} ${trans} ${uni} > > + /bin/true ${FONT} -C ${dev} ${trans} ${uni} > > } > > > > dev_close() { > > > > > > Same as before. > > no.. the real binary. It's called from systemd-vconsole-setup nowadays. The > shell script is only used, if it is a non-systemd system. > > # mv /bin/setfont /bin/setfont.bak > # cp /bin/true /bin/setfont > # dracut --force > # mv /bin/setfont.bak /bin/setfont That works. (In reply to poma from comment #27) > > # mv /bin/setfont /bin/setfont.bak > > # cp /bin/true /bin/setfont > > # dracut --force > > # mv /bin/setfont.bak /bin/setfont > > That works. So, the bug is in the virtualization not being able to handle the setfont. Please reassign! or maybe "vgamem_mb=16" is too small (In reply to Harald Hoyer from comment #29) > or maybe "vgamem_mb=16" is too small /etc/libvirt/qemu/<domain>.xml <model type='qxl' ram='65536' vram='65536' vgamem='32768' heads='1'/> /usr/bin/qemu-system-x86_64 ... -device qxl-vga,id=video0,ram_size=67108864,vram_size=67108864,vgamem_mb=32,bus=pci.0,addr=0x2 and it is still broken. (In reply to Harald Hoyer from comment #28) > (In reply to poma from comment #27) > > > # mv /bin/setfont /bin/setfont.bak > > > # cp /bin/true /bin/setfont > > > # dracut --force > > > # mv /bin/setfont.bak /bin/setfont > > > > That works. > > So, the bug is in the virtualization not being able to handle the setfont. > Please reassign! Christophe? If I followed correctly, this also ends up being related to https://bugzilla.redhat.com/show_bug.cgi?id=1271153#c24 ? (In reply to Christophe Fergeau from comment #32) > If I followed correctly, this also ends up being related to > https://bugzilla.redhat.com/show_bug.cgi?id=1271153#c24 ? Please read https://bugzilla.redhat.com/show_bug.cgi?id=1271153#c27 Why is Component: dracut still selected, is the expected fix coming from dracut? Mister Ziglio care to comment, why "vga=" higher than standard 80x25 mode is fateful for the qxl, thereupon the mouse? # dmesg | grep qxl [ 6.030961] [drm] qxl: 64M of VRAM memory size [ 6.047070] [drm] qxl: 127M of IO pages memory ready (VRAM domain) [ 6.060631] [drm] qxl: 64M of Surface memory size [ 6.235863] fb: switching to qxldrmfb from VESA VGA [ 6.254655] fbcon: qxldrmfb (fb0) is primary device [ 6.267744] [drm:qxl_enc_commit [qxl]] *ERROR* head number too large or missing monitors config: ffffc9000052c000, 0 [ 6.272811] qxl 0000:00:02.0: fb0: qxldrmfb frame buffer device [ 6.334192] [drm] Initialized qxl 0.1.0 20120117 for 0000:00:02.0 on minor 0 [ 24.273480] [drm:qxl_enc_commit [qxl]] *ERROR* head number too large or missing monitors config: ffffc9000052c000, 0 # DISPLAY=:0 xrandr Screen 0: minimum 320 x 200, current 1024 x 768, maximum 8192 x 8192 Virtual-0 disconnected Virtual-1 disconnected Virtual-2 disconnected Virtual-3 disconnected vga and qxl are actually part of the kernel family therefore. In the lack of response, bye. The needinfo request[s] on this closed bug have been removed as they have been unresolved for 1000 days |