| Summary: | Mutter 3.22.2 makes Gnome sessions fail under Xorg | ||
|---|---|---|---|
| Product: | [Fedora] Fedora | Reporter: | David H. Gutteridge <dhgutteridge> |
| Component: | mutter | Assignee: | Florian Müllner <fmuellner> |
| Status: | CLOSED DUPLICATE | QA Contact: | Fedora Extras Quality Assurance <extras-qa> |
| Severity: | high | Docs Contact: | |
| Priority: | unspecified | ||
| Version: | 25 | CC: | BLWedge09, email, fmuellner, james, lucachess, nisutubobi, otaylor, rickatnight11, rmatos, secondary, walters |
| Target Milestone: | --- | ||
| Target Release: | --- | ||
| Hardware: | All | ||
| OS: | Linux | ||
| Whiteboard: | |||
| Fixed In Version: | Doc Type: | If docs needed, set a value | |
| Doc Text: | Story Points: | --- | |
| Clone Of: | Environment: | ||
| Last Closed: | 2016-12-07 16:06:12 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: | |
|
Description
David H. Gutteridge
2016-11-30 16:22:46 UTC
Note there is no such issue on the machine in question running Gnome with Wayland. Only Xorg is affected. I only have the i686 machine to test with at present, so I can't comment if x86_64 is similarly affected (I'm guessing not if no one's already hit this). I tried setting some environment variables as detailed (https://developer.gnome.org/meta/stable/running-mutter.html) but the resulting output doesn't seem useful. It stops outputting after the first five lines are written. I can confirm I get the same issue on x86_64. I'm unable to duplicate this on an x86_64 laptop I just upgraded to Fedora 25, so it appears there's something hardware-specific about this. (General settings are identical on both.) Both the reproducing and non-reproducing laptops have Intel graphics of differing vintages. Interesting, I have the same issue and downgrading mutter alone fixes the problem. I do have Intel graphics card too (and x86_64 system): Intel HD Graphics 3000 on Intel i3-2310M processor (Sandy Bridge). Same issue here on a Dell Lattitude E5470 with Processor: Intel® Core™ i5-6300U CPU @ 2.40GHz × 4 and Graphics: Intel HD Graphics 520 (Skylake GT2). Just run into this on my i7-2760Q notebook with Intel X3100 graphics. Downgrade mutter to fix. My i5-5200U machine is unaffected. I have the same problem on a notebook with Fedora 25 64 bit, Intel(R) Core(TM)2 Quad CPU Q9000 @ 2.00GHz processor and Mobility Radeon HD 4650/5165 video card with radeon open drivers. This is the log of /var/log/messages, taken from ssh connection): [ 75.184974] gnome-session-f[1376]: segfault at 0 ip 00007fd515ddf579 sp 00007ffd2f332670 error 4 in libgtk-3.so.0.2200.4[7fd515b01000+6f0000] Note that gnome-shell do not start, although the desktop icons are visible on a black background, and I can browse folders from them (but no execute any programs: ps -e | grep gnome 999 ? 00:00:00 gnome-keyring-d 1238 tty1 00:00:00 gnome-session-b 1385 tty1 00:00:00 gnome-settings- 1462 tty1 00:00:01 gnome-softwar Note too that another PC with Fedora 25 64 bit, GeForce GTX 650 video card, nouveau driver and Intel(R) Core(TM)2 Quad CPU Q8300 @ 2.50GHz processor is wirking fine with the new version of mutter. This bug seems hardware-dependent. Same here. Hardware: dell precision 7510 with nvidia m1000m and i7-6820HQ. Downgrading to mutter 3.22.1 helps. journalctl: Dez 03 13:45:56 my-laptop gnome-session-binary[1497]: WARNING: App 'org.gnome.Shell.desktop' exited with code 1 Dez 03 13:45:56 my-laptop gnome-session-binary[1497]: Unrecoverable failure in required component org.gnome.Shell.desktop Dez 03 13:45:56 my-laptop gnome-session-binary[1497]: WARNING: App 'org.gnome.Shell.desktop' respawning too quickly Dez 03 13:45:56 my-laptop spice-vdagent[1547]: Cannot access vdagent virtio channel /dev/virtio-ports/com.redhat.spice.0 Dez 03 13:45:56 my-laptop gnome-session-binary[1497]: WARNING: App 'spice-vdagent.desktop' exited with code 1 Dez 03 13:45:56 my-laptop gnome-session[1497]: gnome-session-binary[1497]: WARNING: App 'spice-vdagent.desktop' exited with code 1 Dez 03 13:45:56 my-laptop gnome-session[1497]: Unable to init server: Could not connect: Connection refused Dez 03 13:45:56 my-laptop audit[1546]: ANOM_ABEND auid=4294967295 uid=42 gid=42 ses=4294967295 subj=system_u:system_r:xdm_t:s0-s0:c0.c1023 pid=1546 comm="gnome-session-f" exe="/usr/libexec/gnome-session-fai Dez 03 13:45:56 my-laptop abrt-hook-ccpp[1551]: Process 1546 (gnome-session-failed) of user 42 killed by SIGSEGV - dumping core Looks like nvidia hardware is also affected: Geforce GT 730 using kmod-nvidia-4.8.10-300.fc25.x86_64.x86_64-1:375.20-1.fc25. Downgrading to mutter 3.22.1-8 resolved the issue. I can confirm the same issue in a VirtualBox (5.1.10 on Windows 7 host) VM after upgrading from Fedora 24 to 25. Downgrading mutter with `sudo dnf install mutter-3.22.1-8.fc25.x86_64` resolved this issue for me immediately. I can confirm this bug that appeared last week when I decided to upgrade from Fedora 24 (which works PERFECTLY). I am encountering this error during GNOME startup (core I3 4th generation): [20632.439078] [drm:intel_set_cpu_fifo_underrun_reporting [i915]] *ERROR* uncleared fifo underrun on pipe A [20632.439103] [drm:intel_cpu_fifo_underrun_irq_handler [i915]] *ERROR* CPU pipe A FIFO underrun *** This bug has been marked as a duplicate of bug 1401893 *** Confirming https://bodhi.fedoraproject.org/updates/FEDORA-2016-12b729ad09 fixes the issue for me. (I'm commenting here since when this bug was closed as a duplicate, the CC list wasn't carried over to the other bug. I guess that's a deficiency of Bugzilla.) This bug reappeared again, unlucky... [ 145.610219] gnome-session-f[2606]: segfault at 0 ip 00007fbe2669b579 sp 00007ffeddba0700 error 4 in libgtk-3.so.0.2200.4[7fbe263bd000+6f0000] so this bug is not fixed, again. (In reply to Luca from comment #16) > This bug reappeared again, unlucky... > > [ 145.610219] gnome-session-f[2606]: segfault at 0 ip 00007fbe2669b579 sp > 00007ffeddba0700 error 4 in libgtk-3.so.0.2200.4[7fbe263bd000+6f0000] > > so this bug is not fixed, again. I can't duplicate this, with updates as of last night. There must be something different in your environment versus mine. |