I've already provided the requested crash information directly on the
bug report via the Launchpad web form (see my comment above), including:

- The retraced backtrace with debug symbols
  (libmutter-18-0-dbgsym 50.1-0ubuntu2.2), showing the identical crash
  location to the original report (meta_kms_impl_get_kms /
  meta_kms_impl_device_get_fd).
- journalctl output covering both crash sessions (Sept 4 and Sept 6),
  redacted of identifying details (hostname, MAC/IP addresses,
  disk/network-connection UUIDs, NextDNS account ID).

Regarding the two automated steps in this notification:

1. ubuntu-bug on the .crash file: I attempted this, but apport-gtk
   crashed/failed silently when processing the file (large gnome-shell
   coredump, ~63MB). I worked around this by extracting the backtrace
   manually via gdb and posting it directly as a comment instead.

2. errors.ubuntu.com/user/<whoopsie-id>: this endpoint currently
   returns HTTP 500 (Internal Server Error) for my whoopsie-id. I
   confirmed this is server-side, not a local network/DNS issue:

   $ curl -sI https://errors.ubuntu.com/
   HTTP/2 200

   $ curl -sI https://errors.ubuntu.com/user/<my-whoopsie-id>
   HTTP/2 500

   The main errors.ubuntu.com page loads fine, only the per-user page
   fails.

Let me know if any additional data would be helpful.


poniedziałek, 7 września 2026 04:05, Daniel van Vugt
<[email protected]> napisał(a):

> To help us find the correct crash bucket, please follow these steps:
> 
> 1. Look in /var/crash for crash files and if found run:
>     ubuntu-bug YOURFILE.crash
> Then tell us the ID of the newly-created bug.
> 
> 2. If step 1 failed then look at https://errors.ubuntu.com/user/ID where
> ID is the content of file /var/lib/whoopsie/whoopsie-id on the machine.
> Do you find any links to recent problems on that page? If so then please
> send the links to us.
> 
> Please take care to avoid attaching .crash files to bugs as we are
> unable to process them as file attachments. It would also be a security
> risk for yourself.
> 
> --
> You received this bug notification because you are subscribed to the bug
> report.
> https://bugs.launchpad.net/bugs/2165888
> 
> Title:
>   gnome-shell 50.1 SIGSEGV in KMS thread during mode set on display wake
>   after privacy screen toggle (regression of bug 2085420)
> 
> Status in mutter package in Ubuntu:
>   Incomplete
> 
> Bug description:
>   gnome-shell 50.1 SIGSEGV in KMS thread during mode set on display wake
>   after privacy screen toggle (regression of bug 2085420)
> 
>   gnome-shell crashes with SIGSEGV in mutter's KMS thread when the
>   built-in panel is re-enabled after idle blanking, if the hardware
>   privacy screen was toggled earlier in the session. The crash is in the
>   atomic mode-set path: meta_kms_mode_create_blob_id() calls
>   meta_kms_impl_device_get_fd() with an invalid MetaKmsImplDevice
>   pointer.
> 
>   This is the trigger condition of bug 2085420, closed via bug 2085423
>   with the fix landing upstream in mutter 49.rc (merge request
>   mutter!4742). The fix is present in 50.1, so this is a regression or a
>   new bug with the same trigger.
> 
>   Environment:
>   - libmutter-18-0 50.1-0ubuntu2.2, gnome-shell 50.1-0ubuntu1.2, Ubuntu 26.04
>   - Wayland session, native backend, i915, atomic mode setting
>   - Linux 7.0.0-30-generic
>   - Lenovo ThinkPad T14 Gen 1 (20S00078MX), PrivacyGuard panel, DRM privacy 
> screen provider privacy_screen-thinkpad_acpi
> 
>   Steps to reproduce:
>   1. Fresh boot, Wayland session. Blank the screen on idle and wake it — 
> survives, repeatedly (verified with several cycles at a 60 s blank delay).
>   2. Toggle the privacy screen with Fn+D: on, then off (sw_state and hw_state 
> back to Disabled).
>   3. Let the screen blank, wake with any input.
> 
>   gnome-shell segfaults during the wake mode set and the session is
>   lost. Reproduced 4 times over two days, the last with exactly this
>   protocol. No suspend/resume involved, no monitor hotplug.
> 
>   Backtrace from the apport core dump, retraced with
>   libmutter-18-0-dbgsym 50.1-0ubuntu2.2. Crashing thread is the KMS
>   thread processing META_KMS_UPDATE_FLAG_MODE_SET:
> 
>   #6  meta_kms_impl_get_kms (impl=<error reading variable: Cannot access 
> memory at address 0x638bea15a607>) at 
> ../src/backends/native/meta-kms-impl.c:52
>   #7  meta_kms_impl_device_get_fd (impl_device=0x638bea15a6af) at 
> ../src/backends/native/meta-kms-impl-device.c:1235
>   #8  meta_kms_mode_create_blob_id (mode=0x7dc4c401fc50, 
> error=0x7dc4d91fe118) at ../src/backends/native/meta-kms-mode.c:51
>   #9  process_mode_set (...) at 
> ../src/backends/native/meta-kms-impl-device-atomic.c:409
>   #10 process_entries (...) at 
> ../src/backends/native/meta-kms-impl-device-atomic.c:978
>   #11 meta_kms_impl_device_atomic_process_update (impl_device=0x7dc4c40052c0, 
> ...) at ../src/backends/native/meta-kms-impl-device-atomic.c:1204
>   #12 do_process (impl_device=0x7dc4c40052c0, ..., 
> flags=META_KMS_UPDATE_FLAG_MODE_SET) at 
> ../src/backends/native/meta-kms-impl-device.c:1649
>   #13 process_mode_set_update (...) at 
> ../src/backends/native/meta-kms-impl-device.c:2330
>   #14 meta_kms_impl_device_process_update (...) at 
> ../src/backends/native/meta-kms-impl-device.c:2370
>   #15 process_sync_update_in_impl (...) at 
> ../src/backends/native/meta-kms-device.c:304
>   #16 meta_thread_impl_dispatch (...) at 
> ../src/backends/native/meta-thread-impl.c:543
>   #17 impl_source_dispatch (...) at 
> ../src/backends/native/meta-thread-impl.c:177
> 
>   The impl_device argument in frame #7 (0x638bea15a6af) is unreadable
>   and misaligned, while the impl_device in frames #10-#14
>   (0x7dc4c40052c0) is valid — the MetaKmsMode in frame #8 appears to
>   hold a dangling pointer to a freed or stale MetaKmsImplDevice.
> 
>   Journal at crash:
> 
>   Aug 31 08:29:23 gnome-shell[24522]: Launching DING process
>   Aug 31 08:29:23 gnome-shell[24522]: GNOME Shell crashed with signal 11
> 
> To manage notifications about this bug go to:
> https://bugs.launchpad.net/ubuntu/+source/mutter/+bug/2165888/+subscriptions
> 
> 
>

-- 
You received this bug notification because you are a member of Ubuntu
Bugs, which is subscribed to Ubuntu.
https://bugs.launchpad.net/bugs/2165888

Title:
  gnome-shell 50.1 SIGSEGV in KMS thread during mode set on display wake
  after privacy screen toggle (regression of bug 2085420)

To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/mutter/+bug/2165888/+subscriptions


-- 
ubuntu-bugs mailing list
[email protected]
https://lists.ubuntu.com/mailman/listinfo/ubuntu-bugs

Reply via email to