Arch Linux User Repository

When reporting this package as outdated make sure there is indeed a new version for Linux Desktop. You can have a look at the "Stable updates" tag in Release blog for this.

You can also run this command to obtain the version string for the latest chrome version:

$ curl -sSf https://dl.google.com/linux/chrome/deb/dists/stable/main/binary-amd64/Packages | \
     grep -A1 "Package: google-chrome-stable" | \
     awk '/Version/{print $2}' | \
     cut -d '-' -f1

Do not report updates for ChromeOS, Android or other platforms stable versions as updates here.

@wyf9661

https://gitlab.archlinux.org/archlinux/packaging/packages/chromium/-/commit/fb4ec9fa1d6c8002bdbe3a724042533e159b1e61

Is there any reason this package set ttf-liberation as one of the necessary dependence?

With the newly released chrome version 139.0.7258.66 the issue I had with one machine hanging if using wayland mode has now been resolved, and it will run without issue on all my machines with the latest released kernel 6.16 and earlier kernels. So it looks like there was a bug in chrome in the previous version, that has been fixed in 139.0.7258.66

@mike.cloaked - Thanks, I think you're right. I noticed afterwards the timestamps don't really correlate

I also see the TensorFlow lines in my journal log,

Aug 03 17:28:35 lenovo2 google-chrome-stable[10441]: Created TensorFlow Lite XNNPACK delegate for CPU.
Aug 03 17:21:00 lenovo2 google-chrome-stable[9521]: Attempting to use a delegate that only supports static-sized tensors with a graph that has dynamic-sized tensors (tensor#-1 is a dynamic-sized tensor).

But chrome is running with no problems. So I am not sure that the log lines for TensorFlow are related to the crashes.

My regular crashes are unrelated to Wayland, since Im running under X11. They seem to be related to TensorFlow, but I dont understand what TensorFlow is being used for by Chrome yet

Jul 30 10:43:38 beep google-chrome-stable[3626468]: Opening in existing browser session.
Jul 30 10:44:16 beep google-chrome-stable[3626752]: WARNING: All log messages before absl::InitializeLog() is called are written to STDERR
Jul 30 10:44:16 beep google-chrome-stable[3626752]: I0000 00:00:1753868656.728889 3626771 voice_transcription.cc:58] Registering VoiceTranscriptionCapability
Jul 30 10:44:16 beep google-chrome-stable[3626752]: Created TensorFlow Lite XNNPACK delegate for CPU.
Jul 30 10:44:16 beep google-chrome-stable[3626752]: Attempting to use a delegate that only supports static-sized tensors with a graph that has dynamic-sized tensors (tensor#-1 is a dynamic-sized tensor).
Jul 30 10:44:16 beep google-chrome-stable[3626752]: [3626743:3626775:0730/104416.844466:ERROR:google_apis/gcm/engine/registration_request.cc:291] Registration response error message: DEPRECATED_ENDPOINT
Jul 30 10:44:29 beep google-chrome-stable[3626752]: [3626743:3626743:0730/104429.024283:ERROR:ui/gfx/x/atom_cache.cc:232] Add chromium/from-privileged to kAtomsToCache
Jul 30 10:44:42 beep google-chrome-stable[3626752]: [3626743:3626775:0730/104442.007964:ERROR:google_apis/gcm/engine/registration_request.cc:291] Registration response error message: DEPRECATED_ENDPOINT
Jul 30 10:44:50 beep google-chrome-stable[3626752]: [3626743:3626780:0730/104450.929902:ERROR:content/browser/browser_main_loop.cc:278] GLib: g_main_context_pop_thread_default: assertion 'stack != NULL' failed

After seeing the comments about chrome crashing, I have been testing four machines - all have different hardware. On three machines, chrome runs without issue in wayland mode. However one old Lenovo Ideapad Y510p hangs immediately when I run chrome in wayland mode, but runs fine in x11 mode. So the latest chrome is not compatible with the hardware (presumably the GPU) on that laptop.

The journal on the problematic laptop has these lines:

$ sudo journalctl -b -1 | grep google-chrome-stable
Aug 01 20:23:44 lenovo2 google-chrome-stable[2194]: MESA-INTEL: warning: Haswell Vulkan support is incomplete
Aug 01 20:23:44 lenovo2 google-chrome-stable[2194]: libva error: /usr/lib/dri/iHD_drv_video.so init failed

google_apis/gcm/engine/registration_request.cc:274] Registration URL fetching failed.
Aug 01 20:23:49 lenovo2 google-chrome-stable[2194]:      [2189:2217:0801/202349.669885:ERROR:google_apis/gcm/engine/registration_request.cc:274] Registration URL fetching failed.
Aug 01 20:23:49 lenovo2 google-chrome-stable[2194]: [2189:2217:0801/202349.669941:ERROR:google_apis/gcm/engine/registration_request.cc:274] Registration URL fetching failed.

Perhaps there is somewhere that a bug report for this issue can be filed? Or perhaps a change in the libva/vdpau/mesa packages could get this machine working again?

I did some further investigation on this - and it turns out there is a bug with graphics drivers in the current 6.16 kernel. By simply downgrading the kernel back to 6.15.6 chrome now runs fine in wayland mode. So hopefully there will be a fix in a future kernel release that will resolve the chrome hang issue (for the Intel Haswell series chipsets).

note that the package "google-chrome-stable" has been compromised and will install a RAT on your computer. this package is fine.

@aliu: there is no source available to build from (Google Chrome, while based on Chromium, is closed-source).

The -bin suffix is used for packages that use pre-built binaries while a source you can build from is also available (which would be used for the non-suffixed package).

Hypothetically, if Google opens-source Chrome tomorrow, this package would switch to compile from source, and a new google-chrome-bin could be made using the Google-provided binaries.

Is there a reason this isn't named google-chrome-bin?