Bug 278569

Summary:

[GTK] Apple's Macbook Air page crashes WebKitGTK's web view process

Product:

WebKit

Reporter:

Jeff Fortin <nekohayo>

Component:

WebKitGTK

Assignee:

Nobody <webkit-unassigned>

Status:

RESOLVED DUPLICATE

Severity:

Critical

CC:

bugs-noreply, mcatanzaro, philn

Priority:

Version:

WebKit Nightly Build

Hardware:

OS:

Linux

Attachments:

Description	Flags
Output of about:gpu	none
Backtraces	none

Jeff Fortin

Reported 2024-08-23 11:53:55 PDT

Created attachment 472280 [details] Output of about:gpu On Fedora 40 running Wayland GNOME 46.x with Intel Kabylake graphics, this crash occurs 2 to 3 seconds after loading the page (no scrolling, just looking at the top of the page): $ /usr/libexec/webkitgtk-6.0/MiniBrowser https://www.apple.com/macbook-air/ MESA-INTEL: warning: ../src/intel/vulkan/anv_formats.c:763: FINISHME: support YUV colorspace with DRM format modifiers MESA-INTEL: warning: ../src/intel/vulkan/anv_formats.c:794: FINISHME: support more multi-planar formats with DRM modifiers (WebKitWebProcess:2): GStreamer-WARNING **: 14:45:58.556: External plugin loader failed. This most likely means that the plugin loader helper binary was not found or could not be run. You might need to set the GST_PLUGIN_SCANNER environment variable if your setup is unusual. This should normally not be required though. ** (MiniBrowser:11338): WARNING **: 14:45:59.885: WebProcess CRASHED --------- This happens on Epiphany Nightly (flatpak), Epiphany 46.3 from Fedora, and the minibrowser command above.

Attachments
Output of about:gpu (13.15 KB, text/plain) 2024-08-23 11:53 PDT, Jeff Fortin	no flags	Details
Backtraces (33.96 KB, text/plain) 2024-08-23 13:47 PDT, Jeff Fortin	no flags	Details
View All Add attachment proposed patch, testcase, etc.

Philippe Normand

Comment 1 2024-08-23 12:07:33 PDT

Gonna need crash backtrace... And I don't think this is GStreamer-related, the warning mentioned is not critical.

Jeff Fortin

Comment 2 2024-08-23 13:47:28 PDT

Created attachment 472281 [details] Backtraces Here is my attempt at getting backtraces by running the minibrowser command with gdb and navigating to the affected webpages. I ran a "bt full" and then a "thread apply all bt".

Philippe Normand

Comment 3 2024-08-24 00:49:01 PDT

that seems to be the MiniBrowser stack, not the WebProcess :)

Michael Catanzaro

Comment 4 2024-08-24 06:16:37 PDT

This will probably turn out to be bug #278569, but there's no way to know for sure until we have a web process backtrace. (In reply to Jeff Fortin from comment #2) > Here is my attempt at getting backtraces by running the minibrowser command > with gdb and navigating to the affected webpages. You're wasting your time. Just use 'coredumpctl gdb'.

Jeff Fortin

Comment 5 2024-08-24 06:24:24 PDT

As kdwk was able to reproduce this issue, it really sounds like bug #260455 then. *** This bug has been marked as a duplicate of bug 260455 ***

Note You need to log in before you can comment on or make changes to this bug.