[Bug 290471] NIRI causing a Fatal trap: page fault on FreeBSD 14.3 with NVIDIA drivers

https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=290471

--- Comment #6 from Tomoaki AOKI <[email protected]> ---
(In reply to mfoacs from comment #5)
Versions of graphics/drm-61-kmod and graphics/nvidia-drm-61-kmod-devel looks
matching OK. Not sure freebsd-update handles src distribution (src.txz, IIRC)
or not.

Now I suspect niri somehow directly calls Linux kernel functionality that
LinuxKPI on 14.3 doesn't have (yet), but gave up as niri was written in Rust
and a plenty of crates are used, unable to track for me.

Another thing that came in my mind is that any of crates that manages GPU
functionalities unmatches between upstream and the one specified in ports.

If niri is buildable in this case but crashes in some corner cases only,
the maintainer of x11-wm/niri (jbeich@) doesn't notice the issue in his test
and developer(s) of upstream niri could believe niri is 100% OK in niri side.

And this kind of possibilities makes me hate Rust, besides I like the language
itself or not. (Don't like as of its UNSAFE block and
non-shared-object-friendliness
by default, though.)

-- 
You are receiving this mail because:
You are the assignee for the bug.
You are on the CC list for the bug.