On 2022/03/02 13:01, Mikhail wrote:
> On latest snapshot I see Xorg crash when trying to launch chrome browser
> (firefox works fine), I can reproduce it with 'intel' driver,
> 'modesetting' crashed for me too, but it took about two weeks of uptime
> and wasn't related to chrome launch (no logs for 'modesetting' crash
> unfortunately, but it was something about not being able to access
> address 0x0, this 'modesettnig' crash lead me to changing the driver to
> 'intel', that's how original crash was found, no idea how long it is
> there).

I don't think there's very much in this report that will help tracking
it down, see "How to get a core file out of the X server?" in
https://github.com/openbsd/xenocara/blob/master/README

Though I think there will be little interest in a problem which is
affecting the 'intel' driver on hardware that defaults to using
modsetting.

> vm.malloc_conf=CFGS

imho these flags should only be set if you're willing to experience more
crashes and be able/willing to do the legwork to provide backtraces etc.

> hw.smt=1

this is really not recommended.

> kern.allowkmem=1

also this unless you're actively debugging something which needs it

Reply via email to