Re: [REGRESSION][BISECTED] x86: kernel 6.12 crashes during get_xsave_addr_user on Apple Virtualization

[Ingo Molnar <mingo@kernel.org>]

* Dave Hansen <dave.hansen@intel.com> wrote:

> On 3/11/25 07:35, Akihiro Suda wrote:
> > (...)
> > [    0.477064] Run /init as init process
> > [    0.478585] ------------[ cut here ]------------
> > [    0.478747] WARNING: CPU: 0 PID: 1 at
> > arch/x86/kernel/fpu/xstate.c:1003 get_xsave_addr_user+0x28/0x40
> > [    0.478893] Modules linked in:
> > [    0.478999] CPU: 0 UID: 0 PID: 1 Comm: init Not tainted
> > 6.11.0-rc1-00007-g70044df250d0 #15
> > [    0.479148] RIP: 0010:get_xsave_addr_user+0x28/0x40
> 
> The warning here seems to be because the kernel thinks protections keys
> is enabled (X86_FEATURE_OSPKE) but also thinks XSAVE management of
> protection keys is _dis_abled. That's a valid hardware configuration,
> but it's a configuration the kernel never expects to see.
> 
> New kernels care about this more because we effectively have new code
> that does:
> 
> 	if (cpu_feature_enabled(X86_FEATURE_OSPKE))
> 		get_xsave_addr_user(XFEATURE_PKRU, ...)
> 
> Is there a chance that you have a _full_ dmesg from one of these
> systems? I'm looking for the "x86/fpu: xstate_offset..." lines. It would
> also be interesting to see a CPUID dump on a booted system from before
> the patch that started causing issues and maybe even on one with one of
> the workarounds applied. You can get that with:
> 
> 	cpuid -1 --raw

I'll hold off on the fresh x86/urgent commit I applied earlier today 
until this bug is better understood.

Thanks,

	Ingo