Re: [tip: x86/fpu] x86/fpu/xstate: Define new functions for clearing fpregs and xstates

[Thomas Gleixner <tglx@xxxxxxxxxxxxx>]

On Tue, May 25 2021 at 10:44, Yu-cheng Yu wrote:
> On 5/24/2021 9:34 AM, Andy Lutomirski wrote:
>> So I'm guessing that syzbot may have misattributed the problem.  But
>> we definitely need to clean up the XRSTOR #GP handling before CET
>> lands.
>> 
>  From the crash dump, the system is doing syscall_exit_to_user_mode() 
> for __x64_sys_futex().  The futex syscall does not seem to modify 
> xstates,

Of course does the futex syscall not modify anything, but the task can
schedule out before returning from the syscall so it has to restore the
FPU state.

> but upon returning to user mode, XRSTORS gets a GP.  Can this 
> be some memory corruption?  fpu__clear() is merely helping to clear the 
> mess and seems to be innocent.

What kind of analysis is that?

Thanks,

        tglx