On Fri, Aug 27, 2021 at 11:37:34AM -0700, Yu, Yu-cheng wrote:

Right now, the kernel does lazy restore, and it waits until right before a
task goes back to ring-3 to restore xstates.  If a task needs to write to
any xstate registers before that (e.g. for signals), it restores the whole
xstates first and clears TIF_NEED_FPU_LOAD, which will prevent xstates being
restored again later.

shstk_setup() allocates a shadow stack for the task and puts its pointer
in MSR_IA32_PL3_SSP.

What does that have to do with a task having to write any xstate
registers?

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette