On 10/1/2021 2:29 PM, Thomas Gleixner wrote:

So we'd end up with two XSAVES on context switch. We can simply do:
         XSAVES();
         fpu.state.xtsate.uintr.uinv = 0;

I am a bit confused. Do we need to set UINV to 0 explicitly?

If XSAVES gets called twice during context switch then the UINV in the 
XSTATE buffer automatically gets set to 0. Since XSAVES saves the 
current UINV value in the MISC_MSR which was already set to 0 by the 
previous XSAVES.

Though, this probably happens due to pure luck than intentional design :)

which allows to do as many XRSTORS in a row as we want. Only the final
one on the way to user space will have to restore the real vector if the
register state is not valid:

        if (fpu_state_valid()) {
             if (needs_uinv(current)
                wrmsrl(UINV, vector);
        } else {
             if (needs_uinv(current)
                fpu.state.xtsate.uintr.uinv = vector;
             XRSTORS();
        }

I might have missed some subtle difference. Has this logic changed from 
what you previously suggested for arch_exit_to_user_mode_prepare()?

        if (xrstors_pending)) {
             // Update the saved xstate for xrstors
             // Unconditionally update the UINV since it could have been 
overwritten by calling XSAVES twice.
                current->xstate.uintr.uinv = UINTR_NOTIFICATION_VECTOR;
                 current->xstate.uintr.uirr |= pir;
         } else {
                 // Manually restore UIRR and UINV
                 rdmsrl(IA32_UINTR_RR, uirr);
                 wrmsrl(IA32_UINTR_RR, uirr | pir);

             misc.val64 = 0;
                 misc.uittsz = current->uintr->uittsz;
                 misc.uinv = UINTR_NOTIFICATION_VECTOR;
                 wrmsrl(IA32_UINTR_MISC, misc.val64);
         }

Hmm?

The one case I can see this failing is if there was another XRSTORS 
after the "final" restore in arch_exit_to_user_mode_prepare()? I think 
that is not possible but I am not an expert on this. Did I misunderstand 
something?

Thanks,
Sohil