On Tue, Jun 23, 2020 at 11:45:19AM +0200, Joerg Roedel wrote:

Hi Andy,

On Mon, Apr 27, 2020 at 10:37:41AM -0700, Andy Lutomirski wrote:

quoted

1. Use IST for #VC and deal with all the mess that entails.

With the removal of IST shifting I wonder what you would suggest on how
to best implement an NMI-safe IST handler with nesting support.

My current plan is to implement an IST handler which switches itself off
the IST stack as soon as possible, freeing it for re-use.

The flow would be roughly like this upon entering the handler;

	build_pt_regs();

	RSP = pt_regs->sp;

	if (RSP in VC_IST_stack)
		error("unallowed nesting")

	if (RSP in current_kernel_stack)
		RSP = round_down_to_8(RSP)
	else
		RSP = current_top_of_stack() // non-ist kernel stack

	copy_pt_regs(pt_regs, RSP);
	switch_stack_to(RSP);

To make this NMI safe, the NMI handler needs some logic too. Upon
entering NMI, it needs to check the return RSP, and if it is in the #VC
IST stack, it must do the above flow by itself and update the return RSP
and RIP. It needs to take into account the case when PT_REGS is not
fully populated on the return side.

Alternativly the NMI handler could safe/restore the contents of the #VC
IST stack or just switch to a special #VC-in-NMI IST stack.

All in all it could get complicated, and imho shift_ist would have been
simpler, but who am I anyway...

Or maybe you have a better idea how to implement this, so I'd like to
hear your opinion first before I spend too many days implementing
something.

OK, excuse my ignorance, but I'm not seeing how that IST shifting
nonsense would've helped in the first place.

If I understand correctly the problem is:

	<#VC>
	  shift IST
	  <NMI>
	    ... does stuff
	    <#VC> # again, safe because the shift

But what happens if you get the NMI before your IST adjustment?

	<#VC>
	  <NMI>
	    ... does stuff
	    <#VC> # again, happily wrecks your earlier #VC
	  shift IST # whoopsy, too late

Either way around we get to fix this up in NMI (and any other IST
exception that can happen while in #VC, hello #MC). And more complexity
there is the very last thing we need :-(

There's no way you can fix up the IDT without getting an NMI first.

This entire exception model is fundamentally buggered :-/