Re: [RFC][PATCH 4/7] x86/debug: Move historical SYSENTER junk into exc_debug_kernel()

[Andy Lutomirski]

> On Aug 20, 2020, at 9:35 AM, peterz@xxxxxxxxxxxxx wrote:
> 
> On Thu, Aug 20, 2020 at 11:17:29AM -0500, Josh Poimboeuf wrote:
>> On Thu, Aug 20, 2020 at 05:21:11PM +0200, peterz@xxxxxxxxxxxxx wrote:
> 
>>> qemu-gdb stub should eat the event before it lands in the guest
>> 
>> Are we sure about that?  I triggered the warning just now, stepping
>> through the debug handler.
> 
> Not sure at all, that's what seemed reasonable, but since when did that
> stop virt ;-)
> 
> Also, can you trigger that same with upstream? Because I suspect this
> ought to tickle the old code too.

I’ve lost track of how many bugs QEMU and KVM have in this space.  Let’s keep it as a warning, but a bug. But let’s get rid of the totally bogus TIF_SINGLESTEP manipulation.