Re: [PATCH RFC] x86/bus_lock: Enable bus lock detection

[Fenghua Yu]

Hi, Peter,

On Wed, Jul 29, 2020 at 10:49:47AM +0200, peterz@xxxxxxxxxxxxx wrote:
> On Fri, Jul 17, 2020 at 02:35:00PM -0700, Fenghua Yu wrote:
> 
> > #DB for bus lock detect fixes all issues in #AC for split lock detect:
> > 1) It's architectural ... just need to look at one CPUID bit to know it
> >    exists
> > 2) The IA32_DEBUGCTL MSR, which reports bus lock in #DB, is per-thread.
> >    So each process or guest can have different behavior.
> 
> And it generates a whole new problem due to #DB being an IST, and
> 
> > diff --git a/arch/x86/kernel/traps.c b/arch/x86/kernel/traps.c
> > index b038695f36c5..58725567da39 100644
> > --- a/arch/x86/kernel/traps.c
> > +++ b/arch/x86/kernel/traps.c
> > @@ -812,6 +812,16 @@ static void handle_debug(struct pt_regs *regs, unsigned long dr6, bool user)
> >  	if (!user && !dr6)
> >  		return;
> >  
> > +	/* Handle bus lock. */
> > +	if (!(dr6 & DR_BUS_LOCK)) {
> > +		cond_local_irq_enable(regs);
> > +		if (user)
> > +			handle_user_bus_lock(regs);
> > +		else
> > +			handle_kernel_bus_lock(regs);
> > +		goto out;
> > +	}
> > +
> >  	/*
> >  	 * If dr6 has no reason to give us about the origin of this trap,
> >  	 * then it's very likely the result of an icebp/int01 trap.
> 
> we very much rely on #DB never recursing, which we carefully crafted by
> disallowing hardare breakpoints on noinstr code and clearing DR7 early.
> 
> But now it can... please keep the pieces.

Can we disable Bus Lock Detection before handle it and re-enable it
after handle it? Will that resolve the recursion issue?

Thanks.

-Fenghua