Re: [patch V4 part 4 15/24] x86/db: Split out dr6/7 handling

[Peter Zijlstra]

On Thu, May 07, 2020 at 07:18:45PM +0200, Alexandre Chartre wrote:
> 
> On 5/5/20 3:49 PM, Thomas Gleixner wrote:
> > From: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> > 
> > DR6/7 should be handled before nmi_enter() is invoked and restore after
> > nmi_exit() to minimize the exposure.
> > 
> > Split it out into helper inlines and bring it into the correct order.
> > 
> > Signed-off-by: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
> > Signed-off-by: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
> > ---
> >   arch/x86/kernel/hw_breakpoint.c |    6 ---
> >   arch/x86/kernel/traps.c         |   62 +++++++++++++++++++++++++++-------------
> >   2 files changed, 44 insertions(+), 24 deletions(-)
> > 
> ...
> > --- a/arch/x86/kernel/traps.c
> > +++ b/arch/x86/kernel/traps.c
> > @@ -691,6 +691,44 @@ static bool is_sysenter_singlestep(struc
> >   #endif
> >   }
> > +static __always_inline void debug_enter(unsigned long *dr6, unsigned long *dr7)
> > +{
> > +	/*
> > +	 * Disable breakpoints during exception handling; recursive exceptions
> > +	 * are exceedingly 'fun'.
> > +	 *
> > +	 * Since this function is NOKPROBE, and that also applies to
> > +	 * HW_BREAKPOINT_X, we can't hit a breakpoint before this (XXX except a
> > +	 * HW_BREAKPOINT_W on our stack)
> > +	 *
> > +	 * Entry text is excluded for HW_BP_X and cpu_entry_area, which
> > +	 * includes the entry stack is excluded for everything.
> > +	 */
> > +	get_debugreg(*dr7, 6);
> 
> Do you mean  get_debugreg(*dr7, 7); ?

Shees, I have to go buy a new stack of brown paper bags at this rate,
don't I :/