linux-kernel - Re: [PATCH 4/6] perf: Optimize get_recursion

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 30 Oct 2020 16:22:48 -0400
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Jesper Dangaard Brouer <brouer@...hat.com>
Cc:     Peter Zijlstra <peterz@...radead.org>, mingo@...nel.org,
        tglx@...utronix.de, linux-kernel@...r.kernel.org,
        kan.liang@...ux.intel.com, acme@...nel.org, mark.rutland@....com,
        alexander.shishkin@...ux.intel.com, jolsa@...hat.com,
        namhyung@...nel.org, ak@...ux.intel.com, eranian@...gle.com
Subject: Re: [PATCH 4/6] perf: Optimize get_recursion_context()

On Fri, 30 Oct 2020 18:11:38 +0100
Jesper Dangaard Brouer <brouer@...hat.com> wrote:

> On Fri, 30 Oct 2020 16:13:49 +0100
> Peter Zijlstra <peterz@...radead.org> wrote:
> 
> >   "Look ma, no branches!"
> > 
> > Cc: Jesper Dangaard Brouer <brouer@...hat.com>
> > Cc: Steven Rostedt <rostedt@...dmis.org>
> > Signed-off-by: Peter Zijlstra (Intel) <peterz@...radead.org>
> > ---  
> 
> Cool trick! :-)
> 
> Acked-by: Jesper Dangaard Brouer <brouer@...hat.com>
> 
> >  kernel/events/internal.h |   17 ++++++++---------
> >  1 file changed, 8 insertions(+), 9 deletions(-)
> > 
> > --- a/kernel/events/internal.h
> > +++ b/kernel/events/internal.h
> > @@ -205,16 +205,15 @@ DEFINE_OUTPUT_COPY(__output_copy_user, a
> >  
> >  static inline int get_recursion_context(int *recursion)
> >  {
> > -	int rctx;
> > +	unsigned int pc = preempt_count();
> > +	unsigned int rctx = 0;
> >  
> > -	if (unlikely(in_nmi()))
> > -		rctx = 3;
> > -	else if (in_irq())
> > -		rctx = 2;
> > -	else if (in_serving_softirq())
> > -		rctx = 1;
> > -	else
> > -		rctx = 0;
> > +	if (pc & (NMI_MASK))
> > +		rctx++;
> > +	if (pc & (NMI_MASK | HARDIRQ_MASK))
> > +		rctx++;
> > +	if (pc & (NMI_MASK | HARDIRQ_MASK | SOFTIRQ_OFFSET))
> > +		rctx++;
> >  
> >  	if (recursion[rctx])
> >  		return -1;
> > 
> > 

As this is something that ftrace recursion also does, perhaps we should
move this into interrupt.h so that anyone that needs a counter can get
it quickly, and not keep re-implementing it.

/*
 * Quickly find what context you are in.
 * 0 - normal
 * 1 - softirq
 * 2 - hard interrupt
 * 3 - NMI
 */
static inline int irq_context()
{
	unsigned int pc = preempt_count();
	int rctx = 0;

	if (pc & (NMI_MASK))
		rctx++;
	if (pc & (NMI_MASK | HARDIRQ_MASK))
		rctx++;
	if (pc & (NMI_MASK | HARDIRQ_MASK | SOFTIRQ_OFFSET))
		rctx++;

	return rctx;
}

-- Steve