Re: [PATCH 01/12] perf/breakpoint: Split attribute parse and commit

From: Michael Ellerman <mpe@ellerman.id.au>
To: Frederic Weisbecker <frederic@kernel.org>
Cc: LKML <linux-kernel@vger.kernel.org>, Jiri Olsa <jolsa@redhat.com>,
	Namhyung Kim <namhyung@kernel.org>,
	Joel Fernandes <joel.opensrc@gmail.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Linus Torvalds <torvalds@linux-foundation.org>,
	Yoshinori Sato <ysato@users.sourceforge.jp>,
	Benjamin Herrenschmidt <benh@kernel.crashing.org>,
	Catalin Marinas <catalin.marinas@arm.com>,
	Chris Zankel <chris@zankel.net>,
	Paul Mackerras <paulus@samba.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Will Deacon <will.deacon@arm.com>, Rich Felker <dalias@libc.org>,
	Ingo Molnar <mingo@kernel.org>,
	Mark Rutland <mark.rutland@arm.com>,
	Alexander Shishkin <alexander.shishkin@linux.intel.com>,
	Andy Lutomirski <luto@kernel.org>,
	Arnaldo Carvalho de Melo <acme@kernel.org>,
	Max Filippov <jcmvbkbc@gmail.com>
Subject: Re: [PATCH 01/12] perf/breakpoint: Split attribute parse and commit
Date: Mon, 28 May 2018 21:29:59 +1000	[thread overview]
Message-ID: <87lgc43tmw.fsf@concordia.ellerman.id.au> (raw)
In-Reply-To: <20180525135846.GC22082@lerouge>

Frederic Weisbecker <frederic@kernel.org> writes:

> On Thu, May 24, 2018 at 11:56:01AM +1000, Michael Ellerman wrote:
>> Frederic Weisbecker <frederic@kernel.org> writes:
>> 
>> > diff --git a/kernel/events/hw_breakpoint.c b/kernel/events/hw_breakpoint.c
>> > index 6e28d28..51320c2 100644
>> > --- a/kernel/events/hw_breakpoint.c
>> > +++ b/kernel/events/hw_breakpoint.c
>> > @@ -424,19 +443,22 @@ static int validate_hw_breakpoint(struct perf_event *bp)
>> >  
>> >  int register_perf_hw_breakpoint(struct perf_event *bp)
>> >  {
>> > -	int ret;
>> > +	struct arch_hw_breakpoint hw;
>> > +	int err;
>> >  
>> > -	ret = reserve_bp_slot(bp);
>> > -	if (ret)
>> > -		return ret;
>> > +	err = reserve_bp_slot(bp);
>> > +	if (err)
>> > +		return err;
>> >  
>> > -	ret = validate_hw_breakpoint(bp);
>> > -
>> > -	/* if arch_validate_hwbkpt_settings() fails then release bp slot */
>> > -	if (ret)
>> > +	err = hw_breakpoint_parse(bp, &bp->attr, &hw);
>> 
>> Is there a good reason we pass bp and bp->attr? (I assume so)
>> 
>> That added to the confusion in the existing code I think.
>
> Yes, on breakpoint creation (which is the above function) it's not needed
> but breakpoint modification wants it as we need to pass the attr that are
> to be validated, and those are not yet copied to the breakpoint at this
> stage. This happens in the end of the series.

OK thanks.

cheers