Re: [RFC patch 4/7] timekeeping: Add NMI safe accessor to mono/boot/real clocks

From: Linus Torvalds <torvalds@linux-foundation.org>
To: Steven Rostedt <rostedt@goodmis.org>
Cc: Thomas Gleixner <tglx@linutronix.de>,
	LKML <linux-kernel@vger.kernel.org>,
	Prarit Bhargava <prarit@redhat.com>,
	Mark Salyzyn <salyzyn@android.com>,
	Petr Mladek <pmladek@suse.com>, Ingo Molnar <mingo@kernel.org>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Peter Zijlstra <peterz@infradead.org>,
	Andrew Morton <akpm@linux-foundation.org>,
	Sergey Senozhatsky <sergey.senozhatsky@gmail.com>,
	Joe Perches <joe@perches.com>
Subject: Re: [RFC patch 4/7] timekeeping: Add NMI safe accessor to mono/boot/real clocks
Date: Fri, 17 Nov 2017 15:12:33 -0800	[thread overview]
Message-ID: <CA+55aFzFnQ9n=kfLCnq_mmDD3aqb+Uy_BHy9234UxCYTPPqBig@mail.gmail.com> (raw)
In-Reply-To: <20171117180034.30413b64@gandalf.local.home>

On Fri, Nov 17, 2017 at 3:00 PM, Steven Rostedt <rostedt@goodmis.org> wrote:
>
> Could you indent this better. It looks like timekeeping_delta_to_ns()
> takes 4 arguments, when it really only takes two.

Actually, even better would be to simply not have that ugly three-line
expression from hell.

Maybe it could even be a helper inline function, somethin glike

   static inline u64 tk_delta_ns(struct tk_read_base *tkr)
   {
        u64 cycle = tk_clock_read(tkr);
        u64 delta= clocksource_delta(cycle, tkr->cycle_last, tkr->mask);
        return timekeeping_delta_to_ns(tkr, cycle);
    }

and that whole thing would just become

        now += tk_delta_ns(tkr);

which looks a whole lot more legible. And shouldn't generate any worse code.

(I might have screwed up something, I wrote that in the email client,
maybe I got something wrong.

And my name choice is probably garbage. But you get the idea.

            Linus