linux-kernel - Re: [PATCH 2/2] time: alarmtimer: Add the trcepoints for alarmtimer

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: Windows password security audit tool. GUI, reports in PDF.

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

Message-ID: <20160921093255.5b398d6c@gandalf.local.home>
Date:   Wed, 21 Sep 2016 09:32:55 -0400
From:   Steven Rostedt <rostedt@...dmis.org>
To:     Thomas Gleixner <tglx@...utronix.de>
Cc:     Baolin Wang <baolin.wang@...aro.org>,
        Alessandro Zummo <a.zummo@...ertech.it>,
        Alexandre Belloni <alexandre.belloni@...e-electrons.com>,
        Ingo Molnar <mingo@...hat.com>,
        John Stultz <john.stultz@...aro.org>,
        Mark Brown <broonie@...nel.org>,
        LKML <linux-kernel@...r.kernel.org>, rtc-linux@...glegroups.com
Subject: Re: [PATCH 2/2] time: alarmtimer: Add the trcepoints for alarmtimer

On Wed, 21 Sep 2016 14:36:23 +0200 (CEST)
Thomas Gleixner <tglx@...utronix.de> wrote:

> On Wed, 21 Sep 2016, Steven Rostedt wrote:
> > On Wed, 21 Sep 2016 09:26:20 +0200 (CEST)
> > Thomas Gleixner <tglx@...utronix.de> wrote:  
> > > A single u64 does not take more storage space than this and it's a single
> > > store.  
> > 
> > So to use rtc_tm_to_time64()? Is the work to do the calculations of the
> > conversion faster than a bunch of stores that are going to be in hot
> > cache?  
> 
> Look at the call site. It has already the scalar nsec value and it does a
> conversion to rtc time in order to trace it.

OK. I haven't looked at the callsite. I just did a quick look at the
patch as is, and noticed the wasted space in the buffer for storing a
bunch of ints that will never be bigger than 256.

> 
> Ditto for the other tracepoints where the conversion from scalar nsec is
> done in the tracepoint itself.

This is why I like to have the maintainers review the rest.

-- Steve