linux-kernel - Re: [PATCH v2 14/15] drivers: thermal: tsens: Create function to return sign-extended temperature

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>] [day] [month] [year] [list]

Message-ID: <5d66a168.1c69fb81.570fd.ae07@mx.google.com>
Date:   Wed, 28 Aug 2019 08:44:39 -0700
From:   Stephen Boyd <swboyd@...omium.org>
To:     Amit Kucheria <amit.kucheria@...aro.org>
Cc:     Daniel Lezcano <daniel.lezcano@...aro.org>,
        Mark Rutland <mark.rutland@....com>,
        Rob Herring <robh+dt@...nel.org>,
        Zhang Rui <rui.zhang@...el.com>,
        Andy Gross <agross@...nel.org>,
        Bjorn Andersson <bjorn.andersson@...aro.org>,
        Eduardo Valentin <edubezval@...il.com>,
        linux-arm-msm <linux-arm-msm@...r.kernel.org>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        Marc Gonzalez <marc.w.gonzalez@...e.fr>,
        Brian Masney <masneyb@...tation.org>,
        Linux PM list <linux-pm@...r.kernel.org>
Subject: Re: [PATCH v2 14/15] drivers: thermal: tsens: Create function to return sign-extended temperature

Quoting Amit Kucheria (2019-08-28 03:35:28)
> (Resending, replied only to Stephen by mistake)
> 
> On Wed, Aug 28, 2019 at 6:08 AM Stephen Boyd <swboyd@...omium.org> wrote:
> >
> > Quoting Amit Kucheria (2019-08-27 05:14:10)
> > > @@ -310,6 +328,10 @@ int __init init_common(struct tsens_priv *priv)
> > >                         goto err_put_device;
> > >                 }
> > >         }
> > > +
> > > +       /* Save away resolution of signed temperature value for this IP */
> > > +       priv->tempres = priv->fields[LAST_TEMP_0].msb - priv->fields
> [LAST_TEMP_0].lsb;
> > > +
> >
> > Why not just calculate this in the function that uses it? Is there a
> > reason to stash it away in the struct?
> 
> To avoid recalculating in an often-called function. It doesn't change for an IP
> version.
> 
> We can't make it static either inside that function since the initializer isn't
> constant.
> 

This sounds like a super micro optimization. It's a couple derefs and a
subtraction. If it isn't used anywhere else please just move it into the
function where it's used.