linux-kernel - Re: [PATCH] pwm: tegra: dynamic clk freq configuration by PWM driver

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

Date:   Sat, 18 Apr 2020 10:23:00 +0200
From:   Uwe Kleine-König 
        <u.kleine-koenig@...gutronix.de>
To:     Sandipan Patra <spatra@...dia.com>
Cc:     Thierry Reding <treding@...dia.com>,
        "robh+dt@...nel.org" <robh+dt@...nel.org>,
        Jonathan Hunter <jonathanh@...dia.com>,
        Bibek Basu <bbasu@...dia.com>,
        Laxman Dewangan <ldewangan@...dia.com>,
        "linux-pwm@...r.kernel.org" <linux-pwm@...r.kernel.org>,
        "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
        "linux-tegra@...r.kernel.org" <linux-tegra@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] pwm: tegra: dynamic clk freq configuration by PWM driver

Hello,

On Fri, Apr 17, 2020 at 02:53:22PM +0000, Sandipan Patra wrote:
> > To put my expression in words: pick the maximum of the possible periods that
> > are less or equal to the requested value.  Maybe this is better
> > understandable:
> > 
> >         max { x ∊ implementablePeriods | x <= requestedPeriod }
> > 
> > ?
> 
> I think I got your question.
> Should tegra_pwm_config() not return error (EINVAL) when the requested period is
> invalid but it should configure to a nearest possible value?

If you cannot configure according to the above rule, yes, return an
error code. EINVAL is the usual one I think (some also return ERANGE).

> > > Yes, the output stops as soon as the PWM_ENABLE bit is cleared in
> > > hardware. Then The output is set to 0 (which is inactive).
> > > Once .disable() => tegra_pwm_disable() gets invoked, enable bit is
> > > cleared and hence PWM will possess no output signal.
> > > tegra_pwm_config() will be invoked for any new configuration request.
> > 
> > Some drivers already have a "Limitations" section in their header.
> > Please take a look at the existing examples and provide something similar. (Note
> > you still didn't answer "How does a running PWM behave when the register is
> > updated? Does it complete the currently running period?". I assume the answer
> > to the second question is "No" (and the first is only there for rhetoric reasons).)
> >
>  
> 1. I will add the below comments as Limitations:
> -	When PWM is disabled, the output is driven to 0 and

In fact, this is a good property. So the only problem is, that for both
stop and reconfiguration the currently running period isn't completed.

Best regards
Uwe

-- 
Pengutronix e.K.                           | Uwe Kleine-König            |
Industrial Linux Solutions                 | https://www.pengutronix.de/ |