linux-kernel - Re: [PATCH v4 1/3] pinctrl: Add sleep related configuration

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:   Wed, 12 Jul 2017 14:30:25 +0200
From:   Linus Walleij <linus.walleij@...aro.org>
To:     Rob Herring <robh@...nel.org>,
        Linus Walleij <linus.walleij@...aro.org>,
        Mark Rutland <mark.rutland@....com>,
        "linux-gpio@...r.kernel.org" <linux-gpio@...r.kernel.org>,
        "devicetree@...r.kernel.org" <devicetree@...r.kernel.org>,
        "linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
        Mark Brown <broonie@...nel.org>,
        Baolin Wang <baolin.wang@...aro.org>
Subject: Re: [PATCH v4 1/3] pinctrl: Add sleep related configuration

On Tue, Jun 27, 2017 at 2:06 PM, Baolin Wang <baolin.wang@...eadtrum.com> wrote:

> If we introduce "sleep-input-enable" config, we can set the pin's config
> as below:
>
> vio_sd0_ms_3: regctrl3 {
>         pins = "SC9860_RFCTL30", "SC9860_RFCTL31", "SC9860_RFCTL32";
>         function = "func1";
>         sprd,sleep-mode = <0x3>;
>         sleep-input-enable;
> };

This looks like a "default" mode. Is that correct?

I.e. do you set up this on probe then do not touch it?

It seems some of the problems come from the insistance to use a single
node for all configuration. Compare to this nomadik:

               i2c0 {
                        i2c0_default_mux: i2c0_mux {
                                i2c0_default_mux {
                                        function = "i2c0";
                                        groups = "i2c0_a_1";
                                };
                        };
                        i2c0_default_mode: i2c0_default {
                                i2c0_default_cfg {
                                        pins = "GPIO62_D3", "GPIO63_D2";
                                        input-enable;
                                };
                        };
                };

It is easy to imagine:

               i2c0 {
                        i2c0_default_mux: i2c0_mux {
                                i2c0_default_mux {
                                        function = "i2c0";
                                        groups = "i2c0_a_1";
                                };
                        };
                        i2c0_default_mode: i2c0_default {
                                i2c0_default_cfg {
                                        pins = "GPIO62_D3", "GPIO63_D2";
                                        input-enable;
                                };
                        };
                        i2c0_default_mode_sleep: i2c0_default_sleep {
                                i2c0_default_cfg {
                                        pins = "GPIO62_D3", "GPIO63_D2";
                                        sleep-hardware-state;
                                        input-disable;
                                };
                        };
                };

Notice the new bool property "sleep-hardware-state" that just
indicate that this should be programmed into the registers for
the sleep state.

> But If we create one extra "sleep-xxx" state for sleep-related configs,
> it will be like:
>
> grp1: regctrl3 {
>         pins = "SC9860_RFCTL30", "SC9860_RFCTL31";
>         function = "func1";
>         sprd,sleep-mode = <0x3>;
> };
>
> sleep-input: input_grp {
>         pins = "SC9860_RFCTL30", "SC9860_RFCTL31", "SC9860_RFCTL32";
>         input-enable;
> };
>
> pinctrl-names = "sleep-input";
> pinctrl-0 = <&sleep-input>;
>
> "sleep-input" state will be selected when initializing pinctrl driver,

The state you should use for initial configuration should be called
just "init".

> "grp1"
> will be selected by user to set other pin configuration.

Like "default"?

> Then we need config "SC9860_RFCTL30" pin in 2 different places, which is
> more inconvenient for users.

I'm not so sure about that. Having a lot more sleep,* config options
may be even more inconvenient for users, and especially for the
community of developers as a whole.

Several config nodes on the other hand, we have had in the pin
control subsystem since day 1.

Yours,
Linus Walleij