linux-kernel - Re: [PATCH v2 2/2] drivers: bus: simple-pm-bus: Add support for probing simple bus only devices

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]

Message-ID: <CAPDyKFr7z6W3ghoAmTVKHaM8saycbhYyBcDXaTuT_Daz+djx2A@mail.gmail.com>
Date:   Fri, 3 Sep 2021 11:14:31 +0200
From:   Ulf Hansson <ulf.hansson@...aro.org>
To:     Saravana Kannan <saravanak@...gle.com>
Cc:     Rob Herring <robh+dt@...nel.org>,
        Frank Rowand <frowand.list@...il.com>,
        Geert Uytterhoeven <geert+renesas@...der.be>,
        Android Kernel Team <kernel-team@...roid.com>,
        Linux Kernel Mailing List <linux-kernel@...r.kernel.org>,
        DTML <devicetree@...r.kernel.org>
Subject: Re: [PATCH v2 2/2] drivers: bus: simple-pm-bus: Add support for
 probing simple bus only devices

On Fri, 3 Sept 2021 at 01:04, Saravana Kannan <saravanak@...gle.com> wrote:
>
> The OF platform code sets up simple bus only devices (i.e. devices that
> won't match with any other driver) to get probed by the simple-pm-bus to
> keep fw_devlink from blocking probe() or sync_state() [1] callbacks of
> other devices. There's no need to populate the child devices since the
> OF platform code would do that anyway, so return early for these simple
> bus only devices.

This looks like a neat solution to our problem. Although, a few comments below.

>
> [1] - https://lore.kernel.org/lkml/CAPDyKFo9Bxremkb1dDrr4OcXSpE0keVze94Cm=zrkOVxHHxBmQ@mail.gmail.com/
> Signed-off-by: Saravana Kannan <saravanak@...gle.com>
> Tested-by: Saravana Kannan <saravanak@...gle.com>
> ---
>  drivers/bus/simple-pm-bus.c | 7 +++++++
>  1 file changed, 7 insertions(+)
>
> diff --git a/drivers/bus/simple-pm-bus.c b/drivers/bus/simple-pm-bus.c
> index 01a3d0cd08ed..91d52021b7f9 100644
> --- a/drivers/bus/simple-pm-bus.c
> +++ b/drivers/bus/simple-pm-bus.c
> @@ -19,6 +19,13 @@ static int simple_pm_bus_probe(struct platform_device *pdev)
>         const struct of_dev_auxdata *lookup = dev_get_platdata(&pdev->dev);
>         struct device_node *np = pdev->dev.of_node;
>
> +       /*
> +        * These are transparent bus devices (not simple-pm-bus matches) that
> +        * get populated automatically.  So, don't need to do anything more.
> +        */
> +       if (pdev->driver_override)
> +               return 0;

You need the same kind of check in simple_pm_bus_remove(), to avoid
pm_runtime_disable(). At least for consistency.

> +
>         dev_dbg(&pdev->dev, "%s\n", __func__);
>
>         pm_runtime_enable(&pdev->dev);
> --
> 2.33.0.153.gba50c8fa24-goog
>

It also looks like we should flip the order of the patches in the
series, to keep things bisectable.

Kind regards
Uffe