linux-kernel - Re: [PATCH] clk: Add support for enabling/disabling clocks from debugfs

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: <159558161499.3847286.16032712002329424951@swboyd.mtv.corp.google.com>
Date:   Fri, 24 Jul 2020 02:06:54 -0700
From:   Stephen Boyd <sboyd@...nel.org>
To:     Mike Tipton <mdtipton@...eaurora.org>, mturquette@...libre.com
Cc:     linux-clk@...r.kernel.org, linux-kernel@...r.kernel.org,
        tdas@...eaurora.org, Mike Tipton <mdtipton@...eaurora.org>
Subject: Re: [PATCH] clk: Add support for enabling/disabling clocks from debugfs

Quoting Mike Tipton (2020-06-29 17:30:24)
> For test and debug purposes, it's often necessary to enable or disable
> clocks from shell. Add a new debugfs file (clk_prepare_enable) that
> calls clk_prepare_enable() when writing "1" and clk_disable_unprepare()
> when writing "0".
> 
> This can have security implications, so only support it when the code
> has been modified to #define CLOCK_ALLOW_WRITE_DEBUGS.

DEBUGFS

> 
> Signed-off-by: Mike Tipton <mdtipton@...eaurora.org>
> ---
>  drivers/clk/clk.c | 38 ++++++++++++++++++++++++++++++++++----
>  1 file changed, 34 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c
> index 236923b25543..3640bbc56870 100644
> --- a/drivers/clk/clk.c
> +++ b/drivers/clk/clk.c
> @@ -3034,13 +3034,14 @@ static int clk_dump_show(struct seq_file *s, void *data)
>  }
>  DEFINE_SHOW_ATTRIBUTE(clk_dump);
>  
> -#undef CLOCK_ALLOW_WRITE_DEBUGFS
> -#ifdef CLOCK_ALLOW_WRITE_DEBUGFS

This didn't need to change.

>  /*
> - * This can be dangerous, therefore don't provide any real compile time
> - * configuration option for this feature.
> + * These features can be dangerous, therefore don't provide any real compile
> + * time configuration option for them.
>   * People who want to use this will need to modify the source code directly.
>   */
> +#undef CLOCK_ALLOW_WRITE_DEBUGFS
> +#ifdef CLOCK_ALLOW_WRITE_DEBUGFS

So I put it back and dropped your rewording.

> +
>  static int clk_rate_set(void *data, u64 val)
>  {
>         struct clk_core *core = data;
> @@ -3054,6 +3055,31 @@ static int clk_rate_set(void *data, u64 val)
>  }
>  
>  #define clk_rate_mode  0644
> +
> +static int clk_prepare_enable_set(void *data, u64 val)
> +{
> +       struct clk_core *core = data;
> +       int ret = 0;
> +
> +       if (val)
> +               ret = clk_prepare_enable(core->hw->clk);
> +       else
> +               clk_disable_unprepare(core->hw->clk);
> +
> +       return ret;
> +}
> +
> +static int clk_prepare_enable_get(void *data, u64 *val)
> +{
> +       struct clk_core *core = data;
> +
> +       *val = core->enable_count && core->prepare_count;
> +       return 0;
> +}
> +
> +DEFINE_DEBUGFS_ATTRIBUTE(clk_prepare_enable_fops, clk_prepare_enable_get,
> +                        clk_prepare_enable_set, "%llu\n");
> +
>  #else
>  #define clk_rate_set   NULL
>  #define clk_rate_mode  0444
> @@ -3231,6 +3257,10 @@ static void clk_debug_create_one(struct clk_core *core, struct dentry *pdentry)
>         debugfs_create_u32("clk_notifier_count", 0444, root, &core->notifier_count);
>         debugfs_create_file("clk_duty_cycle", 0444, root, core,
>                             &clk_duty_cycle_fops);
> +#ifdef CLOCK_ALLOW_WRITE_DEBUGFS

This is ugly but I guess we can't do much.

> +       debugfs_create_file("clk_prepare_enable", 0644, root, core,
> +                           &clk_prepare_enable_fops);
> +#endif
>  

Applied to clk-next.