linux-kernel - Re: [PATCH] drivers: firmware: psci: add support for warm reset

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, 3 Apr 2019 04:33:15 +0100
From:   Mark Rutland <mark.rutland@....com>
To:     Aaro Koskinen <aaro.koskinen@....fi>
Cc:     Lorenzo Pieralisi <lorenzo.pieralisi@....com>,
        linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        Aaro Koskinen <aaro.koskinen@...ia.com>
Subject: Re: [PATCH] drivers: firmware: psci: add support for warm reset

On Mon, Apr 01, 2019 at 09:14:43PM +0300, Aaro Koskinen wrote:
> From: Aaro Koskinen <aaro.koskinen@...ia.com>
> 
> Add support for warm reset using SYSTEM_RESET2 introduced in PSCI
> 1.1 specification.

For reference, how do you intend to use this?

> 
> Signed-off-by: Aaro Koskinen <aaro.koskinen@...ia.com>
> ---
>  drivers/firmware/psci.c   | 25 +++++++++++++++++++++++++
>  include/uapi/linux/psci.h |  3 +++
>  2 files changed, 28 insertions(+)
> 
> diff --git a/drivers/firmware/psci.c b/drivers/firmware/psci.c
> index c80ec1d03274..1d95b5f6d403 100644
> --- a/drivers/firmware/psci.c
> +++ b/drivers/firmware/psci.c
> @@ -73,6 +73,7 @@ enum psci_function {
>  	PSCI_FN_CPU_ON,
>  	PSCI_FN_CPU_OFF,
>  	PSCI_FN_MIGRATE,
> +	PSCI_FN_SYSTEM_RESET2,
>  	PSCI_FN_MAX,
>  };
>  
> @@ -256,6 +257,14 @@ static void psci_sys_reset(enum reboot_mode reboot_mode, const char *cmd)
>  	invoke_psci_fn(PSCI_0_2_FN_SYSTEM_RESET, 0, 0, 0);
>  }
>  
> +static void psci_sys_reset2(enum reboot_mode reboot_mode, const char *cmd)
> +{
> +	if (reboot_mode == REBOOT_WARM)
> +		invoke_psci_fn(psci_function_id[PSCI_FN_SYSTEM_RESET2], 0, 0, 0);
> +	else
> +		psci_sys_reset(reboot_mode, cmd);
> +}

What happens when RESET2 isn't available, but the user requests s REBOOT_WARM
reset? It looks like we'd invoke a bogus SMC.

Could we please add a preprocessor definition for SYSTEM_RESET_WARM so that
it's clear that we're passing a parameter here?

> +
>  static void psci_sys_poweroff(void)
>  {
>  	invoke_psci_fn(PSCI_0_2_FN_SYSTEM_OFF, 0, 0, 0);
> @@ -564,6 +573,20 @@ static void __init psci_0_2_set_functions(void)
>  	pm_power_off = psci_sys_poweroff;
>  }
>  
> +static void __init psci_1_1_set_functions(void)
> +{
> +	int sys_reset2;
> +	int feature;
> +
> +	sys_reset2 = PSCI_FN_NATIVE(1_1, SYSTEM_RESET2);
> +	feature	= psci_features(sys_reset2);
> +
> +	if (feature != PSCI_RET_NOT_SUPPORTED) {
> +		psci_function_id[PSCI_FN_SYSTEM_RESET2] = sys_reset2;
> +		arm_pm_restart = psci_sys_reset2;
> +	}
> +}
> +
>  /*
>   * Probe function for PSCI firmware versions >= 0.2
>   */
> @@ -588,6 +611,8 @@ static int __init psci_probe(void)
>  		psci_init_smccc();
>  		psci_init_cpu_suspend();
>  		psci_init_system_suspend();
> +		if (PSCI_VERSION_MINOR(ver) >= 1)
> +			psci_1_1_set_functions();
>  	}

If we're going to add more things to this chain, can we please refactor this as:

	if (ver < PSCI_VERSION(1, 0))
		return;

	/* PSCI 1.0 setup here */

	if (ver < PSCI_VERSION(1,1))
		return;

	/* PSCI 1.1 setup here */

... as that saves on indentation and is easier to extend in future.

Thanks,
Mark