Message-ID: <ZzdR1HuTpnU1OL/i@lpieralisi>
Date: Fri, 15 Nov 2024 14:51:16 +0100
From: Lorenzo Pieralisi <lpieralisi@...nel.org>
To: Elliot Berman <quic_eberman@...cinc.com>
Cc: Bjorn Andersson <andersson@...nel.org>,
	Sebastian Reichel <sre@...nel.org>, Rob Herring <robh@...nel.org>,
	Conor Dooley <conor+dt@...nel.org>, Vinod Koul <vkoul@...nel.org>,
	Andy Yan <andy.yan@...k-chips.com>,
	Mark Rutland <mark.rutland@....com>,
	Bartosz Golaszewski <bartosz.golaszewski@...aro.org>,
	Arnd Bergmann <arnd@...db.de>, Olof Johansson <olof@...om.net>,
	Catalin Marinas <catalin.marinas@....com>,
	Will Deacon <will@...nel.org>, cros-qcom-dts-watchers@...omium.org,
	Krzysztof Kozlowski <krzk+dt@...nel.org>,
	Konrad Dybcio <konradybcio@...nel.org>,
	Satya Durga Srinivasu Prabhala <quic_satyap@...cinc.com>,
	Melody Olvera <quic_molvera@...cinc.com>,
	Shivendra Pratap <quic_spratap@...cinc.com>,
	devicetree@...r.kernel.org, linux-kernel@...r.kernel.org,
	linux-arm-kernel@...ts.infradead.org,
	Florian Fainelli <florian.fainelli@...adcom.com>,
	Stephen Boyd <swboyd@...omium.org>, linux-pm@...r.kernel.org,
	linux-arm-msm@...r.kernel.org
Subject: Re: [PATCH v8 3/6] firmware: psci: Read and use vendor reset types

On Thu, Nov 07, 2024 at 03:38:27PM -0800, Elliot Berman wrote:
> SoC vendors have different types of resets and are controlled through
> various registers. For instance, Qualcomm chipsets can reboot to a
> "download mode" that allows a RAM dump to be collected. Another example
> is they also support writing a cookie that can be read by bootloader
> during next boot. PSCI offers a mechanism, SYSTEM_RESET2, for these
> vendor reset types to be implemented without requiring drivers for every
> register/cookie.
> 
> Add support in PSCI to statically map reboot mode commands from
> userspace to a vendor reset and cookie value using the device tree.
> 
> A separate initcall is needed to parse the devicetree, instead of using
> psci_dt_init because mm isn't sufficiently set up to allocate memory.

Nit: information below this point is more a cover letter than for the
commit log.

> Reboot mode framework is close but doesn't quite fit with the
> design and requirements for PSCI SYSTEM_RESET2. Some of these issues can
> be solved but doesn't seem reasonable in sum:
>  1. reboot mode registers against the reboot_notifier_list, which is too
>     early to call SYSTEM_RESET2. PSCI would need to remember the reset
>     type from the reboot-mode framework callback and use it
>     psci_sys_reset.
>  2. reboot mode assumes only one cookie/parameter is described in the
>     device tree. SYSTEM_RESET2 uses 2: one for the type and one for
>     cookie.
>  3. psci cpuidle driver already registers a driver against the
>     arm,psci-1.0 compatible. Refactoring would be needed to have both a
>     cpuidle and reboot-mode driver.
> 
> Tested-by: Florian Fainelli <florian.fainelli@...adcom.com>
> Signed-off-by: Elliot Berman <quic_eberman@...cinc.com>
> ---
>  drivers/firmware/psci/psci.c | 104 +++++++++++++++++++++++++++++++++++++++++++
>  1 file changed, 104 insertions(+)
> 
> diff --git a/drivers/firmware/psci/psci.c b/drivers/firmware/psci/psci.c
> index 2328ca58bba61fdb677ac20a1a7447882cd0cf22..e60e3f8749c5a6732c51d23a2c1f453361132d9a 100644
> --- a/drivers/firmware/psci/psci.c
> +++ b/drivers/firmware/psci/psci.c
> @@ -79,6 +79,14 @@ struct psci_0_1_function_ids get_psci_0_1_function_ids(void)
>  static u32 psci_cpu_suspend_feature;
>  static bool psci_system_reset2_supported;
>  
> +struct psci_reset_param {
> +	const char *mode;
> +	u32 reset_type;
> +	u32 cookie;
> +};
> +static struct psci_reset_param *psci_reset_params __ro_after_init;
> +static size_t num_psci_reset_params __ro_after_init;
> +
>  static inline bool psci_has_ext_power_state(void)
>  {
>  	return psci_cpu_suspend_feature &
> @@ -305,9 +313,38 @@ static int get_set_conduit_method(const struct device_node *np)
>  	return 0;
>  }
>  
> +static void psci_vendor_system_reset2(const char *cmd)
> +{
> +	unsigned long ret;
> +	size_t i;
> +
> +	for (i = 0; i < num_psci_reset_params; i++) {
> +		if (!strcmp(psci_reset_params[i].mode, cmd)) {
> +			ret = invoke_psci_fn(PSCI_FN_NATIVE(1_1, SYSTEM_RESET2),
> +					     psci_reset_params[i].reset_type,
> +					     psci_reset_params[i].cookie, 0);
> +			/*
> +			 * if vendor reset fails, log it and fall back to
> +			 * architecture reset types
> +			 */
> +			pr_err("failed to perform reset \"%s\": %ld\n", cmd,
> +			       (long)ret);
> +			return;
> +		}
> +	}
> +}
> +
>  static int psci_sys_reset(struct notifier_block *nb, unsigned long action,
>  			  void *data)
>  {
> +	/*
> +	 * try to do the vendor system_reset2
> +	 * If the reset fails or there wasn't a match on the command,
> +	 * fall back to architectural resets
> +	 */
> +	if (data && num_psci_reset_params)
> +		psci_vendor_system_reset2(data);
> +
>  	if ((reboot_mode == REBOOT_WARM || reboot_mode == REBOOT_SOFT) &&
>  	    psci_system_reset2_supported) {

This is a mess. To issue architectural warm reset we check reboot_mode,
for vendor resets we ignore it - there is no rationale, that's the point
I am making.

Also see my question on the other thread re: user space and reset
"modes".

I appreciate we are not making progress but I don't want to pick up
the pieces later after merging this code - it is unclear to me what's
the best path forward - I would like to understand how other
platforms/arches behave in this respect.

>  		/*
> @@ -750,6 +787,73 @@ static const struct of_device_id psci_of_match[] __initconst = {
>  	{},
>  };
>  
> +#define REBOOT_PREFIX "mode-"
> +
> +static int __init psci_init_system_reset2_modes(void)
> +{
> +	const size_t len = strlen(REBOOT_PREFIX);
> +	struct psci_reset_param *param;
> +	struct device_node *psci_np __free(device_node) = NULL;
> +	struct device_node *np __free(device_node) = NULL;
> +	struct property *prop;
> +	size_t count = 0;
> +	u32 magic[2];
> +	int num;
> +
> +	if (!psci_system_reset2_supported)
> +		return 0;
> +
> +	psci_np = of_find_matching_node(NULL, psci_of_match);
> +	if (!psci_np)
> +		return 0;
> +
> +	np = of_find_node_by_name(psci_np, "reset-types");
> +	if (!np)
> +		return 0;
> +
> +	for_each_property_of_node(np, prop) {
> +		if (strncmp(prop->name, REBOOT_PREFIX, len))
> +			continue;
> +		num = of_property_count_u32_elems(np, prop->name);
> +		if (num != 1 && num != 2)
> +			continue;
> +
> +		count++;
> +	}
> +
> +	param = psci_reset_params =
> +		kcalloc(count, sizeof(*psci_reset_params), GFP_KERNEL);
> +	if (!psci_reset_params)
> +		return -ENOMEM;
> +
> +	for_each_property_of_node(np, prop) {
> +		if (strncmp(prop->name, REBOOT_PREFIX, len))
> +			continue;
> +
> +		param->mode = kstrdup_const(prop->name + len, GFP_KERNEL);

FWIW - I think you need to keep the logic in the previous loop into account
because that's what is used to allocate param, it is not a given that
param is valid at this stage if I am not mistaken - the previous loop
checked:

	num = of_property_count_u32_elems(np, prop->name);
	if (num != 1 && num != 2)
		continue;

Lorenzo

> +		if (!param->mode)
> +			continue;
> +
> +		num = of_property_read_variable_u32_array(np, prop->name, magic,
> +							  1, ARRAY_SIZE(magic));
> +		if (num < 0) {
> +			pr_warn("Failed to parse vendor reboot mode %s\n",
> +				param->mode);
> +			kfree_const(param->mode);
> +			continue;
> +		}
> +
> +		/* Force reset type to be in vendor space */
> +		param->reset_type = PSCI_1_1_RESET_TYPE_VENDOR_START | magic[0];
> +		param->cookie = num > 1 ? magic[1] : 0;
> +		param++;
> +		num_psci_reset_params++;
> +	}
> +
> +	return 0;
> +}
> +arch_initcall(psci_init_system_reset2_modes);
> +
>  int __init psci_dt_init(void)
>  {
>  	struct device_node *np;
> 
> -- 
> 2.34.1
> 

Powered by blists - more mailing lists