[PATCH 06/19] ARM64 / ACPI: Parse FADT table to get PSCI flags for PSCI init
Hanjun Guo
hanjun.guo at linaro.org
Thu Jul 31 03:23:18 PDT 2014
On 2014-7-31 12:22, Olof Johansson wrote:
> Hi,
Hi Olof,
>
> On Thu, Jul 24, 2014 at 6:00 AM, Hanjun Guo <hanjun.guo at linaro.org> wrote:
>> There are two flags: PSCI_COMPLIANT and PSCI_USE_HVC. When set,
>> the former signals to the OS that the hardware is PSCI compliant.
>> The latter selects the appropriate conduit for PSCI calls by
>> toggling between Hypervisor Calls (HVC) and Secure Monitor Calls
>> (SMC).
>>
>> FADT table contains such information, parse FADT to get the flags
>> for PSCI init. Since ACPI 5.1 doesn't support self defined PSCI
>> function IDs, which means that only PSCI 0.2+ is supported in ACPI.
>>
>> At the same time, only ACPI 5.1 or higher verison supports PSCI,
>> and FADT Major.Minor version was introduced in ACPI 5.1, so we
>> will check the version and only parse FADT table with version >= 5.1.
>>
>> If firmware provides ACPI tables with ACPI version less than 5.1,
>> OS will be messed up with those information and have no way to
>> bring up secondery CPUs, so disable ACPI if we get an FADT table
>> with version less that 5.1.
>>
>> Signed-off-by: Hanjun Guo <hanjun.guo at linaro.org>
>> Signed-off-by: Graeme Gregory <graeme.gregory at linaro.org>
>> ---
>> arch/arm64/include/asm/acpi.h | 2 +
>> arch/arm64/kernel/acpi.c | 50 ++++++++++++++++++++++
>> arch/arm64/kernel/psci.c | 95 +++++++++++++++++++++++++++++------------
>> arch/arm64/kernel/setup.c | 2 +
>> 4 files changed, 121 insertions(+), 28 deletions(-)
>>
>> diff --git a/arch/arm64/include/asm/acpi.h b/arch/arm64/include/asm/acpi.h
>> index 44b617f..67dac90 100644
>> --- a/arch/arm64/include/asm/acpi.h
>> +++ b/arch/arm64/include/asm/acpi.h
>> @@ -18,6 +18,8 @@ extern int acpi_disabled;
>> extern int acpi_noirq;
>> extern int acpi_pci_disabled;
>> extern int acpi_strict;
>> +extern int acpi_psci_present;
>> +extern int acpi_psci_use_hvc;
>>
>> static inline void disable_acpi(void)
>> {
>> diff --git a/arch/arm64/kernel/acpi.c b/arch/arm64/kernel/acpi.c
>> index f5a10b5..374926f 100644
>> --- a/arch/arm64/kernel/acpi.c
>> +++ b/arch/arm64/kernel/acpi.c
>> @@ -11,6 +11,8 @@
>> * published by the Free Software Foundation.
>> */
>>
>> +#define pr_fmt(fmt) "ACPI: " fmt
>> +
>> #include <linux/init.h>
>> #include <linux/acpi.h>
>> #include <linux/cpumask.h>
>> @@ -34,6 +36,12 @@ EXPORT_SYMBOL(acpi_disabled);
>> int acpi_pci_disabled; /* skip ACPI PCI scan and IRQ initialization */
>> EXPORT_SYMBOL(acpi_pci_disabled);
>>
>> +/* 1 to indicate PSCI is implemented */
>> +int acpi_psci_present;
>> +
>> +/* 1 to indicate HVC must be used instead of SMC as the PSCI conduit */
>> +int acpi_psci_use_hvc;
>
> Here's a prime example of where it would just make more sense to
> populate DT based on what's in the ACPI info.
>
> Have a acpi_parse_fadt() that, if needed, creates a /psci node in the
> system-wide DT and populates it with the few properties needed.
>
> That way, the rest of the code path in the kernel setup is identical,
> instead of dealing with separate functions for setup, two exported
> variables just to communicate the state, and so on. It's just extra
> complexity for no good reason. The ACPi side code isn't even adding
> significant complexity compared to this. We'll need to add an
> of_add_node() property though.
Yes, this will make the code path in the kernel setup is identical, but
I think mixture of ACPI and DT (converting ACPI into DT at run-time) is
not a good solution, and this had been discussed last year:
http://lists.infradead.org/pipermail/linux-arm-kernel/2013-November/211662.html
>
>> /*
>> * __acpi_map_table() will be called before page_init(), so early_ioremap()
>> * or early_memremap() should be called here to for ACPI table mapping.
>> @@ -54,6 +62,33 @@ void __init __acpi_unmap_table(char *map, unsigned long size)
>> early_iounmap(map, size);
>> }
>>
>> +static int __init acpi_parse_fadt(struct acpi_table_header *table)
>> +{
>> + struct acpi_table_fadt *fadt = (struct acpi_table_fadt *)table;
>> +
>> + /*
>> + * Revision in table header is the FADT Major version,
>> + * and there is a minor version of FADT which was introduced
>> + * by ACPI 5.1, we only deal with ACPI 5.1 or higher version
>> + * to get arm boot flags, or we will disable ACPI.
>> + */
>> + if (table->revision < 5 || fadt->minor_version < 1) {
>> + pr_info("FADT version is %d.%d, no PSCI support, should be 5.1 or higher\n",
>> + table->revision, fadt->minor_version);
>> + acpi_psci_present = 0;
>> + disable_acpi();
>
> This does more than just check for FADT/PSCI, it disables ACPI
> alltogether. That kind of side effect from parse_fadt() doesn't seem
> appropriate.
As the this version of patch set only handle ACPI 5.1 version or higher,
and when we test this patch set when we passing ACPI tables with version
5.0, it messed up the kernel at boot-time, so we disable ACPI when
we got tables with ACPI version less that 5.1.
>
>> + return -EINVAL;
>> + }
>> +
>> + if (acpi_gbl_FADT.arm_boot_flags & ACPI_FADT_PSCI_COMPLIANT)
>> + acpi_psci_present = 1;
>> +
>> + if (acpi_gbl_FADT.arm_boot_flags & ACPI_FADT_PSCI_USE_HVC)
>> + acpi_psci_use_hvc = 1;
>> +
>> + return 0;
>> +}
>> +
>> /*
>> * acpi_boot_table_init() called from setup_arch(), always.
>> * 1. find RSDP and get its address, and then find XSDT
>> @@ -75,6 +110,21 @@ void __init acpi_boot_table_init(void)
>> }
>> }
>>
>> +int __init acpi_boot_init(void)
>> +{
>> + int err = 0;
>> +
>> + /* If acpi_disabled, bail out */
>> + if (acpi_disabled)
>> + return -ENODEV;
>> +
>> + err = acpi_table_parse(ACPI_SIG_FADT, acpi_parse_fadt);
>> + if (err)
>> + pr_err("Can't find FADT\n");
>> +
>> + return err;
>> +}
>> +
>> static int __init parse_acpi(char *arg)
>> {
>> if (!arg)
>> diff --git a/arch/arm64/kernel/psci.c b/arch/arm64/kernel/psci.c
>> index 9e9798f..adb25f3 100644
>> --- a/arch/arm64/kernel/psci.c
>> +++ b/arch/arm64/kernel/psci.c
>> @@ -15,6 +15,7 @@
>>
>> #define pr_fmt(fmt) "psci: " fmt
>>
>> +#include <linux/acpi.h>
>> #include <linux/init.h>
>> #include <linux/of.h>
>> #include <linux/smp.h>
>> @@ -231,6 +232,33 @@ static void psci_sys_poweroff(void)
>> invoke_psci_fn(PSCI_0_2_FN_SYSTEM_OFF, 0, 0, 0);
>> }
>>
>> +static void psci_0_2_set_functions(void)
>> +{
>> + pr_info("Using standard PSCI v0.2 function IDs\n");
>> + psci_function_id[PSCI_FN_CPU_SUSPEND] = PSCI_0_2_FN64_CPU_SUSPEND;
>> + psci_ops.cpu_suspend = psci_cpu_suspend;
>> +
>> + psci_function_id[PSCI_FN_CPU_OFF] = PSCI_0_2_FN_CPU_OFF;
>> + psci_ops.cpu_off = psci_cpu_off;
>> +
>> + psci_function_id[PSCI_FN_CPU_ON] = PSCI_0_2_FN64_CPU_ON;
>> + psci_ops.cpu_on = psci_cpu_on;
>> +
>> + psci_function_id[PSCI_FN_MIGRATE] = PSCI_0_2_FN64_MIGRATE;
>> + psci_ops.migrate = psci_migrate;
>> +
>> + psci_function_id[PSCI_FN_AFFINITY_INFO] = PSCI_0_2_FN64_AFFINITY_INFO;
>> + psci_ops.affinity_info = psci_affinity_info;
>> +
>> + psci_function_id[PSCI_FN_MIGRATE_INFO_TYPE] =
>> + PSCI_0_2_FN_MIGRATE_INFO_TYPE;
>> + psci_ops.migrate_info_type = psci_migrate_info_type;
>> +
>> + arm_pm_restart = psci_sys_reset;
>> +
>> + pm_power_off = psci_sys_poweroff;
>> +}
>> +
>> /*
>> * PSCI Function IDs for v0.2+ are well defined so use
>> * standard values.
>> @@ -264,29 +292,7 @@ static int psci_0_2_init(struct device_node *np)
>> }
>> }
>>
>> - pr_info("Using standard PSCI v0.2 function IDs\n");
>> - psci_function_id[PSCI_FN_CPU_SUSPEND] = PSCI_0_2_FN64_CPU_SUSPEND;
>> - psci_ops.cpu_suspend = psci_cpu_suspend;
>> -
>> - psci_function_id[PSCI_FN_CPU_OFF] = PSCI_0_2_FN_CPU_OFF;
>> - psci_ops.cpu_off = psci_cpu_off;
>> -
>> - psci_function_id[PSCI_FN_CPU_ON] = PSCI_0_2_FN64_CPU_ON;
>> - psci_ops.cpu_on = psci_cpu_on;
>> -
>> - psci_function_id[PSCI_FN_MIGRATE] = PSCI_0_2_FN64_MIGRATE;
>> - psci_ops.migrate = psci_migrate;
>> -
>> - psci_function_id[PSCI_FN_AFFINITY_INFO] = PSCI_0_2_FN64_AFFINITY_INFO;
>> - psci_ops.affinity_info = psci_affinity_info;
>> -
>> - psci_function_id[PSCI_FN_MIGRATE_INFO_TYPE] =
>> - PSCI_0_2_FN_MIGRATE_INFO_TYPE;
>> - psci_ops.migrate_info_type = psci_migrate_info_type;
>> -
>> - arm_pm_restart = psci_sys_reset;
>> -
>> - pm_power_off = psci_sys_poweroff;
>> + psci_0_2_set_functions();
>>
>> out_put_node:
>> of_node_put(np);
>> @@ -333,6 +339,33 @@ out_put_node:
>> return err;
>> }
>>
>> +#ifdef CONFIG_ACPI
>> +static int get_set_conduit_method_acpi(void)
>> +{
>> + if (acpi_psci_use_hvc)
>> + invoke_psci_fn = __invoke_psci_fn_hvc;
>> + else
>> + invoke_psci_fn = __invoke_psci_fn_smc;
>> +
>> + return 0;
>> +}
>> +
>> +/* We use PSCI 0.2+ when ACPI is deployed */
>> +static int psci_0_2_init_acpi(void)
>> +{
>> + get_set_conduit_method_acpi();
>> +
>> + psci_0_2_set_functions();
>
> Here's a good example of the two code paths doing different things.
> The DT code paths does a test call into PSCI to make sure it's
> actually supported, while you blindly trust it. Keeping the code paths
> common has value w.r.t these kind of things.
I should check acpi_psci_present first, and I will check if I can reuse
the DT code.
>
>> + return 0;
>> +}
>> +#else
>> +static inline int psci_0_2_init_acpi(void)
>> +{
>> + return -ENODEV;
>> +}
>> +#endif
>> +
>> static const struct of_device_id psci_of_match[] __initconst = {
>> { .compatible = "arm,psci", .data = psci_0_1_init},
>> { .compatible = "arm,psci-0.2", .data = psci_0_2_init},
>> @@ -345,13 +378,19 @@ int __init psci_init(void)
>> const struct of_device_id *matched_np;
>> psci_initcall_t init_fn;
>>
>> - np = of_find_matching_node_and_match(NULL, psci_of_match, &matched_np);
>> + if (acpi_disabled) {
>> + np = of_find_matching_node_and_match(NULL,
>> + psci_of_match, &matched_np);
>
> Ideally this code should go away by changing the rest of it, but for
> future cases: It'd be a lot cleaner to do this as:
>
> if (!acpi_disabled)
> return psci_0_2_init_acpi();
>
> ... then fall through to the current implementation instead.
I prefer this one :)
Thanks
Hanjun
>
> But sharing the whole code path is even better, so let's do that.
>
>>
>> - if (!np)
>> - return -ENODEV;
>> + if (!np)
>> + return -ENODEV;
>> +
>> + init_fn = (psci_initcall_t)matched_np->data;
>> +
>> + return init_fn(np);
>> + }
>>
>> - init_fn = (psci_initcall_t)matched_np->data;
>> - return init_fn(np);
>> + return psci_0_2_init_acpi();
>> }
>>
>> #ifdef CONFIG_SMP
>> diff --git a/arch/arm64/kernel/setup.c b/arch/arm64/kernel/setup.c
>> index 9dedb0b..e00d40c 100644
>> --- a/arch/arm64/kernel/setup.c
>> +++ b/arch/arm64/kernel/setup.c
>> @@ -394,6 +394,8 @@ void __init setup_arch(char **cmdline_p)
>>
>> efi_idmap_init();
>>
>> + acpi_boot_init();
>> +
>> unflatten_device_tree();
>
> The above two will need to be swapped when you move over to populating
> the DT instead.
>
>>
>> psci_init();
>
>
> -Olof
>
More information about the linux-arm-kernel
mailing list