linux-arm-kernel.lists.infradead.org archive mirror
 help / color / mirror / Atom feed
From: geoff@infradead.org (Geoff Levand)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH 10/19] ARM64 / ACPI: Get the enable method for SMP initialization in ACPI way
Date: Thu, 31 Jul 2014 11:52:04 -0700	[thread overview]
Message-ID: <1406832724.8971.29.camel@smoke> (raw)
In-Reply-To: <1406206825-15590-11-git-send-email-hanjun.guo@linaro.org>

Hi Hanjun,

On Thu, 2014-07-24 at 21:00 +0800, Hanjun Guo wrote:
> ACPI 5.1 only has two explicit methods to boot up SMP,
> PSCI and Parking protocol, but the Parking protocol is
> only suitable for ARMv7 now, so make PSCI as the only way
> for the SMP boot protocol before some updates for the
> ACPI spec or the Parking protocol spec.
> 
> Signed-off-by: Hanjun Guo <hanjun.guo@linaro.org>
> Signed-off-by: Tomasz Nowicki <tomasz.nowicki@linaro.org>
> ---
>  arch/arm64/include/asm/acpi.h    |   21 +++++++++++++++
>  arch/arm64/include/asm/cpu_ops.h |    9 ++++++-
>  arch/arm64/include/asm/smp.h     |    2 +-
>  arch/arm64/kernel/acpi.c         |    9 +++++++
>  arch/arm64/kernel/cpu_ops.c      |   52 ++++++++++++++++++++++++++++++++++----
>  arch/arm64/kernel/smp.c          |   29 +++++++++++++++++++--
>  6 files changed, 113 insertions(+), 9 deletions(-)
> 
> diff --git a/arch/arm64/include/asm/acpi.h b/arch/arm64/include/asm/acpi.h
> index 5ce85f8..6240327 100644
> --- a/arch/arm64/include/asm/acpi.h
> +++ b/arch/arm64/include/asm/acpi.h
> @@ -14,6 +14,27 @@
>  
>  /* Basic configuration for ACPI */
>  #ifdef	CONFIG_ACPI

^^ This seems to be a tab (\t) character here, which is a strange thing
for me to see...

> +/*
> + * ACPI 5.1 only has two explicit methods to
> + * boot up SMP, PSCI and Parking protocol,
> + * but the Parking protocol is only defined
> + * for ARMv7 now, so make PSCI as the only
> + * way for the SMP boot protocol before some
> + * updates for the ACPI spec or the Parking
> + * protocol spec.
> + *
> + * This enum is intend to make the boot method
> + * scalable when above updates are happended,
> + * which NOT means to support all of them.
> + */

This comment will become out of date soon (I hope), and it is often the
case that these short term comments are not removed, so I think it
better to put this kind of note into the commit message, not the code.

> +enum acpi_smp_boot_protocol {
> +	ACPI_SMP_BOOT_PSCI,
> +	ACPI_SMP_BOOT_PARKING_PROTOCOL,
> +	ACPI_SMP_BOOT_PROTOCOL_MAX
> +};
> +
> +enum acpi_smp_boot_protocol smp_boot_protocol(void);
> +
>  extern int acpi_disabled;
>  extern int acpi_noirq;
>  extern int acpi_pci_disabled;
> diff --git a/arch/arm64/include/asm/cpu_ops.h b/arch/arm64/include/asm/cpu_ops.h
> index d7b4b38..2a7c6fd 100644
> --- a/arch/arm64/include/asm/cpu_ops.h
> +++ b/arch/arm64/include/asm/cpu_ops.h
> @@ -61,7 +61,14 @@ struct cpu_operations {
>  };
>  
>  extern const struct cpu_operations *cpu_ops[NR_CPUS];
> -extern int __init cpu_read_ops(struct device_node *dn, int cpu);
> +extern int __init cpu_of_read_ops(struct device_node *dn, int cpu);
> +
> +#ifdef CONFIG_ACPI
> +extern int __init cpu_acpi_read_ops(int cpu);
> +#else
> +static inline int __init cpu_acpi_read_ops(int cpu) { return -ENODEV; }
> +#endif

This looks messy and not scalable for new enable methods.  It
seems a better way is to retain cpu_read_ops() and its functionality,
which is to return the proper enable method for that cpu in a generic
way.

Is there some reason you can't integrate acpi into the existing
cpu_ops and need to make this completely parallel method?

>  extern void __init cpu_read_bootcpu_ops(void);
>  
>  #endif /* ifndef __ASM_CPU_OPS_H */
> diff --git a/arch/arm64/include/asm/smp.h b/arch/arm64/include/asm/smp.h
> index a498f2c..a5cea56 100644
> --- a/arch/arm64/include/asm/smp.h
> +++ b/arch/arm64/include/asm/smp.h
> @@ -39,7 +39,7 @@ extern void show_ipi_list(struct seq_file *p, int prec);
>  extern void handle_IPI(int ipinr, struct pt_regs *regs);
>  
>  /*
> - * Setup the set of possible CPUs (via set_cpu_possible)
> + * Platform specific SMP operations
>   */
>  extern void smp_init_cpus(void);
>  
> diff --git a/arch/arm64/kernel/acpi.c b/arch/arm64/kernel/acpi.c
> index ff0f6a0..2af6662 100644
> --- a/arch/arm64/kernel/acpi.c
> +++ b/arch/arm64/kernel/acpi.c
> @@ -184,6 +184,15 @@ static int __init acpi_parse_madt_gic_cpu_interface_entries(void)
>  	return 0;
>  }
>  
> +/* Protocol to bring up secondary CPUs */
> +enum acpi_smp_boot_protocol smp_boot_protocol(void)
> +{
> +	if (acpi_psci_present)
> +		return ACPI_SMP_BOOT_PSCI;
> +	else
> +		return ACPI_SMP_BOOT_PARKING_PROTOCOL;
> +}
> +
>  static int __init acpi_parse_fadt(struct acpi_table_header *table)
>  {
>  	struct acpi_table_fadt *fadt = (struct acpi_table_fadt *)table;
> diff --git a/arch/arm64/kernel/cpu_ops.c b/arch/arm64/kernel/cpu_ops.c
> index d62d12f..4d9b3cf 100644
> --- a/arch/arm64/kernel/cpu_ops.c
> +++ b/arch/arm64/kernel/cpu_ops.c
> @@ -16,11 +16,13 @@
>   * along with this program.  If not, see <http://www.gnu.org/licenses/>.
>   */
>  
> -#include <asm/cpu_ops.h>
> -#include <asm/smp_plat.h>
>  #include <linux/errno.h>
>  #include <linux/of.h>
>  #include <linux/string.h>
> +#include <linux/acpi.h>
> +
> +#include <asm/cpu_ops.h>
> +#include <asm/smp_plat.h>
>  
>  extern const struct cpu_operations smp_spin_table_ops;
>  extern const struct cpu_operations cpu_psci_ops;
> @@ -52,7 +54,7 @@ static const struct cpu_operations * __init cpu_get_ops(const char *name)
>  /*
>   * Read a cpu's enable method from the device tree and record it in cpu_ops.
>   */
> -int __init cpu_read_ops(struct device_node *dn, int cpu)
> +int __init cpu_of_read_ops(struct device_node *dn, int cpu)
>  {
>  	const char *enable_method = of_get_property(dn, "enable-method", NULL);
>  	if (!enable_method) {
> @@ -76,12 +78,52 @@ int __init cpu_read_ops(struct device_node *dn, int cpu)
>  	return 0;
>  }
>  
> +#ifdef CONFIG_ACPI
> +/*
> + * Read a cpu's enable method in the ACPI way and record it in cpu_ops.
> + */
> +int __init cpu_acpi_read_ops(int cpu)
> +{
> +	/*
> +	 * For ACPI 5.1, only two kind of methods are provided,
> +	 * Parking protocol and PSCI, but Parking protocol is
> +	 * used on ARMv7 only, so make PSCI as the only method
> +	 * for SMP initialization before the ACPI spec or Parking
> +	 * protocol spec is updated.
> +	 */

Again, this comment will get old fast (I hope).

> +	switch (smp_boot_protocol()) {
> +	case ACPI_SMP_BOOT_PSCI:
> +		cpu_ops[cpu] = cpu_get_ops("psci");
> +		break;
> +	case ACPI_SMP_BOOT_PARKING_PROTOCOL:
> +	default:
> +		cpu_ops[cpu] = NULL;
> +		break;
> +	}
> +
> +	if (!cpu_ops[cpu]) {
> +		pr_warn("CPU %d: unsupported enable-method, only PSCI is supported\n",
> +			cpu);
> +		return -EOPNOTSUPP;
> +	}
> +
> +	return 0;
> +}
> +#endif
> +
>  void __init cpu_read_bootcpu_ops(void)
>  {
> -	struct device_node *dn = of_get_cpu_node(0, NULL);
> +	struct device_node *dn;
> +
> +	if (!acpi_disabled) {
> +		cpu_acpi_read_ops(0);
> +		return;
> +	}
> +
> +	dn = of_get_cpu_node(0, NULL);
>  	if (!dn) {
>  		pr_err("Failed to find device node for boot cpu\n");
>  		return;
>  	}
> -	cpu_read_ops(dn, 0);
> +	cpu_of_read_ops(dn, 0);
>  }
> diff --git a/arch/arm64/kernel/smp.c b/arch/arm64/kernel/smp.c
> index 8f1d37c..cb71662 100644
> --- a/arch/arm64/kernel/smp.c
> +++ b/arch/arm64/kernel/smp.c
> @@ -315,7 +315,7 @@ static void (*smp_cross_call)(const struct cpumask *, unsigned int);
>   * cpu logical map array containing MPIDR values related to logical
>   * cpus. Assumes that cpu_logical_map(0) has already been initialized.
>   */
> -void __init smp_init_cpus(void)
> +static void __init of_smp_init_cpus(void)
>  {
>  	struct device_node *dn = NULL;
>  	unsigned int i, cpu = 1;
> @@ -387,7 +387,7 @@ void __init smp_init_cpus(void)
>  		if (cpu >= NR_CPUS)
>  			goto next;
>  
> -		if (cpu_read_ops(dn, cpu) != 0)
> +		if (cpu_of_read_ops(dn, cpu) != 0)
>  			goto next;
>  
>  		if (cpu_ops[cpu]->cpu_init(dn, cpu))
> @@ -418,6 +418,31 @@ next:
>  			set_cpu_possible(i, true);
>  }
>  
> +/*
> + * In ACPI mode, the cpu possible map was enumerated before SMP
> + * initialization when MADT table was parsed, so we can get the
> + * possible map here to initialize CPUs.
> + */
> +static void __init acpi_smp_init_cpus(void)
> +{
> +	int cpu;
> +
> +	for_each_possible_cpu(cpu) {
> +		if (cpu_acpi_read_ops(cpu) != 0)
> +			continue;
> +
> +		cpu_ops[cpu]->cpu_init(NULL, cpu);
> +	}
> +}
> +
> +void __init smp_init_cpus(void)
> +{
> +	if (acpi_disabled)
> +		of_smp_init_cpus();
> +	else
> +		acpi_smp_init_cpus();
> +}

This is the same as cpu_ops, is acpi so special we need a completely
parallel method of initializing secondary cpus?

-Geoff

  parent reply	other threads:[~2014-07-31 18:52 UTC|newest]

Thread overview: 131+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-07-24 13:00 [PATCH 00/19] Introduce ACPI for ARM64 based on ACPI 5.1 Hanjun Guo
2014-07-24 13:00 ` [PATCH 01/19] ARM64 / ACPI: Get RSDP and ACPI boot-time tables Hanjun Guo
2014-07-28 18:29   ` Sudeep Holla
2014-07-28 22:49     ` Graeme Gregory
2014-07-29  8:49       ` Sudeep Holla
2014-07-29 13:08     ` Hanjun Guo
2014-07-29 13:50       ` Sudeep Holla
2014-07-29 14:07         ` Hanjun Guo
2014-07-28 18:30   ` Sudeep Holla
2014-07-24 13:00 ` [PATCH 02/19] ARM64 / ACPI: Introduce early_param for "acpi" Hanjun Guo
2014-07-28 18:35   ` Sudeep Holla
2014-07-29 13:10     ` Hanjun Guo
2014-07-24 13:00 ` [PATCH 03/19] ARM64 / ACPI: Introduce lowlevel suspend function Hanjun Guo
2014-07-24 15:34   ` Mark Rutland
2014-07-25 10:42     ` Hanjun Guo
2014-07-28 18:28   ` Sudeep Holla
2014-07-29 13:00     ` Hanjun Guo
2014-07-24 13:00 ` [PATCH 04/19] ARM64 / ACPI: Introduce arch_fix_phys_package_id() for cpu topology Hanjun Guo
2014-07-24 14:43   ` Mark Brown
2014-07-25 10:32     ` Hanjun Guo
2014-07-28 18:51   ` Sudeep Holla
2014-08-01  6:35     ` Hanjun Guo
2014-08-01 10:48       ` Sudeep Holla
2014-07-24 13:00 ` [PATCH 05/19] ARM64 / ACPI: Make PCI optional for ACPI on ARM64 Hanjun Guo
2014-07-24 21:57   ` Naresh Bhat
2014-07-29 16:40   ` Sudeep Holla
2014-07-24 13:00 ` [PATCH 06/19] ARM64 / ACPI: Parse FADT table to get PSCI flags for PSCI init Hanjun Guo
2014-07-29 16:40   ` Sudeep Holla
2014-07-31  3:53     ` Hanjun Guo
2014-07-31  4:22   ` Olof Johansson
2014-07-31 10:23     ` Hanjun Guo
2014-08-20 15:02       ` Grant Likely
2014-08-20 15:00   ` Grant Likely
2014-08-20 15:29     ` Catalin Marinas
2014-08-20 15:43       ` graeme.gregory at linaro.org
2014-07-24 13:00 ` [PATCH 07/19] ARM64 / ACPI: Parse MADT to map logical cpu to MPIDR and get cpu_possible/present_map Hanjun Guo
2014-07-24 23:06   ` Naresh Bhat
2014-07-25 11:11     ` Hanjun Guo
2014-07-30 18:20   ` Sudeep Holla
2014-07-31  8:14     ` Hanjun Guo
2014-08-20 15:14   ` Grant Likely
2014-07-24 13:00 ` [PATCH 08/19] ACPI / table: Print GIC information when MADT is parsed Hanjun Guo
2014-07-30 18:21   ` Sudeep Holla
2014-07-31  8:15     ` Hanjun Guo
2014-07-24 13:00 ` [PATCH 09/19] ARM64 / ACPI: Move the initialization of cpu_logical_map(0) before acpi_boot_init() Hanjun Guo
2014-07-24 15:21   ` Mark Rutland
2014-07-25 10:39     ` Hanjun Guo
2014-07-25 12:18       ` Mark Rutland
2014-07-24 13:00 ` [PATCH 10/19] ARM64 / ACPI: Get the enable method for SMP initialization in ACPI way Hanjun Guo
2014-07-24 15:47   ` Mark Rutland
2014-07-25 10:51     ` Hanjun Guo
2014-07-25 12:24       ` Mark Rutland
2014-07-29  8:12         ` Hanjun Guo
2014-07-31  6:54   ` Olof Johansson
2014-07-31 10:57     ` Hanjun Guo
2014-08-04  9:56       ` Hanjun Guo
2014-07-31 18:52   ` Geoff Levand [this message]
2014-08-01  6:49     ` Hanjun Guo
2014-07-24 13:00 ` [PATCH 11/19] ACPI / processor: Make it possible to get CPU hardware ID via GICC Hanjun Guo
2014-07-24 13:00 ` [PATCH 12/19] ARM64 / ACPI: Introduce ACPI_IRQ_MODEL_GIC and register device's gsi Hanjun Guo
2014-07-24 13:00 ` [PATCH 13/19] ACPI / table: Add new function to get table entries Hanjun Guo
2014-07-24 13:00 ` [PATCH 14/19] ARM64 / ACPI: Add GICv2 specific ACPI boot support Hanjun Guo
2014-07-24 13:00 ` [PATCH 15/19] ARM64 / ACPI: Parse GTDT to initialize arch timer Hanjun Guo
2014-07-24 13:00 ` [PATCH 16/19] ARM64 / ACPI: Select ACPI_REDUCED_HARDWARE_ONLY if ACPI is enabled on ARM64 Hanjun Guo
2014-07-24 13:00 ` [PATCH 17/19] ARM64 / ACPI: If we chose to boot from acpi then disable FDT Hanjun Guo
2014-07-24 13:00 ` [PATCH 18/19] ARM64 / ACPI: Enable ARM64 in Kconfig Hanjun Guo
2014-07-24 13:00 ` [PATCH 19/19] Documentation: ACPI for ARM64 Hanjun Guo
2014-07-24 20:42   ` Randy Dunlap
2014-07-25 10:55     ` Hanjun Guo
     [not found]   ` <CAFoFrHaWWxRPRYM5+bWj0tGnz05SokqwVGejUCUi+U-KChFBdQ@mail.gmail.com>
2014-07-24 21:19     ` Randy Dunlap
2014-07-29 10:07       ` Christoffer Dall
2014-07-27  2:34   ` Olof Johansson
2014-07-28  8:42     ` Graeme Gregory
2014-07-28 16:23       ` Olof Johansson
2014-07-28 17:44         ` Mark Brown
2014-07-28  9:07     ` Arnd Bergmann
2014-07-28  9:23       ` Graeme Gregory
2014-07-28 10:46         ` Arnd Bergmann
2014-07-28 14:20           ` Andre Przywara
2014-07-28 15:23             ` Arnd Bergmann
2014-07-28 16:14               ` Andre Przywara
2014-07-29  9:17                 ` Graeme Gregory
2014-07-29 10:07                   ` Arnd Bergmann
2014-07-28 10:12       ` Mark Rutland
2014-07-28 16:33         ` Olof Johansson
2014-07-28 18:37           ` Mark Rutland
2014-07-28 18:44             ` Olof Johansson
2014-07-28 16:27       ` Olof Johansson
2014-07-28 17:00         ` Mark Rutland
2014-07-28 18:27           ` Olof Johansson
2014-08-12 18:23             ` Catalin Marinas
2014-08-13 23:41               ` Rafael J. Wysocki
2014-08-14  3:21                 ` Hanjun Guo
2014-08-14 10:27                   ` Catalin Marinas
2014-08-14 20:53                     ` Arnd Bergmann
2014-08-15  1:02                       ` Olof Johansson
2014-08-15 19:49                         ` Arnd Bergmann
2014-08-15 23:19                           ` Mark Brown
2014-08-16 12:51                           ` graeme.gregory at linaro.org
2014-08-15  9:09                     ` Hanjun Guo
2014-08-15 10:01                       ` Catalin Marinas
2014-08-18  9:29                         ` Hanjun Guo
2014-08-18 12:49                           ` Mark Rutland
2014-08-20 22:17                           ` Olof Johansson
2014-08-21  4:00                             ` Hanjun Guo
2014-07-29  9:01       ` Hanjun Guo
2014-07-28 10:06     ` Mark Rutland
2014-07-28 16:44       ` Olof Johansson
2014-07-28 17:36         ` Mark Rutland
2014-07-28 18:34           ` Olof Johansson
2014-07-29 10:29         ` Christoffer Dall
2014-07-29 10:41           ` Arnd Bergmann
2014-07-29 10:55           ` Mark Rutland
2014-07-29 11:28             ` Mark Rutland
2014-07-29 12:37               ` Christoffer Dall
2014-07-29 12:52                 ` Arnd Bergmann
2014-07-29 13:08                   ` Mark Rutland
2014-07-29 13:31                     ` Christoffer Dall
2014-07-29 14:04                       ` Mark Rutland
2014-07-29 14:41                       ` Arnd Bergmann
2014-07-29 15:01                         ` Christoffer Dall
2014-07-30  6:47                       ` Hanjun Guo
2014-07-30  7:14                         ` Christoffer Dall
2014-07-30  9:36                           ` Hanjun Guo
2014-07-29 13:33                   ` Christoffer Dall
2014-07-29  7:58     ` Hanjun Guo
2014-07-29 10:30   ` Christoffer Dall
2014-08-15 22:43   ` Len Brown
2014-08-16 12:45     ` Graeme Gregory
2014-08-20 16:42   ` Grant Likely
2014-07-25  0:46 ` [PATCH 00/19] Introduce ACPI for ARM64 based on ACPI 5.1 Hanjun Guo

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=1406832724.8971.29.camel@smoke \
    --to=geoff@infradead.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).