From: Julien Grall <julien.grall@linaro.org>
To: parth.dixit@linaro.org, xen-devel@lists.xen.org
Cc: ian.campbell@citrix.com, Naresh Bhat <naresh.bhat@linaro.org>,
tim@xen.org, Tomasz Nowicki <tomasz.nowicki@linaro.org>,
stefano.stabellini@citrix.com, Hanjun Guo <hanjun.guo@linaro.org>,
jbeulich@suse.com, christoffer.dall@linaro.org
Subject: Re: [PATCH RFC 11/35] ARM64 / ACPI: Parse MADT to map logical cpu to MPIDR and get cpu_possible/present_map
Date: Wed, 04 Feb 2015 21:44:56 +0000 [thread overview]
Message-ID: <54D292D8.9080808@linaro.org> (raw)
In-Reply-To: <1423058539-26403-12-git-send-email-parth.dixit@linaro.org>
Hi Parth,
On 04/02/2015 14:01, parth.dixit@linaro.org wrote:
> From: Naresh Bhat <naresh.bhat@linaro.org>
>
> MADT contains the information for MPIDR which is essential for
> SMP initialization, parse the GIC cpu interface structures to
> get the MPIDR value and map it to cpu_logical_map(), and add
> enabled cpu with valid MPIDR into cpu_possible_map and
> cpu_present_map.
>
> Signed-off-by: Hanjun Guo <hanjun.guo@linaro.org>
> Signed-off-by: Tomasz Nowicki <tomasz.nowicki@linaro.org>
> Signed-off-by: Naresh Bhat <naresh.bhat@linaro.org>
> ---
> xen/arch/arm/arm64/acpi/arm-core.c | 139 +++++++++++++++++++++++++++++++++++++
> xen/include/asm-arm/acpi.h | 2 +
> xen/include/xen/acpi.h | 5 ++
> 3 files changed, 146 insertions(+)
>
> diff --git a/xen/arch/arm/arm64/acpi/arm-core.c b/xen/arch/arm/arm64/acpi/arm-core.c
> index 2b7e2ef..84b0032 100644
> --- a/xen/arch/arm/arm64/acpi/arm-core.c
> +++ b/xen/arch/arm/arm64/acpi/arm-core.c
> @@ -26,7 +26,10 @@
> #include <xen/init.h>
> #include <xen/acpi.h>
> #include <xen/errno.h>
> +#include <xen/stdbool.h>
> +#include <xen/cpumask.h>
>
> +#include <asm/cputype.h>
> #include <asm/acpi.h>
>
> /*
> @@ -49,10 +52,141 @@ int acpi_psci_present;
> /* 1 to indicate HVC must be used instead of SMC as the PSCI conduit */
> int acpi_psci_use_hvc;
>
> +/* available_cpus means enabled cpu in MADT */
> +static int available_cpus;
> +
Why adding available_cpus here, and total_cpus a bit below?
> enum acpi_irq_model_id acpi_irq_model = ACPI_IRQ_MODEL_PLATFORM;
>
> struct acpi_arm_root acpi_arm_rsdp_info; /* info about RSDP from FDT */
>
> +/* arch-optional setting to enable display of offline cpus >= nr_cpu_ids */
> +unsigned int total_cpus = 0;
Set but never used, neither exported...
> +/*
> + * acpi_register_gic_cpu_interface - register a gic cpu interface and
> + * generates a logic cpu number
> + * @mpidr: CPU's hardware id to register, MPIDR represented in MADT
> + * @enabled: this cpu is enabled or not
> + *
> + * Returns the logic cpu number which maps to the gic cpu interface
> + */
> +static int acpi_register_gic_cpu_interface(u64 mpidr, u8 enabled)
> +{
> + int cpu;
> +
> + if ( mpidr == INVALID_HWID )
> + {
> + printk("Skip invalid cpu hardware ID\n");
> + return -EINVAL;
> + }
> +
> + total_cpus++;
> + if ( !enabled )
> + return -EINVAL;
> +
> + if ( available_cpus >= NR_CPUS )
> + {
> + printk("NR_CPUS limit of %d reached, Processor %d/0x%llx ignored.\n",
> + NR_CPUS, total_cpus, (long long unsigned int)mpidr);
Why the (long long unsigned int) cast? Please use PRIx64.
> + return -EINVAL;
> + }
> +
> + /* If it is the first CPU, no need to check duplicate MPIDRs */
> + if ( !available_cpus )
> + goto skip_mpidr_check;
I don't think it's necessary, for_each_present_cpu should not contain
the first CPU...
> +
> + /*
> + * Duplicate MPIDRs are a recipe for disaster. Scan
> + * all initialized entries and check for
> + * duplicates. If any is found just ignore the CPU.
> + */
> + for_each_present_cpu(cpu)
> + {
> + if ( cpu_logical_map(cpu) == mpidr )
> + {
> + printk("Firmware bug, duplicate CPU MPIDR: 0x%llx in MADT\n",
> + (long long unsigned int)mpidr);
Ditto for the cast.
> + return -EINVAL;
> + }
> + }
> +
> +skip_mpidr_check:
> + available_cpus++;
> +
> + /* allocate a logic cpu id for the new comer */
> + if ( cpu_logical_map(0) == mpidr )
> + {
> + /*
> + * boot_cpu_init() already hold bit 0 in cpu_present_mask
> + * for BSP, no need to allocte again.
allocate
> + */
> + cpu = 0;
> + }
> + else
> + cpu = cpumask_next_zero(-1, &cpu_present_map);
> +
> + /* map the logic cpu id to cpu MPIDR */
> + cpu_logical_map(cpu) = mpidr;
> +
> + set_cpu_possible(cpu, true);
> + set_cpu_present(cpu, true);
> +
> + return cpu;
> +}
> +
> +static int __init
> +acpi_parse_gic_cpu_interface(struct acpi_subtable_header *header,
> + const unsigned long end)
> +{
> + struct acpi_madt_generic_interrupt *processor;
> +
> + processor = (struct acpi_madt_generic_interrupt *)header;
> +
> + if ( BAD_MADT_ENTRY(processor, end) )
> + return -EINVAL;
> +
> + acpi_table_print_madt_entry(header);
> +
> + acpi_register_gic_cpu_interface(processor->mpidr,
> + processor->flags & ACPI_MADT_ENABLED);
> +
> + return 0;
> +}
> +
> +/*
> + * Parse GIC cpu interface related entries in MADT
> + * returns 0 on success, < 0 on error
> + */
> +static int __init acpi_parse_madt_gic_cpu_interface_entries(void)
> +{
> + int count;
> +
> + /*
> + * do a partial walk of MADT to determine how many CPUs
> + * we have including disabled CPUs, and get information
> + * we need for SMP init
> + */
> + count = acpi_table_parse_madt(ACPI_MADT_TYPE_GENERIC_INTERRUPT,
> + acpi_parse_gic_cpu_interface, MAX_GIC_CPU_INTERFACE);
> +
> + if ( !count )
> + {
> + printk("No GIC CPU interface entries present\n");
> + return -ENODEV;
> + }
> + else if ( count < 0 )
> + {
> + printk("Error parsing GIC CPU interface entry\n");
> + return count;
> + }
> +
> + /* Make boot-up look pretty */
> + printk("%d CPUs available, %d CPUs total\n", available_cpus,
> + total_cpus);
> +
> + return 0;
> +}
> +
> int acpi_gsi_to_irq(u32 gsi, unsigned int *irq)
> {
> *irq = -1;
> @@ -140,6 +274,11 @@ int __init acpi_boot_init(void)
> if ( err )
> printk("Can't find FADT\n");
If we don't find the FADT, we should give up rather than trying to parse
the GIC ACPI table.
>
> + /* Get the boot CPU's MPIDR before MADT parsing */
> + cpu_logical_map(0) = read_cpuid_mpidr() & MPIDR_HWID_BITMASK;
> +
> + err = acpi_parse_madt_gic_cpu_interface_entries();
> +
> return err;
> }
> #endif
> diff --git a/xen/include/asm-arm/acpi.h b/xen/include/asm-arm/acpi.h
> index 03051ef..c2d25db 100644
> --- a/xen/include/asm-arm/acpi.h
> +++ b/xen/include/asm-arm/acpi.h
> @@ -105,4 +105,6 @@ static inline void acpi_disable_pci(void)
> #define acpi_strict 1 /* no ACPI spec workarounds on ARM */
> #endif
>
> +#define MAX_GIC_CPU_INTERFACE 65535
> +
> #endif /*_ASM_ARM_ACPI_H*/
> diff --git a/xen/include/xen/acpi.h b/xen/include/xen/acpi.h
> index ff96336..9387b36 100644
> --- a/xen/include/xen/acpi.h
> +++ b/xen/include/xen/acpi.h
> @@ -67,6 +67,11 @@ typedef int (*acpi_table_handler) (struct acpi_table_header *table);
> typedef int (*acpi_table_entry_handler) (struct acpi_subtable_header *header, const unsigned long end);
>
> unsigned int acpi_get_processor_id (unsigned int cpu);
> +
> +#define BAD_MADT_ENTRY(entry, end) ( \
> + (!entry) || (unsigned long)entry + sizeof(*entry) > end || \
> + ((struct acpi_subtable_header *)entry)->length < sizeof(*entry))
> +
IHMO, this should belong to a separate patch.
> char * __acpi_map_table (paddr_t phys_addr, unsigned long size);
> int acpi_boot_init (void);
> int acpi_boot_table_init (void);
>
Regards,
--
Julien Grall
next prev parent reply other threads:[~2015-02-04 21:44 UTC|newest]
Thread overview: 166+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-02-04 14:01 [PATCH RFC 00/35] Add ACPI support for arm64 on Xen parth.dixit
2015-02-04 14:01 ` [PATCH RFC 01/35] xen: acpi: Build numa and pmstate x86 only parth.dixit
2015-02-04 17:03 ` Julien Grall
2015-02-04 14:01 ` [PATCH RFC 02/35] xen: arm64: ACPI: Support common ACPI drivers parth.dixit
2015-02-04 17:34 ` Stefano Stabellini
2015-02-04 17:36 ` Julien Grall
2015-02-05 11:04 ` Ian Campbell
2015-02-05 11:35 ` Jan Beulich
2015-02-05 11:57 ` Ian Campbell
2015-02-05 12:01 ` Jan Beulich
2015-02-05 14:05 ` Julien Grall
2015-02-05 11:34 ` Jan Beulich
2015-02-05 11:56 ` Ian Campbell
2015-02-04 14:01 ` [PATCH RFC 03/35] xen: arm64: ACPI: Add basic ACPI initialization parth.dixit
2015-02-04 17:40 ` Stefano Stabellini
2015-02-04 21:00 ` Julien Grall
2015-02-04 14:01 ` [PATCH RFC 04/35] ACPI / ACPICA: Introduce ARM Boot Architecture Flags in FADT parth.dixit
2015-02-04 17:42 ` Stefano Stabellini
2015-02-04 21:03 ` Julien Grall
2015-02-05 11:06 ` Ian Campbell
2015-02-05 14:09 ` Julien Grall
2015-02-05 14:10 ` Julien Grall
2015-02-04 14:01 ` [PATCH RFC 05/35] ARM64 / ACPI: Parse FADT table to get PSCI flags parth.dixit
2015-02-04 17:45 ` Stefano Stabellini
2015-02-05 3:56 ` Hanjun Guo
2015-02-05 11:09 ` Ian Campbell
2015-02-04 21:14 ` Julien Grall
2015-02-04 14:01 ` [PATCH RFC 06/35] ACPI: Add Generic Interrupt and Distributor struct parth.dixit
2015-02-04 17:52 ` Stefano Stabellini
2015-02-04 21:16 ` Julien Grall
2015-02-04 14:01 ` [PATCH RFC 07/35] ACPI / ACPICA: Add new features for MADT which introduced by ACPI 5.1 parth.dixit
2015-02-04 17:52 ` Stefano Stabellini
2015-02-08 14:27 ` Tomasz Nowicki
2015-02-04 14:01 ` [PATCH RFC 08/35] ACPI / table: Print GIC information when MADT is parsed parth.dixit
2015-02-04 14:01 ` [PATCH RFC 09/35] Add cpumask_next_zero set_cpu_present and possible parth.dixit
2015-02-04 18:47 ` Stefano Stabellini
2015-02-05 11:47 ` Jan Beulich
2015-02-04 21:28 ` Julien Grall
2015-02-04 14:01 ` [PATCH RFC 10/35] asm / arm: Introduce cputype.h parth.dixit
2015-02-04 18:56 ` Stefano Stabellini
2015-02-04 21:33 ` Julien Grall
2015-02-04 14:01 ` [PATCH RFC 11/35] ARM64 / ACPI: Parse MADT to map logical cpu to MPIDR and get cpu_possible/present_map parth.dixit
2015-02-04 21:44 ` Julien Grall [this message]
2015-02-04 14:01 ` [PATCH RFC 12/35] ARM64: Initialization of cpu_logical_map(0) parth.dixit
2015-02-04 21:45 ` Julien Grall
2015-02-05 10:26 ` Stefano Stabellini
2015-02-11 5:09 ` Julien Grall
2015-02-04 14:01 ` [PATCH RFC 13/35] ACPI: Introduce acpi_parse_entries parth.dixit
2015-02-05 10:29 ` Stefano Stabellini
2015-02-11 5:26 ` Julien Grall
2015-02-04 14:01 ` [PATCH RFC 14/35] ACPI / ACPICA: Add GTDT support updated by ACPI 5.1 parth.dixit
2015-02-05 13:22 ` Stefano Stabellini
2015-02-04 14:01 ` [PATCH RFC 15/35] ARM64 / ACPI: Define ACPI_IRQ_MODEL_GIC needed for arm parth.dixit
2015-02-05 14:39 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 16/35] ARM64 / ACPI: Parse GTDT to initialize timer parth.dixit
2015-02-04 21:51 ` Julien Grall
2015-02-05 11:39 ` Ian Campbell
2015-02-05 14:26 ` Julien Grall
2015-02-05 14:51 ` Stefano Stabellini
2015-02-05 14:55 ` Ian Campbell
2015-02-05 14:46 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 17/35] pl011: Initialize serial from ACPI SPCR table parth.dixit
2015-02-04 21:57 ` Julien Grall
2015-02-05 11:42 ` Ian Campbell
2015-02-05 14:29 ` Julien Grall
2015-02-05 14:52 ` Ian Campbell
2015-02-11 6:10 ` Julien Grall
2015-02-05 15:27 ` Stefano Stabellini
2015-02-05 15:32 ` Ian Campbell
2015-02-04 14:02 ` [PATCH RFC 18/35] arm : add helper function for setting interrupt type parth.dixit
2015-02-04 21:59 ` Julien Grall
2015-02-05 15:33 ` Stefano Stabellini
2015-02-11 6:12 ` Julien Grall
2015-02-04 14:02 ` [PATCH RFC 19/35] ACPI / GICv2: Add GIC specific ACPI boot support parth.dixit
2015-02-04 14:43 ` G Gregory
2015-02-05 6:26 ` Parth Dixit
2015-02-05 3:41 ` Julien Grall
2015-02-05 15:54 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 20/35] xen/arm: Prepare a min DT for DOM0 parth.dixit
2015-02-05 3:48 ` Julien Grall
2015-02-05 15:58 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 21/35] xen/arm: Create memory node " parth.dixit
2015-02-05 3:51 ` Julien Grall
2015-02-05 16:01 ` Stefano Stabellini
2015-02-11 6:27 ` Julien Grall
2015-02-04 14:02 ` [PATCH RFC 22/35] xen/arm: Create chosen " parth.dixit
2015-02-05 16:09 ` Stefano Stabellini
2015-02-06 0:29 ` Julien Grall
2015-02-06 14:09 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 23/35] arm: acpi add status override table parth.dixit
2015-02-05 16:14 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 24/35] arm : acpi add xen environment table parth.dixit
2015-02-05 16:16 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 25/35] arm: acpi add helper functions to map memory regions parth.dixit
2015-02-05 4:03 ` Julien Grall
2015-02-05 16:21 ` Stefano Stabellini
2015-02-06 0:35 ` Julien Grall
2015-02-06 14:12 ` Stefano Stabellini
2015-02-11 6:49 ` Julien Grall
2015-02-04 14:02 ` [PATCH RFC 26/35] arm : acpi read mmio tables from uefi parth.dixit
2015-02-05 4:17 ` Julien Grall
2015-02-05 16:34 ` Stefano Stabellini
2015-02-06 0:38 ` Julien Grall
2015-02-06 14:17 ` Stefano Stabellini
2015-02-11 9:14 ` Julien Grall
2015-02-04 14:02 ` [PATCH RFC 27/35] arm: acpi map mmio regions to dom0 parth.dixit
2015-02-05 16:49 ` Stefano Stabellini
2015-02-05 19:40 ` Parth Dixit
2015-02-06 0:44 ` Julien Grall
2015-02-06 14:21 ` Stefano Stabellini
2015-02-11 9:26 ` Julien Grall
2015-02-04 14:02 ` [PATCH RFC 28/35] arm: acpi map acpi tables in dom0 parth.dixit
2015-02-05 4:29 ` Julien Grall
2015-02-05 16:55 ` Stefano Stabellini
2015-02-05 19:38 ` Parth Dixit
2015-02-06 14:23 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 29/35] arm : acpi enable PSCI and hvc in acpi FADT table parth.dixit
2015-02-05 4:33 ` Julien Grall
2015-02-05 17:12 ` Stefano Stabellini
2015-02-06 0:47 ` Julien Grall
2015-02-06 15:13 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 30/35] arm : acpi map XSDT table to dom0 parth.dixit
2015-02-05 4:46 ` Julien Grall
2015-02-05 17:24 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 31/35] arm : acpi map status override " parth.dixit
2015-02-05 5:24 ` Julien Grall
2015-02-05 10:57 ` Parth Dixit
2015-02-05 11:47 ` Ian Campbell
2015-02-11 9:45 ` Julien Grall
2015-02-12 6:50 ` Stefano Stabellini
2015-02-05 14:39 ` Julien Grall
2015-02-05 17:39 ` Stefano Stabellini
2015-02-06 0:54 ` Julien Grall
2015-02-06 14:32 ` Stefano Stabellini
2015-02-05 17:27 ` Stefano Stabellini
2015-02-04 14:02 ` [PATCH RFC 32/35] arm : acpi map xen environment " parth.dixit
2015-02-05 5:29 ` Julien Grall
2015-02-05 10:49 ` Parth Dixit
2015-02-05 17:36 ` Stefano Stabellini
2015-02-06 0:57 ` Julien Grall
2015-02-04 14:02 ` [PATCH RFC 33/35] arm : acpi enable efi for acpi parth.dixit
2015-02-05 5:31 ` Julien Grall
2015-02-05 10:32 ` Parth Dixit
2015-02-05 11:58 ` Jan Beulich
2015-02-05 12:05 ` Ian Campbell
2015-02-11 9:57 ` Julien Grall
2015-02-11 10:31 ` Jan Beulich
2015-02-11 14:34 ` Julien Grall
2015-02-11 9:51 ` Usage of efi_enabled - Was: " Julien Grall
2015-02-11 10:28 ` Jan Beulich
2015-02-11 10:49 ` Ian Campbell
2015-02-11 11:22 ` Jan Beulich
2015-02-12 4:18 ` Ian Campbell
2015-02-04 14:02 ` [PATCH RFC 34/35] arm : acpi workarounds for firmware/linux dependencies parth.dixit
2015-02-05 5:38 ` Julien Grall
2015-02-05 10:30 ` Parth Dixit
2015-02-05 14:59 ` Julien Grall
2015-02-10 9:38 ` Julien Grall
2015-02-10 10:01 ` Jan Beulich
2015-02-10 10:26 ` Julien Grall
2015-02-05 17:48 ` Stefano Stabellini
2015-02-05 19:30 ` Parth Dixit
2015-02-06 14:38 ` Stefano Stabellini
2015-02-06 14:49 ` Jan Beulich
2015-02-04 14:02 ` [PATCH RFC 35/35] xen: arm64: Add ACPI support parth.dixit
2015-02-04 16:38 ` [PATCH RFC 00/35] Add ACPI support for arm64 on Xen Julien Grall
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=54D292D8.9080808@linaro.org \
--to=julien.grall@linaro.org \
--cc=christoffer.dall@linaro.org \
--cc=hanjun.guo@linaro.org \
--cc=ian.campbell@citrix.com \
--cc=jbeulich@suse.com \
--cc=naresh.bhat@linaro.org \
--cc=parth.dixit@linaro.org \
--cc=stefano.stabellini@citrix.com \
--cc=tim@xen.org \
--cc=tomasz.nowicki@linaro.org \
--cc=xen-devel@lists.xen.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.