Raspi/03-QNX_Drivers
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[Init] First commit

Browse Source
This commit is contained in:
gaoyang3513
2025-05-22 22:52:09 +08:00
commit ff2f239df2
565 changed files with 116309 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

543
prebuilt/usr/include/sys/syspage.h Normal file
View File

@ -0,0 +1,543 @@
/*
* $QNXLicenseC:
* Copyright 2007, 2009, QNX Software Systems. All Rights Reserved.
*
* You must obtain a written license from and pay applicable license fees to QNX
* Software Systems before you may reproduce, modify or distribute this software,
* or any work that includes all or part of this software. Free development
* licenses are available for evaluation and non-commercial purposes. For more
* information visit http://licensing.qnx.com or email licensing@qnx.com.
*
* This file may contain contributions from others. Please review this entire
* file for other proprietary rights or license notices, as well as the QNX
* Development Suite License Guide at http://licensing.qnx.com/license-guide/
* for other information.
* $
*/
#ifndef __SYSPAGE_H_INCLUDED
#define __SYSPAGE_H_INCLUDED
#ifndef __PLATFORM_H_INCLUDED
#include <sys/platform.h>
#endif
#define SYSPAGE_VERSION_MAJOR 2
#define SYSPAGE_VERSION_MINOR 0
__BEGIN_DECLS
#if defined(__CLOCKADJUST)
struct _clockadjust __CLOCKADJUST;
#undef __CLOCKADJUST
#endif
struct _process_local_storage;
struct syspage_entry;
extern struct syspage_entry *_syspage_ptr;
typedef struct {
_Uint16t entry_off;
_Uint16t entry_size;
} syspage_entry_info;
typedef struct {
_Uint16t entry_off;
_Uint16t entry_size;
_Uint16t element_size;
} syspage_array_info;
/*
* Include CPU specific definition for network queries.
*/
#if !defined(SYSPAGE_TARGET_ALL) \
&& !defined(SYSPAGE_TARGET_X86_64) \
&& !defined(SYSPAGE_TARGET_AARCH64)
#if defined(__X86_64__)
#define SYSPAGE_TARGET_X86_64
#elif defined(__aarch64__)
#define SYSPAGE_TARGET_AARCH64
#else
#error not configured for system
#endif
#endif
#if defined(SYSPAGE_TARGET_ALL) || defined(SYSPAGE_TARGET_X86_64)
#include _NTO_HDR_(x86_64/syspage.h)
#endif
#if defined(SYSPAGE_TARGET_ALL) || defined(SYSPAGE_TARGET_AARCH64)
#include _NTO_HDR_(aarch64/syspage.h)
#endif
struct cpupage_entry {
_Uint32t cpu;
_Uint32t state;
union {
#if defined(SYSPAGE_TARGET_ALL) || defined(SYSPAGE_TARGET_AARCH64)
struct aarch64_cpupage_entry aarch64;
#endif
struct {
_Uint32t dummy[8];
} dummy;
} un;
struct _process_local_storage *pls;
_Uint32t spare[8];
};
enum {
CPUPAGE_ADDR = -1, /* R */
CPUPAGE_TLS = 0, /* R */
CPUPAGE_PLS, /* RW */
CPUPAGE_SYSPAGE, /* R */
CPUPAGE_CPU, /* R */
CPUPAGE_STATE, /* R */
CPUPAGE_MAX
};
/*
* Architecture independent system page definitions
*/
/*
* System page types
*/
enum {
/* technically not needed */
SYSPAGE_64BIT = 0x100,
/* syspage now has a version field */
SYSPAGE_VERSIONED = 0x1000,
SYSPAGE_X86_64 = (_Uint16t)SYSPAGE_VERSIONED | (_Uint16t)SYSPAGE_64BIT,
SYSPAGE_AARCH64,
};
#define QTIME_FLAG_CHECK_STABLE 0x00000002u
#define QTIME_FLAG_TICKLESS 0x00000004u
#define QTIME_FLAG_GLOBAL_CLOCKCYCLES 0x00000008u /* ClockCycles() is synchronized between all processors */
struct qtime_entry {
/**
* Cycles-per-second, for ClockCycles()
*/
_Uint64t cycles_per_sec;
/**
* Time-of-day adjustment for CLOCK_REALTIME/CLOCK_SOFTTIME and their timers.
*/
_Uint64t volatile nsec_tod_adjust;
/**
* Number of nanoseconds-per-tick: the tick period.
*/
_Uint32t nsec_inc;
/**
* UTC seconds when machine booted.
*/
_Uint32t boot_time;
/**
* Incremental CLOCK_REALTIME adjustment in progress (ClockAdjust()).
*/
struct _clockadjust adjust;
/**
* times 10^timer_scale
*/
_Uint32t timer_period;
/**
* Exponent for timer_period.
*/
_Int32t timer_scale;
/**
* Interrupt number for CPU0.
*/
_Int32t intr;
/**
* 1970.
*/
_Uint32t epoch;
/**
* QTIME_FLAG_* flags.
*/
_Uint32t flags;
/**
* Round-robin interval in ticks.
*/
_Uint32t rr_interval_mul;
/**
* The max absolute value that can be written in the the hardware timer's
* comparator. For correct operation, this value must never be reached
* during the system's lifetime.
*/
_Uint64t timer_load_max;
/**
* Boot time in clock cycles. This is the epoch for all calculations done
* on time in clock cycles.
*/
_Uint64t boot_cc;
/**
* Exact tick period used by the kernel, in clock cycles.
*
* The kernel computes the closest tick period value that can be
* represented in clock cycles from the tick frequency passed to procnto.
* Depending on the frequency requested and the hardware CPS, it is
* possible that corresponding value in nanoseconds cannot be represented
* as an integer number of nanoseconds. When that is the case, 'nsec_inc'
* is only an approximation in nanoseconds of tick_period_cc.
* This can happen if the kernel is configured with a weird tick frequency
* (e.g. the system configuration passing -C 347 to procnto) that is not a
* factor of a sensible CPS, or if the CPS itself is weird (e.g.
* startup-x86 calibration finding the CPS to be 1.804807028 GHz).
*/
_Uint64t tick_period_cc;
_Uint64t spare[3];
};
struct intrspin;
struct debug_callout {
void (*display_char)(struct syspage_entry *__syspage, char __char);
int (*poll_key)(struct syspage_entry *__syspage);
int (*break_detect)(struct syspage_entry *__syspage);
void *spare;
};
typedef enum {
DEBUG_WATCHDOG_STOP = 0,
DEBUG_WATCHDOG_CONTINUE,
DEBUG_WATCHDOG_FEED
} debug_watchdog_cmd;
struct callout_entry {
void (*reboot)(struct syspage_entry *__syspage, int __flags);
struct debug_callout debug[2];
void (*debug_watchdog)(struct syspage_entry *__syspage, debug_watchdog_cmd __cmd);
_Uint64t (*custom)(struct syspage_entry *__syspage, _Uint32t __type_id, _Uint64t *__regs);
_Uint64t spare[2];
};
#define AS_NULL_OFF ((_Uint16t)-1)
#define AS_ATTR_KIDS 0x0010u
#define AS_ATTR_CONTINUED 0x0020u
#define AS_PRIORITY_DEFAULT 100
struct asinfo_entry {
_Uint64t start;
_Uint64t end;
_Uint16t owner;
_Uint16t name;
_Uint16t attr;
_Uint16t priority;
};
struct hwinfo_entry {
__FLEXARY(_Uint32t, data); /* data[] variable sized, see <hw/sysinfo.h> */
};
/**
* Reboot in case of an abnormal shutdown.
* This flag is set by passing the -A option to startup.
* While not the default behaviour of startup, it should be
* used for safe systems.
*/
#define SYSTEM_PRIVATE_FLAG_ABNORMAL_REBOOT 0x00000001u
#define SYSTEM_PRIVATE_FLAG_TRACE_ACTIVE 0x00000004u
#define SYSTEM_PRIVATE_FLAG_ETB_RUNNING 0x00000008u
struct system_private_entry {
_Uint32t cpupage_spacing;
_Uint32t private_flags;
_Uint32t pagesize;
_Uint32t num_ready;
_Uint32t traced_classes;
_Uint32t spare;
_Paddr64t kdump_info;
struct cpupage_entry *user_cpupageptr;
struct syspage_entry *user_syspageptr;
struct cpupage_entry *kern_cpupageptr;
struct syspage_entry *kern_syspageptr;
struct kdebug_info *kdebug_info;
struct kdebug_callback *kdebug_call;
_Uint64t entropy[2];
};
/*
* System independent CPU flags (system dependent grow from bottom up)
*/
#define CPU_FLAG_FPU (1U << 31) /* CPU has floating point support */
#define CPU_FLAG_MMU (1U << 30) /* MMU is active */
struct cpuinfo_entry {
_Uint32t cpu;
_Uint32t speed;
_Uint64t flags;
_Uint64t smp_hwcoreid;
_Uint32t spare;
_Uint16t name;
_Uint8t ins_cache;
_Uint8t data_cache;
};
#define CACHE_LIST_END 0xffU /* terminate a cache hierarchy list */
/*
* Cache flags
*/
#define CACHE_FLAG_INSTR 0x0001U /* cache holds instructions */
#define CACHE_FLAG_DATA 0x0002U /* cache holds data */
#define CACHE_FLAG_UNIFIED 0x0003U /* cache holds both ins & data */
#define CACHE_FLAG_SHARED 0x0004U /* cache is shared between multiple */
/* processors in an SMP system */
#define CACHE_FLAG_SNOOPED 0x0008U /* cache implements a bus snooping */
/* protocol */
#define CACHE_FLAG_VIRT_TAG 0x0010U /* cache is virtually tagged */
#define CACHE_FLAG_VIRTUAL 0x0010U /* backwards compatability flag for above */
#define CACHE_FLAG_WRITEBACK 0x0020U/* cache does writeback, not writethru */
#define CACHE_FLAG_CTRL_PHYS 0x0040U/* control function takes 32-bit paddrs */
#define CACHE_FLAG_SUBSET 0x0080U /* this cache obeys the 'subset' property */
#define CACHE_FLAG_NONCOHERENT 0x0100U /* cache is non-coherent on SMP */
#define CACHE_FLAG_NONISA 0x0200U /* cache doesn't obey ISA cache instr */
#define CACHE_FLAG_NOBROADCAST 0x0400U /* cache ops aren't broadcast on bus for SMP */
#define CACHE_FLAG_VIRT_IDX 0x0800U /* cache is virtually indexed */
#define CACHE_FLAG_CTRL_PHYS64 0x1000U /* control function takes 64-bit paddrs */
struct cacheattr_entry;
typedef unsigned __cache_rtn(_Paddr64t, unsigned, int, struct cacheattr_entry *, volatile struct syspage_entry *);
struct cacheattr_entry {
_Uint32t next;
_Uint32t line_size;
_Uint32t num_lines;
_Uint32t flags;
_Uint16t ways;
_Uint16t spare0[3];
__cache_rtn *control;
};
struct typed_strings_entry {
char data[4]; /* variable size */
};
struct strings_entry {
char data[4]; /* variable size */
};
/*
* System independent interrupt flags
* (system dependent grow from bottom up).
*/
#define INTR_FLAG_CPU_FAULT 0x2000U
/* INTR_FLAG_MSI allows a range of vectors to be flagged for use as MSI vectors */
#define INTR_FLAG_MSI 0x0400U
/*
* System independent interrupt code gen flags
* (system dependent grow from bottom up).
*/
#define INTR_GENFLAG_LOAD_SYSPAGE 0x8000U
#define INTR_GENFLAG_LOAD_INTRINFO 0x4000U
#define INTR_GENFLAG_LOAD_LEVEL 0x2000U
#define INTR_GENFLAG_LOAD_INTRMASK 0x1000U
#define INTR_GENFLAG_NOGLITCH 0x0800U
#define INTR_GENFLAG_LOAD_CPUNUM 0x0400U
#define INTR_GENFLAG_ID_LOOP 0x0200U
struct __intrgen_data {
_Uint16t genflags;
_Uint16t size;
_Uint32t spare;
void (*rtn)(void);
};
#define INTR_CONFIG_FLAG_PREATTACH 0x0001U
#define INTR_CONFIG_FLAG_DISALLOWED 0x0002U
#define INTR_CONFIG_FLAG_IPI 0x0004U
/* system wide interrupt, not per-cpu*/
#define INTR_CONFIG_FLAG_SYS 0x0008U
/**
* Describes an interrupt controller and its relationship to other interrupt
* controllers. Controllers may be cascaded and form a tree-like structure.
*
* An instance of this structure is initialized by startup for every interrupt
* controller attached to the system.
*/
struct intrinfo_entry {
/**
* Logical interrupt vector used as the lower bound for the set of
* interrupts handled by this controller.
*/
_Uint32t vector_base;
/**
* Number of logical interrupt vectors handled by this controller. The upper
* bound of interrupts handled by this controller is `vector_base +
* num_vectors`.
*/
_Uint32t num_vectors;
/**
* Logical interupt vector that this controller cascades into, i.e. the
* logical interrupt vector that this controller is a child of.
*/
_Uint32t cascade_vector;
/**
* Base hardware interrupt vector for this controller. The meaning of this
* value is architecture dependent; for example, on x86-64 it corresponds to
* the base IDT entry for this controller while on aarch64 it is unused.
*/
_Uint32t cpu_intr_base;
/**
* Spacing between interrupt vector entries when using autovectoring.
*/
_Uint16t cpu_intr_stride;
/**
* An OR of `INTR_FLAG_*` values.
*/
_Uint16t flags;
/**
* Code fragment that identifies the source of this interrupt.
*/
struct __intrgen_data id;
/**
* Code fragment that signals the end of interrupt handling.
*/
struct __intrgen_data eoi;
/**
* A kernel callout to mask the interrupt source at the controller.
*/
int (*mask)(struct syspage_entry * spp, int id);
/**
* A kernel callout to unmask the interrupt source at the controller.
*/
int (*unmask)(struct syspage_entry * spp, int id);
/**
* A kernel callout to query the configuration of the interrupt source at the controller.
*/
unsigned (*config)(struct syspage_entry * spp, struct intrinfo_entry * iip, int id);
/**
* The per-CPU stride of this interrupt controller. If set to zero then this
* controller does not describe per-CPU interrupts. If non-zero then this
* controller does describe per-CPU interrupts and the value is the stride
* to be added to `vector_base` per-CPU. Example: a system with four CPU's, a
* `vector_base` of 0x300, and `local_stride` set to 0x100 would have the
* per-CPU interrupts for CPU3 start at logical interrupt vector 0x600.
*/
_Uint32t local_stride;
/**
* Reserved.
*/
_Uint32t spare[3];
};
struct smp_entry {
_Uint32t pending;
_Uint32t spare;
void (*send_ipi)(struct syspage_entry * __syspage,
_Uint64t __target, unsigned __cmd, unsigned *__cmdsp);
void *start_addr;
};
struct hypinfo_entry {
_Uint64t flags;
};
#define HYPINFO_ENABLED 0x1UL
struct cluster_entry {
_Uint64t cpumask; /* bitmask of the CPUs included in a cluster */
_Uint16t name;
_Uint16t spare[3];
};
#define SYSTEM_CLUSTER_NAME_PREFIX '_'
struct syspage_entry {
_Uint16t size; /* size of syspage_entry */
_Uint16t total_size; /* size of system page */
_Uint16t type;
_Uint16t num_cpu;
syspage_entry_info system_private;
struct {
_Uint16t major;
_Uint16t minor;
} version;
syspage_entry_info hwinfo;
syspage_entry_info qtime;
syspage_entry_info callout;
syspage_entry_info typed_strings;
syspage_entry_info strings;
syspage_entry_info smp;
union {
#if defined(SYSPAGE_TARGET_ALL) || defined(SYSPAGE_TARGET_X86_64)
struct x86_64_syspage_entry x86_64;
#endif
#if defined(SYSPAGE_TARGET_ALL) || defined(SYSPAGE_TARGET_AARCH64)
struct aarch64_syspage_entry aarch64;
#endif
_Uint64t filler[20];
} un;
syspage_array_info asinfo;
syspage_array_info cpuinfo;
syspage_array_info cacheattr;
syspage_array_info intrinfo;
syspage_entry_info hypinfo;
syspage_array_info cluster;
};
/* <STAN_MACRO5> */
#define _SYSPAGE_ENTRY_SIZE( __base, __field ) \
((__base)->__field.entry_size)
#define _SYSPAGE_ELEMENT_SIZE( __base, __field ) \
((__base)->__field.element_size)
#define SYSPAGE_ENTRY_SIZE( __field ) (_SYSPAGE_ENTRY_SIZE( _syspage_ptr, __field ))
#define SYSPAGE_ELEMENT_SIZE(__field) (_SYSPAGE_ELEMENT_SIZE(_syspage_ptr, __field))
#define _SYSPAGE_ENTRY( __base, __field ) \
((struct __field##_entry *)(void *)((char *)(__base) + (__base)->__field.entry_off))
#define _SYSPAGE_CPU_ENTRY_SIZE( __base, __cpu, __field ) \
((__base)->un.__cpu.__field.entry_size)
#define _SYSPAGE_CPU_ENTRY( __base, __cpu, __field ) \
((struct __cpu##_##__field##_entry *)(void *)((char *)(__base) + (__base)->un.__cpu.__field.entry_off))
#define SYSPAGE_ENTRY( __field ) (_SYSPAGE_ENTRY( _syspage_ptr, __field ))
#define SYSPAGE_CPU_ENTRY( __cpu, __field ) (_SYSPAGE_CPU_ENTRY( _syspage_ptr, __cpu, __field ))
#define SYSPAGE_CPU_ENTRY_SIZE(__cpu, __field) (_SYSPAGE_CPU_ENTRY_SIZE(_syspage_ptr, __cpu, __field))
#define SYSPAGE_ARRAY_ADJ_OFFSET(__field, __p, __adj_off) \
((struct __field##_entry *)((_Uintptrt)(__p) + (__adj_off)))
#define _SYSPAGE_ARRAY_IDX(__base, __field, __idx) \
(SYSPAGE_ARRAY_ADJ_OFFSET(__field, _SYSPAGE_ENTRY((__base), __field), (__idx) * _SYSPAGE_ELEMENT_SIZE((__base), __field)))
#define SYSPAGE_ARRAY_IDX(__field, __idx) (_SYSPAGE_ARRAY_IDX(_syspage_ptr, __field,( __idx)))
/* </STAN_MACRO5> */
__END_DECLS
#endif /* __SYSPAGE_H_INCLUDED */