/*
* %CopyrightBegin%
*
* Copyright Ericsson AB 1996-2010. All Rights Reserved.
*
* The contents of this file are subject to the Erlang Public License,
* Version 1.1, (the "License"); you may not use this file except in
* compliance with the License. You should have received a copy of the
* Erlang Public License along with this software. If not, it can be
* retrieved online at http://www.erlang.org/.
*
* Software distributed under the License is distributed on an "AS IS"
* basis, WITHOUT WARRANTY OF ANY KIND, either express or implied. See
* the License for the specific language governing rights and limitations
* under the License.
*
* %CopyrightEnd%
*/
#ifndef __PROCESS_H__
#define __PROCESS_H__
#undef ERTS_INCLUDE_SCHEDULER_INTERNALS
#if (defined(ERL_PROCESS_C__) \
|| defined(ERL_PORT_TASK_C__) \
|| (ERTS_GLB_INLINE_INCL_FUNC_DEF \
&& defined(ERTS_DO_INCL_GLB_INLINE_FUNC_DEF)))
#define ERTS_INCLUDE_SCHEDULER_INTERNALS
#endif
typedef struct process Process;
#include "sys.h"
#define ERTS_PROCESS_LOCK_ONLY_PROC_LOCK_TYPE__
#include "erl_process_lock.h" /* Only pull out important types... */
#undef ERTS_PROCESS_LOCK_ONLY_PROC_LOCK_TYPE__
#include "erl_vm.h"
#include "erl_smp.h"
#include "erl_message.h"
#include "erl_process_dict.h"
#include "erl_node_container_utils.h"
#include "erl_node_tables.h"
#include "erl_monitors.h"
#include "erl_bif_timer.h"
#include "erl_time.h"
#include "erl_atom_table.h"
#include "external.h"
#ifdef HIPE
#include "hipe_process.h"
#endif
struct ErtsNodesMonitor_;
struct port;
#define ERTS_MAX_NO_OF_SCHEDULERS 1024
#define ERTS_DEFAULT_MAX_PROCESSES (1 << 15)
#define ERTS_HEAP_ALLOC(Type, Size) \
erts_alloc((Type), (Size))
#define ERTS_HEAP_REALLOC(Type, Ptr, OldSize, NewSize) \
erts_realloc((Type), (Ptr), (NewSize))
#define ERTS_HEAP_FREE(Type, Ptr, Size) \
erts_free((Type), (Ptr))
#define INITIAL_MOD 0
#define INITIAL_FUN 1
#define INITIAL_ARI 2
#include "export.h"
struct saved_calls {
int len;
int n;
int cur;
Export *ct[1];
};
extern Export exp_send, exp_receive, exp_timeout;
extern Uint erts_no_schedulers;
extern Uint erts_no_run_queues;
extern int erts_sched_thread_suggested_stack_size;
#define ERTS_SCHED_THREAD_MIN_STACK_SIZE 4 /* Kilo words */
#define ERTS_SCHED_THREAD_MAX_STACK_SIZE 8192 /* Kilo words */
#ifdef ERTS_SMP
extern Uint erts_max_main_threads;
#include "erl_bits.h"
#endif
/* process priorities */
#define PRIORITY_MAX 0
#define PRIORITY_HIGH 1
#define PRIORITY_NORMAL 2
#define PRIORITY_LOW 3
#define ERTS_NO_PROC_PRIO_LEVELS 4
#define ERTS_PORT_PRIO_LEVEL ERTS_NO_PROC_PRIO_LEVELS
#define ERTS_RUNQ_FLGS_PROCS_QMASK \
((((Uint32) 1) << ERTS_NO_PROC_PRIO_LEVELS) - 1)
#define ERTS_NO_PRIO_LEVELS (ERTS_NO_PROC_PRIO_LEVELS + 1)
#define ERTS_RUNQ_FLGS_MIGRATE_QMASK \
((((Uint32) 1) << ERTS_NO_PRIO_LEVELS) - 1)
#define ERTS_RUNQ_FLGS_EMIGRATE_SHFT \
ERTS_NO_PROC_PRIO_LEVELS
#define ERTS_RUNQ_FLGS_IMMIGRATE_SHFT \
(ERTS_RUNQ_FLGS_EMIGRATE_SHFT + ERTS_NO_PRIO_LEVELS)
#define ERTS_RUNQ_FLGS_EVACUATE_SHFT \
(ERTS_RUNQ_FLGS_IMMIGRATE_SHFT + ERTS_NO_PRIO_LEVELS)
#define ERTS_RUNQ_FLGS_EMIGRATE_QMASK \
(ERTS_RUNQ_FLGS_MIGRATE_QMASK << ERTS_RUNQ_FLGS_EMIGRATE_SHFT)
#define ERTS_RUNQ_FLGS_IMMIGRATE_QMASK \
(ERTS_RUNQ_FLGS_MIGRATE_QMASK << ERTS_RUNQ_FLGS_IMMIGRATE_SHFT)
#define ERTS_RUNQ_FLGS_EVACUATE_QMASK \
(ERTS_RUNQ_FLGS_MIGRATE_QMASK << ERTS_RUNQ_FLGS_EVACUATE_SHFT)
#define ERTS_RUNQ_FLG_BASE2 \
(ERTS_RUNQ_FLGS_EVACUATE_SHFT + ERTS_NO_PRIO_LEVELS)
#define ERTS_RUNQ_FLG_OUT_OF_WORK \
(((Uint32) 1) << (ERTS_RUNQ_FLG_BASE2 + 0))
#define ERTS_RUNQ_FLG_HALFTIME_OUT_OF_WORK \
(((Uint32) 1) << (ERTS_RUNQ_FLG_BASE2 + 1))
#define ERTS_RUNQ_FLG_SUSPENDED \
(((Uint32) 1) << (ERTS_RUNQ_FLG_BASE2 + 2))
#define ERTS_RUNQ_FLG_SHARED_RUNQ \
(((Uint32) 1) << (ERTS_RUNQ_FLG_BASE2 + 3))
#define ERTS_RUNQ_FLG_CHK_CPU_BIND \
(((Uint32) 1) << (ERTS_RUNQ_FLG_BASE2 + 4))
#define ERTS_RUNQ_FLG_INACTIVE \
(((Uint32) 1) << (ERTS_RUNQ_FLG_BASE2 + 5))
#define ERTS_RUNQ_FLGS_MIGRATION_QMASKS \
(ERTS_RUNQ_FLGS_EMIGRATE_QMASK \
| ERTS_RUNQ_FLGS_IMMIGRATE_QMASK \
| ERTS_RUNQ_FLGS_EVACUATE_QMASK)
#define ERTS_RUNQ_FLGS_MIGRATION_INFO \
(ERTS_RUNQ_FLGS_MIGRATION_QMASKS \
| ERTS_RUNQ_FLG_INACTIVE \
| ERTS_RUNQ_FLG_OUT_OF_WORK \
| ERTS_RUNQ_FLG_HALFTIME_OUT_OF_WORK)
#define ERTS_RUNQ_FLG_EMIGRATE(PRIO) \
(((Uint32) 1) << (ERTS_RUNQ_FLGS_EMIGRATE_SHFT + (PRIO)))
#define ERTS_CHK_RUNQ_FLG_EMIGRATE(FLGS, PRIO) \
((FLGS) & ERTS_RUNQ_FLG_EMIGRATE((PRIO)))
#define ERTS_SET_RUNQ_FLG_EMIGRATE(FLGS, PRIO) \
((FLGS) |= ERTS_RUNQ_FLG_EMIGRATE((PRIO)))
#define ERTS_UNSET_RUNQ_FLG_EMIGRATE(FLGS, PRIO) \
((FLGS) &= ~ERTS_RUNQ_FLG_EMIGRATE((PRIO)))
#define ERTS_RUNQ_FLG_IMMIGRATE(PRIO) \
(((Uint32) 1) << (ERTS_RUNQ_FLGS_IMMIGRATE_SHFT + (PRIO)))
#define ERTS_CHK_RUNQ_FLG_IMMIGRATE(FLGS, PRIO) \
((FLGS) & ERTS_RUNQ_FLG_IMMIGRATE((PRIO)))
#define ERTS_SET_RUNQ_FLG_IMMIGRATE(FLGS, PRIO) \
((FLGS) |= ERTS_RUNQ_FLG_IMMIGRATE((PRIO)))
#define ERTS_UNSET_RUNQ_FLG_IMMIGRATE(FLGS, PRIO) \
((FLGS) &= ~ERTS_RUNQ_FLG_IMMIGRATE((PRIO)))
#define ERTS_RUNQ_FLG_EVACUATE(PRIO) \
(((Uint32) 1) << (ERTS_RUNQ_FLGS_EVACUATE_SHFT + (PRIO)))
#define ERTS_CHK_RUNQ_FLG_EVACUATE(FLGS, PRIO) \
((FLGS) & ERTS_RUNQ_FLG_EVACUATE((PRIO)))
#define ERTS_SET_RUNQ_FLG_EVACUATE(FLGS, PRIO) \
((FLGS) |= ERTS_RUNQ_FLG_EVACUATE((PRIO)))
#define ERTS_UNSET_RUNQ_FLG_EVACUATE(FLGS, PRIO) \
((FLGS) &= ~ERTS_RUNQ_FLG_EVACUATE((PRIO)))
#define ERTS_RUNQ_IFLG_SUSPENDED (((long) 1) << 0)
#define ERTS_RUNQ_IFLG_NONEMPTY (((long) 1) << 1)
#ifdef DEBUG
# if defined(ARCH_64) && !HALFWORD_HEAP
# define ERTS_DBG_SET_INVALID_RUNQP(RQP, N) \
(*((char **) &(RQP)) = (char *) (0xdeadbeefdead0003 | ((N) << 4)))
# define ERTS_DBG_VERIFY_VALID_RUNQP(RQP) \
do { \
ASSERT((RQP) != NULL); \
ASSERT(((((Uint) (RQP)) & ((Uint) 0x3))) == ((Uint) 0)); \
ASSERT((((Uint) (RQP)) & ~((Uint) 0xffff)) != ((Uint) 0xdeadbeefdead0000));\
} while (0)
# else
# define ERTS_DBG_SET_INVALID_RUNQP(RQP, N) \
(*((char **) &(RQP)) = (char *) (0xdead0003 | ((N) << 4)))
# define ERTS_DBG_VERIFY_VALID_RUNQP(RQP) \
do { \
ASSERT((RQP) != NULL); \
ASSERT(((((UWord) (RQP)) & ((UWord) 1))) == ((UWord) 0)); \
ASSERT((((UWord) (RQP)) & ~((UWord) 0xffff)) != ((UWord) 0xdead0000)); \
} while (0)
# endif
#else
# define ERTS_DBG_SET_INVALID_RUNQP(RQP, N)
# define ERTS_DBG_VERIFY_VALID_RUNQP(RQP)
#endif
typedef enum {
ERTS_SCHDLR_SSPND_DONE_MSCHED_BLOCKED,
ERTS_SCHDLR_SSPND_YIELD_DONE_MSCHED_BLOCKED,
ERTS_SCHDLR_SSPND_DONE,
ERTS_SCHDLR_SSPND_YIELD_RESTART,
ERTS_SCHDLR_SSPND_YIELD_DONE,
ERTS_SCHDLR_SSPND_EINVAL
} ErtsSchedSuspendResult;
typedef enum {
ERTS_MIGRATE_SUCCESS,
ERTS_MIGRATE_FAILED_NOT_IN_RUNQ,
ERTS_MIGRATE_FAILED_RUNQ_CHANGED,
ERTS_MIGRATE_FAILED_RUNQ_SUSPENDED
} ErtsMigrateResult;
#define ERTS_SSI_FLG_SLEEPING (((long) 1) << 0)
#define ERTS_SSI_FLG_POLL_SLEEPING (((long) 1) << 1)
#define ERTS_SSI_FLG_TSE_SLEEPING (((long) 1) << 2)
#define ERTS_SSI_FLG_WAITING (((long) 1) << 3)
#define ERTS_SSI_FLG_SUSPENDED (((long) 1) << 4)
#define ERTS_SSI_FLGS_SLEEP_TYPE \
(ERTS_SSI_FLG_TSE_SLEEPING|ERTS_SSI_FLG_POLL_SLEEPING)
#define ERTS_SSI_FLGS_SLEEP \
(ERTS_SSI_FLG_SLEEPING|ERTS_SSI_FLGS_SLEEP_TYPE)
#define ERTS_SSI_FLGS_ALL \
(ERTS_SSI_FLGS_SLEEP \
| ERTS_SSI_FLG_WAITING \
| ERTS_SSI_FLG_SUSPENDED)
#if !defined(ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK) \
&& defined(ERTS_SMP_SCHEDULERS_NEED_TO_CHECK_CHILDREN)
#define ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK
#endif
#define ERTS_SSI_AUX_WORK_CHECK_CHILDREN (((long) 1) << 0)
#define ERTS_SSI_BLOCKABLE_AUX_WORK_MASK \
(ERTS_SSI_AUX_WORK_CHECK_CHILDREN)
#define ERTS_SSI_NONBLOCKABLE_AUX_WORK_MASK \
(0)
typedef struct ErtsSchedulerSleepInfo_ ErtsSchedulerSleepInfo;
typedef struct {
erts_smp_spinlock_t lock;
ErtsSchedulerSleepInfo *list;
} ErtsSchedulerSleepList;
struct ErtsSchedulerSleepInfo_ {
ErtsSchedulerSleepInfo *next;
ErtsSchedulerSleepInfo *prev;
erts_smp_atomic_t flags;
erts_tse_t *event;
erts_smp_atomic_t aux_work;
};
/* times to reschedule low prio process before running */
#define RESCHEDULE_LOW 8
#define ERTS_MAX_MISC_OPS 5
#define ERTS_FULL_REDS_HISTORY_AVG_SHFT 3
#define ERTS_FULL_REDS_HISTORY_SIZE \
((1 << ERTS_FULL_REDS_HISTORY_AVG_SHFT) - 1)
typedef struct ErtsProcList_ ErtsProcList;
struct ErtsProcList_ {
Eterm pid;
SysTimeval started;
ErtsProcList* next;
};
typedef struct ErtsMiscOpList_ ErtsMiscOpList;
struct ErtsMiscOpList_ {
ErtsMiscOpList *next;
void (*func)(void *arg);
void *arg;
};
typedef struct {
Process* first;
Process* last;
} ErtsRunPrioQueue;
typedef struct ErtsSchedulerData_ ErtsSchedulerData;
typedef struct ErtsRunQueue_ ErtsRunQueue;
typedef struct {
int len;
int max_len;
int reds;
struct {
struct {
int this;
int other;
} limit;
ErtsRunQueue *runq;
} migrate;
} ErtsRunQueueInfo;
struct ErtsRunQueue_ {
int ix;
erts_smp_atomic_t info_flags;
erts_smp_mtx_t mtx;
erts_smp_cnd_t cnd;
#ifdef ERTS_SMP
ErtsSchedulerSleepList sleepers;
#endif
ErtsSchedulerData *scheduler;
int waiting; /* < 0 in sys schedule; > 0 on cnd variable */
int woken;
Uint32 flags;
int check_balance_reds;
int full_reds_history_sum;
int full_reds_history[ERTS_FULL_REDS_HISTORY_SIZE];
int out_of_work_count;
int max_len;
int len;
int wakeup_other;
int wakeup_other_reds;
struct {
int len;
ErtsProcList *pending_exiters;
Uint context_switches;
Uint reductions;
ErtsRunQueueInfo prio_info[ERTS_NO_PROC_PRIO_LEVELS];
/* We use the same prio queue for low and
normal prio processes */
ErtsRunPrioQueue prio[ERTS_NO_PROC_PRIO_LEVELS-1];
} procs;
struct {
ErtsMiscOpList *start;
ErtsMiscOpList *end;
ErtsRunQueue *evac_runq;
} misc;
struct {
ErtsRunQueueInfo info;
struct port *start;
struct port *end;
} ports;
};
typedef union {
ErtsRunQueue runq;
char align[ERTS_ALC_CACHE_LINE_ALIGN_SIZE(sizeof(ErtsRunQueue))];
} ErtsAlignedRunQueue;
extern ErtsAlignedRunQueue *erts_aligned_run_queues;
extern ErtsRunQueue *erts_common_run_queue;
#define ERTS_PROC_REDUCTIONS_EXECUTED(RQ, PRIO, REDS, AREDS) \
do { \
(RQ)->procs.reductions += (AREDS); \
(RQ)->procs.prio_info[p->prio].reds += (REDS); \
(RQ)->check_balance_reds -= (REDS); \
(RQ)->wakeup_other_reds += (AREDS); \
} while (0)
#define ERTS_PORT_REDUCTIONS_EXECUTED(RQ, REDS) \
do { \
(RQ)->ports.info.reds += (REDS); \
(RQ)->check_balance_reds -= (REDS); \
(RQ)->wakeup_other_reds += (REDS); \
} while (0)
struct ErtsSchedulerData_ {
#ifdef ERTS_SMP
/*
* Keep X registers first (so we get as many low
* numbered registers as possible in the same cache
* line).
*/
#if !HALFWORD_HEAP
Eterm save_reg[ERTS_X_REGS_ALLOCATED]; /* X registers */
#else
Eterm *save_reg;
#endif
FloatDef freg[MAX_REG]; /* Floating point registers. */
ethr_tid tid; /* Thread id */
struct erl_bits_state erl_bits_state; /* erl_bits.c state */
void *match_pseudo_process; /* erl_db_util.c:db_prog_match() */
ErtsSchedulerSleepInfo *ssi;
Process *free_process;
#endif
#if !HEAP_ON_C_STACK
Eterm tmp_heap[TMP_HEAP_SIZE];
int num_tmp_heap_used;
Eterm beam_emu_tmp_heap[BEAM_EMU_TMP_HEAP_SIZE];
Eterm cmp_tmp_heap[CMP_TMP_HEAP_SIZE];
Eterm erl_arith_tmp_heap[ERL_ARITH_TMP_HEAP_SIZE];
#endif
Process *current_process;
Uint no; /* Scheduler number */
struct port *current_port;
ErtsRunQueue *run_queue;
int virtual_reds;
int cpu_id; /* >= 0 when bound */
ErtsAtomCacheMap atom_cache_map;
#ifdef ERTS_SMP
/* NOTE: These fields are modified under held mutexes by other threads */
erts_smp_atomic_t chk_cpu_bind; /* Only used when common run queue */
#endif
};
#ifndef ERTS_SMP
extern ErtsSchedulerData *erts_scheduler_data;
#endif
/*
* Process Specific Data.
*
* NOTE: Only use PSD for very rarely used data.
*/
#define ERTS_PSD_ERROR_HANDLER 0
#define ERTS_PSD_SAVED_CALLS_BUF 1
#define ERTS_PSD_SCHED_ID 2
#define ERTS_PSD_DIST_ENTRY 3
#define ERTS_PSD_CALL_TIME_BP 4
#define ERTS_PSD_SIZE 5
typedef struct {
void *data[ERTS_PSD_SIZE];
} ErtsPSD;
#ifdef ERTS_ENABLE_LOCK_CHECK
#define ERTS_LC_PSD_ANY_LOCK (~ERTS_PROC_LOCKS_ALL)
#define ERTS_PSD_ERROR_HANDLER_BUF_GET_LOCKS ERTS_PROC_LOCK_MAIN
#define ERTS_PSD_ERROR_HANDLER_BUF_SET_LOCKS ERTS_PROC_LOCK_MAIN
#define ERTS_PSD_SAVED_CALLS_BUF_GET_LOCKS ERTS_PROC_LOCK_MAIN
#define ERTS_PSD_SAVED_CALLS_BUF_SET_LOCKS ERTS_PROC_LOCK_MAIN
#define ERTS_PSD_SCHED_ID_GET_LOCKS ERTS_PROC_LOCK_STATUS
#define ERTS_PSD_SCHED_ID_SET_LOCKS ERTS_PROC_LOCK_STATUS
#define ERTS_PSD_DIST_ENTRY_GET_LOCKS ERTS_PROC_LOCK_MAIN
#define ERTS_PSD_DIST_ENTRY_SET_LOCKS ERTS_PROC_LOCK_MAIN
#define ERTS_PSD_CALL_TIME_BP_GET_LOCKS ERTS_PROC_LOCK_MAIN
#define ERTS_PSD_CALL_TIME_BP_SET_LOCKS ERTS_PROC_LOCK_MAIN
typedef struct {
ErtsProcLocks get_locks;
ErtsProcLocks set_locks;
} ErtsLcPSDLocks;
extern ErtsLcPSDLocks erts_psd_required_locks[ERTS_PSD_SIZE];
#endif
#define ERTS_SCHED_STAT_MODIFY_DISABLE 1
#define ERTS_SCHED_STAT_MODIFY_ENABLE 2
#define ERTS_SCHED_STAT_MODIFY_CLEAR 3
typedef struct {
erts_smp_spinlock_t lock;
int enabled;
struct {
Eterm name;
Uint total_executed;
Uint executed;
Uint total_migrated;
Uint migrated;
} prio[ERTS_NO_PRIO_LEVELS];
} erts_sched_stat_t;
extern erts_sched_stat_t erts_sched_stat;
typedef struct {
Eterm reason;
ErlHeapFragment *bp;
} ErtsPendExit;
#ifdef ERTS_SMP
typedef struct ErtsPendingSuspend_ ErtsPendingSuspend;
struct ErtsPendingSuspend_ {
ErtsPendingSuspend *next;
ErtsPendingSuspend *end;
Eterm pid;
void (*handle_func)(Process *suspendee,
ErtsProcLocks suspendee_locks,
int suspendee_alive,
Eterm pid);
};
#endif
/* Defines to ease the change of memory architecture */
# define HEAP_START(p) (p)->heap
# define HEAP_TOP(p) (p)->htop
# define HEAP_LIMIT(p) (p)->stop
# define HEAP_END(p) (p)->hend
# define HEAP_SIZE(p) (p)->heap_sz
# define STACK_START(p) (p)->hend
# define STACK_TOP(p) (p)->stop
# define STACK_END(p) (p)->htop
# define HIGH_WATER(p) (p)->high_water
# define OLD_HEND(p) (p)->old_hend
# define OLD_HTOP(p) (p)->old_htop
# define OLD_HEAP(p) (p)->old_heap
# define GEN_GCS(p) (p)->gen_gcs
# define MAX_GEN_GCS(p) (p)->max_gen_gcs
# define FLAGS(p) (p)->flags
# define MBUF(p) (p)->mbuf
# define HALLOC_MBUF(p) (p)->halloc_mbuf
# define MBUF_SIZE(p) (p)->mbuf_sz
# define MSO(p) (p)->off_heap
# define MIN_HEAP_SIZE(p) (p)->min_heap_size
# define MIN_VHEAP_SIZE(p) (p)->min_vheap_size
# define BIN_VHEAP_SZ(p) (p)->bin_vheap_sz
# define BIN_OLD_VHEAP_SZ(p) (p)->bin_old_vheap_sz
# define BIN_OLD_VHEAP(p) (p)->bin_old_vheap
struct process {
/* All fields in the PCB that differs between different heap
* architectures, have been moved to the end of this struct to
* make sure that as few offsets as possible differ. Different
* offsets between memory architectures in this struct, means that
* native code have to use functions instead of constants.
*/
Eterm* htop; /* Heap top */
Eterm* stop; /* Stack top */
Eterm* heap; /* Heap start */
Eterm* hend; /* Heap end */
Uint heap_sz; /* Size of heap in words */
Uint min_heap_size; /* Minimum size of heap (in words). */
Uint min_vheap_size; /* Minimum size of virtual heap (in words). */
#if !defined(NO_FPE_SIGNALS)
volatile unsigned long fp_exception;
#endif
#ifdef HIPE
/* HiPE-specific process fields. Put it early in struct process,
to enable smaller & faster addressing modes on the x86. */
struct hipe_process_state hipe;
#endif
/*
* Saved x registers.
*/
Uint arity; /* Number of live argument registers (only valid
* when process is *not* running).
*/
Eterm* arg_reg; /* Pointer to argument registers. */
unsigned max_arg_reg; /* Maximum number of argument registers available. */
Eterm def_arg_reg[6]; /* Default array for argument registers. */
BeamInstr* cp; /* (untagged) Continuation pointer (for threaded code). */
BeamInstr* i; /* Program counter for threaded code. */
Sint catches; /* Number of catches on stack */
Sint fcalls; /*
* Number of reductions left to execute.
* Only valid for the current process.
*/
Uint32 status; /* process STATE */
Uint32 gcstatus; /* process gc STATE */
Uint32 rstatus; /* process resume STATE */
Uint32 rcount; /* suspend count */
Eterm id; /* The pid of this process */
int prio; /* Priority of process */
int skipped; /* Times a low prio process has been rescheduled */
Uint reds; /* No of reductions for this process */
Eterm tracer_proc; /* If proc is traced, this is the tracer
(can NOT be boxed) */
Uint trace_flags; /* Trace flags (used to be in flags) */
Eterm group_leader; /* Pid in charge
(can be boxed) */
Uint flags; /* Trap exit, etc (no trace flags anymore) */
Eterm fvalue; /* Exit & Throw value (failure reason) */
Uint freason; /* Reason for detected failure */
Eterm ftrace; /* Latest exception stack trace dump */
Process *next; /* Pointer to next process in run queue */
Process *prev; /* Pointer to prev process in run queue */
struct reg_proc *reg; /* NULL iff not registered */
ErtsLink *nlinks;
ErtsMonitor *monitors; /* The process monitors, both ends */
struct ErtsNodesMonitor_ *nodes_monitors;
ErtsSuspendMonitor *suspend_monitors; /* Processes suspended by
this process via
erlang:suspend_process/1 */
ErlMessageQueue msg; /* Message queue */
ErtsBifTimer *bif_timers; /* Bif timers aiming at this process */
ProcDict *dictionary; /* Process dictionary, may be NULL */
Uint seq_trace_clock;
Uint seq_trace_lastcnt;
Eterm seq_trace_token; /* Sequential trace token (tuple size 5 see below) */
BeamInstr initial[3]; /* Initial module(0), function(1), arity(2), often used instead
of pointer to funcinfo instruction, hence the BeamInstr datatype */
BeamInstr* current; /* Current Erlang function, part of the funcinfo:
* module(0), function(1), arity(2)
* (module and functions are tagged atoms;
* arity an untagged integer). BeamInstr * because it references code
*/
/*
* Information mainly for post-mortem use (erl crash dump).
*/
Eterm parent; /* Pid of process that created this process. */
SysTimeval started; /* Time when started. */
/* This is the place, where all fields that differs between memory
* architectures, have gone to.
*/
Eterm *high_water;
Eterm *old_hend; /* Heap pointers for generational GC. */
Eterm *old_htop;
Eterm *old_heap;
Uint16 gen_gcs; /* Number of (minor) generational GCs. */
Uint16 max_gen_gcs; /* Max minor gen GCs before fullsweep. */
ErlOffHeap off_heap; /* Off-heap data updated by copy_struct(). */
ErlHeapFragment* mbuf; /* Pointer to message buffer list */
Uint mbuf_sz; /* Size of all message buffers */
ErtsPSD *psd; /* Rarely used process specific data */
Uint bin_vheap_sz; /* Virtual heap block size for binaries */
Uint bin_old_vheap_sz; /* Virtual old heap block size for binaries */
Uint bin_old_vheap; /* Virtual old heap size for binaries */
union {
#ifdef ERTS_SMP
ErtsSmpPTimer *ptimer;
#else
ErlTimer tm; /* Timer entry */
#endif
void *exit_data; /* Misc data referred during termination */
} u;
ErtsRunQueue *bound_runq;
#ifdef ERTS_SMP
erts_proc_lock_t lock;
ErtsSchedulerData *scheduler_data;
int is_exiting;
Uint32 runq_flags;
Uint32 status_flags;
ErlMessageInQueue msg_inq;
Eterm suspendee;
ErtsPendingSuspend *pending_suspenders;
ErtsPendExit pending_exit;
ErtsRunQueue *run_queue;
#ifdef HIPE
struct hipe_process_state_smp hipe_smp;
#endif
#endif
#ifdef HYBRID
Eterm *rrma; /* Remembered roots to Message Area */
Eterm **rrsrc; /* The source of the root */
Uint nrr; /* Number of remembered roots */
Uint rrsz; /* Size of root array */
#endif
#ifdef HYBRID
Uint active; /* Active since last major collection? */
Uint active_index; /* Index in the active process array */
#endif
#ifdef INCREMENTAL
Process *active_next; /* Active processes to scan for roots */
Process *active_prev; /* in collection of the message area */
Eterm *scan_top;
#endif
#ifdef CHECK_FOR_HOLES
Eterm* last_htop; /* No need to scan the heap below this point. */
ErlHeapFragment* last_mbuf; /* No need to scan beyond this mbuf. */
#endif
#ifdef DEBUG
Eterm* last_old_htop; /*
* No need to scan the old heap below this point
* when looking for invalid pointers into the new heap or
* heap fragments.
*/
#endif
#ifdef FORCE_HEAP_FRAGS
Uint space_verified; /* Avoid HAlloc forcing heap fragments when */
Eterm* space_verified_from; /* we rely on available heap space (TestHeap) */
#endif
};
#ifdef CHECK_FOR_HOLES
# define INIT_HOLE_CHECK(p) \
do { \
(p)->last_htop = 0; \
(p)->last_mbuf = 0; \
} while (0)
# define ERTS_HOLE_CHECK(p) erts_check_for_holes((p))
void erts_check_for_holes(Process* p);
#else
# define INIT_HOLE_CHECK(p)
# define ERTS_HOLE_CHECK(p)
#endif
/*
* The MBUF_GC_FACTOR decides how easily a process is subject to GC
* due to message buffers allocated outside the heap.
* The larger the factor, the easier the process gets GCed.
* On a small memory system with lots of processes, this makes a significant
* difference, especially since the GCs help fragmentation quite a bit too.
*/
#if defined(SMALL_MEMORY)
#define MBUF_GC_FACTOR 4
#else
#define MBUF_GC_FACTOR 1
#endif
#define SEQ_TRACE_TOKEN(p) ((p)->seq_trace_token)
/* The sequential tracing token is a tuple of size 5:
*
* {Flags, Label, Serial, Sender}
*/
#define SEQ_TRACE_TOKEN_ARITY(p) (arityval(*(tuple_val(SEQ_TRACE_TOKEN(p)))))
#define SEQ_TRACE_TOKEN_FLAGS(p) (*(tuple_val(SEQ_TRACE_TOKEN(p)) + 1))
#define SEQ_TRACE_TOKEN_LABEL(p) (*(tuple_val(SEQ_TRACE_TOKEN(p)) + 2))
#define SEQ_TRACE_TOKEN_SERIAL(p) (*(tuple_val(SEQ_TRACE_TOKEN(p)) + 3))
#define SEQ_TRACE_TOKEN_SENDER(p) (*(tuple_val(SEQ_TRACE_TOKEN(p)) + 4))
#define SEQ_TRACE_TOKEN_LASTCNT(p) (*(tuple_val(SEQ_TRACE_TOKEN(p)) + 5))
/* used when we have unit32 token */
#define SEQ_TRACE_T_ARITY(token) (arityval(*(tuple_val(token))))
#define SEQ_TRACE_T_FLAGS(token) (*(tuple_val(token) + 1))
#define SEQ_TRACE_T_LABEL(token) (*(tuple_val(token) + 2))
#define SEQ_TRACE_T_SERIAL(token) (*(tuple_val(token) + 3))
#define SEQ_TRACE_T_SENDER(token) (*(tuple_val(token) + 4))
#define SEQ_TRACE_T_LASTCNT(token) (*(tuple_val(token) + 5))
/*
* Possible flags for the flags field in ErlSpawnOpts below.
*/
#define SPO_LINK 1
#define SPO_USE_ARGS 2
#define SPO_MONITOR 4
/*
* The following struct contains options for a process to be spawned.
*/
typedef struct {
Uint flags;
int error_code; /* Error code returned from create_process(). */
Eterm mref; /* Monitor ref returned (if SPO_MONITOR was given). */
/*
* The following items are only initialized if the SPO_USE_ARGS flag is set.
*/
Uint min_heap_size; /* Minimum heap size (must be a valued returned
* from next_heap_size()). */
Uint min_vheap_size; /* Minimum virtual heap size */
int priority; /* Priority for process. */
Uint16 max_gen_gcs; /* Maximum number of gen GCs before fullsweep. */
int scheduler;
} ErlSpawnOpts;
/*
* The KILL_CATCHES(p) macro kills pending catches for process p.
*/
#define KILL_CATCHES(p) (p)->catches = -1
/* Shrink heap fragment from _last_ HAlloc.
*/
ERTS_GLB_INLINE void erts_heap_frag_shrink(Process* p, Eterm* hp);
#if ERTS_GLB_INLINE_INCL_FUNC_DEF
ERTS_GLB_INLINE void erts_heap_frag_shrink(Process* p, Eterm* hp)
{
ErlHeapFragment* hf = MBUF(p);
ASSERT(hf!=NULL && (hp - hf->mem < (unsigned long)hf->alloc_size));
hf->used_size = hp - hf->mem;
}
#endif /* inline */
Eterm* erts_heap_alloc(Process* p, Uint need);
#ifdef CHECK_FOR_HOLES
Eterm* erts_set_hole_marker(Eterm* ptr, Uint sz);
#endif
extern Process** process_tab;
#ifdef HYBRID
extern Uint erts_num_active_procs;
extern Process** erts_active_procs;
#endif
extern Uint erts_max_processes;
extern Uint erts_process_tab_index_mask;
extern Uint erts_default_process_flags;
extern erts_smp_rwmtx_t erts_cpu_bind_rwmtx;
/* If any of the erts_system_monitor_* variables are set (enabled),
** erts_system_monitor must be != NIL, to allow testing on just
** the erts_system_monitor_* variables.
*/
extern Eterm erts_system_monitor;
extern Uint erts_system_monitor_long_gc;
extern Uint erts_system_monitor_large_heap;
struct erts_system_monitor_flags_t {
unsigned int busy_port : 1;
unsigned int busy_dist_port : 1;
};
extern struct erts_system_monitor_flags_t erts_system_monitor_flags;
/* system_profile, same rules as for system_monitor.
erts_profile must be != NIL when
erts_profile_* is set. */
extern Eterm erts_system_profile;
struct erts_system_profile_flags_t {
unsigned int scheduler : 1;
unsigned int runnable_procs : 1;
unsigned int runnable_ports : 1;
unsigned int exclusive : 1;
};
extern struct erts_system_profile_flags_t erts_system_profile_flags;
#define INVALID_PID(p, pid) ((p) == NULL \
|| (p)->id != (pid) \
|| (p)->status == P_EXITING)
#define IS_TRACED(p) ( (p)->tracer_proc != NIL )
#define ARE_TRACE_FLAGS_ON(p,tf) ( ((p)->trace_flags & (tf|F_SENSITIVE)) == (tf) )
#define IS_TRACED_FL(p,tf) ( IS_TRACED(p) && ARE_TRACE_FLAGS_ON(p,tf) )
/* process flags */
#define F_TRAPEXIT (1 << 0)
#define F_INSLPQUEUE (1 << 1) /* Set if in timer queue */
#define F_TIMO (1 << 2) /* Set if timeout */
#define F_HEAP_GROW (1 << 3)
#define F_NEED_FULLSWEEP (1 << 4)
#define F_USING_DB (1 << 5) /* If have created tables */
#define F_DISTRIBUTION (1 << 6) /* Process used in distribution */
#define F_USING_DDLL (1 << 7) /* Process has used the DDLL interface */
#define F_HAVE_BLCKD_MSCHED (1 << 8) /* Process has blocked multi-scheduling */
#define F_P2PNR_RESCHED (1 << 9) /* Process has been rescheduled via erts_pid2proc_not_running() */
#define F_FORCE_GC (1 << 10) /* Force gc at process in-scheduling */
/* process trace_flags */
#define F_SENSITIVE (1 << 0)
#define F_TRACE_SEND (1 << 1)
#define F_TRACE_RECEIVE (1 << 2)
#define F_TRACE_SOS (1 << 3) /* Set on spawn */
#define F_TRACE_SOS1 (1 << 4) /* Set on first spawn */
#define F_TRACE_SOL (1 << 5) /* Set on link */
#define F_TRACE_SOL1 (1 << 6) /* Set on first link */
#define F_TRACE_CALLS (1 << 7)
#define F_TIMESTAMP (1 << 8)
#define F_TRACE_PROCS (1 << 9)
#define F_TRACE_FIRST_CHILD (1 << 10)
#define F_TRACE_SCHED (1 << 11)
#define F_TRACE_GC (1 << 12)
#define F_TRACE_ARITY_ONLY (1 << 13)
#define F_TRACE_RETURN_TO (1 << 14) /* Return_to trace when breakpoint tracing */
#define F_TRACE_SILENT (1 << 15) /* No call trace msg suppress */
#define F_TRACER (1 << 16) /* May be (has been) tracer */
#define F_EXCEPTION_TRACE (1 << 17) /* May have exception trace on stack */
/* port trace flags, currently the same as process trace flags */
#define F_TRACE_SCHED_PORTS (1 << 18) /* Trace of port scheduling */
#define F_TRACE_SCHED_PROCS (1 << 19) /* With virtual scheduling */
#define F_TRACE_PORTS (1 << 20) /* Ports equivalent to F_TRACE_PROCS */
#define F_TRACE_SCHED_NO (1 << 21) /* Trace with scheduler id */
#define F_TRACE_SCHED_EXIT (1 << 22)
#define F_NUM_FLAGS 23
#ifdef DEBUG
# define F_INITIAL_TRACE_FLAGS (5 << F_NUM_FLAGS)
#else
# define F_INITIAL_TRACE_FLAGS 0
#endif
#define TRACEE_FLAGS ( F_TRACE_PROCS | F_TRACE_CALLS \
| F_TRACE_SOS | F_TRACE_SOS1| F_TRACE_RECEIVE \
| F_TRACE_SOL | F_TRACE_SOL1 | F_TRACE_SEND \
| F_TRACE_SCHED | F_TIMESTAMP | F_TRACE_GC \
| F_TRACE_ARITY_ONLY | F_TRACE_RETURN_TO \
| F_TRACE_SILENT | F_TRACE_SCHED_PROCS | F_TRACE_PORTS \
| F_TRACE_SCHED_PORTS | F_TRACE_SCHED_NO \
| F_TRACE_SCHED_EXIT)
#define ERTS_TRACEE_MODIFIER_FLAGS \
(F_TRACE_SILENT | F_TIMESTAMP | F_TRACE_SCHED_NO)
#define ERTS_PORT_TRACEE_FLAGS \
(ERTS_TRACEE_MODIFIER_FLAGS | F_TRACE_PORTS | F_TRACE_SCHED_PORTS)
#define ERTS_PROC_TRACEE_FLAGS \
((TRACEE_FLAGS & ~ERTS_PORT_TRACEE_FLAGS) | ERTS_TRACEE_MODIFIER_FLAGS)
/* Sequential trace flags */
#define SEQ_TRACE_SEND (1 << 0)
#define SEQ_TRACE_RECEIVE (1 << 1)
#define SEQ_TRACE_PRINT (1 << 2)
#define SEQ_TRACE_TIMESTAMP (1 << 3)
#ifdef ERTS_SMP
/* Status flags ... */
#define ERTS_PROC_SFLG_PENDADD2SCHEDQ (((Uint32) 1) << 0) /* Pending
add to
schedule q */
#define ERTS_PROC_SFLG_INRUNQ (((Uint32) 1) << 1) /* Process is
in run q */
#define ERTS_PROC_SFLG_TRAPEXIT (((Uint32) 1) << 2) /* Process is
trapping
exit */
#define ERTS_PROC_SFLG_RUNNING (((Uint32) 1) << 3) /* Process is
running */
/* Scheduler flags in process struct... */
#define ERTS_PROC_RUNQ_FLG_RUNNING (((Uint32) 1) << 0) /* Process is
running */
#endif
#ifdef ERTS_SMP
#define ERTS_PROC_IS_TRAPPING_EXITS(P) \
(ERTS_SMP_LC_ASSERT(erts_proc_lc_my_proc_locks((P)) \
& ERTS_PROC_LOCK_STATUS), \
(P)->status_flags & ERTS_PROC_SFLG_TRAPEXIT)
#define ERTS_PROC_SET_TRAP_EXIT(P) \
(ERTS_SMP_LC_ASSERT(((ERTS_PROC_LOCK_MAIN|ERTS_PROC_LOCK_STATUS) \
& erts_proc_lc_my_proc_locks((P))) \
== (ERTS_PROC_LOCK_MAIN|ERTS_PROC_LOCK_STATUS)), \
(P)->status_flags |= ERTS_PROC_SFLG_TRAPEXIT, \
(P)->flags |= F_TRAPEXIT, \
1)
#define ERTS_PROC_UNSET_TRAP_EXIT(P) \
(ERTS_SMP_LC_ASSERT(((ERTS_PROC_LOCK_MAIN|ERTS_PROC_LOCK_STATUS) \
& erts_proc_lc_my_proc_locks((P))) \
== (ERTS_PROC_LOCK_MAIN|ERTS_PROC_LOCK_STATUS)), \
(P)->status_flags &= ~ERTS_PROC_SFLG_TRAPEXIT, \
(P)->flags &= ~F_TRAPEXIT, \
0)
#else
#define ERTS_PROC_IS_TRAPPING_EXITS(P) ((P)->flags & F_TRAPEXIT)
#define ERTS_PROC_SET_TRAP_EXIT(P) ((P)->flags |= F_TRAPEXIT, 1)
#define ERTS_PROC_UNSET_TRAP_EXIT(P) ((P)->flags &= ~F_TRAPEXIT, 0)
#endif
/* Option flags to erts_send_exit_signal() */
#define ERTS_XSIG_FLG_IGN_KILL (((Uint32) 1) << 0)
#define ERTS_XSIG_FLG_NO_IGN_NORMAL (((Uint32) 1) << 1)
/* Process status values */
#define P_FREE 0
#define P_RUNABLE 1
#define P_WAITING 2
#define P_RUNNING 3
#define P_EXITING 4
#define P_GARBING 5
#define P_SUSPENDED 6
#define CANCEL_TIMER(p) \
do { \
if ((p)->flags & (F_INSLPQUEUE)) \
cancel_timer(p); \
else \
(p)->flags &= ~F_TIMO; \
} while (0)
#define ERTS_INIT_SCHED_BIND_TYPE_SUCCESS 0
#define ERTS_INIT_SCHED_BIND_TYPE_NOT_SUPPORTED 1
#define ERTS_INIT_SCHED_BIND_TYPE_ERROR_NO_CPU_TOPOLOGY 2
#define ERTS_INIT_SCHED_BIND_TYPE_ERROR_NO_BAD_TYPE 3
int erts_init_scheduler_bind_type(char *how);
#define ERTS_INIT_CPU_TOPOLOGY_OK 0
#define ERTS_INIT_CPU_TOPOLOGY_INVALID_ID 1
#define ERTS_INIT_CPU_TOPOLOGY_INVALID_ID_RANGE 2
#define ERTS_INIT_CPU_TOPOLOGY_INVALID_HIERARCHY 3
#define ERTS_INIT_CPU_TOPOLOGY_INVALID_ID_TYPE 4
#define ERTS_INIT_CPU_TOPOLOGY_INVALID_NODES 5
#define ERTS_INIT_CPU_TOPOLOGY_MISSING_LID 6
#define ERTS_INIT_CPU_TOPOLOGY_NOT_UNIQUE_LIDS 7
#define ERTS_INIT_CPU_TOPOLOGY_NOT_UNIQUE_ENTITIES 8
#define ERTS_INIT_CPU_TOPOLOGY_MISSING 9
int erts_init_cpu_topology(char *topology_str);
void erts_pre_init_process(void);
void erts_late_init_process(void);
void erts_early_init_scheduling(void);
void erts_init_scheduling(int, int, int);
ErtsProcList *erts_proclist_create(Process *);
void erts_proclist_destroy(ErtsProcList *);
int erts_proclist_same(ErtsProcList *, Process *);
#ifdef DEBUG
void erts_dbg_multi_scheduling_return_trap(Process *, Eterm);
#endif
#ifdef ERTS_SMP
ErtsSchedSuspendResult
erts_schedulers_state(Uint *, Uint *, Uint *, int);
ErtsSchedSuspendResult
erts_set_schedulers_online(Process *p,
ErtsProcLocks plocks,
Sint new_no,
Sint *old_no);
ErtsSchedSuspendResult
erts_block_multi_scheduling(Process *, ErtsProcLocks, int, int);
int erts_is_multi_scheduling_blocked(void);
Eterm erts_multi_scheduling_blockers(Process *);
void erts_start_schedulers(void);
void erts_smp_notify_check_children_needed(void);
#endif
Uint erts_active_schedulers(void);
void erts_init_process(void);
Eterm erts_process_status(Process *, ErtsProcLocks, Process *, Eterm);
Uint erts_run_queues_len(Uint *);
void erts_add_to_runq(Process *);
Eterm erts_bound_schedulers_term(Process *c_p);
Eterm erts_get_cpu_topology_term(Process *c_p, Eterm which);
Eterm erts_get_schedulers_binds(Process *c_p);
Eterm erts_set_cpu_topology(Process *c_p, Eterm term);
Eterm erts_bind_schedulers(Process *c_p, Eterm how);
ErtsRunQueue *erts_schedid2runq(Uint);
#ifdef ERTS_SMP
ErtsMigrateResult erts_proc_migrate(Process *,
ErtsProcLocks *,
ErtsRunQueue *,
int *,
ErtsRunQueue *,
int *);
#endif
Process *schedule(Process*, int);
void erts_schedule_misc_op(void (*)(void *), void *);
Eterm erl_create_process(Process*, Eterm, Eterm, Eterm, ErlSpawnOpts*);
void erts_do_exit_process(Process*, Eterm);
void erts_continue_exit_process(Process *);
void set_timer(Process*, Uint);
void cancel_timer(Process*);
/* Begin System profile */
Uint erts_runnable_process_count(void);
Uint erts_process_count(void);
/* End System profile */
void erts_init_empty_process(Process *p);
void erts_cleanup_empty_process(Process* p);
#ifdef DEBUG
void erts_debug_verify_clean_empty_process(Process* p);
#endif
void erts_stack_dump(int to, void *to_arg, Process *);
void erts_program_counter_info(int to, void *to_arg, Process *);
Eterm erts_get_process_priority(Process *p);
Eterm erts_set_process_priority(Process *p, Eterm prio);
Uint erts_get_total_context_switches(void);
void erts_get_total_reductions(Uint *, Uint *);
void erts_get_exact_total_reductions(Process *, Uint *, Uint *);
Eterm erts_fake_scheduler_bindings(Process *p, Eterm how);
void erts_sched_stat_modify(int what);
Eterm erts_sched_stat_term(Process *p, int total);
void erts_free_proc(Process *);
void erts_suspend(Process*, ErtsProcLocks, struct port*);
void erts_resume(Process*, ErtsProcLocks);
int erts_resume_processes(ErtsProcList *);
int erts_send_exit_signal(Process *,
Eterm,
Process *,
ErtsProcLocks *,
Eterm,
Eterm,
Process *,
Uint32);
#ifdef ERTS_SMP
void erts_handle_pending_exit(Process *, ErtsProcLocks);
#define ERTS_PROC_PENDING_EXIT(P) \
(ERTS_SMP_LC_ASSERT(erts_proc_lc_my_proc_locks((P)) & ERTS_PROC_LOCK_STATUS),\
(P)->pending_exit.reason != THE_NON_VALUE)
#else
#define ERTS_PROC_PENDING_EXIT(P) 0
#endif
void erts_deep_process_dump(int, void *);
Eterm erts_get_reader_groups_map(Process *c_p);
Eterm erts_debug_reader_groups_map(Process *c_p, int groups);
Sint erts_test_next_pid(int, Uint);
Eterm erts_debug_processes(Process *c_p);
Eterm erts_debug_processes_bif_info(Process *c_p);
Uint erts_debug_nbalance(void);
#ifdef ERTS_SMP
# define ERTS_GET_SCHEDULER_DATA_FROM_PROC(PROC) ((PROC)->scheduler_data)
# define ERTS_PROC_GET_SCHDATA(PROC) ((PROC)->scheduler_data)
#else
# define ERTS_GET_SCHEDULER_DATA_FROM_PROC(PROC) (erts_scheduler_data)
# define ERTS_PROC_GET_SCHDATA(PROC) (erts_scheduler_data)
#endif
#if defined(ERTS_SMP) || defined(USE_THREADS)
ErtsSchedulerData *erts_get_scheduler_data(void);
#else
ERTS_GLB_INLINE ErtsSchedulerData *erts_get_scheduler_data(void);
#if ERTS_GLB_INLINE_INCL_FUNC_DEF
ERTS_GLB_INLINE
ErtsSchedulerData *erts_get_scheduler_data(void)
{
return erts_scheduler_data;
}
#endif
#endif
#if defined(ERTS_SMP) && defined(ERTS_ENABLE_LOCK_CHECK)
#define ERTS_PROCESS_LOCK_ONLY_LOCK_CHECK_PROTO__
#include "erl_process_lock.h"
#undef ERTS_PROCESS_LOCK_ONLY_LOCK_CHECK_PROTO__
int erts_smp_lc_runq_is_locked(ErtsRunQueue *);
#define ERTS_SMP_LC_CHK_RUNQ_LOCK(RQ, L) \
do { \
if ((L)) \
ERTS_SMP_LC_ASSERT(erts_smp_lc_runq_is_locked((RQ))); \
else \
ERTS_SMP_LC_ASSERT(!erts_smp_lc_runq_is_locked((RQ))); \
} while (0)
#else
#define ERTS_SMP_LC_CHK_RUNQ_LOCK(RQ, L)
#endif
void *erts_psd_set_init(Process *p, ErtsProcLocks plocks, int ix, void *data);
ERTS_GLB_INLINE void *
erts_psd_get(Process *p, int ix);
ERTS_GLB_INLINE void *
erts_psd_set(Process *p, ErtsProcLocks plocks, int ix, void *new);
#if ERTS_GLB_INLINE_INCL_FUNC_DEF
ERTS_GLB_INLINE void *
erts_psd_get(Process *p, int ix)
{
#if defined(ERTS_SMP) && defined(ERTS_ENABLE_LOCK_CHECK)
ErtsProcLocks locks = erts_proc_lc_my_proc_locks(p);
if (ERTS_LC_PSD_ANY_LOCK == erts_psd_required_locks[ix].get_locks)
ERTS_SMP_LC_ASSERT(locks
|| erts_is_system_blocked(0)
|| (ERTS_IS_CRASH_DUMPING
&& erts_is_system_blocked(ERTS_BS_FLG_ALLOW_GC)));
else {
locks &= erts_psd_required_locks[ix].get_locks;
ERTS_SMP_LC_ASSERT(erts_psd_required_locks[ix].get_locks == locks
|| erts_is_system_blocked(0)
|| (ERTS_IS_CRASH_DUMPING
&& erts_is_system_blocked(ERTS_BS_FLG_ALLOW_GC)));
}
#endif
ASSERT(0 <= ix && ix < ERTS_PSD_SIZE);
return p->psd ? p->psd->data[ix] : NULL;
}
/*
* NOTE: erts_psd_set() might release and reacquire locks on 'p'.
*/
ERTS_GLB_INLINE void *
erts_psd_set(Process *p, ErtsProcLocks plocks, int ix, void *data)
{
#if defined(ERTS_SMP) && defined(ERTS_ENABLE_LOCK_CHECK)
ErtsProcLocks locks = erts_proc_lc_my_proc_locks(p);
if (ERTS_LC_PSD_ANY_LOCK == erts_psd_required_locks[ix].set_locks)
ERTS_SMP_LC_ASSERT(locks
|| erts_is_system_blocked(0)
|| (ERTS_IS_CRASH_DUMPING
&& erts_is_system_blocked(ERTS_BS_FLG_ALLOW_GC)));
else {
locks &= erts_psd_required_locks[ix].set_locks;
ERTS_SMP_LC_ASSERT(erts_psd_required_locks[ix].set_locks == locks
|| erts_is_system_blocked(0)
|| (ERTS_IS_CRASH_DUMPING
&& erts_is_system_blocked(ERTS_BS_FLG_ALLOW_GC)));
}
#endif
ASSERT(0 <= ix && ix < ERTS_PSD_SIZE);
if (p->psd) {
void *old = p->psd->data[ix];
p->psd->data[ix] = data;
return old;
}
else {
if (!data)
return NULL;
else
return erts_psd_set_init(p, plocks, ix, data);
}
}
#endif
#define ERTS_PROC_SCHED_ID(P, L, ID) \
((UWord) erts_psd_set((P), (L), ERTS_PSD_SCHED_ID, (void *) (ID)))
#define ERTS_PROC_GET_DIST_ENTRY(P) \
((DistEntry *) erts_psd_get((P), ERTS_PSD_DIST_ENTRY))
#define ERTS_PROC_SET_DIST_ENTRY(P, L, D) \
((DistEntry *) erts_psd_set((P), (L), ERTS_PSD_DIST_ENTRY, (void *) (D)))
#define ERTS_PROC_GET_SAVED_CALLS_BUF(P) \
((struct saved_calls *) erts_psd_get((P), ERTS_PSD_SAVED_CALLS_BUF))
#define ERTS_PROC_SET_SAVED_CALLS_BUF(P, L, SCB) \
((struct saved_calls *) erts_psd_set((P), (L), ERTS_PSD_SAVED_CALLS_BUF, (void *) (SCB)))
#define ERTS_PROC_GET_CALL_TIME(P) \
((process_breakpoint_time_t *) erts_psd_get((P), ERTS_PSD_CALL_TIME_BP))
#define ERTS_PROC_SET_CALL_TIME(P, L, PBT) \
((process_breakpoint_time_t *) erts_psd_set((P), (L), ERTS_PSD_CALL_TIME_BP, (void *) (PBT)))
ERTS_GLB_INLINE Eterm erts_proc_get_error_handler(Process *p);
ERTS_GLB_INLINE Eterm erts_proc_set_error_handler(Process *p,
ErtsProcLocks plocks,
Eterm handler);
#if ERTS_GLB_INLINE_INCL_FUNC_DEF
ERTS_GLB_INLINE Eterm
erts_proc_get_error_handler(Process *p)
{
void *val = erts_psd_get(p, ERTS_PSD_ERROR_HANDLER);
if (!val)
return am_error_handler;
else {
ASSERT(is_atom(((Eterm) (UWord) val)));
return (Eterm) (UWord) val;
}
}
ERTS_GLB_INLINE Eterm
erts_proc_set_error_handler(Process *p, ErtsProcLocks plocks, Eterm handler)
{
void *old_val;
void *new_val;
ASSERT(is_atom(handler));
new_val = (handler == am_error_handler) ? NULL : (void *) (UWord) handler;
old_val = erts_psd_set(p, plocks, ERTS_PSD_ERROR_HANDLER, new_val);
if (!old_val)
return am_error_handler;
else {
ASSERT(is_atom(((Eterm) (UWord) old_val)));
return (Eterm) (UWord) old_val;
}
}
#endif
#ifdef ERTS_SMP
ErtsRunQueue *erts_prepare_emigrate(ErtsRunQueue *c_rq,
ErtsRunQueueInfo *c_rqi,
int prio);
ERTS_GLB_INLINE ErtsRunQueue *erts_check_emigration_need(ErtsRunQueue *c_rq,
int prio);
#endif
ERTS_GLB_INLINE int erts_is_scheduler_bound(ErtsSchedulerData *esdp);
ERTS_GLB_INLINE Process *erts_get_current_process(void);
ERTS_GLB_INLINE Eterm erts_get_current_pid(void);
ERTS_GLB_INLINE Uint erts_get_scheduler_id(void);
ERTS_GLB_INLINE ErtsRunQueue *erts_get_runq_proc(Process *p);
ERTS_GLB_INLINE ErtsRunQueue *erts_get_runq_current(ErtsSchedulerData *esdp);
ERTS_GLB_INLINE void erts_smp_runq_lock(ErtsRunQueue *rq);
ERTS_GLB_INLINE int erts_smp_runq_trylock(ErtsRunQueue *rq);
ERTS_GLB_INLINE void erts_smp_runq_unlock(ErtsRunQueue *rq);
ERTS_GLB_INLINE void erts_smp_xrunq_lock(ErtsRunQueue *rq, ErtsRunQueue *xrq);
ERTS_GLB_INLINE void erts_smp_xrunq_unlock(ErtsRunQueue *rq, ErtsRunQueue *xrq);
ERTS_GLB_INLINE void erts_smp_runqs_lock(ErtsRunQueue *rq1, ErtsRunQueue *rq2);
ERTS_GLB_INLINE void erts_smp_runqs_unlock(ErtsRunQueue *rq1, ErtsRunQueue *rq2);
#if ERTS_GLB_INLINE_INCL_FUNC_DEF
#ifdef ERTS_SMP
ERTS_GLB_INLINE ErtsRunQueue *
erts_check_emigration_need(ErtsRunQueue *c_rq, int prio)
{
ErtsRunQueueInfo *c_rqi;
if (!ERTS_CHK_RUNQ_FLG_EMIGRATE(c_rq->flags, prio))
return NULL;
if (prio == ERTS_PORT_PRIO_LEVEL)
c_rqi = &c_rq->ports.info;
else
c_rqi = &c_rq->procs.prio_info[prio];
if (!ERTS_CHK_RUNQ_FLG_EVACUATE(c_rq->flags, prio)
&& !(c_rq->flags & ERTS_RUNQ_FLG_INACTIVE)
&& c_rqi->len <= c_rqi->migrate.limit.this)
return NULL;
return erts_prepare_emigrate(c_rq, c_rqi, prio);
}
#endif
ERTS_GLB_INLINE
int erts_is_scheduler_bound(ErtsSchedulerData *esdp)
{
if (!esdp)
esdp = erts_get_scheduler_data();
ASSERT(esdp);
return esdp->cpu_id >= 0;
}
ERTS_GLB_INLINE
Process *erts_get_current_process(void)
{
ErtsSchedulerData *esdp = erts_get_scheduler_data();
return esdp ? esdp->current_process : NULL;
}
ERTS_GLB_INLINE
Eterm erts_get_current_pid(void)
{
Process *proc = erts_get_current_process();
return proc ? proc->id : THE_NON_VALUE;
}
ERTS_GLB_INLINE
Uint erts_get_scheduler_id(void)
{
#ifdef ERTS_SMP
ErtsSchedulerData *esdp = erts_get_scheduler_data();
return esdp ? esdp->no : (Uint) 0;
#else
return erts_get_scheduler_data() ? (Uint) 1 : (Uint) 0;
#endif
}
ERTS_GLB_INLINE ErtsRunQueue *
erts_get_runq_proc(Process *p)
{
ERTS_SMP_LC_ASSERT(ERTS_PROC_LOCK_STATUS & erts_proc_lc_my_proc_locks(p));
#ifdef ERTS_SMP
ASSERT(p->run_queue);
return p->run_queue;
#else
ASSERT(erts_common_run_queue);
return erts_common_run_queue;
#endif
}
ERTS_GLB_INLINE ErtsRunQueue *
erts_get_runq_current(ErtsSchedulerData *esdp)
{
ASSERT(!esdp || esdp == erts_get_scheduler_data());
#ifdef ERTS_SMP
if (!esdp)
esdp = erts_get_scheduler_data();
return esdp->run_queue;
#else
ASSERT(erts_common_run_queue);
return erts_common_run_queue;
#endif
}
ERTS_GLB_INLINE void
erts_smp_runq_lock(ErtsRunQueue *rq)
{
#ifdef ERTS_SMP
erts_smp_mtx_lock(&rq->mtx);
#endif
}
ERTS_GLB_INLINE int
erts_smp_runq_trylock(ErtsRunQueue *rq)
{
#ifdef ERTS_SMP
return erts_smp_mtx_trylock(&rq->mtx);
#else
return 0;
#endif
}
ERTS_GLB_INLINE void
erts_smp_runq_unlock(ErtsRunQueue *rq)
{
#ifdef ERTS_SMP
erts_smp_mtx_unlock(&rq->mtx);
#endif
}
ERTS_GLB_INLINE void
erts_smp_xrunq_lock(ErtsRunQueue *rq, ErtsRunQueue *xrq)
{
#ifdef ERTS_SMP
ERTS_SMP_LC_ASSERT(erts_smp_lc_mtx_is_locked(&rq->mtx));
if (xrq != rq) {
if (erts_smp_mtx_trylock(&xrq->mtx) == EBUSY) {
if (rq < xrq)
erts_smp_mtx_lock(&xrq->mtx);
else {
erts_smp_mtx_unlock(&rq->mtx);
erts_smp_mtx_lock(&xrq->mtx);
erts_smp_mtx_lock(&rq->mtx);
}
}
}
#endif
}
ERTS_GLB_INLINE void
erts_smp_xrunq_unlock(ErtsRunQueue *rq, ErtsRunQueue *xrq)
{
#ifdef ERTS_SMP
if (xrq != rq)
erts_smp_mtx_unlock(&xrq->mtx);
#endif
}
ERTS_GLB_INLINE void
erts_smp_runqs_lock(ErtsRunQueue *rq1, ErtsRunQueue *rq2)
{
#ifdef ERTS_SMP
ASSERT(rq1 && rq2);
if (rq1 == rq2)
erts_smp_mtx_lock(&rq1->mtx);
else if (rq1 < rq2) {
erts_smp_mtx_lock(&rq1->mtx);
erts_smp_mtx_lock(&rq2->mtx);
}
else {
erts_smp_mtx_lock(&rq2->mtx);
erts_smp_mtx_lock(&rq1->mtx);
}
#endif
}
ERTS_GLB_INLINE void
erts_smp_runqs_unlock(ErtsRunQueue *rq1, ErtsRunQueue *rq2)
{
#ifdef ERTS_SMP
ASSERT(rq1 && rq2);
erts_smp_mtx_unlock(&rq1->mtx);
if (rq1 != rq2)
erts_smp_mtx_unlock(&rq2->mtx);
#endif
}
#endif /* #if ERTS_GLB_INLINE_INCL_FUNC_DEF */
ERTS_GLB_INLINE ErtsAtomCacheMap *erts_get_atom_cache_map(Process *c_p);
#if ERTS_GLB_INLINE_INCL_FUNC_DEF
ERTS_GLB_INLINE ErtsAtomCacheMap *
erts_get_atom_cache_map(Process *c_p)
{
ErtsSchedulerData *esdp = (c_p
? ERTS_PROC_GET_SCHDATA(c_p)
: erts_get_scheduler_data());
ASSERT(esdp);
return &esdp->atom_cache_map;
}
#endif
#ifdef ERTS_SMP
Process *erts_pid2proc_not_running(Process *,
ErtsProcLocks,
Eterm,
ErtsProcLocks);
Process *erts_pid2proc_nropt(Process *c_p,
ErtsProcLocks c_p_locks,
Eterm pid,
ErtsProcLocks pid_locks);
extern int erts_disable_proc_not_running_opt;
#ifdef DEBUG
#define ERTS_SMP_ASSERT_IS_NOT_EXITING(P) \
do { ASSERT(!(P)->is_exiting); } while (0)
#else
#define ERTS_SMP_ASSERT_IS_NOT_EXITING(P)
#endif
/* NOTE: At least one process lock has to be held on P! */
#ifdef ERTS_ENABLE_LOCK_CHECK
#define ERTS_PROC_IS_EXITING(P) \
(ERTS_SMP_LC_ASSERT(erts_proc_lc_my_proc_locks((P)) != 0 \
|| erts_lc_pix_lock_is_locked(ERTS_PID2PIXLOCK((P)->id))),\
(P)->is_exiting)
#else
#define ERTS_PROC_IS_EXITING(P) ((P)->is_exiting)
#endif
#else /* !ERTS_SMP */
#define ERTS_PROC_IS_EXITING(P) ((P)->status == P_EXITING)
#define ERTS_SMP_ASSERT_IS_NOT_EXITING(P)
#define erts_pid2proc_not_running erts_pid2proc
#define erts_pid2proc_nropt erts_pid2proc
#endif
/* Minimum NUMBER of processes for a small system to start */
#ifdef ERTS_SMP
#define ERTS_MIN_PROCESSES ERTS_NO_OF_PIX_LOCKS
#else
#define ERTS_MIN_PROCESSES 16
#endif
void erts_smp_notify_inc_runq(ErtsRunQueue *runq);
#ifdef ERTS_SMP
void erts_sched_finish_poke(ErtsSchedulerSleepInfo *, long);
ERTS_GLB_INLINE void erts_sched_poke(ErtsSchedulerSleepInfo *ssi);
#if ERTS_GLB_INLINE_INCL_FUNC_DEF
ERTS_GLB_INLINE void
erts_sched_poke(ErtsSchedulerSleepInfo *ssi)
{
long flags = erts_smp_atomic_read(&ssi->flags);
ASSERT(!(flags & ERTS_SSI_FLG_SLEEPING)
|| (flags & ERTS_SSI_FLG_WAITING));
if (flags & ERTS_SSI_FLG_SLEEPING) {
flags = erts_smp_atomic_band(&ssi->flags, ~ERTS_SSI_FLGS_SLEEP);
erts_sched_finish_poke(ssi, flags);
}
}
#endif /* #if ERTS_GLB_INLINE_INCL_FUNC_DEF */
#endif /* #ifdef ERTS_SMP */
#include "erl_process_lock.h"
#undef ERTS_INCLUDE_SCHEDULER_INTERNALS
#endif