diff options
Diffstat (limited to 'erts/emulator')
29 files changed, 3207 insertions, 1196 deletions
diff --git a/erts/emulator/Makefile.in b/erts/emulator/Makefile.in index eca6121a1e..2b9f70b0f4 100644 --- a/erts/emulator/Makefile.in +++ b/erts/emulator/Makefile.in @@ -199,6 +199,14 @@ MKDIR = @MKDIR@ USING_MINGW=@MIXED_CYGWIN_MINGW@ +ifeq ($(TARGET),win32) +LIB_PREFIX= +LIB_SUFFIX=.lib +else +LIB_PREFIX=lib +LIB_SUFFIX=.a +endif + OMIT_OMIT_FP=no ifeq (@EMU_LOCK_CHECKING@,yes) @@ -279,15 +287,12 @@ endif ifeq ($(TARGET),win32) LIBS += -L$(ERL_TOP)/erts/emulator/pcre/obj/$(TARGET)/$(TYPE) -lepcre -DEPLIBS += $(ERL_TOP)/erts/emulator/pcre/obj/$(TARGET)/$(TYPE)/epcre.lib else -LIBS += $(ERL_TOP)/erts/emulator/pcre/obj/$(TARGET)/$(TYPE)/libepcre.a -DEPLIBS += \ - $(ERL_TOP)/erts/emulator/pcre/obj/$(TARGET)/$(TYPE)/libepcre.a \ - $(ERL_TOP)/erts/lib/internal/$(TARGET)/liberts_internal.a -# rem liberts_internal.a +LIBS += $(ERL_TOP)/erts/emulator/pcre/obj/$(TARGET)/$(TYPE)/$(LIB_PREFIX)epcre$(LIB_SUFFIX) endif +DEPLIBS += $(ERL_TOP)/erts/emulator/pcre/obj/$(TARGET)/$(TYPE)/$(LIB_PREFIX)epcre$(LIB_SUFFIX) + ELIB_FLAGS = -DENABLE_ELIB_MALLOC -DELIB_ALLOC_IS_CLIB -DELIB_HEAP_SBRK PERFCTR_PATH=@PERFCTR_PATH@ @@ -305,7 +310,19 @@ LIBSCTP = @LIBSCTP@ ORG_THR_LIBS=@EMU_THR_LIBS@ THR_LIB_NAME=@EMU_THR_LIB_NAME@ -THR_LIBS=$(subst -l$(THR_LIB_NAME),-l$(THR_LIB_NAME)$(TYPEMARKER),$(ORG_THR_LIBS)) +ifneq ($(strip $(THR_LIB_NAME)),) +DEPLIBS += $(ERL_TOP)/erts/lib/internal/$(TARGET)/$(LIB_PREFIX)erts_internal_r$(TYPEMARKER)$(LIB_SUFFIX) \ + $(ERL_TOP)/erts/lib/internal/$(TARGET)/$(LIB_PREFIX)ethread$(TYPEMARKER)$(LIB_SUFFIX) +else +DEPLIBS += $(ERL_TOP)/erts/lib/internal/$(TARGET)/$(LIB_PREFIX)erts_internal$(TYPEMARKER)$(LIB_SUFFIX) +endif + +THR_LIBS=$(subst -l$(THR_LIB_NAME),-l$(THR_LIB_NAME)$(TYPEMARKER), \ + $(subst -lerts_internal_r,-lerts_internal_r$(TYPEMARKER),$(ORG_THR_LIBS))) + +LIBS += $(THR_LIBS) + +ifneq ($(findstring erts_internal_r, $(THR_LIBS)),erts_internal_r) ifeq ($(findstring vxworks,$(TARGET)),vxworks) ERTS_INTERNAL_LIB=erts_internal @@ -317,7 +334,9 @@ ERTS_INTERNAL_LIB=erts_internal endif endif -LIBS += $(THR_LIBS) -l$(ERTS_INTERNAL_LIB)$(TYPEMARKER) +LIBS += -l$(ERTS_INTERNAL_LIB)$(TYPEMARKER) + +endif # erts_internal_r LIBS += @LIBRT@ diff --git a/erts/emulator/beam/atom.c b/erts/emulator/beam/atom.c index e2a79d6e4f..6b3c106a97 100644 --- a/erts/emulator/beam/atom.c +++ b/erts/emulator/beam/atom.c @@ -41,8 +41,7 @@ static erts_smp_rwmtx_t atom_table_lock; #define atom_read_unlock() erts_smp_rwmtx_runlock(&atom_table_lock) #define atom_write_lock() erts_smp_rwmtx_rwlock(&atom_table_lock) #define atom_write_unlock() erts_smp_rwmtx_rwunlock(&atom_table_lock) -#define atom_init_lock() erts_smp_rwmtx_init(&atom_table_lock, \ - "atom_tab") + #if 0 #define ERTS_ATOM_PUT_OPS_STAT #endif @@ -304,12 +303,17 @@ init_atom_table(void) HashFunctions f; int i; Atom a; + erts_smp_rwmtx_opt_t rwmtx_opt = ERTS_SMP_THR_OPTS_DEFAULT_INITER; + + rwmtx_opt.type = ERTS_SMP_RWMTX_TYPE_FREQUENT_READ; + rwmtx_opt.lived = ERTS_SMP_RWMTX_LONG_LIVED; #ifdef ERTS_ATOM_PUT_OPS_STAT erts_smp_atomic_init(&atom_put_ops, 0); #endif - atom_init_lock(); + erts_smp_rwmtx_init_opt(&atom_table_lock, &rwmtx_opt, "atom_tab"); + f.hash = (H_FUN) atom_hash; f.cmp = (HCMP_FUN) atom_cmp; f.alloc = (HALLOC_FUN) atom_alloc; diff --git a/erts/emulator/beam/atom.names b/erts/emulator/beam/atom.names index 28f69b9460..0815cdbc7f 100644 --- a/erts/emulator/beam/atom.names +++ b/erts/emulator/beam/atom.names @@ -432,6 +432,7 @@ atom raw atom re atom re_pattern atom re_run_trap +atom read_concurrency atom ready_input atom ready_output atom ready_async diff --git a/erts/emulator/beam/erl_alloc.c b/erts/emulator/beam/erl_alloc.c index 16ae643ed9..87503af7d5 100644 --- a/erts/emulator/beam/erl_alloc.c +++ b/erts/emulator/beam/erl_alloc.c @@ -64,6 +64,14 @@ #ifdef DEBUG static Uint install_debug_functions(void); +#if 0 +#define HARD_DEBUG +#ifdef __GNUC__ +#warning "* * * * * * * * * * * * * *" +#warning "* HARD DEBUG IS ENABLED! *" +#warning "* * * * * * * * * * * * * *" +#endif +#endif #endif extern void elib_ensure_initialized(void); @@ -391,6 +399,10 @@ refuse_af_strategy(struct au_init *init) static void init_thr_ix(int static_ixs); +#ifdef HARD_DEBUG +static void hdbg_init(void); +#endif + void erts_alloc_init(int *argc, char **argv, ErtsAllocInitOpts *eaiop) { @@ -406,6 +418,10 @@ erts_alloc_init(int *argc, char **argv, ErtsAllocInitOpts *eaiop) ERTS_DEFAULT_ALCU_INIT }; +#ifdef HARD_DEBUG + hdbg_init(); +#endif + erts_sys_alloc_init(); init_thr_ix(erts_no_schedulers); erts_init_utils_mem(); @@ -2862,12 +2878,10 @@ unsigned long erts_alc_test(unsigned long op, break; } case 0xf0a: - if (ethr_mutex_lock((ethr_mutex *) a1) != 0) - ERTS_ALC_TEST_ABORT; + ethr_mutex_lock((ethr_mutex *) a1); break; case 0xf0b: - if (ethr_mutex_unlock((ethr_mutex *) a1) != 0) - ERTS_ALC_TEST_ABORT; + ethr_mutex_unlock((ethr_mutex *) a1); break; case 0xf0c: { ethr_cond *cnd = erts_alloc(ERTS_ALC_T_UNDEF, sizeof(ethr_cond)); @@ -2883,16 +2897,13 @@ unsigned long erts_alc_test(unsigned long op, break; } case 0xf0e: - if (ethr_cond_broadcast((ethr_cond *) a1) != 0) - ERTS_ALC_TEST_ABORT; + ethr_cond_broadcast((ethr_cond *) a1); break; case 0xf0f: { int res; do { res = ethr_cond_wait((ethr_cond *) a1, (ethr_mutex *) a2); } while (res == EINTR); - if (res != 0) - ERTS_ALC_TEST_ABORT; break; } case 0xf10: { @@ -2939,8 +2950,11 @@ unsigned long erts_alc_test(unsigned long op, #undef PRINT_OPS #endif - +#ifdef HARD_DEBUG +#define FENCE_SZ (4*sizeof(UWord)) +#else #define FENCE_SZ (3*sizeof(UWord)) +#endif #if defined(ARCH_64) #define FENCE_PATTERN 0xABCDEF97ABCDEF97 @@ -2962,6 +2976,104 @@ unsigned long erts_alc_test(unsigned long op, #define GET_TYPE_OF_PATTERN(P) \ (((P) >> TYPE_PATTERN_SHIFT) & TYPE_PATTERN_MASK) +#ifdef HARD_DEBUG + +#define ERL_ALC_HDBG_MAX_MBLK 100000 +#define ERTS_ALC_O_CHECK -1 + +typedef struct hdbg_mblk_ hdbg_mblk; +struct hdbg_mblk_ { + hdbg_mblk *next; + hdbg_mblk *prev; + void *p; + Uint s; + ErtsAlcType_t n; +}; + +static hdbg_mblk hdbg_mblks[ERL_ALC_HDBG_MAX_MBLK]; + +static hdbg_mblk *free_hdbg_mblks; +static hdbg_mblk *used_hdbg_mblks; +static erts_mtx_t hdbg_mblk_mtx; + +static void +hdbg_init(void) +{ + int i; + for (i = 0; i < ERL_ALC_HDBG_MAX_MBLK-1; i++) + hdbg_mblks[i].next = &hdbg_mblks[i+1]; + hdbg_mblks[ERL_ALC_HDBG_MAX_MBLK-1].next = NULL; + free_hdbg_mblks = &hdbg_mblks[0]; + used_hdbg_mblks = NULL; + erts_mtx_init(&hdbg_mblk_mtx, "erts_alloc_hard_debug"); +} + +static void *check_memory_fence(void *ptr, + Uint *size, + ErtsAlcType_t n, + int func); +void erts_hdbg_chk_blks(void); + +void +erts_hdbg_chk_blks(void) +{ + hdbg_mblk *mblk; + + erts_mtx_lock(&hdbg_mblk_mtx); + for (mblk = used_hdbg_mblks; mblk; mblk = mblk->next) { + Uint sz; + check_memory_fence(mblk->p, &sz, mblk->n, ERTS_ALC_O_CHECK); + ASSERT(sz == mblk->s); + } + erts_mtx_unlock(&hdbg_mblk_mtx); +} + +static hdbg_mblk * +hdbg_alloc(void *p, Uint s, ErtsAlcType_t n) +{ + hdbg_mblk *mblk; + + erts_mtx_lock(&hdbg_mblk_mtx); + mblk = free_hdbg_mblks; + if (!mblk) { + erts_fprintf(stderr, + "Ran out of debug blocks; please increase " + "ERL_ALC_HDBG_MAX_MBLK=%d and recompile!\n", + ERL_ALC_HDBG_MAX_MBLK); + abort(); + } + free_hdbg_mblks = mblk->next; + + mblk->p = p; + mblk->s = s; + mblk->n = n; + + mblk->next = used_hdbg_mblks; + mblk->prev = NULL; + if (used_hdbg_mblks) + used_hdbg_mblks->prev = mblk; + used_hdbg_mblks = mblk; + erts_mtx_unlock(&hdbg_mblk_mtx); + return (void *) mblk; +} + +static void +hdbg_free(hdbg_mblk *mblk) +{ + erts_mtx_lock(&hdbg_mblk_mtx); + if (mblk->next) + mblk->next->prev = mblk->prev; + if (mblk->prev) + mblk->prev->next = mblk->next; + else + used_hdbg_mblks = mblk->next; + + mblk->next = free_hdbg_mblks; + free_hdbg_mblks = mblk; + erts_mtx_unlock(&hdbg_mblk_mtx); +} + +#endif #ifdef ERTS_ALLOC_UTIL_HARD_DEBUG static void *check_memory_fence(void *ptr, Uint *size, ErtsAlcType_t n, int func); @@ -3007,17 +3119,28 @@ set_memory_fence(void *ptr, Uint sz, ErtsAlcType_t n) { UWord *ui_ptr; UWord pattern; +#ifdef HARD_DEBUG + hdbg_mblk **mblkpp; +#endif if (!ptr) return NULL; ui_ptr = (UWord *) ptr; pattern = MK_PATTERN(n); - + +#ifdef HARD_DEBUG + mblkpp = (hdbg_mblk **) ui_ptr++; +#endif + *(ui_ptr++) = sz; *(ui_ptr++) = pattern; memcpy((void *) (((char *) ui_ptr)+sz), (void *) &pattern, sizeof(UWord)); +#ifdef HARD_DEBUG + *mblkpp = hdbg_alloc((void *) ui_ptr, sz, n); +#endif + return (void *) ui_ptr; } @@ -3029,6 +3152,9 @@ check_memory_fence(void *ptr, Uint *size, ErtsAlcType_t n, int func) UWord pre_pattern; UWord post_pattern; UWord *ui_ptr; +#ifdef HARD_DEBUG + hdbg_mblk *mblk; +#endif if (!ptr) return NULL; @@ -3036,6 +3162,9 @@ check_memory_fence(void *ptr, Uint *size, ErtsAlcType_t n, int func) ui_ptr = (UWord *) ptr; pre_pattern = *(--ui_ptr); *size = sz = *(--ui_ptr); +#ifdef HARD_DEBUG + mblk = (hdbg_mblk *) *(--ui_ptr); +#endif found_type = GET_TYPE_OF_PATTERN(pre_pattern); if (pre_pattern != MK_PATTERN(n)) { @@ -3091,6 +3220,17 @@ check_memory_fence(void *ptr, Uint *size, ErtsAlcType_t n, int func) (unsigned long) ptr, (unsigned long) sz, ftype, op_str, otype); } +#ifdef HARD_DEBUG + switch (func) { + case ERTS_ALC_O_REALLOC: + case ERTS_ALC_O_FREE: + hdbg_free(mblk); + break; + default: + break; + } +#endif + return (void *) ui_ptr; } @@ -3103,6 +3243,10 @@ debug_alloc(ErtsAlcType_t n, void *extra, Uint size) Uint dsize; void *res; +#ifdef HARD_DEBUG + erts_hdbg_chk_blks(); +#endif + ASSERT(ERTS_ALC_N_MIN <= n && n <= ERTS_ALC_N_MAX); dsize = size + FENCE_SZ; res = (*real_af->alloc)(n, real_af->extra, dsize); @@ -3132,13 +3276,17 @@ debug_realloc(ErtsAlcType_t n, void *extra, void *ptr, Uint size) dsize = size + FENCE_SZ; dptr = check_memory_fence(ptr, &old_size, n, ERTS_ALC_O_REALLOC); +#ifdef HARD_DEBUG + erts_hdbg_chk_blks(); +#endif + if (old_size > size) sys_memset((void *) (((char *) ptr) + size), 0xf, sizeof(Uint) + old_size - size); res = (*real_af->realloc)(n, real_af->extra, dptr, dsize); - + res = set_memory_fence(res, size, n); #ifdef PRINT_OPS @@ -3168,6 +3316,10 @@ debug_free(ErtsAlcType_t n, void *extra, void *ptr) fprintf(stderr, "free(%s, 0x%lx)\r\n", ERTS_ALC_N2TD(n), (Uint) ptr); #endif +#ifdef HARD_DEBUG + erts_hdbg_chk_blks(); +#endif + } static Uint diff --git a/erts/emulator/beam/erl_alloc.types b/erts/emulator/beam/erl_alloc.types index 6f88bbe5b8..7df9f19af0 100644 --- a/erts/emulator/beam/erl_alloc.types +++ b/erts/emulator/beam/erl_alloc.types @@ -212,7 +212,8 @@ type LOGGER_DSBUF TEMPORARY SYSTEM logger_dsbuf type TMP_DSBUF TEMPORARY SYSTEM tmp_dsbuf type INFO_DSBUF SYSTEM SYSTEM info_dsbuf # INFO_DSBUF have to use the SYSTEM allocator; otherwise, a deadlock might occur -type SCHDLR_DATA LONG_LIVED PROCESSES scheduler_data +type SCHDLR_DATA LONG_LIVED SYSTEM scheduler_data +type SCHDLR_SLP_INFO LONG_LIVED SYSTEM scheduler_sleep_info type RUNQS LONG_LIVED SYSTEM run_queues type DDLL_PROCESS STANDARD SYSTEM ddll_processes type DDLL_HANDLE STANDARD SYSTEM ddll_handle @@ -246,6 +247,7 @@ type CPUDATA LONG_LIVED SYSTEM cpu_data type TMP_CPU_IDS SHORT_LIVED SYSTEM tmp_cpu_ids type EXT_TERM_DATA SHORT_LIVED PROCESSES external_term_data type ZLIB STANDARD SYSTEM zlib +type RDR_GRPS_MAP LONG_LIVED SYSTEM reader_groups_map +if smp type ASYNC SHORT_LIVED SYSTEM async @@ -269,7 +271,9 @@ type RUNQ_BLNS LONG_LIVED SYSTEM run_queue_balancing +if threads -type ETHR_INTERNAL SYSTEM SYSTEM ethread_internal +type ETHR_STD STANDARD SYSTEM ethread_standard +type ETHR_SL SHORT_LIVED SYSTEM ethread_short_lived +type ETHR_LL LONG_LIVED SYSTEM ethread_long_lived +ifnot smp diff --git a/erts/emulator/beam/erl_bif_info.c b/erts/emulator/beam/erl_bif_info.c index ca1ab44012..5128776c47 100644 --- a/erts/emulator/beam/erl_bif_info.c +++ b/erts/emulator/beam/erl_bif_info.c @@ -2565,6 +2565,8 @@ BIF_RETTYPE system_info_1(BIF_ALIST_1) BIF_RET(erts_sched_stat_term(BIF_P, 1)); } else if (ERTS_IS_ATOM_STR("taints", BIF_ARG_1)) { BIF_RET(erts_nif_taints(BIF_P)); + } else if (ERTS_IS_ATOM_STR("reader_groups_map", BIF_ARG_1)) { + BIF_RET(erts_get_reader_groups_map(BIF_P)); } BIF_ERROR(BIF_P, BADARG); @@ -3428,6 +3430,16 @@ BIF_RETTYPE erts_debug_get_internal_state_1(BIF_ALIST_1) else if (ERTS_IS_ATOM_STR("fake_scheduler_bindings", tp[1])) { return erts_fake_scheduler_bindings(BIF_P, tp[2]); } + else if (ERTS_IS_ATOM_STR("reader_groups_map", tp[1])) { + Sint groups; + if (is_not_small(tp[2])) + BIF_ERROR(BIF_P, BADARG); + groups = signed_val(tp[2]); + if (groups < (Sint) 1 || groups > (Sint) INT_MAX) + BIF_ERROR(BIF_P, BADARG); + + BIF_RET(erts_debug_reader_groups_map(BIF_P, (int) groups)); + } break; } default: @@ -3732,8 +3744,8 @@ static Eterm lcnt_build_lock_stats_term(Eterm **hpp, Uint *szp, erts_lcnt_lock_s * [{{file, line}, {tries, colls, {seconds, nanoseconds, n_blocks}}}] */ - ethr_atomic_read(&stats->tries, (long *)&tries); - ethr_atomic_read(&stats->colls, (long *)&colls); + tries = (unsigned long) ethr_atomic_read(&stats->tries); + colls = (unsigned long) ethr_atomic_read(&stats->colls); line = stats->line; timer_s = stats->timer.s; diff --git a/erts/emulator/beam/erl_db.c b/erts/emulator/beam/erl_db.c index cbdaa459de..b0369a402b 100644 --- a/erts/emulator/beam/erl_db.c +++ b/erts/emulator/beam/erl_db.c @@ -78,11 +78,19 @@ enum DbIterSafety { ** The main meta table, containing all ets tables. */ #ifdef ERTS_SMP -# define META_MAIN_TAB_LOCK_CNT 16 -static union { - erts_smp_spinlock_t lck; - byte _cache_line_alignment[64]; -}meta_main_tab_locks[META_MAIN_TAB_LOCK_CNT]; + +#define ERTS_META_MAIN_TAB_LOCK_TAB_BITS 8 +#define ERTS_META_MAIN_TAB_LOCK_TAB_SIZE (1 << ERTS_META_MAIN_TAB_LOCK_TAB_BITS) +#define ERTS_META_MAIN_TAB_LOCK_TAB_MASK (ERTS_META_MAIN_TAB_LOCK_TAB_SIZE - 1) + +typedef union { + erts_smp_rwmtx_t rwmtx; + byte cache_line_align__[ERTS_ALC_CACHE_LINE_ALIGN_SIZE( + sizeof(erts_smp_rwmtx_t))]; +} erts_meta_main_tab_lock_t; + +static erts_meta_main_tab_lock_t *meta_main_tab_locks; + #endif static struct { union { @@ -104,17 +112,13 @@ static struct { #define MARK_SLOT_DEAD(i) (meta_main_tab[(i)].u.next_free |= 2) #define GET_ANY_SLOT_TAB(i) ((DbTable*)(meta_main_tab[(i)].u.next_free & ~(1|2))) /* dead or alive */ -static ERTS_INLINE void meta_main_tab_lock(unsigned slot) +static ERTS_INLINE erts_smp_rwmtx_t * +get_meta_main_tab_lock(unsigned slot) { #ifdef ERTS_SMP - erts_smp_spin_lock(&meta_main_tab_locks[slot % META_MAIN_TAB_LOCK_CNT].lck); -#endif -} - -static ERTS_INLINE void meta_main_tab_unlock(unsigned slot) -{ -#ifdef ERTS_SMP - erts_smp_spin_unlock(&meta_main_tab_locks[slot % META_MAIN_TAB_LOCK_CNT].lck); + return &meta_main_tab_locks[slot & ERTS_META_MAIN_TAB_LOCK_TAB_MASK].rwmtx; +#else + return NULL; #endif } @@ -166,7 +170,8 @@ struct meta_name_tab_entry* meta_name_tab_bucket(Eterm name, typedef enum { LCK_READ=1, /* read only access */ LCK_WRITE=2, /* exclusive table write access */ - LCK_WRITE_REC=3 /* record write access */ + LCK_WRITE_REC=3, /* record write access */ + LCK_NONE=4 } db_lock_kind_t; extern DbTableMethod db_hash; @@ -214,17 +219,17 @@ Export ets_select_continue_exp; */ static Export ets_delete_continue_exp; -static ERTS_INLINE DbTable* db_ref(DbTable* tb) +static ERTS_INLINE DbTable* db_ref(DbTable* tb, db_lock_kind_t kind) { - if (tb != NULL) { + if (tb != NULL && kind != LCK_READ) { erts_refc_inc(&tb->common.ref, 2); } return tb; } -static ERTS_INLINE DbTable* db_unref(DbTable* tb) +static ERTS_INLINE DbTable* db_unref(DbTable* tb, db_lock_kind_t kind) { - if (!erts_refc_dectest(&tb->common.ref, 0)) { + if (kind != LCK_READ && !erts_refc_dectest(&tb->common.ref, 0)) { #ifdef HARDDEBUG if (erts_smp_atomic_read(&tb->common.memory_size) != sizeof(DbTable)) { erts_fprintf(stderr, "ets: db_unref memory remain=%ld fix=%x\n", @@ -256,12 +261,19 @@ static ERTS_INLINE DbTable* db_unref(DbTable* tb) return tb; } -static ERTS_INLINE void db_init_lock(DbTable* tb, char *rwname, char* fixname) +static ERTS_INLINE void db_init_lock(DbTable* tb, int use_frequent_read_lock, + char *rwname, char* fixname) { +#ifdef ERTS_SMP + erts_smp_rwmtx_opt_t rwmtx_opt = ERTS_SMP_THR_OPTS_DEFAULT_INITER; + if (use_frequent_read_lock) + rwmtx_opt.type = ERTS_SMP_RWMTX_TYPE_FREQUENT_READ; +#endif erts_refc_init(&tb->common.ref, 1); erts_refc_init(&tb->common.fixref, 0); #ifdef ERTS_SMP - erts_smp_rwmtx_init_x(&tb->common.rwlock, rwname, tb->common.the_name); + erts_smp_rwmtx_init_opt_x(&tb->common.rwlock, &rwmtx_opt, + rwname, tb->common.the_name); erts_smp_mtx_init_x(&tb->common.fixlock, fixname, tb->common.the_name); tb->common.is_thread_safe = !(tb->common.status & DB_FINE_LOCKED); #endif @@ -297,7 +309,7 @@ static ERTS_INLINE void db_lock_take_over_ref(DbTable* tb, db_lock_kind_t kind) static ERTS_INLINE void db_lock(DbTable* tb, db_lock_kind_t kind) { - (void) db_ref(tb); + (void) db_ref(tb, kind); #ifdef ERTS_SMP db_lock_take_over_ref(tb, kind); #endif @@ -331,7 +343,7 @@ static ERTS_INLINE void db_unlock(DbTable* tb, db_lock_kind_t kind) } } #endif - (void) db_unref(tb); /* May delete table... */ + (void) db_unref(tb, kind); /* May delete table... */ } @@ -349,32 +361,49 @@ static ERTS_INLINE void db_meta_unlock(DbTable* tb, db_lock_kind_t kind) } static ERTS_INLINE -DbTable* db_get_table(Process *p, - Eterm id, - int what, - db_lock_kind_t kind) +DbTable* db_get_table_aux(Process *p, + Eterm id, + int what, + db_lock_kind_t kind, + int meta_already_locked) { DbTable *tb = NULL; + erts_smp_rwmtx_t *mtl = NULL; if (is_small(id)) { Uint slot = unsigned_val(id) & meta_main_tab_slot_mask; - meta_main_tab_lock(slot); + if (!meta_already_locked) { + mtl = get_meta_main_tab_lock(slot); + erts_smp_rwmtx_rlock(mtl); + } +#if defined(ERTS_SMP) && defined(ERTS_ENABLE_LOCK_CHECK) + else { + erts_smp_rwmtx_t *test_mtl = get_meta_main_tab_lock(slot); + ERTS_SMP_LC_ASSERT(erts_lc_rwmtx_is_rlocked(test_mtl) + || erts_lc_rwmtx_is_rwlocked(test_mtl)); + } +#endif if (slot < db_max_tabs && IS_SLOT_ALIVE(slot)) { /* SMP: inc to prevent race, between unlock of meta_main_tab_lock * and the table locking outside the meta_main_tab_lock */ - tb = db_ref(meta_main_tab[slot].u.tb); + tb = db_ref(meta_main_tab[slot].u.tb, kind); } - meta_main_tab_unlock(slot); } else if (is_atom(id)) { - erts_smp_rwmtx_t* rwlock; - struct meta_name_tab_entry* bucket = meta_name_tab_bucket(id,&rwlock); - erts_smp_rwmtx_rlock(rwlock); + struct meta_name_tab_entry* bucket = meta_name_tab_bucket(id,&mtl); + if (!meta_already_locked) + erts_smp_rwmtx_rlock(mtl); + else{ + ERTS_SMP_LC_ASSERT(erts_lc_rwmtx_is_rlocked(mtl) + || erts_lc_rwmtx_is_rwlocked(mtl)); + mtl = NULL; + } + if (bucket->pu.tb != NULL) { if (is_atom(bucket->u.name_atom)) { /* single */ if (bucket->u.name_atom == id) { - tb = db_ref(bucket->pu.tb); + tb = db_ref(bucket->pu.tb, kind); } } else { /* multi */ @@ -382,23 +411,33 @@ DbTable* db_get_table(Process *p, Uint i; for (i=0; i<cnt; i++) { if (bucket->pu.mvec[i].u.name_atom == id) { - tb = db_ref(bucket->pu.mvec[i].pu.tb); + tb = db_ref(bucket->pu.mvec[i].pu.tb, kind); break; } } } } - erts_smp_rwmtx_runlock(rwlock); } if (tb) { db_lock_take_over_ref(tb, kind); - if (tb->common.id == id && ((tb->common.status & what) != 0 || - p->id == tb->common.owner)) { - return tb; + if (tb->common.id != id + || ((tb->common.status & what) == 0 && p->id != tb->common.owner)) { + db_unlock(tb, kind); + tb = NULL; } - db_unlock(tb, kind); } - return NULL; + if (mtl) + erts_smp_rwmtx_runlock(mtl); + return tb; +} + +static ERTS_INLINE +DbTable* db_get_table(Process *p, + Eterm id, + int what, + db_lock_kind_t kind) +{ + return db_get_table_aux(p, id, what, kind, 0); } /* Requires meta_main_tab_locks[slot] locked. @@ -413,15 +452,15 @@ static ERTS_INLINE void free_slot(int slot) erts_smp_spin_unlock(&meta_main_tab_main_lock); } -static int insert_named_tab(Eterm name_atom, DbTable* tb) +static int insert_named_tab(Eterm name_atom, DbTable* tb, int have_lock) { int ret = 0; erts_smp_rwmtx_t* rwlock; struct meta_name_tab_entry* new_entry; struct meta_name_tab_entry* bucket = meta_name_tab_bucket(name_atom, &rwlock); - - erts_smp_rwmtx_rwlock(rwlock); + if (!have_lock) + erts_smp_rwmtx_rwlock(rwlock); if (bucket->pu.tb == NULL) { /* empty */ new_entry = bucket; @@ -468,17 +507,32 @@ static int insert_named_tab(Eterm name_atom, DbTable* tb) ret = 1; /* Ok */ done: - erts_smp_rwmtx_rwunlock(rwlock); + if (!have_lock) + erts_smp_rwmtx_rwunlock(rwlock); return ret; } -static int remove_named_tab(Eterm name_atom) +static int remove_named_tab(DbTable *tb, int have_lock) { int ret = 0; erts_smp_rwmtx_t* rwlock; + Eterm name_atom = tb->common.id; struct meta_name_tab_entry* bucket = meta_name_tab_bucket(name_atom, &rwlock); - erts_smp_rwmtx_rwlock(rwlock); +#ifdef ERTS_SMP + if (!have_lock && erts_smp_rwmtx_tryrwlock(rwlock) == EBUSY) { + /* + * We keep our increased refc over this op in order to + * prevent the table from disapearing. + */ + erts_smp_rwmtx_rwunlock(&tb->common.rwlock); + erts_smp_rwmtx_rwlock(rwlock); + erts_smp_rwmtx_rwlock(&tb->common.rwlock); + } +#endif + + ERTS_SMP_LC_ASSERT(erts_lc_rwmtx_is_rwlocked(rwlock)); + if (bucket->pu.tb == NULL) { goto done; } @@ -529,7 +583,8 @@ static int remove_named_tab(Eterm name_atom) ret = 1; /* Ok */ done: - erts_smp_rwmtx_rwunlock(rwlock); + if (!have_lock) + erts_smp_rwmtx_rwunlock(rwlock); return ret; } @@ -1133,6 +1188,7 @@ BIF_RETTYPE ets_rename_2(BIF_ALIST_2) { DbTable* tb; Eterm ret; + erts_smp_rwmtx_t *lck1, *lck2; #ifdef HARDDEBUG erts_fprintf(stderr, @@ -1141,34 +1197,65 @@ BIF_RETTYPE ets_rename_2(BIF_ALIST_2) BIF_P->initial[0], BIF_P->initial[1], BIF_P->initial[2]); #endif - if ((tb = db_get_table(BIF_P, BIF_ARG_1, DB_WRITE, LCK_WRITE)) == NULL) { + + if (is_not_atom(BIF_ARG_2)) { BIF_ERROR(BIF_P, BADARG); } - if (is_not_atom(BIF_ARG_2)) { - goto badarg; + (void) meta_name_tab_bucket(BIF_ARG_2, &lck1); + + if (is_small(BIF_ARG_1)) { + Uint slot = unsigned_val(BIF_ARG_1) & meta_main_tab_slot_mask; + lck2 = get_meta_main_tab_lock(slot); + } + else if (is_atom(BIF_ARG_1)) { + (void) meta_name_tab_bucket(BIF_ARG_1, &lck2); + if (lck1 == lck2) + lck2 = NULL; + else if (lck1 > lck2) { + erts_smp_rwmtx_t *tmp = lck1; + lck1 = lck2; + lck2 = tmp; + } + } + else { + BIF_ERROR(BIF_P, BADARG); } + erts_smp_rwmtx_rwlock(lck1); + if (lck2) + erts_smp_rwmtx_rwlock(lck2); + + tb = db_get_table_aux(BIF_P, BIF_ARG_1, DB_WRITE, LCK_WRITE, 1); + if (!tb) + goto badarg; + if (is_not_atom(tb->common.id)) { /* Not a named table */ tb->common.the_name = BIF_ARG_2; goto done; } - if (!insert_named_tab(BIF_ARG_2,tb)) { + if (!insert_named_tab(BIF_ARG_2, tb, 1)) goto badarg; - } - if (!remove_named_tab(tb->common.id)) { + + if (!remove_named_tab(tb, 1)) erl_exit(1,"Could not find named tab %s", tb->common.id); - } tb->common.id = tb->common.the_name = BIF_ARG_2; done: ret = tb->common.id; db_unlock(tb, LCK_WRITE); + erts_smp_rwmtx_rwunlock(lck1); + if (lck2) + erts_smp_rwmtx_rwunlock(lck2); BIF_RET(ret); badarg: - db_unlock(tb, LCK_WRITE); + if (tb) + db_unlock(tb, LCK_WRITE); + erts_smp_rwmtx_rwunlock(lck1); + if (lck2) + erts_smp_rwmtx_rwunlock(lck2); BIF_ERROR(BIF_P, BADARG); } @@ -1189,10 +1276,11 @@ BIF_RETTYPE ets_new_2(BIF_ALIST_2) UWord heir_data; Uint32 status; Sint keypos; - int is_named, is_fine_locked; + int is_named, is_fine_locked, frequent_read; int cret; DeclareTmpHeap(meta_tuple,3,BIF_P); DbTableMethod* meth; + erts_smp_rwmtx_t *mmtl; if (is_not_atom(BIF_ARG_1)) { BIF_ERROR(BIF_P, BADARG); @@ -1205,6 +1293,7 @@ BIF_RETTYPE ets_new_2(BIF_ALIST_2) keypos = 1; is_named = 0; is_fine_locked = 0; + frequent_read = 0; heir = am_none; heir_data = (UWord) am_undefined; @@ -1238,6 +1327,13 @@ BIF_RETTYPE ets_new_2(BIF_ALIST_2) is_fine_locked = 0; } else break; } + else if (tp[1] == am_read_concurrency) { + if (tp[2] == am_true) { + frequent_read = 1; + } else if (tp[2] == am_false) { + frequent_read = 0; + } else break; + } else if (tp[1] == am_heir && tp[2] == am_none) { heir = am_none; heir_data = am_undefined; @@ -1286,6 +1382,11 @@ BIF_RETTYPE ets_new_2(BIF_ALIST_2) BIF_ERROR(BIF_P, BADARG); } +#ifdef ERTS_SMP + if (frequent_read && !(status & DB_PRIVATE)) + status |= DB_FREQ_READ; +#endif + /* we create table outside any table lock * and take the unusal cost of destroy table if it * fails to find a slot @@ -1308,7 +1409,8 @@ BIF_RETTYPE ets_new_2(BIF_ALIST_2) tb->common.type = status & ERTS_ETS_TABLE_TYPES; /* Note, 'type' is *read only* from now on... */ #endif - db_init_lock(tb, "db_tab", "db_tab_fix"); + db_init_lock(tb, status & (DB_FINE_LOCKED|DB_FREQ_READ), + "db_tab", "db_tab_fix"); tb->common.keypos = keypos; tb->common.owner = BIF_P->id; set_heir(BIF_P, tb, heir, heir_data); @@ -1351,15 +1453,17 @@ BIF_RETTYPE ets_new_2(BIF_ALIST_2) tb->common.id = ret; tb->common.slot = slot; /* store slot for erase */ - meta_main_tab_lock(slot); + mmtl = get_meta_main_tab_lock(slot); + erts_smp_rwmtx_rwlock(mmtl); meta_main_tab[slot].u.tb = tb; ASSERT(IS_SLOT_ALIVE(slot)); - meta_main_tab_unlock(slot); + erts_smp_rwmtx_rwunlock(mmtl); - if (is_named && !insert_named_tab(BIF_ARG_1, tb)) { - meta_main_tab_lock(slot); + if (is_named && !insert_named_tab(BIF_ARG_1, tb, 0)) { + mmtl = get_meta_main_tab_lock(slot); + erts_smp_rwmtx_rwlock(mmtl); free_slot(slot); - meta_main_tab_unlock(slot); + erts_smp_rwmtx_rwunlock(mmtl); db_lock_take_over_ref(tb,LCK_WRITE); free_heir_data(tb); @@ -1499,6 +1603,7 @@ BIF_RETTYPE ets_delete_1(BIF_ALIST_1) { int trap; DbTable* tb; + erts_smp_rwmtx_t *mmtl; #ifdef HARDDEBUG erts_fprintf(stderr, @@ -1520,13 +1625,23 @@ BIF_RETTYPE ets_delete_1(BIF_ALIST_1) tb->common.status &= ~(DB_PROTECTED|DB_PUBLIC|DB_PRIVATE); tb->common.status |= DB_DELETE; - meta_main_tab_lock(tb->common.slot); + mmtl = get_meta_main_tab_lock(tb->common.slot); +#ifdef ERTS_SMP + if (erts_smp_rwmtx_tryrwlock(mmtl) == EBUSY) { + /* + * We keep our increased refc over this op in order to + * prevent the table from disapearing. + */ + erts_smp_rwmtx_rwunlock(&tb->common.rwlock); + erts_smp_rwmtx_rwlock(mmtl); + erts_smp_rwmtx_rwlock(&tb->common.rwlock); + } +#endif /* We must keep the slot, to be found by db_proc_dead() if process dies */ MARK_SLOT_DEAD(tb->common.slot); - meta_main_tab_unlock(tb->common.slot); - if (is_atom(tb->common.id)) { - remove_named_tab(tb->common.id); - } + erts_smp_rwmtx_rwunlock(mmtl); + if (is_atom(tb->common.id)) + remove_named_tab(tb, 0); if (tb->common.owner != BIF_P->id) { DeclareTmpHeap(meta_tuple,3,BIF_P); @@ -1919,14 +2034,15 @@ BIF_RETTYPE ets_all_0(BIF_ALIST_0) previous = NIL; j = 0; for(i = 0; (i < t_max_tabs && j < t_tabs_cnt); i++) { - meta_main_tab_lock(i); + erts_smp_rwmtx_t *mmtl = get_meta_main_tab_lock(i); + erts_smp_rwmtx_rlock(mmtl); if (IS_SLOT_ALIVE(i)) { j++; tb = meta_main_tab[i].u.tb; previous = CONS(hp, tb->common.id, previous); hp += 2; } - meta_main_tab_unlock(i); + erts_smp_rwmtx_runlock(mmtl); } HRelease(BIF_P, hendp, hp); BIF_RET(previous); @@ -2630,12 +2746,30 @@ void init_db(void) size_t size; #ifdef ERTS_SMP - for (i=0; i<META_MAIN_TAB_LOCK_CNT; i++) { - erts_smp_spinlock_init_x(&meta_main_tab_locks[i].lck, "meta_main_tab_slot", make_small(i)); + erts_smp_rwmtx_opt_t rwmtx_opt = ERTS_SMP_THR_OPTS_DEFAULT_INITER; + rwmtx_opt.type = ERTS_SMP_RWMTX_TYPE_FREQUENT_READ; + rwmtx_opt.lived = ERTS_SMP_RWMTX_LONG_LIVED; + + meta_main_tab_locks = erts_alloc(ERTS_ALC_T_DB_TABLES, + (sizeof(erts_meta_main_tab_lock_t) + * (ERTS_META_MAIN_TAB_LOCK_TAB_SIZE+1))); + + if ((((Uint) meta_main_tab_locks) & ERTS_CACHE_LINE_MASK) != 0) + meta_main_tab_locks = ((erts_meta_main_tab_lock_t *) + ((((Uint) meta_main_tab_locks) + & ~ERTS_CACHE_LINE_MASK) + + ERTS_CACHE_LINE_SIZE)); + + ASSERT((((Uint) meta_main_tab_locks) & ERTS_CACHE_LINE_MASK) == 0); + + for (i = 0; i < ERTS_META_MAIN_TAB_LOCK_TAB_SIZE; i++) { + erts_smp_rwmtx_init_opt_x(&meta_main_tab_locks[i].rwmtx, &rwmtx_opt, + "meta_main_tab_slot", make_small(i)); } erts_smp_spinlock_init(&meta_main_tab_main_lock, "meta_main_tab_main"); for (i=0; i<META_NAME_TAB_LOCK_CNT; i++) { - erts_smp_rwmtx_init_x(&meta_name_tab_rwlocks[i].lck, "meta_name_tab", make_small(i)); + erts_smp_rwmtx_init_opt_x(&meta_name_tab_rwlocks[i].lck, &rwmtx_opt, + "meta_name_tab", make_small(i)); } #endif @@ -2895,12 +3029,12 @@ retry: to_pid, to_locks, ERTS_P2P_FLG_TRY_LOCK); if (to_proc == ERTS_PROC_LOCK_BUSY) { - db_ref(tb); /* while unlocked */ + db_ref(tb, LCK_NONE); /* while unlocked */ db_unlock(tb,LCK_WRITE); to_proc = erts_pid2proc(p, ERTS_PROC_LOCK_MAIN, to_pid, to_locks); db_lock(tb,LCK_WRITE); - tb = db_unref(tb); + tb = db_unref(tb, LCK_NONE); ASSERT(tb != NULL); if (tb->common.owner != p->id) { @@ -3008,12 +3142,13 @@ erts_db_process_exiting(Process *c_p, ErtsProcLocks c_p_locks) while (state->slots.ix < state->slots.size) { DbTable *tb = NULL; Sint ix = unsigned_val(state->slots.arr[state->slots.ix]); - meta_main_tab_lock(ix); + erts_smp_rwmtx_t *mmtl = get_meta_main_tab_lock(ix); + erts_smp_rwmtx_rlock(mmtl); if (!IS_SLOT_FREE(ix)) { - tb = db_ref(GET_ANY_SLOT_TAB(ix)); + tb = db_ref(GET_ANY_SLOT_TAB(ix), LCK_WRITE); ASSERT(tb); } - meta_main_tab_unlock(ix); + erts_smp_rwmtx_runlock(mmtl); if (tb) { int do_yield; db_lock_take_over_ref(tb, LCK_WRITE); @@ -3045,7 +3180,7 @@ erts_db_process_exiting(Process *c_p, ErtsProcLocks c_p_locks) tb->common.status |= DB_DELETE; if (is_atom(tb->common.id)) - remove_named_tab(tb->common.id); + remove_named_tab(tb, 0); free_heir_data(tb); free_fixations_locked(tb); @@ -3095,12 +3230,13 @@ erts_db_process_exiting(Process *c_p, ErtsProcLocks c_p_locks) while (state->slots.ix < state->slots.size) { DbTable *tb = NULL; Sint ix = unsigned_val(state->slots.arr[state->slots.ix]); - meta_main_tab_lock(ix); + erts_smp_rwmtx_t *mmtl = get_meta_main_tab_lock(ix); + erts_smp_rwmtx_rlock(mmtl); if (IS_SLOT_ALIVE(ix)) { - tb = db_ref(meta_main_tab[ix].u.tb); + tb = db_ref(meta_main_tab[ix].u.tb, LCK_WRITE_REC); ASSERT(tb); } - meta_main_tab_unlock(ix); + erts_smp_rwmtx_runlock(mmtl); if (tb) { int reds; DbFixation** pp; @@ -3274,6 +3410,7 @@ unlocked: #ifdef ERTS_SMP if (*kind_p == LCK_READ && tb->common.is_thread_safe) { /* Must have write lock while purging pseudo-deleted (OTP-8166) */ + db_ref(tb, LCK_WRITE); /* LCK_WRITE need it, but not LCK_READ */ erts_smp_rwmtx_runlock(&tb->common.rwlock); erts_smp_rwmtx_rwlock(&tb->common.rwlock); *kind_p = LCK_WRITE; @@ -3386,6 +3523,7 @@ static int free_table_cont(Process *p, int clean_meta_tab) { Eterm result; + erts_smp_rwmtx_t *mmtl; #ifdef HARDDEBUG if (!first) { @@ -3411,9 +3549,20 @@ static int free_table_cont(Process *p, tb->common.id); #endif /* Completely done - we will not get called again. */ - meta_main_tab_lock(tb->common.slot); + mmtl = get_meta_main_tab_lock(tb->common.slot); +#ifdef ERTS_SMP + if (erts_smp_rwmtx_tryrwlock(mmtl) == EBUSY) { + /* + * We keep our increased refc over this op in order to + * prevent the table from disapearing. + */ + erts_smp_rwmtx_rwunlock(&tb->common.rwlock); + erts_smp_rwmtx_rwlock(mmtl); + erts_smp_rwmtx_rwlock(&tb->common.rwlock); + } +#endif free_slot(tb->common.slot); - meta_main_tab_unlock(tb->common.slot); + erts_smp_rwmtx_rwunlock(mmtl); if (clean_meta_tab) { db_meta_lock(meta_pid_to_tab, LCK_WRITE_REC); @@ -3421,7 +3570,7 @@ static int free_table_cont(Process *p, make_small(tb->common.slot)); db_meta_unlock(meta_pid_to_tab, LCK_WRITE_REC); } - db_unref(tb); + db_unref(tb, LCK_NONE); BUMP_REDS(p, 100); return 0; } diff --git a/erts/emulator/beam/erl_db_hash.c b/erts/emulator/beam/erl_db_hash.c index 96008ccef0..08117bb6e5 100644 --- a/erts/emulator/beam/erl_db_hash.c +++ b/erts/emulator/beam/erl_db_hash.c @@ -623,12 +623,16 @@ int db_create_hash(Process *p, DbTable *tbl) erts_smp_atomic_init(&tb->is_resizing, 0); #ifdef ERTS_SMP if (tb->common.type & DB_FINE_LOCKED) { + erts_smp_rwmtx_opt_t rwmtx_opt = ERTS_SMP_THR_OPTS_DEFAULT_INITER; int i; + if (tb->common.type & DB_FREQ_READ) + rwmtx_opt.type = ERTS_SMP_RWMTX_TYPE_FREQUENT_READ; tb->locks = (DbTableHashFineLocks*) erts_db_alloc_fnf(ERTS_ALC_T_DB_SEG, /* Other type maybe? */ (DbTable *) tb, sizeof(DbTableHashFineLocks)); for (i=0; i<DB_HASH_LOCK_CNT; ++i) { - erts_rwmtx_init_x(&tb->locks->lck_vec[i].lck, "db_hash_slot", make_small(i)); + erts_smp_rwmtx_init_opt_x(&tb->locks->lck_vec[i].lck, &rwmtx_opt, + "db_hash_slot", make_small(i)); } /* This important property is needed to guarantee that the buckets * involved in a grow/shrink operation it protected by the same lock: diff --git a/erts/emulator/beam/erl_db_util.h b/erts/emulator/beam/erl_db_util.h index 382e5dceb5..672c5f2cd1 100644 --- a/erts/emulator/beam/erl_db_util.h +++ b/erts/emulator/beam/erl_db_util.h @@ -240,8 +240,9 @@ typedef struct db_table_common { #define DB_DUPLICATE_BAG (1 << 8) #define DB_ORDERED_SET (1 << 9) #define DB_DELETE (1 << 10) /* table is being deleted */ +#define DB_FREQ_READ (1 << 11) -#define ERTS_ETS_TABLE_TYPES (DB_BAG|DB_SET|DB_DUPLICATE_BAG|DB_ORDERED_SET|DB_FINE_LOCKED) +#define ERTS_ETS_TABLE_TYPES (DB_BAG|DB_SET|DB_DUPLICATE_BAG|DB_ORDERED_SET|DB_FINE_LOCKED|DB_FREQ_READ) #define IS_HASH_TABLE(Status) (!!((Status) & \ (DB_BAG | DB_SET | DB_DUPLICATE_BAG))) diff --git a/erts/emulator/beam/erl_drv_thread.c b/erts/emulator/beam/erl_drv_thread.c index aa37edafd1..d42820ddf3 100644 --- a/erts/emulator/beam/erl_drv_thread.c +++ b/erts/emulator/beam/erl_drv_thread.c @@ -186,10 +186,9 @@ int erl_drv_mutex_trylock(ErlDrvMutex *dmtx) { #ifdef USE_THREADS - int res = dmtx ? ethr_mutex_trylock(&dmtx->mtx) : EINVAL; - if (res != 0 && res != EBUSY) - fatal_error(res, "erl_drv_mutex_trylock()"); - return res; + if (!dmtx) + fatal_error(EINVAL, "erl_drv_mutex_trylock()"); + return ethr_mutex_trylock(&dmtx->mtx); #else return 0; #endif @@ -199,9 +198,9 @@ void erl_drv_mutex_lock(ErlDrvMutex *dmtx) { #ifdef USE_THREADS - int res = dmtx ? ethr_mutex_lock(&dmtx->mtx) : EINVAL; - if (res != 0) - fatal_error(res, "erl_drv_mutex_lock()"); + if (!dmtx) + fatal_error(EINVAL, "erl_drv_mutex_lock()"); + ethr_mutex_lock(&dmtx->mtx); #endif } @@ -209,9 +208,9 @@ void erl_drv_mutex_unlock(ErlDrvMutex *dmtx) { #ifdef USE_THREADS - int res = dmtx ? ethr_mutex_unlock(&dmtx->mtx) : EINVAL; - if (res != 0) - fatal_error(res, "erl_drv_mutex_unlock()"); + if (!dmtx) + fatal_error(EINVAL, "erl_drv_mutex_unlock()"); + ethr_mutex_unlock(&dmtx->mtx); #endif } @@ -256,9 +255,9 @@ void erl_drv_cond_signal(ErlDrvCond *dcnd) { #ifdef USE_THREADS - int res = dcnd ? ethr_cond_signal(&dcnd->cnd) : EINVAL; - if (res != 0) - fatal_error(res, "erl_drv_cond_signal()"); + if (!dcnd) + fatal_error(EINVAL, "erl_drv_cond_signal()"); + ethr_cond_signal(&dcnd->cnd); #endif } @@ -266,9 +265,9 @@ void erl_drv_cond_broadcast(ErlDrvCond *dcnd) { #ifdef USE_THREADS - int res = dcnd ? ethr_cond_broadcast(&dcnd->cnd) : EINVAL; - if (res != 0) - fatal_error(res, "erl_drv_cond_broadcast()"); + if (!dcnd) + fatal_error(EINVAL, "erl_drv_cond_broadcast()"); + ethr_cond_broadcast(&dcnd->cnd); #endif } @@ -277,18 +276,13 @@ void erl_drv_cond_wait(ErlDrvCond *dcnd, ErlDrvMutex *dmtx) { #ifdef USE_THREADS - int res; if (!dcnd || !dmtx) { - res = EINVAL; - error: - fatal_error(res, "erl_drv_cond_wait()"); + fatal_error(EINVAL, "erl_drv_cond_wait()"); } while (1) { - res = ethr_cond_wait(&dcnd->cnd, &dmtx->mtx); + int res = ethr_cond_wait(&dcnd->cnd, &dmtx->mtx); if (res == 0) break; - if (res != EINTR) - goto error; } #endif } @@ -333,10 +327,9 @@ int erl_drv_rwlock_tryrlock(ErlDrvRWLock *drwlck) { #ifdef USE_THREADS - int res = drwlck ? ethr_rwmutex_tryrlock(&drwlck->rwmtx) : EINVAL; - if (res != 0 && res != EBUSY) - fatal_error(res, "erl_drv_rwlock_tryrlock()"); - return res; + if (!drwlck) + fatal_error(EINVAL, "erl_drv_rwlock_tryrlock()"); + return ethr_rwmutex_tryrlock(&drwlck->rwmtx); #else return 0; #endif @@ -346,9 +339,9 @@ void erl_drv_rwlock_rlock(ErlDrvRWLock *drwlck) { #ifdef USE_THREADS - int res = drwlck ? ethr_rwmutex_rlock(&drwlck->rwmtx) : EINVAL; - if (res != 0) - fatal_error(res, "erl_drv_rwlock_rlock()"); + if (!drwlck) + fatal_error(EINVAL, "erl_drv_rwlock_rlock()"); + ethr_rwmutex_rlock(&drwlck->rwmtx); #endif } @@ -356,9 +349,9 @@ void erl_drv_rwlock_runlock(ErlDrvRWLock *drwlck) { #ifdef USE_THREADS - int res = drwlck ? ethr_rwmutex_runlock(&drwlck->rwmtx) : EINVAL; - if (res != 0) - fatal_error(res, "erl_drv_rwlock_runlock()"); + if (!drwlck) + fatal_error(EINVAL, "erl_drv_rwlock_runlock()"); + ethr_rwmutex_runlock(&drwlck->rwmtx); #endif } @@ -366,10 +359,9 @@ int erl_drv_rwlock_tryrwlock(ErlDrvRWLock *drwlck) { #ifdef USE_THREADS - int res = drwlck ? ethr_rwmutex_tryrwlock(&drwlck->rwmtx) : EINVAL; - if (res != 0 && res != EBUSY) - fatal_error(res, "erl_drv_rwlock_tryrwlock()"); - return res; + if (!drwlck) + fatal_error(EINVAL, "erl_drv_rwlock_tryrwlock()"); + return ethr_rwmutex_tryrwlock(&drwlck->rwmtx); #else return 0; #endif @@ -379,9 +371,9 @@ void erl_drv_rwlock_rwlock(ErlDrvRWLock *drwlck) { #ifdef USE_THREADS - int res = drwlck ? ethr_rwmutex_rwlock(&drwlck->rwmtx) : EINVAL; - if (res != 0) - fatal_error(res, "erl_drv_rwlock_rwlock()"); + if (!drwlck) + fatal_error(EINVAL, "erl_drv_rwlock_rwlock()"); + ethr_rwmutex_rwlock(&drwlck->rwmtx); #endif } @@ -389,9 +381,9 @@ void erl_drv_rwlock_rwunlock(ErlDrvRWLock *drwlck) { #ifdef USE_THREADS - int res = drwlck ? ethr_rwmutex_rwunlock(&drwlck->rwmtx) : EINVAL; - if (res != 0) - fatal_error(res, "erl_drv_rwlock_rwunlock()"); + if (!drwlck) + fatal_error(EINVAL, "erl_drv_rwlock_rwunlock()"); + ethr_rwmutex_rwunlock(&drwlck->rwmtx); #endif } diff --git a/erts/emulator/beam/erl_fun.c b/erts/emulator/beam/erl_fun.c index 1f74393a96..5dce5ad262 100644 --- a/erts/emulator/beam/erl_fun.c +++ b/erts/emulator/beam/erl_fun.c @@ -37,8 +37,6 @@ static erts_smp_rwmtx_t erts_fun_table_lock; #define erts_fun_read_unlock() erts_smp_rwmtx_runlock(&erts_fun_table_lock) #define erts_fun_write_lock() erts_smp_rwmtx_rwlock(&erts_fun_table_lock) #define erts_fun_write_unlock() erts_smp_rwmtx_rwunlock(&erts_fun_table_lock) -#define erts_fun_init_lock() erts_smp_rwmtx_init(&erts_fun_table_lock, \ - "fun_tab") static HashValue fun_hash(ErlFunEntry* obj); static int fun_cmp(ErlFunEntry* obj1, ErlFunEntry* obj2); @@ -57,8 +55,12 @@ void erts_init_fun_table(void) { HashFunctions f; + erts_smp_rwmtx_opt_t rwmtx_opt = ERTS_SMP_THR_OPTS_DEFAULT_INITER; + rwmtx_opt.type = ERTS_SMP_RWMTX_TYPE_FREQUENT_READ; + rwmtx_opt.lived = ERTS_SMP_RWMTX_LONG_LIVED; + + erts_smp_rwmtx_init_opt(&erts_fun_table_lock, &rwmtx_opt, "fun_tab"); - erts_fun_init_lock(); f.hash = (H_FUN) fun_hash; f.cmp = (HCMP_FUN) fun_cmp; f.alloc = (HALLOC_FUN) fun_alloc; diff --git a/erts/emulator/beam/erl_init.c b/erts/emulator/beam/erl_init.c index 4a4507b212..14bd10b42c 100644 --- a/erts/emulator/beam/erl_init.c +++ b/erts/emulator/beam/erl_init.c @@ -78,6 +78,7 @@ static erts_tid_t main_thread; erts_cpu_info_t *erts_cpuinfo; +int erts_reader_groups; int erts_use_sender_punish; /* @@ -110,6 +111,7 @@ int erts_compat_rel; static int use_multi_run_queue; static int no_schedulers; static int no_schedulers_online; +static int max_reader_groups; #ifdef DEBUG Uint32 verbose; /* See erl_debug.h for information about verbose */ @@ -505,6 +507,7 @@ void erts_usage(void) ERTS_MIN_COMPAT_REL, this_rel_num()); erts_fprintf(stderr, "-r force ets memory block to be moved on realloc\n"); + erts_fprintf(stderr, "-rg amount set reader groups limit\n"); erts_fprintf(stderr, "-sbt type set scheduler bind type, valid types are:\n"); erts_fprintf(stderr, " u|ns|ts|ps|s|nnts|nnps|tnnps|db\n"); erts_fprintf(stderr, "-sct cput set cpu topology,\n"); @@ -538,6 +541,50 @@ void erts_usage(void) erl_exit(-1, ""); } +#ifdef USE_THREADS +/* + * allocators for thread lib + */ + +static void *ethr_std_alloc(size_t size) +{ + return erts_alloc_fnf(ERTS_ALC_T_ETHR_STD, (Uint) size); +} +static void *ethr_std_realloc(void *ptr, size_t size) +{ + return erts_realloc_fnf(ERTS_ALC_T_ETHR_STD, ptr, (Uint) size); +} +static void ethr_std_free(void *ptr) +{ + erts_free(ERTS_ALC_T_ETHR_STD, ptr); +} +static void *ethr_sl_alloc(size_t size) +{ + return erts_alloc_fnf(ERTS_ALC_T_ETHR_SL, (Uint) size); +} +static void *ethr_sl_realloc(void *ptr, size_t size) +{ + return erts_realloc_fnf(ERTS_ALC_T_ETHR_SL, ptr, (Uint) size); +} +static void ethr_sl_free(void *ptr) +{ + erts_free(ERTS_ALC_T_ETHR_SL, ptr); +} +static void *ethr_ll_alloc(size_t size) +{ + return erts_alloc_fnf(ERTS_ALC_T_ETHR_LL, (Uint) size); +} +static void *ethr_ll_realloc(void *ptr, size_t size) +{ + return erts_realloc_fnf(ERTS_ALC_T_ETHR_LL, ptr, (Uint) size); +} +static void ethr_ll_free(void *ptr) +{ + erts_free(ERTS_ALC_T_ETHR_LL, ptr); +} + +#endif + static void early_init(int *argc, char **argv) /* * Only put things here which are @@ -615,9 +662,15 @@ early_init(int *argc, char **argv) /* ? ncpuavail : (ncpuonln > 0 ? ncpuonln : no_schedulers)); +#ifdef ERTS_SMP + erts_max_main_threads = no_schedulers_online; +#endif + schdlrs = no_schedulers; schdlrs_onln = no_schedulers_online; + max_reader_groups = ERTS_MAX_READER_GROUPS; + if (argc && argv) { int i = 1; while (i < *argc) { @@ -627,6 +680,24 @@ early_init(int *argc, char **argv) /* } if (argv[i][0] == '-') { switch (argv[i][1]) { + case 'r': { + char *sub_param = argv[i]+2; + if (has_prefix("g", sub_param)) { + char *arg = get_arg(sub_param+1, argv[i+1], &i); + if (sscanf(arg, "%d", &max_reader_groups) != 1) { + erts_fprintf(stderr, + "bad reader groups limit: %s\n", arg); + erts_usage(); + } + if (max_reader_groups < 0) { + erts_fprintf(stderr, + "bad reader groups limit: %d\n", + max_reader_groups); + erts_usage(); + } + } + break; + } case 'S' : { int tot, onln; char *arg = get_arg(argv[i]+2, argv[i+1], &i); @@ -699,6 +770,36 @@ early_init(int *argc, char **argv) /* erts_early_init_scheduling(); /* Require allocators */ erts_init_utils(); /* Require allocators */ +#ifdef USE_THREADS + { + erts_thr_late_init_data_t elid = ERTS_THR_LATE_INIT_DATA_DEF_INITER; + elid.mem.std.alloc = ethr_std_alloc; + elid.mem.std.realloc = ethr_std_realloc; + elid.mem.std.free = ethr_std_free; + elid.mem.sl.alloc = ethr_sl_alloc; + elid.mem.sl.realloc = ethr_sl_realloc; + elid.mem.sl.free = ethr_sl_free; + elid.mem.ll.alloc = ethr_ll_alloc; + elid.mem.ll.realloc = ethr_ll_realloc; + elid.mem.ll.free = ethr_ll_free; + +#ifdef ERTS_SMP + elid.main_threads = erts_max_main_threads; +#else + elid.main_threads = 1; +#endif + elid.reader_groups = (elid.main_threads > 1 + ? elid.main_threads + : 0); + if (max_reader_groups <= 1) + elid.reader_groups = 0; + if (elid.reader_groups > max_reader_groups) + elid.reader_groups = max_reader_groups; + erts_reader_groups = elid.reader_groups; + + erts_thr_late_init(&elid); + } +#endif #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_late_init(); #endif @@ -1193,9 +1294,17 @@ erl_start(int argc, char **argv) erts_async_thread_suggested_stack_size)); break; - case 'r': - erts_ets_realloc_always_moves = 1; + case 'r': { + char *sub_param = argv[i]+2; + if (has_prefix("g", sub_param)) { + get_arg(sub_param+1, argv[i+1], &i); + /* already handled */ + } + else { + erts_ets_realloc_always_moves = 1; + } break; + } case 'n': /* XXX obsolete */ break; case 'c': @@ -1280,6 +1389,7 @@ erl_start(int argc, char **argv) erts_sys_main_thread(); /* May or may not return! */ #else + erts_thr_set_main_status(1, 1); set_main_stack_size(); process_main(); #endif @@ -1353,7 +1463,7 @@ system_cleanup(int exit_code) erts_cleanup_incgc(); #endif -#if defined(USE_THREADS) && !defined(ERTS_SMP) +#if defined(USE_THREADS) exit_async(); #endif #if HAVE_ERTS_MSEG diff --git a/erts/emulator/beam/erl_lock_check.c b/erts/emulator/beam/erl_lock_check.c index cee470ae37..99cc80e259 100644 --- a/erts/emulator/beam/erl_lock_check.c +++ b/erts/emulator/beam/erl_lock_check.c @@ -96,10 +96,10 @@ static erts_lc_lock_order_t erts_lock_order[] = { { "proc_status", "pid" }, { "proc_tab", NULL }, { "ports_snapshot", NULL }, - { "db_tab", "address" }, - { "db_tab_fix", "address" }, { "meta_name_tab", "address" }, { "meta_main_tab_slot", "address" }, + { "db_tab", "address" }, + { "db_tab_fix", "address" }, { "meta_main_tab_main", NULL }, { "db_hash_slot", "address" }, { "node_table", NULL }, @@ -119,9 +119,9 @@ static erts_lc_lock_order_t erts_lock_order[] = { { "child_status", NULL }, #endif #ifdef __WIN32__ - { "sys_driver_data_lock", NULL }, + { "sys_driver_data_lock", NULL }, #endif - { "drv_ev_state_grow", NULL, }, + { "drv_ev_state_grow", NULL, }, { "drv_ev_state", "address" }, { "safe_hash", "address" }, { "pollset_rm_list", NULL }, @@ -153,6 +153,7 @@ static erts_lc_lock_order_t erts_lock_order[] = { { "instr", NULL }, { "fix_alloc", "index" }, { "alcu_allocator", "index" }, + { "alcu_delayed_free", "index" }, { "mseg", NULL }, #ifdef HALFWORD_HEAP { "pmmap", NULL }, @@ -177,17 +178,19 @@ static erts_lc_lock_order_t erts_lock_order[] = { { "pix_lock", "address" }, { "run_queues_lists", NULL }, { "sched_stat", NULL }, + { "run_queue_sleep_list", "address" }, #endif { "alloc_thr_ix_lock", NULL }, #ifdef ERTS_SMP - { "proc_lck_wtr_alloc", NULL }, + { "proc_lck_qs_alloc", NULL }, #endif #ifdef __WIN32__ #ifdef DEBUG { "save_ops_lock", NULL }, #endif #endif - { "mtrace_buf", NULL } + { "mtrace_buf", NULL }, + { "erts_alloc_hard_debug", NULL } }; #define ERTS_LOCK_ORDER_SIZE \ @@ -199,6 +202,8 @@ static erts_lc_lock_order_t erts_lock_order[] = { & ERTS_LC_FLG_LT_ALL \ & ~(ERTS_LC_FLG_LT_SPINLOCK|ERTS_LC_FLG_LT_RWSPINLOCK))) +static __decl_noreturn void __noreturn lc_abort(void); + static char * lock_type(Uint16 flags) { @@ -222,7 +227,7 @@ rw_op_str(Uint16 flags) return " (r)"; case ERTS_LC_FLG_LO_WRITE: erts_fprintf(stderr, "\nInternal error\n"); - abort(); + lc_abort(); default: break; } @@ -271,28 +276,18 @@ static erts_lc_free_block_t *free_blocks; #define ERTS_LC_FB_CHUNK_SIZE 10 #endif -#ifdef ETHR_HAVE_NATIVE_LOCKS static ethr_spinlock_t free_blocks_lock; -#define ERTS_LC_LOCK ethr_spin_lock -#define ERTS_LC_UNLOCK ethr_spin_unlock -#else -static ethr_mutex free_blocks_lock; -#define ERTS_LC_LOCK ethr_mutex_lock -#define ERTS_LC_UNLOCK ethr_mutex_unlock -#endif static ERTS_INLINE void lc_lock(void) { - if (ERTS_LC_LOCK(&free_blocks_lock) != 0) - abort(); + ethr_spin_lock(&free_blocks_lock); } static ERTS_INLINE void lc_unlock(void) { - if (ERTS_LC_UNLOCK(&free_blocks_lock) != 0) - abort(); + ethr_spin_unlock(&free_blocks_lock); } static ERTS_INLINE void lc_free(void *p) @@ -313,7 +308,7 @@ static void *lc_core_alloc(void) { lc_unlock(); erts_fprintf(stderr, "Lock checker out of memory!\n"); - abort(); + lc_abort(); } #else @@ -327,7 +322,7 @@ static void *lc_core_alloc(void) * ERTS_LC_FB_CHUNK_SIZE); if (!fbs) { erts_fprintf(stderr, "Lock checker failed to allocate memory!\n"); - abort(); + lc_abort(); } for (i = 1; i < ERTS_LC_FB_CHUNK_SIZE - 1; i++) { #ifdef DEBUG @@ -367,11 +362,11 @@ create_locked_locks(char *thread_name) { erts_lc_locked_locks_t *l_lcks = malloc(sizeof(erts_lc_locked_locks_t)); if (!l_lcks) - abort(); + lc_abort(); l_lcks->thread_name = strdup(thread_name ? thread_name : "unknown"); if (!l_lcks->thread_name) - abort(); + lc_abort(); l_lcks->tid = erts_thr_self(); l_lcks->required.first = NULL; @@ -513,7 +508,7 @@ uninitialized_lock(void) { erts_fprintf(stderr, "Performing operations on uninitialized lock!\n"); print_curr_locks(get_my_locked_locks()); - abort(); + lc_abort(); } static void @@ -523,7 +518,7 @@ lock_twice(char *prefix, erts_lc_locked_locks_t *l_lcks, erts_lc_lock_t *lck, erts_fprintf(stderr, "%s%s", prefix, rw_op_str(op_flags)); print_lock(" ", lck, " lock which is already locked by thread!\n"); print_curr_locks(l_lcks); - abort(); + lc_abort(); } static void @@ -533,7 +528,7 @@ unlock_op_mismatch(erts_lc_locked_locks_t *l_lcks, erts_lc_lock_t *lck, erts_fprintf(stderr, "Unlocking%s ", rw_op_str(op_flags)); print_lock("", lck, " lock which mismatch previous lock operation!\n"); print_curr_locks(l_lcks); - abort(); + lc_abort(); } static void @@ -541,7 +536,7 @@ unlock_of_not_locked(erts_lc_locked_locks_t *l_lcks, erts_lc_lock_t *lck) { print_lock("Unlocking ", lck, " lock which is not locked by thread!\n"); print_curr_locks(l_lcks); - abort(); + lc_abort(); } static void @@ -550,7 +545,7 @@ lock_order_violation(erts_lc_locked_locks_t *l_lcks, erts_lc_lock_t *lck) print_lock("Lock order violation occured when locking ", lck, "!\n"); print_curr_locks(l_lcks); print_lock_order(); - abort(); + lc_abort(); } static void @@ -561,7 +556,7 @@ type_order_violation(char *op, erts_lc_locked_locks_t *l_lcks, print_lock(op, lck, "!\n"); ASSERT(l_lcks); print_curr_locks(l_lcks); - abort(); + lc_abort(); } static void @@ -613,7 +608,7 @@ lock_mismatch(erts_lc_locked_locks_t *l_lcks, int exact, } } print_curr_locks(l_lcks); - abort(); + lc_abort(); } static void @@ -621,7 +616,7 @@ unlock_of_required_lock(erts_lc_locked_locks_t *l_lcks, erts_lc_lock_t *lck) { print_lock("Unlocking required ", lck, " lock!\n"); print_curr_locks(l_lcks); - abort(); + lc_abort(); } static void @@ -629,7 +624,7 @@ unrequire_of_not_required_lock(erts_lc_locked_locks_t *l_lcks, erts_lc_lock_t *l { print_lock("Unrequire on ", lck, " lock not required!\n"); print_curr_locks(l_lcks); - abort(); + lc_abort(); } static void @@ -637,7 +632,7 @@ require_twice(erts_lc_locked_locks_t *l_lcks, erts_lc_lock_t *lck) { print_lock("Require on ", lck, " lock already required!\n"); print_curr_locks(l_lcks); - abort(); + lc_abort(); } static void @@ -645,7 +640,7 @@ required_not_locked(erts_lc_locked_locks_t *l_lcks, erts_lc_lock_t *lck) { print_lock("Required ", lck, " lock not locked!\n"); print_curr_locks(l_lcks); - abort(); + lc_abort(); } @@ -658,13 +653,23 @@ thread_exit_handler(void) erts_fprintf(stderr, "Thread exiting while having locked locks!\n"); print_curr_locks(l_lcks); - abort(); + lc_abort(); } destroy_locked_locks(l_lcks); /* erts_tsd_set(locks_key, NULL); */ } } +static __decl_noreturn void +lc_abort(void) +{ +#ifdef __WIN32__ + DebugBreak(); +#else + abort(); +#endif +} + void erts_lc_set_thread_name(char *thread_name) { @@ -676,7 +681,7 @@ erts_lc_set_thread_name(char *thread_name) free((void *) l_lcks->thread_name); l_lcks->thread_name = strdup(thread_name ? thread_name : "unknown"); if (!l_lcks->thread_name) - abort(); + lc_abort(); } } @@ -686,7 +691,7 @@ erts_lc_assert_failed(char *file, int line, char *assertion) erts_fprintf(stderr, "%s:%d: Lock check assertion \"%s\" failed!\n", file, line, assertion); print_curr_locks(get_my_locked_locks()); - abort(); + lc_abort(); return 0; } @@ -699,7 +704,7 @@ void erts_lc_fail(char *fmt, ...) va_end(args); erts_fprintf(stderr, "\n"); print_curr_locks(get_my_locked_locks()); - abort(); + lc_abort(); } @@ -719,7 +724,7 @@ erts_lc_get_lock_order_id(char *name) "(update erl_lock_check.c)\n", name); } - abort(); + lc_abort(); return (Sint16) -1; } @@ -895,6 +900,25 @@ erts_lc_check_exact(erts_lc_lock_t *have, int have_len) } } +void +erts_lc_check_no_locked_of_type(Uint16 flags) +{ + erts_lc_locked_locks_t *l_lcks = get_my_locked_locks(); + if (l_lcks) { + erts_lc_locked_lock_t *l_lck = l_lcks->locked.first; + for (l_lck = l_lcks->locked.first; l_lck; l_lck = l_lck->next) { + if (l_lck->flags & flags) { + erts_fprintf(stderr, + "Locked lock of type %s found which isn't " + "allowed here!\n", + lock_type(l_lck->flags)); + print_curr_locks(l_lcks); + lc_abort(); + } + } + } +} + int erts_lc_trylock_force_busy_flg(erts_lc_lock_t *lck, Uint16 op_flags) { @@ -1283,13 +1307,8 @@ erts_lc_init(void) free_blocks = NULL; #endif /* #ifdef ERTS_LC_STATIC_ALLOC */ -#ifdef ETHR_HAVE_NATIVE_LOCKS if (ethr_spinlock_init(&free_blocks_lock) != 0) - abort(); -#else - if (ethr_mutex_init(&free_blocks_lock) != 0) - abort(); -#endif + lc_abort(); erts_tsd_key_create(&locks_key); } diff --git a/erts/emulator/beam/erl_lock_check.h b/erts/emulator/beam/erl_lock_check.h index d5e2ede9ac..0372e6850d 100644 --- a/erts/emulator/beam/erl_lock_check.h +++ b/erts/emulator/beam/erl_lock_check.h @@ -77,6 +77,7 @@ void erts_lc_check(erts_lc_lock_t *have, int have_len, void erts_lc_check_exact(erts_lc_lock_t *have, int have_len); void erts_lc_have_locks(int *resv, erts_lc_lock_t *lcks, int len); void erts_lc_have_lock_ids(int *resv, int *ids, int len); +void erts_lc_check_no_locked_of_type(Uint16 flags); int erts_lc_trylock_force_busy_flg(erts_lc_lock_t *lck, Uint16 op_flags); void erts_lc_trylock_flg(int locked, erts_lc_lock_t *lck, Uint16 op_flags); void erts_lc_lock_flg(erts_lc_lock_t *lck, Uint16 op_flags); diff --git a/erts/emulator/beam/erl_lock_count.c b/erts/emulator/beam/erl_lock_count.c index 26028aeefc..239773f366 100644 --- a/erts/emulator/beam/erl_lock_count.c +++ b/erts/emulator/beam/erl_lock_count.c @@ -166,8 +166,8 @@ static void print_lock_x(erts_lcnt_lock_t *lock, Uint16 flag, char *action, char int i; type = lcnt_lock_type(lock->flag); - ethr_atomic_read(&lock->r_state, &r_state); - ethr_atomic_read(&lock->w_state, &w_state); + r_state = ethr_atomic_read(&lock->r_state); + w_state = ethr_atomic_read(&lock->w_state); if (lock->flag & flag) { @@ -394,10 +394,10 @@ void erts_lcnt_lock_opt(erts_lcnt_lock_t *lock, Uint16 option) { ASSERT(eltd); - ethr_atomic_read(&lock->w_state, &w_state); + w_state = ethr_atomic_read(&lock->w_state); if (option & ERTS_LCNT_LO_WRITE) { - ethr_atomic_read(&lock->r_state, &r_state); + r_state = ethr_atomic_read(&lock->r_state); ethr_atomic_inc( &lock->w_state); } if (option & ERTS_LCNT_LO_READ) { @@ -423,7 +423,7 @@ void erts_lcnt_lock(erts_lcnt_lock_t *lock) { if (erts_lcnt_rt_options & ERTS_LCNT_OPT_SUSPEND) return; - ethr_atomic_read(&lock->w_state, &w_state); + w_state = ethr_atomic_read(&lock->w_state); ethr_atomic_inc( &lock->w_state); eltd = lcnt_get_thread_data(); @@ -478,7 +478,7 @@ void erts_lcnt_lock_post_x(erts_lcnt_lock_t *lock, char *file, unsigned int line #ifdef DEBUG if (!(lock->flag & (ERTS_LCNT_LT_RWMUTEX | ERTS_LCNT_LT_RWSPINLOCK))) { - ethr_atomic_read(&lock->flowstate, &flowstate); + flowstate = ethr_atomic_read(&lock->flowstate); ASSERT(flowstate == 0); ethr_atomic_inc( &lock->flowstate); } @@ -522,12 +522,12 @@ void erts_lcnt_unlock(erts_lcnt_lock_t *lock) { if (erts_lcnt_rt_options & ERTS_LCNT_OPT_SUSPEND) return; #ifdef DEBUG /* flowstate */ - ethr_atomic_read(&lock->flowstate, &flowstate); + flowstate = ethr_atomic_read(&lock->flowstate); ASSERT(flowstate == 1); ethr_atomic_dec( &lock->flowstate); /* write state */ - ethr_atomic_read(&lock->w_state, &w_state); + w_state = ethr_atomic_read(&lock->w_state); ASSERT(w_state > 0) #endif ethr_atomic_dec(&lock->w_state); @@ -558,7 +558,7 @@ void erts_lcnt_trylock(erts_lcnt_lock_t *lock, int res) { if (res != EBUSY) { #ifdef DEBUG - ethr_atomic_read(&lock->flowstate, &flowstate); + flowstate = ethr_atomic_read(&lock->flowstate); ASSERT(flowstate == 0); ethr_atomic_inc( &lock->flowstate); #endif diff --git a/erts/emulator/beam/erl_mtrace.c b/erts/emulator/beam/erl_mtrace.c index 9cf55ee319..b1478758a1 100644 --- a/erts/emulator/beam/erl_mtrace.c +++ b/erts/emulator/beam/erl_mtrace.c @@ -585,9 +585,7 @@ void erts_mtrace_init(char *receiver, char *nodename) Uint16 port; erts_mtx_init(&mtrace_buf_mutex, "mtrace_buf"); - erts_mtx_set_forksafe(&mtrace_buf_mutex); erts_mtx_init(&mtrace_op_mutex, "mtrace_op"); - erts_mtx_set_forksafe(&mtrace_op_mutex); socket_desc = erts_sock_open(); if (socket_desc == ERTS_SOCK_INVALID_SOCKET) { diff --git a/erts/emulator/beam/erl_node_tables.c b/erts/emulator/beam/erl_node_tables.c index 7482da251e..e430b4ad77 100644 --- a/erts/emulator/beam/erl_node_tables.c +++ b/erts/emulator/beam/erl_node_tables.c @@ -80,6 +80,8 @@ dist_table_alloc(void *dep_tmpl) Eterm chnl_nr; Eterm sysname; DistEntry *dep; + erts_smp_rwmtx_opt_t rwmtx_opt = ERTS_SMP_THR_OPTS_DEFAULT_INITER; + rwmtx_opt.type = ERTS_SMP_RWMTX_TYPE_FREQUENT_READ; if(((DistEntry *) dep_tmpl) == erts_this_dist_entry) return dep_tmpl; @@ -92,7 +94,7 @@ dist_table_alloc(void *dep_tmpl) dep->prev = NULL; erts_refc_init(&dep->refc, -1); - erts_smp_rwmtx_init_x(&dep->rwmtx, "dist_entry", chnl_nr); + erts_smp_rwmtx_init_opt_x(&dep->rwmtx, &rwmtx_opt, "dist_entry", chnl_nr); dep->sysname = sysname; dep->cid = NIL; dep->connection_id = 0; @@ -580,6 +582,18 @@ ErlNode *erts_find_or_insert_node(Eterm sysname, Uint creation) ErlNode ne; ne.sysname = sysname; ne.creation = creation; + + erts_smp_rwmtx_rlock(&erts_node_table_rwmtx); + res = hash_get(&erts_node_table, (void *) &ne); + if (res && res != erts_this_node) { + long refc = erts_refc_inctest(&res->refc, 0); + if (refc < 2) /* New or pending delete */ + erts_refc_inc(&res->refc, 1); + } + erts_smp_rwmtx_runlock(&erts_node_table_rwmtx); + if (res) + return res; + erts_smp_rwmtx_rwlock(&erts_node_table_rwmtx); res = hash_put(&erts_node_table, (void *) &ne); ASSERT(res); @@ -696,8 +710,12 @@ erts_set_this_node(Eterm sysname, Uint creation) void erts_init_node_tables(void) { + erts_smp_rwmtx_opt_t rwmtx_opt = ERTS_SMP_THR_OPTS_DEFAULT_INITER; HashFunctions f; + rwmtx_opt.type = ERTS_SMP_RWMTX_TYPE_FREQUENT_READ; + rwmtx_opt.lived = ERTS_SMP_RWMTX_LONG_LIVED; + f.hash = (H_FUN) dist_table_hash; f.cmp = (HCMP_FUN) dist_table_cmp; f.alloc = (HALLOC_FUN) dist_table_alloc; @@ -719,9 +737,10 @@ void erts_init_node_tables(void) erts_this_dist_entry->prev = NULL; erts_refc_init(&erts_this_dist_entry->refc, 1); /* erts_this_node */ - erts_smp_rwmtx_init_x(&erts_this_dist_entry->rwmtx, - "dist_entry", - make_small(ERST_INTERNAL_CHANNEL_NO)); + erts_smp_rwmtx_init_opt_x(&erts_this_dist_entry->rwmtx, + &rwmtx_opt, + "dist_entry", + make_small(ERST_INTERNAL_CHANNEL_NO)); erts_this_dist_entry->sysname = am_Noname; erts_this_dist_entry->cid = NIL; erts_this_dist_entry->connection_id = 0; @@ -772,8 +791,8 @@ void erts_init_node_tables(void) (void) hash_put(&erts_node_table, (void *) erts_this_node); - erts_smp_rwmtx_init(&erts_node_table_rwmtx, "node_table"); - erts_smp_rwmtx_init(&erts_dist_table_rwmtx, "dist_table"); + erts_smp_rwmtx_init_opt(&erts_node_table_rwmtx, &rwmtx_opt, "node_table"); + erts_smp_rwmtx_init_opt(&erts_dist_table_rwmtx, &rwmtx_opt, "dist_table"); references_atoms_need_init = 1; } diff --git a/erts/emulator/beam/erl_port_task.c b/erts/emulator/beam/erl_port_task.c index 967a14f0d1..c10724b951 100644 --- a/erts/emulator/beam/erl_port_task.c +++ b/erts/emulator/beam/erl_port_task.c @@ -625,6 +625,7 @@ erts_port_task_schedule(Eterm id, if (!enq_port) { ERTS_PT_CHK_PRES_PORTQ(runq, pp); + erts_smp_runq_unlock(runq); } else { enqueue_port(runq, pp); @@ -634,9 +635,10 @@ erts_port_task_schedule(Eterm id, profile_runnable_port(pp, am_active); } + erts_smp_runq_unlock(runq); + erts_smp_notify_inc_runq(runq); } - erts_smp_runq_unlock(runq); return 0; } @@ -944,8 +946,8 @@ erts_port_task_execute(ErtsRunQueue *runq, Port **curr_port_pp) enqueue_port(xrunq, pp); ASSERT(pp->sched.exe_taskq); pp->sched.exe_taskq = NULL; - erts_smp_notify_inc_runq(xrunq); erts_smp_runq_unlock(xrunq); + erts_smp_notify_inc_runq(xrunq); } #endif port_was_enqueued = 1; @@ -1112,7 +1114,6 @@ erts_port_migrate(Port *prt, int *prt_locked, dequeue_port(from_rq, prt); erts_smp_atomic_set(&prt->run_queue, (long) to_rq); enqueue_port(to_rq, prt); - erts_smp_notify_inc_runq(to_rq); return ERTS_MIGRATE_SUCCESS; } diff --git a/erts/emulator/beam/erl_process.c b/erts/emulator/beam/erl_process.c index 8c88353593..a644520442 100644 --- a/erts/emulator/beam/erl_process.c +++ b/erts/emulator/beam/erl_process.c @@ -46,7 +46,13 @@ #define ERTS_PROC_MIN_CONTEXT_SWITCH_REDS_COST (CONTEXT_REDS/10) -#define ERTS_SCHED_SLEEP_SPINCOUNT 10000 +#define ERTS_SCHED_SPIN_UNTIL_YIELD 100 + +#define ERTS_SCHED_SYS_SLEEP_SPINCOUNT 10 +#define ERTS_SCHED_TSE_SLEEP_SPINCOUNT_FACT 1000 +#define ERTS_SCHED_TSE_SLEEP_SPINCOUNT \ + (ERTS_SCHED_SYS_SLEEP_SPINCOUNT*ERTS_SCHED_TSE_SLEEP_SPINCOUNT_FACT) +#define ERTS_SCHED_SUSPEND_SLEEP_SPINCOUNT 0 #define ERTS_WAKEUP_OTHER_LIMIT (100*CONTEXT_REDS/2) #define ERTS_WAKEUP_OTHER_DEC 10 @@ -106,6 +112,10 @@ Uint erts_no_schedulers; Uint erts_max_processes = ERTS_DEFAULT_MAX_PROCESSES; Uint erts_process_tab_index_mask; +#ifdef ERTS_SMP +Uint erts_max_main_threads; +#endif + int erts_sched_thread_suggested_stack_size = -1; #ifdef ERTS_ENABLE_LOCK_CHECK @@ -116,16 +126,34 @@ ErtsLcPSDLocks erts_psd_required_locks[ERTS_PSD_SIZE]; int erts_disable_proc_not_running_opt; -#define ERTS_SCHED_CHANGING_ONLINE 1 -#define ERTS_SCHED_CHANGING_MULTI_SCHED 2 +#define ERTS_SCHDLR_SSPND_CHNG_WAITER (((long) 1) << 0) +#define ERTS_SCHDLR_SSPND_CHNG_MSB (((long) 1) << 1) +#define ERTS_SCHDLR_SSPND_CHNG_ONLN (((long) 1) << 2) + +#ifndef DEBUG + +#define ERTS_SCHDLR_SSPND_CHNG_SET(VAL, OLD_VAL) \ + erts_smp_atomic_set(&schdlr_sspnd.changing, (VAL)) + +#else + +#define ERTS_SCHDLR_SSPND_CHNG_SET(VAL, OLD_VAL) \ +do { \ + long old_val__ = erts_smp_atomic_xchg(&schdlr_sspnd.changing, \ + (VAL)); \ + ASSERT(old_val__ == (OLD_VAL)); \ +} while (0) + +#endif + static struct { erts_smp_mtx_t mtx; erts_smp_cnd_t cnd; - int changing; int online; int curr_online; int wait_curr_online; + erts_smp_atomic_t changing; erts_smp_atomic_t active; struct { erts_smp_atomic_t ongoing; @@ -231,6 +259,17 @@ typedef union { ErtsAlignedSchedulerData *erts_aligned_scheduler_data; +#ifdef ERTS_SMP + +typedef union { + ErtsSchedulerSleepInfo ssi; + char align[ERTS_ALC_CACHE_LINE_ALIGN_SIZE(sizeof(ErtsSchedulerSleepInfo))]; +} ErtsAlignedSchedulerSleepInfo; + +static ErtsAlignedSchedulerSleepInfo *aligned_sched_sleep_info; + +#endif + #ifndef BM_COUNTERS static int processes_busy; #endif @@ -288,8 +327,15 @@ ERTS_SCHED_PREF_QUICK_ALLOC_IMPL(proclist, 200, ERTS_ALC_T_PROC_LIST) -#define ERTS_RUNQ_IX(IX) (&erts_aligned_run_queues[(IX)].runq) -#define ERTS_SCHEDULER_IX(IX) (&erts_aligned_scheduler_data[(IX)].esd) +#define ERTS_RUNQ_IX(IX) \ + (ASSERT_EXPR(0 <= (IX) && (IX) < erts_no_run_queues), \ + &erts_aligned_run_queues[(IX)].runq) +#define ERTS_SCHEDULER_IX(IX) \ + (ASSERT_EXPR(0 <= (IX) && (IX) < erts_no_schedulers), \ + &erts_aligned_scheduler_data[(IX)].esd) +#define ERTS_SCHED_SLEEP_INFO_IX(IX) \ + (ASSERT_EXPR(0 <= (IX) && (IX) < erts_no_schedulers), \ + &aligned_sched_sleep_info[(IX)].ssi) #define ERTS_FOREACH_RUNQ(RQVAR, DO) \ do { \ @@ -353,6 +399,11 @@ static void signal_schedulers_bind_change(erts_cpu_topology_t *cpudata, int size #endif +static int reader_group_lookup(int logical); +static void create_tmp_cpu_topology_copy(erts_cpu_topology_t **cpudata, + int *cpudata_size); +static void destroy_tmp_cpu_topology_copy(erts_cpu_topology_t *cpudata); + static void early_cpu_bind_init(void); static void late_cpu_bind_init(void); @@ -582,6 +633,76 @@ erts_psd_set_init(Process *p, ErtsProcLocks plocks, int ix, void *data) #ifdef ERTS_SMP +void +erts_sched_finish_poke(ErtsSchedulerSleepInfo *ssi, long flags) +{ + switch (flags & ERTS_SSI_FLGS_SLEEP_TYPE) { + case ERTS_SSI_FLG_POLL_SLEEPING: + erts_sys_schedule_interrupt(1); + break; + case ERTS_SSI_FLG_TSE_SLEEPING: + erts_tse_set(ssi->event); + break; + case 0: + break; + default: + erl_exit(ERTS_ABORT_EXIT, "%s:%d: Internal error\n", + __FILE__, __LINE__); + break; + } +} + +#ifdef ERTS_SMP_SCHEDULERS_NEED_TO_CHECK_CHILDREN +void +erts_smp_notify_check_children_needed(void) +{ + int i; + + for (i = 0; i < erts_no_schedulers; i++) { + long aux_work; + ErtsSchedulerSleepInfo *ssi; + ssi = ERTS_SCHED_SLEEP_INFO_IX(i); + aux_work = erts_smp_atomic_bor(&ssi->aux_work, + ERTS_SSI_AUX_WORK_CHECK_CHILDREN); + if (!(aux_work & ERTS_SSI_AUX_WORK_CHECK_CHILDREN)) + erts_sched_poke(ssi); + } +} +#endif + +#ifdef ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK +static ERTS_INLINE long +blockable_aux_work(ErtsSchedulerData *esdp, + ErtsSchedulerSleepInfo *ssi, + long aux_work) +{ + if (aux_work & ERTS_SSI_BLOCKABLE_AUX_WORK_MASK) { +#ifdef ERTS_SMP_SCHEDULERS_NEED_TO_CHECK_CHILDREN + if (aux_work & ERTS_SSI_AUX_WORK_CHECK_CHILDREN) { + aux_work = erts_smp_atomic_band(&ssi->aux_work, + ~ERTS_SSI_AUX_WORK_CHECK_CHILDREN); + aux_work &= ~ERTS_SSI_AUX_WORK_CHECK_CHILDREN; + erts_check_children(); + } +#endif + } + return aux_work; +} + +#endif + +#ifdef ERTS_SCHED_NEED_NONBLOCKABLE_AUX_WORK +static ERTS_INLINE long +nonblockable_aux_work(ErtsSchedulerData *esdp, + ErtsSchedulerSleepInfo *ssi, + long aux_work) +{ + if (aux_work & ERTS_SSI_NONBLOCKABLE_AUX_WORK_MASK) { + + } +} +#endif + static void prepare_for_block(void *vrq) { @@ -634,7 +755,31 @@ erts_active_schedulers(void) return as; } +static ERTS_INLINE int +prepare_for_sys_schedule(void) +{ #ifdef ERTS_SMP + while (!erts_port_task_have_outstanding_io_tasks() + && !erts_smp_atomic_xchg(&doing_sys_schedule, 1)) { + if (!erts_port_task_have_outstanding_io_tasks()) + return 1; + erts_smp_atomic_set(&doing_sys_schedule, 0); + } + return 0; +#else + return !erts_port_task_have_outstanding_io_tasks(); +#endif +} + +#ifdef ERTS_SMP + +static ERTS_INLINE void +sched_change_waiting_sys_to_waiting(Uint no, ErtsRunQueue *rq) +{ + ERTS_SMP_LC_ASSERT(erts_smp_lc_runq_is_locked(rq)); + ASSERT(rq->waiting < 0); + rq->waiting *= -1; +} static ERTS_INLINE void sched_waiting(Uint no, ErtsRunQueue *rq) @@ -695,101 +840,304 @@ non_empty_runq(ErtsRunQueue *rq) } } -static ERTS_INLINE int -sched_spin_wake(ErtsRunQueue *rq) +static long +sched_prep_spin_wait(ErtsSchedulerSleepInfo *ssi) { -#if ERTS_SCHED_SLEEP_SPINCOUNT == 0 - return 0; -#else - long val; - ERTS_SMP_LC_ASSERT(erts_smp_lc_runq_is_locked(rq)); + long oflgs; + long nflgs = (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_WAITING); + long xflgs = 0; - val = erts_smp_atomic_read(&rq->spin_waiter); - ASSERT(val >= 0); - if (val != 0) { - erts_smp_atomic_inc(&rq->spin_wake); - return 1; - } - return 0; -#endif + do { + oflgs = erts_smp_atomic_cmpxchg(&ssi->flags, nflgs, xflgs); + if (oflgs == xflgs) + return nflgs; + xflgs = oflgs; + } while (!(oflgs & ERTS_SSI_FLG_SUSPENDED)); + return oflgs; } -static ERTS_INLINE int -sched_spin_wake_all(ErtsRunQueue *rq) +static long +sched_prep_cont_spin_wait(ErtsSchedulerSleepInfo *ssi) { -#if ERTS_SCHED_SLEEP_SPINCOUNT == 0 - return 0; -#else - long val; - ERTS_SMP_LC_ASSERT(erts_smp_lc_runq_is_locked(rq)); + long oflgs; + long nflgs = (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_WAITING); + long xflgs = ERTS_SSI_FLG_WAITING; - val = erts_smp_atomic_read(&rq->spin_waiter); - ASSERT(val >= 0); - if (val != 0) - erts_smp_atomic_add(&rq->spin_wake, val); - return val; -#endif + do { + oflgs = erts_smp_atomic_cmpxchg(&ssi->flags, nflgs, xflgs); + if (oflgs == xflgs) + return nflgs; + xflgs = oflgs; + nflgs |= oflgs & ERTS_SSI_FLG_SUSPENDED; + } while (oflgs & ERTS_SSI_FLG_WAITING); + return oflgs; +} + +static long +sched_spin_wait(ErtsSchedulerSleepInfo *ssi, int spincount) +{ + long until_yield = ERTS_SCHED_SPIN_UNTIL_YIELD; + int sc = spincount; + long flgs; + + do { + flgs = erts_smp_atomic_read(&ssi->flags); + if ((flgs & (ERTS_SSI_FLG_SLEEPING|ERTS_SSI_FLG_WAITING)) + != (ERTS_SSI_FLG_SLEEPING|ERTS_SSI_FLG_WAITING)) { + break; + } + ERTS_SPIN_BODY; + if (--until_yield == 0) { + until_yield = ERTS_SCHED_SPIN_UNTIL_YIELD; + erts_thr_yield(); + } + } while (--sc > 0); + return flgs; } +static long +sched_set_sleeptype(ErtsSchedulerSleepInfo *ssi, long sleep_type) +{ + long oflgs; + long nflgs = ERTS_SSI_FLG_SLEEPING|ERTS_SSI_FLG_WAITING|sleep_type; + long xflgs = ERTS_SSI_FLG_SLEEPING|ERTS_SSI_FLG_WAITING; + + if (sleep_type == ERTS_SSI_FLG_TSE_SLEEPING) + erts_tse_reset(ssi->event); + + while (1) { + oflgs = erts_smp_atomic_cmpxchg(&ssi->flags, nflgs, xflgs); + if (oflgs == xflgs) + return nflgs; + if ((oflgs & (ERTS_SSI_FLG_SLEEPING|ERTS_SSI_FLG_WAITING)) + != (ERTS_SSI_FLG_SLEEPING|ERTS_SSI_FLG_WAITING)) { + return oflgs; + } + xflgs = oflgs; + nflgs |= oflgs & ERTS_SSI_FLG_SUSPENDED; + } +} + +#define ERTS_SCHED_WAIT_WOKEN(FLGS) \ + (((FLGS) & (ERTS_SSI_FLG_WAITING|ERTS_SSI_FLG_SUSPENDED)) \ + != ERTS_SSI_FLG_WAITING) + static void -sched_sys_wait(Uint no, ErtsRunQueue *rq) +scheduler_wait(long *fcalls, ErtsSchedulerData *esdp, ErtsRunQueue *rq) { - long dt; -#if ERTS_SCHED_SLEEP_SPINCOUNT != 0 - int val; - int spincount = ERTS_SCHED_SLEEP_SPINCOUNT; + ErtsSchedulerSleepInfo *ssi = esdp->ssi; + int spincount; + long flgs; +#if defined(ERTS_SCHED_NEED_NONBLOCKABLE_AUX_WORK) \ + || defined(ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK) + long aux_work; +#endif + ERTS_SMP_LC_ASSERT(erts_smp_lc_runq_is_locked(rq)); + erts_smp_spin_lock(&rq->sleepers.lock); + flgs = sched_prep_spin_wait(ssi); + if (flgs & ERTS_SSI_FLG_SUSPENDED) { + /* Go suspend instead... */ + erts_smp_spin_unlock(&rq->sleepers.lock); + return; + } + + ssi->prev = NULL; + ssi->next = rq->sleepers.list; + if (rq->sleepers.list) + rq->sleepers.list->prev = ssi; + rq->sleepers.list = ssi; + erts_smp_spin_unlock(&rq->sleepers.lock); + + /* + * If all schedulers are waiting, one of them *should* + * be waiting in erl_sys_schedule() + */ + + if (!prepare_for_sys_schedule()) { + + sched_waiting(esdp->no, rq); + + erts_smp_runq_unlock(rq); + + spincount = ERTS_SCHED_TSE_SLEEP_SPINCOUNT; + + tse_wait: + +#ifdef ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK + aux_work = erts_smp_atomic_read(&ssi->aux_work); + tse_blockable_aux_work: + aux_work = blockable_aux_work(esdp, ssi, aux_work); #endif + erts_smp_activity_begin(ERTS_ACTIVITY_WAIT, NULL, NULL, NULL); - sched_waiting_sys(no, rq); + while (1) { -#if ERTS_SCHED_SLEEP_SPINCOUNT != 0 - erts_smp_atomic_inc(&rq->spin_waiter); - erts_smp_runq_unlock(rq); +#ifdef ERTS_SCHED_NEED_NONBLOCKABLE_AUX_WORK +#ifndef ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK + aux_work = erts_smp_atomic_read(&ssi->aux_work); +#endif + nonblockable_aux_work(esdp, ssi, aux_work); +#endif - erl_sys_schedule(1); /* Might give us something to do */ + flgs = sched_spin_wait(ssi, spincount); + if (flgs & ERTS_SSI_FLG_SLEEPING) { + ASSERT(flgs & ERTS_SSI_FLG_WAITING); + flgs = sched_set_sleeptype(ssi, ERTS_SSI_FLG_TSE_SLEEPING); + if (flgs & ERTS_SSI_FLG_SLEEPING) { + int res; + ASSERT(flgs & ERTS_SSI_FLG_TSE_SLEEPING); + ASSERT(flgs & ERTS_SSI_FLG_WAITING); + do { + res = erts_tse_wait(ssi->event); + } while (res == EINTR); + } + } - dt = do_time_read_and_reset(); - if (dt) bump_timer(dt); + if (!(flgs & ERTS_SSI_FLG_WAITING)) { + ASSERT(!(flgs & ERTS_SSI_FLG_SLEEPING)); + break; + } + + flgs = sched_prep_cont_spin_wait(ssi); + spincount = ERTS_SCHED_TSE_SLEEP_SPINCOUNT; + + if (!(flgs & ERTS_SSI_FLG_WAITING)) { + ASSERT(!(flgs & ERTS_SSI_FLG_SLEEPING)); + break; + } + +#ifdef ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK + aux_work = erts_smp_atomic_read(&ssi->aux_work); + if (aux_work & ERTS_SSI_BLOCKABLE_AUX_WORK_MASK) { + erts_smp_activity_end(ERTS_ACTIVITY_WAIT, NULL, NULL, NULL); + goto tse_blockable_aux_work; + } +#endif - while (spincount-- > 0) { - val = erts_smp_atomic_read(&rq->spin_wake); - ASSERT(val >= 0); - if (val != 0) { - erts_smp_runq_lock(rq); - val = erts_smp_atomic_read(&rq->spin_wake); - ASSERT(val >= 0); - if (val != 0) - goto woken; - if (spincount == 0) - goto sleep; - erts_smp_runq_unlock(rq); } - } - erts_smp_runq_lock(rq); - val = erts_smp_atomic_read(&rq->spin_wake); - ASSERT(val >= 0); - if (val != 0) { - woken: - erts_smp_atomic_dec(&rq->spin_wake); - ASSERT(erts_smp_atomic_read(&rq->spin_wake) >= 0); - erts_smp_atomic_dec(&rq->spin_waiter); - ASSERT(erts_smp_atomic_read(&rq->spin_waiter) >= 0); + erts_smp_activity_end(ERTS_ACTIVITY_WAIT, NULL, NULL, NULL); + + if (flgs & ~ERTS_SSI_FLG_SUSPENDED) + erts_smp_atomic_band(&ssi->flags, ERTS_SSI_FLG_SUSPENDED); + + erts_smp_runq_lock(rq); + sched_active(esdp->no, rq); + } else { - sleep: - erts_smp_atomic_dec(&rq->spin_waiter); - ASSERT(erts_smp_atomic_read(&rq->spin_waiter) >= 0); + long dt; + + erts_smp_atomic_set(&function_calls, 0); + *fcalls = 0; + + sched_waiting_sys(esdp->no, rq); + + erts_smp_runq_unlock(rq); + + spincount = ERTS_SCHED_SYS_SLEEP_SPINCOUNT; + + while (spincount-- > 0) { + + sys_poll_aux_work: + + erl_sys_schedule(1); /* Might give us something to do */ + + dt = do_time_read_and_reset(); + if (dt) bump_timer(dt); + + sys_aux_work: + +#ifdef ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK + aux_work = erts_smp_atomic_read(&ssi->aux_work); + aux_work = blockable_aux_work(esdp, ssi, aux_work); +#endif +#ifdef ERTS_SCHED_NEED_NONBLOCKABLE_AUX_WORK +#ifndef ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK + aux_work = erts_smp_atomic_read(&ssi->aux_work); +#endif + nonblockable_aux_work(esdp, ssi, aux_work); +#endif + + flgs = erts_smp_atomic_read(&ssi->flags); + if (!(flgs & ERTS_SSI_FLG_WAITING)) { + ASSERT(!(flgs & ERTS_SSI_FLG_SLEEPING)); + goto sys_woken; + } + if (!(flgs & ERTS_SSI_FLG_SLEEPING)) { + flgs = sched_prep_cont_spin_wait(ssi); + if (!(flgs & ERTS_SSI_FLG_WAITING)) { + ASSERT(!(flgs & ERTS_SSI_FLG_SLEEPING)); + goto sys_woken; + } + } + + /* + * If we got new I/O tasks we aren't allowed to + * call erl_sys_schedule() until it is handled. + */ + if (erts_port_task_have_outstanding_io_tasks()) { + erts_smp_atomic_set(&doing_sys_schedule, 0); + /* + * Got to check that we still got I/O tasks; otherwise + * we have to continue checking for I/O... + */ + if (!prepare_for_sys_schedule()) { + spincount *= ERTS_SCHED_TSE_SLEEP_SPINCOUNT_FACT; + goto tse_wait; + } + } + } + + erts_smp_runq_lock(rq); + /* * If we got new I/O tasks we aren't allowed to * sleep in erl_sys_schedule(). */ - if (!erts_port_task_have_outstanding_io_tasks()) { -#endif + if (erts_port_task_have_outstanding_io_tasks()) { + erts_smp_atomic_set(&doing_sys_schedule, 0); + /* + * Got to check that we still got I/O tasks; otherwise + * we have to wait in erl_sys_schedule() after all... + */ + if (prepare_for_sys_schedule()) + goto do_sys_schedule; + + /* + * Not allowed to wait in erl_sys_schedule; + * do tse wait instead... + */ + sched_change_waiting_sys_to_waiting(esdp->no, rq); + erts_smp_runq_unlock(rq); + spincount = 0; + goto tse_wait; + } + else { + do_sys_schedule: erts_sys_schedule_interrupt(0); + flgs = sched_set_sleeptype(ssi, ERTS_SSI_FLG_POLL_SLEEPING); + if (!(flgs & ERTS_SSI_FLG_SLEEPING)) { + if (!(flgs & ERTS_SSI_FLG_WAITING)) + goto sys_locked_woken; + erts_smp_runq_unlock(rq); + flgs = sched_prep_cont_spin_wait(ssi); + if (!(flgs & ERTS_SSI_FLG_WAITING)) { + ASSERT(!(flgs & ERTS_SSI_FLG_SLEEPING)); + goto sys_woken; + } + ASSERT(!erts_port_task_have_outstanding_io_tasks()); + goto sys_poll_aux_work; + } + + ASSERT(flgs & ERTS_SSI_FLG_POLL_SLEEPING); + ASSERT(flgs & ERTS_SSI_FLG_WAITING); + erts_smp_runq_unlock(rq); erl_sys_schedule(0); @@ -797,134 +1145,103 @@ sched_sys_wait(Uint no, ErtsRunQueue *rq) dt = do_time_read_and_reset(); if (dt) bump_timer(dt); - erts_smp_runq_lock(rq); + flgs = sched_prep_cont_spin_wait(ssi); + if (flgs & ERTS_SSI_FLG_WAITING) + goto sys_aux_work; -#if ERTS_SCHED_SLEEP_SPINCOUNT != 0 + sys_woken: + erts_smp_runq_lock(rq); + sys_locked_woken: + erts_smp_atomic_set(&doing_sys_schedule, 0); + if (flgs & ~ERTS_SSI_FLG_SUSPENDED) + erts_smp_atomic_band(&ssi->flags, ERTS_SSI_FLG_SUSPENDED); + sched_active_sys(esdp->no, rq); } } -#endif - sched_active_sys(no, rq); -} - -static void -sched_cnd_wait(Uint no, ErtsRunQueue *rq) -{ -#if ERTS_SCHED_SLEEP_SPINCOUNT != 0 - int val; - int spincount = ERTS_SCHED_SLEEP_SPINCOUNT; ERTS_SMP_LC_ASSERT(erts_smp_lc_runq_is_locked(rq)); -#endif - - sched_waiting(no, rq); - erts_smp_activity_begin(ERTS_ACTIVITY_WAIT, - prepare_for_block, - resume_after_block, - (void *) rq); - -#if ERTS_SCHED_SLEEP_SPINCOUNT == 0 - erts_smp_cnd_wait(&rq->cnd, &rq->mtx); -#else - erts_smp_atomic_inc(&rq->spin_waiter); - erts_smp_mtx_unlock(&rq->mtx); - - while (spincount-- > 0) { - val = erts_smp_atomic_read(&rq->spin_wake); - ASSERT(val >= 0); - if (val != 0) { - erts_smp_mtx_lock(&rq->mtx); - val = erts_smp_atomic_read(&rq->spin_wake); - ASSERT(val >= 0); - if (val != 0) - goto woken; - if (spincount == 0) - goto sleep; - erts_smp_mtx_unlock(&rq->mtx); - } - } - - erts_smp_mtx_lock(&rq->mtx); - val = erts_smp_atomic_read(&rq->spin_wake); - ASSERT(val >= 0); - if (val == 0) { - sleep: - erts_smp_atomic_dec(&rq->spin_waiter); - ASSERT(erts_smp_atomic_read(&rq->spin_waiter) >= 0); - erts_smp_cnd_wait(&rq->cnd, &rq->mtx); - } - else { - woken: - erts_smp_atomic_dec(&rq->spin_wake); - ASSERT(erts_smp_atomic_read(&rq->spin_wake) >= 0); - erts_smp_atomic_dec(&rq->spin_waiter); - ASSERT(erts_smp_atomic_read(&rq->spin_waiter) >= 0); - } -#endif - - erts_smp_activity_end(ERTS_ACTIVITY_WAIT, - prepare_for_block, - resume_after_block, - (void *) rq); - - sched_active(no, rq); } -static void -wake_one_scheduler(void) -{ - ASSERT(erts_common_run_queue); - ERTS_SMP_LC_ASSERT(erts_smp_lc_runq_is_locked(erts_common_run_queue)); - if (erts_common_run_queue->waiting) { - if (!sched_spin_wake(erts_common_run_queue)) { - if (erts_common_run_queue->waiting == -1) /* One scheduler waiting - and doing so in - sys_schedule */ - erts_sys_schedule_interrupt(1); - else - erts_smp_cnd_signal(&erts_common_run_queue->cnd); - } +static ERTS_INLINE long +ssi_flags_set_wake(ErtsSchedulerSleepInfo *ssi) +{ + /* reset all flags but suspended */ + long oflgs; + long nflgs = 0; + long xflgs = ERTS_SSI_FLG_SLEEPING|ERTS_SSI_FLG_WAITING; + while (1) { + oflgs = erts_smp_atomic_cmpxchg(&ssi->flags, nflgs, xflgs); + if (oflgs == xflgs) + return oflgs; + nflgs = oflgs & ERTS_SSI_FLG_SUSPENDED; + xflgs = oflgs; } } static void -wake_scheduler(ErtsRunQueue *rq, int incq) +wake_scheduler(ErtsRunQueue *rq, int incq, int one) { - ASSERT(!erts_common_run_queue); - ASSERT(-1 <= rq->waiting && rq->waiting <= 1); - ERTS_SMP_LC_ASSERT(erts_smp_lc_runq_is_locked(rq)); - if (rq->waiting && !rq->woken) { - if (!sched_spin_wake(rq)) { - if (rq->waiting < 0) - erts_sys_schedule_interrupt(1); - else - erts_smp_cnd_signal(&rq->cnd); + int res; + ErtsSchedulerSleepInfo *ssi; + ErtsSchedulerSleepList *sl; + + /* + * The unlocked run queue is not strictly necessary + * from a thread safety or deadlock prevention + * perspective. It will, however, cost us performance + * if it is locked during wakup of another scheduler, + * so all code *should* handle this without having + * the lock on the run queue. + */ + ERTS_SMP_LC_ASSERT(!erts_smp_lc_runq_is_locked(rq)); + + sl = &rq->sleepers; + + erts_smp_spin_lock(&sl->lock); + ssi = sl->list; + if (!ssi) + erts_smp_spin_unlock(&sl->lock); + else if (one) { + long flgs; + if (ssi->prev) + ssi->prev->next = ssi->next; + else { + ASSERT(sl->list == ssi); + sl->list = ssi->next; } - rq->woken = 1; - if (incq) + if (ssi->next) + ssi->next->prev = ssi->prev; + + res = sl->list != NULL; + erts_smp_spin_unlock(&sl->lock); + + flgs = ssi_flags_set_wake(ssi); + erts_sched_finish_poke(ssi, flgs); + + if (incq && !erts_common_run_queue && (flgs & ERTS_SSI_FLG_WAITING)) non_empty_runq(rq); } + else { + sl->list = NULL; + erts_smp_spin_unlock(&sl->lock); + do { + ErtsSchedulerSleepInfo *wake_ssi = ssi; + ssi = ssi->next; + erts_sched_finish_poke(ssi, ssi_flags_set_wake(wake_ssi)); + } while (ssi); + } } static void wake_all_schedulers(void) { - if (erts_common_run_queue) { - erts_smp_runq_lock(erts_common_run_queue); - if (erts_common_run_queue->waiting) { - if (erts_common_run_queue->waiting < 0) - erts_sys_schedule_interrupt(1); - sched_spin_wake_all(erts_common_run_queue); - erts_smp_cnd_broadcast(&erts_common_run_queue->cnd); - } - erts_smp_runq_unlock(erts_common_run_queue); - } + if (erts_common_run_queue) + wake_scheduler(erts_common_run_queue, 0, 0); else { int ix; for (ix = 0; ix < erts_no_run_queues; ix++) { ErtsRunQueue *rq = ERTS_RUNQ_IX(ix); - erts_smp_runq_lock(rq); - wake_scheduler(rq, 0); - erts_smp_runq_unlock(rq); + wake_scheduler(rq, 0, 1); } } } @@ -939,14 +1256,14 @@ chk_wake_sched(ErtsRunQueue *crq, int ix, int activate) wrq = ERTS_RUNQ_IX(ix); iflgs = erts_smp_atomic_read(&wrq->info_flags); if (!(iflgs & (ERTS_RUNQ_IFLG_SUSPENDED|ERTS_RUNQ_IFLG_NONEMPTY))) { - erts_smp_xrunq_lock(crq, wrq); if (activate) { if (ix == erts_smp_atomic_cmpxchg(&balance_info.active_runqs, ix+1, ix)) { + erts_smp_xrunq_lock(crq, wrq); wrq->flags &= ~ERTS_RUNQ_FLG_INACTIVE; + erts_smp_xrunq_unlock(crq, wrq); } } - wake_scheduler(wrq, 0); - erts_smp_xrunq_unlock(crq, wrq); + wake_scheduler(wrq, 0, 1); return 1; } return 0; @@ -992,15 +1309,13 @@ static ERTS_INLINE void smp_notify_inc_runq(ErtsRunQueue *runq) { #ifdef ERTS_SMP - if (erts_common_run_queue) - wake_one_scheduler(); - else - wake_scheduler(runq, 1); + if (runq) + wake_scheduler(runq, 1, 1); #endif } void -erts_smp_notify_inc_runq__(ErtsRunQueue *runq) +erts_smp_notify_inc_runq(ErtsRunQueue *runq) { smp_notify_inc_runq(runq); } @@ -1146,20 +1461,23 @@ static void evacuate_run_queue(ErtsRunQueue *evac_rq, ErtsRunQueue *rq) { Port *prt; + int notify_to_rq = 0; int prio; int prt_locked = 0; int rq_locked = 0; int evac_rq_locked = 1; + ErtsMigrateResult mres; erts_smp_runq_lock(evac_rq); + erts_smp_atomic_bor(&evac_rq->scheduler->ssi->flags, ERTS_SSI_FLG_SUSPENDED); + evac_rq->flags &= ~ERTS_RUNQ_FLGS_IMMIGRATE_QMASK; evac_rq->flags |= (ERTS_RUNQ_FLGS_EMIGRATE_QMASK | ERTS_RUNQ_FLGS_EVACUATE_QMASK | ERTS_RUNQ_FLG_SUSPENDED); erts_smp_atomic_bor(&evac_rq->info_flags, ERTS_RUNQ_IFLG_SUSPENDED); - /* * Need to set up evacuation paths first since we * may release the run queue lock on evac_rq @@ -1187,9 +1505,11 @@ evacuate_run_queue(ErtsRunQueue *evac_rq, ErtsRunQueue *rq) /* Evacuate scheduled ports */ prt = evac_rq->ports.start; while (prt) { - (void) erts_port_migrate(prt, &prt_locked, + mres = erts_port_migrate(prt, &prt_locked, evac_rq, &evac_rq_locked, rq, &rq_locked); + if (mres == ERTS_MIGRATE_SUCCESS) + notify_to_rq = 1; if (prt_locked) erts_smp_port_unlock(prt); if (!evac_rq_locked) { @@ -1218,9 +1538,11 @@ evacuate_run_queue(ErtsRunQueue *evac_rq, ErtsRunQueue *rq) goto end_of_proc; } - (void) erts_proc_migrate(proc, &proc_locks, + mres = erts_proc_migrate(proc, &proc_locks, evac_rq, &evac_rq_locked, rq, &rq_locked); + if (mres == ERTS_MIGRATE_SUCCESS) + notify_to_rq = 1; if (proc_locks) erts_smp_proc_unlock(proc, proc_locks); if (!evac_rq_locked) { @@ -1252,10 +1574,13 @@ evacuate_run_queue(ErtsRunQueue *evac_rq, ErtsRunQueue *rq) if (rq_locked) erts_smp_runq_unlock(rq); - if (!evac_rq_locked) - erts_smp_runq_lock(evac_rq); - wake_scheduler(evac_rq, 0); - erts_smp_runq_unlock(evac_rq); + if (evac_rq_locked) + erts_smp_runq_unlock(evac_rq); + + if (notify_to_rq) + smp_notify_inc_runq(rq); + + wake_scheduler(evac_rq, 0, 1); } static int @@ -1483,31 +1808,6 @@ try_steal_task(ErtsRunQueue *rq) return res; } -#ifdef ERTS_SMP_SCHEDULERS_NEED_TO_CHECK_CHILDREN -void -erts_smp_notify_check_children_needed(void) -{ - int i; - for (i = 0; i < erts_no_schedulers; i++) { - erts_smp_runq_lock(ERTS_SCHEDULER_IX(i)->run_queue); - ERTS_SCHEDULER_IX(i)->check_children = 1; - if (!erts_common_run_queue) - wake_scheduler(ERTS_SCHEDULER_IX(i)->run_queue, 0); - erts_smp_runq_unlock(ERTS_SCHEDULER_IX(i)->run_queue); - } - if (ongoing_multi_scheduling_block()) { - /* Also blocked schedulers need to check children */ - erts_smp_mtx_lock(&schdlr_sspnd.mtx); - for (i = 0; i < erts_no_schedulers; i++) - ERTS_SCHEDULER_IX(i)->blocked_check_children = 1; - erts_smp_cnd_broadcast(&schdlr_sspnd.cnd); - erts_smp_mtx_unlock(&schdlr_sspnd.mtx); - } - if (erts_common_run_queue) - wake_all_schedulers(); -} -#endif - /* Run queue balancing */ typedef struct { @@ -2100,6 +2400,8 @@ erts_init_scheduling(int mrq, int no_schedulers, int no_schedulers_online) erts_smp_atomic_init(&no_empty_run_queues, 0); #endif + erts_no_run_queues = n; + for (ix = 0; ix < n; ix++) { int pix, rix; ErtsRunQueue *rq = ERTS_RUNQ_IX(ix); @@ -2114,8 +2416,10 @@ erts_init_scheduling(int mrq, int no_schedulers, int no_schedulers_online) erts_smp_mtx_init_x(&rq->mtx, "run_queue", make_small(ix + 1)); erts_smp_cnd_init(&rq->cnd); - erts_smp_atomic_init(&rq->spin_waiter, 0); - erts_smp_atomic_init(&rq->spin_wake, 0); +#ifdef ERTS_SMP + erts_smp_spinlock_init(&rq->sleepers.lock, "run_queue_sleep_list"); + rq->sleepers.list = NULL; +#endif rq->waiting = 0; rq->woken = 0; @@ -2166,7 +2470,6 @@ erts_init_scheduling(int mrq, int no_schedulers, int no_schedulers_online) } erts_common_run_queue = !mrq ? ERTS_RUNQ_IX(0) : NULL; - erts_no_run_queues = n; #ifdef ERTS_SMP @@ -2181,9 +2484,34 @@ erts_init_scheduling(int mrq, int no_schedulers, int no_schedulers_online) #endif + n = (int) no_schedulers; + erts_no_schedulers = n; + +#ifdef ERTS_SMP + /* Create and initialize scheduler sleep info */ + + aligned_sched_sleep_info = erts_alloc(ERTS_ALC_T_SCHDLR_SLP_INFO, + (sizeof(ErtsAlignedSchedulerSleepInfo) + *(n+1))); + if ((((Uint) aligned_sched_sleep_info) & ERTS_CACHE_LINE_MASK) == 0) + aligned_sched_sleep_info = ((ErtsAlignedSchedulerSleepInfo *) + ((((Uint) aligned_sched_sleep_info) + & ~ERTS_CACHE_LINE_MASK) + + ERTS_CACHE_LINE_SIZE)); + for (ix = 0; ix < n; ix++) { + ErtsSchedulerSleepInfo *ssi = ERTS_SCHED_SLEEP_INFO_IX(ix); +#if 0 /* no need to initialize these... */ + ssi->next = NULL; + ssi->prev = NULL; +#endif + erts_smp_atomic_init(&ssi->flags, 0); + ssi->event = NULL; /* initialized in sched_thread_func */ + erts_smp_atomic_init(&ssi->aux_work, 0); + } +#endif + /* Create and initialize scheduler specific data */ - n = (int) no_schedulers; erts_aligned_scheduler_data = erts_alloc(ERTS_ALC_T_SCHDLR_DATA, (sizeof(ErtsAlignedSchedulerData) *(n+1))); @@ -2200,6 +2528,7 @@ erts_init_scheduling(int mrq, int no_schedulers, int no_schedulers_online) #ifdef ERTS_SMP erts_bits_init_state(&esdp->erl_bits_state); esdp->match_pseudo_process = NULL; + esdp->ssi = ERTS_SCHED_SLEEP_INFO_IX(ix); esdp->free_process = NULL; #if HALFWORD_HEAP /* Registers need to be heap allocated (correct memory range) for tracing to work */ @@ -2228,11 +2557,6 @@ erts_init_scheduling(int mrq, int no_schedulers, int no_schedulers_online) } #ifdef ERTS_SMP -#ifdef ERTS_SMP_SCHEDULERS_NEED_TO_CHECK_CHILDREN - esdp->check_children = 0; - esdp->blocked_check_children = 0; -#endif - erts_smp_atomic_init(&esdp->suspended, 0); erts_smp_atomic_init(&esdp->chk_cpu_bind, 0); #endif } @@ -2241,7 +2565,7 @@ erts_init_scheduling(int mrq, int no_schedulers, int no_schedulers_online) erts_smp_mtx_init(&schdlr_sspnd.mtx, "schdlr_sspnd"); erts_smp_cnd_init(&schdlr_sspnd.cnd); - schdlr_sspnd.changing = 0; + erts_smp_atomic_init(&schdlr_sspnd.changing, 0); schdlr_sspnd.online = no_schedulers_online; schdlr_sspnd.curr_online = no_schedulers; erts_smp_atomic_init(&schdlr_sspnd.msb.ongoing, 0); @@ -2264,7 +2588,8 @@ erts_init_scheduling(int mrq, int no_schedulers, int no_schedulers_online) if (no_schedulers_online < no_schedulers) { if (erts_common_run_queue) { for (ix = no_schedulers_online; ix < no_schedulers; ix++) - erts_smp_atomic_set(&(ERTS_SCHEDULER_IX(ix)->suspended), 1); + erts_smp_atomic_bor(&ERTS_SCHED_SLEEP_INFO_IX(ix)->flags, + ERTS_SSI_FLG_SUSPENDED); } else { for (ix = no_schedulers_online; ix < erts_no_run_queues; ix++) @@ -2275,7 +2600,8 @@ erts_init_scheduling(int mrq, int no_schedulers, int no_schedulers_online) schdlr_sspnd.wait_curr_online = no_schedulers_online; schdlr_sspnd.curr_online *= 2; /* Boot strapping... */ - schdlr_sspnd.changing = ERTS_SCHED_CHANGING_ONLINE; + ERTS_SCHDLR_SSPND_CHNG_SET((ERTS_SCHDLR_SSPND_CHNG_ONLN + | ERTS_SCHDLR_SSPND_CHNG_WAITER), 0); erts_smp_atomic_init(&doing_sys_schedule, 0); @@ -2423,13 +2749,115 @@ susp_sched_resume_block(void *unused) } static void +scheduler_ix_resume_wake(Uint ix) +{ + ErtsSchedulerSleepInfo *ssi = ERTS_SCHED_SLEEP_INFO_IX(ix); + long xflgs = (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_TSE_SLEEPING + | ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED); + long oflgs; + do { + oflgs = erts_smp_atomic_cmpxchg(&ssi->flags, 0, xflgs); + if (oflgs == xflgs) { + erts_sched_finish_poke(ssi, oflgs); + break; + } + xflgs = oflgs; + } while (oflgs & ERTS_SSI_FLG_SUSPENDED); +} + +static long +sched_prep_spin_suspended(ErtsSchedulerSleepInfo *ssi, long xpct) +{ + long oflgs; + long nflgs = (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED); + long xflgs = xpct; + + do { + oflgs = erts_smp_atomic_cmpxchg(&ssi->flags, nflgs, xflgs); + if (oflgs == xflgs) + return nflgs; + xflgs = oflgs; + } while (oflgs & ERTS_SSI_FLG_SUSPENDED); + + return oflgs; +} + +static long +sched_spin_suspended(ErtsSchedulerSleepInfo *ssi, int spincount) +{ + int until_yield = ERTS_SCHED_SPIN_UNTIL_YIELD; + int sc = spincount; + long flgs; + + do { + flgs = erts_smp_atomic_read(&ssi->flags); + if ((flgs & (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED)) + != (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED)) { + break; + } + ERTS_SPIN_BODY; + if (--until_yield == 0) { + until_yield = ERTS_SCHED_SPIN_UNTIL_YIELD; + erts_thr_yield(); + } + } while (--sc > 0); + return flgs; +} + +static long +sched_set_suspended_sleeptype(ErtsSchedulerSleepInfo *ssi) +{ + long oflgs; + long nflgs = (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_TSE_SLEEPING + | ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED); + long xflgs = (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED); + + erts_tse_reset(ssi->event); + + while (1) { + oflgs = erts_smp_atomic_cmpxchg(&ssi->flags, nflgs, xflgs); + if (oflgs == xflgs) + return nflgs; + if ((oflgs & (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED)) + != (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED)) { + return oflgs; + } + xflgs = oflgs; + } +} + +static void suspend_scheduler(ErtsSchedulerData *esdp) { + long flgs; + int changing; long no = (long) esdp->no; ErtsRunQueue *rq = esdp->run_queue; + ErtsSchedulerSleepInfo *ssi = esdp->ssi; long active_schedulers; int curr_online = 1; int wake = 0; + int reset_read_group = 0; +#if defined(ERTS_SCHED_NEED_NONBLOCKABLE_AUX_WORK) \ + || defined(ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK) + long aux_work; +#endif /* * Schedulers may be suspended in two different ways: @@ -2451,113 +2879,151 @@ suspend_scheduler(ErtsSchedulerData *esdp) if (scheduler2cpu_map[esdp->no].bound_id >= 0 && erts_unbind_from_cpu(erts_cpuinfo) == 0) { esdp->cpu_id = scheduler2cpu_map[esdp->no].bound_id = -1; + reset_read_group = 1; } erts_smp_rwmtx_rwunlock(&erts_cpu_bind_rwmtx); + if (reset_read_group) + erts_smp_rwmtx_set_reader_group(0); + + if (esdp->no <= erts_max_main_threads) + erts_thr_set_main_status(0, 0); + if (erts_system_profile_flags.scheduler) profile_scheduler(make_small(esdp->no), am_inactive); erts_smp_mtx_lock(&schdlr_sspnd.mtx); - active_schedulers = erts_smp_atomic_dectest(&schdlr_sspnd.active); - ASSERT(active_schedulers >= 1); - if (schdlr_sspnd.changing == ERTS_SCHED_CHANGING_MULTI_SCHED) { - if (active_schedulers == schdlr_sspnd.msb.wait_active) - wake = 1; - if (active_schedulers == 1) - schdlr_sspnd.changing = 0; - } - - while (1) { + flgs = sched_prep_spin_suspended(ssi, ERTS_SSI_FLG_SUSPENDED); + if (flgs & ERTS_SSI_FLG_SUSPENDED) { -#ifdef ERTS_SMP_SCHEDULERS_NEED_TO_CHECK_CHILDREN - int check_children; - erts_smp_runq_lock(esdp->run_queue); - check_children = esdp->check_children; - esdp->check_children = 0; - erts_smp_runq_unlock(esdp->run_queue); - if (check_children) { - erts_smp_mtx_unlock(&schdlr_sspnd.mtx); - erts_check_children(); - erts_smp_mtx_lock(&schdlr_sspnd.mtx); + active_schedulers = erts_smp_atomic_dectest(&schdlr_sspnd.active); + ASSERT(active_schedulers >= 1); + changing = erts_smp_atomic_read(&schdlr_sspnd.changing); + if (changing & ERTS_SCHDLR_SSPND_CHNG_MSB) { + if (active_schedulers == schdlr_sspnd.msb.wait_active) + wake = 1; + if (active_schedulers == 1) { + changing = erts_smp_atomic_band(&schdlr_sspnd.changing, + ~ERTS_SCHDLR_SSPND_CHNG_MSB); + changing &= ~ERTS_SCHDLR_SSPND_CHNG_MSB; + } } -#endif - if (schdlr_sspnd.changing == ERTS_SCHED_CHANGING_ONLINE) { - int changed = 0; - if (no > schdlr_sspnd.online && curr_online) { - schdlr_sspnd.curr_online--; - curr_online = 0; - changed = 1; + while (1) { + if (changing & ERTS_SCHDLR_SSPND_CHNG_ONLN) { + int changed = 0; + if (no > schdlr_sspnd.online && curr_online) { + schdlr_sspnd.curr_online--; + curr_online = 0; + changed = 1; + } + else if (no <= schdlr_sspnd.online && !curr_online) { + schdlr_sspnd.curr_online++; + curr_online = 1; + changed = 1; + } + if (changed + && schdlr_sspnd.curr_online == schdlr_sspnd.wait_curr_online) + wake = 1; + if (schdlr_sspnd.online == schdlr_sspnd.curr_online) { + changing = erts_smp_atomic_band(&schdlr_sspnd.changing, + ~ERTS_SCHDLR_SSPND_CHNG_ONLN); + changing &= ~ERTS_SCHDLR_SSPND_CHNG_ONLN; + } } - else if (no <= schdlr_sspnd.online && !curr_online) { - schdlr_sspnd.curr_online++; - curr_online = 1; - changed = 1; + + if (wake) { + erts_smp_cnd_signal(&schdlr_sspnd.cnd); + wake = 0; } - if (changed - && schdlr_sspnd.curr_online == schdlr_sspnd.wait_curr_online) - wake = 1; - if (schdlr_sspnd.online == schdlr_sspnd.curr_online) - schdlr_sspnd.changing = 0; - } - if (wake) { - erts_smp_cnd_broadcast(&schdlr_sspnd.cnd); - wake = 0; - } + flgs = erts_smp_atomic_read(&ssi->flags); + if (!(flgs & ERTS_SSI_FLG_SUSPENDED)) + break; + erts_smp_mtx_unlock(&schdlr_sspnd.mtx); - if (!(rq->flags & (ERTS_RUNQ_FLG_SHARED_RUNQ|ERTS_RUNQ_FLG_SUSPENDED))) - break; - if ((rq->flags & ERTS_RUNQ_FLG_SHARED_RUNQ) - && !erts_smp_atomic_read(&esdp->suspended)) - break; +#ifdef ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK + aux_work = erts_smp_atomic_read(&ssi->aux_work); + blockable_aux_work: + blockable_aux_work(esdp, ssi, aux_work); +#endif - erts_smp_activity_begin(ERTS_ACTIVITY_WAIT, - susp_sched_prep_block, - susp_sched_resume_block, - NULL); - while (1) { + erts_smp_activity_begin(ERTS_ACTIVITY_WAIT, NULL, NULL, NULL); + while (1) { + long flgs; +#ifdef ERTS_SCHED_NEED_NONBLOCKABLE_AUX_WORK +#ifndef ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK + aux_work = erts_smp_atomic_read(&ssi->aux_work); +#endif + nonblockable_aux_work(esdp, ssi, aux_work); +#endif -#ifdef ERTS_SMP_SCHEDULERS_NEED_TO_CHECK_CHILDREN - if (esdp->blocked_check_children) - break; + flgs = sched_spin_suspended(ssi, ERTS_SCHED_SUSPEND_SLEEP_SPINCOUNT); + if (flgs == (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED)) { + flgs = sched_set_suspended_sleeptype(ssi); + if (flgs == (ERTS_SSI_FLG_SLEEPING + | ERTS_SSI_FLG_TSE_SLEEPING + | ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED)) { + int res; + do { + res = erts_tse_wait(ssi->event); + } while (res == EINTR); + } + } + + flgs = sched_prep_spin_suspended(ssi, (ERTS_SSI_FLG_WAITING + | ERTS_SSI_FLG_SUSPENDED)); + if (!(flgs & ERTS_SSI_FLG_SUSPENDED)) + break; + changing = erts_smp_atomic_read(&schdlr_sspnd.changing); + if (changing & ~ERTS_SCHDLR_SSPND_CHNG_WAITER) + break; + + +#ifdef ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK + aux_work = erts_smp_atomic_read(&ssi->aux_work); + if (aux_work & ERTS_SSI_BLOCKABLE_AUX_WORK_MASK) { + erts_smp_activity_end(ERTS_ACTIVITY_WAIT, NULL, NULL, NULL); + goto blockable_aux_work; + } #endif - erts_smp_cnd_wait(&schdlr_sspnd.cnd, &schdlr_sspnd.mtx); + } + + erts_smp_activity_end(ERTS_ACTIVITY_WAIT, NULL, NULL, NULL); - if (schdlr_sspnd.changing == ERTS_SCHED_CHANGING_ONLINE) - break; + erts_smp_mtx_lock(&schdlr_sspnd.mtx); + changing = erts_smp_atomic_read(&schdlr_sspnd.changing); + } - if (!(rq->flags & (ERTS_RUNQ_FLG_SHARED_RUNQ - | ERTS_RUNQ_FLG_SUSPENDED))) - break; - if ((rq->flags & ERTS_RUNQ_FLG_SHARED_RUNQ) - && !erts_smp_atomic_read(&esdp->suspended)) - break; + active_schedulers = erts_smp_atomic_inctest(&schdlr_sspnd.active); + changing = erts_smp_atomic_read(&schdlr_sspnd.changing); + if ((changing & ERTS_SCHDLR_SSPND_CHNG_MSB) + && schdlr_sspnd.online == active_schedulers) { + erts_smp_atomic_band(&schdlr_sspnd.changing, + ~ERTS_SCHDLR_SSPND_CHNG_MSB); } -#ifdef ERTS_SMP_SCHEDULERS_NEED_TO_CHECK_CHILDREN - esdp->blocked_check_children = 0; -#endif + ASSERT(no <= schdlr_sspnd.online); + ASSERT(!erts_smp_atomic_read(&schdlr_sspnd.msb.ongoing)); - erts_smp_activity_end(ERTS_ACTIVITY_WAIT, - susp_sched_prep_block, - susp_sched_resume_block, - NULL); } - active_schedulers = erts_smp_atomic_inctest(&schdlr_sspnd.active); - if (schdlr_sspnd.changing == ERTS_SCHED_CHANGING_MULTI_SCHED - && schdlr_sspnd.online == active_schedulers) { - schdlr_sspnd.changing = 0; - } erts_smp_mtx_unlock(&schdlr_sspnd.mtx); + ASSERT(curr_online); + if (erts_system_profile_flags.scheduler) profile_scheduler(make_small(esdp->no), am_active); + if (esdp->no <= erts_max_main_threads) + erts_thr_set_main_status(1, (int) esdp->no); + erts_smp_runq_lock(esdp->run_queue); non_empty_runq(esdp->run_queue); @@ -2622,8 +3088,10 @@ erts_schedulers_state(Uint *total, int yield_allowed) { int res; + long changing; erts_smp_mtx_lock(&schdlr_sspnd.mtx); - if (yield_allowed && schdlr_sspnd.changing) + changing = erts_smp_atomic_read(&schdlr_sspnd.changing); + if (yield_allowed && (changing & ~ERTS_SCHDLR_SSPND_CHNG_WAITER)) res = ERTS_SCHDLR_SSPND_YIELD_RESTART; else { *active = *online = schdlr_sspnd.online; @@ -2643,6 +3111,7 @@ erts_set_schedulers_online(Process *p, Sint *old_no) { int ix, res, no, have_unlocked_plocks; + long changing; if (new_no < 1 || erts_no_schedulers < new_no) return ERTS_SCHDLR_SSPND_EINVAL; @@ -2652,7 +3121,8 @@ erts_set_schedulers_online(Process *p, have_unlocked_plocks = 0; no = (int) new_no; - if (schdlr_sspnd.changing) { + changing = erts_smp_atomic_read(&schdlr_sspnd.changing); + if (changing) { res = ERTS_SCHDLR_SSPND_YIELD_RESTART; } else { @@ -2661,17 +3131,19 @@ erts_set_schedulers_online(Process *p, res = ERTS_SCHDLR_SSPND_DONE; } else { - schdlr_sspnd.changing = ERTS_SCHED_CHANGING_ONLINE; + ERTS_SCHDLR_SSPND_CHNG_SET((ERTS_SCHDLR_SSPND_CHNG_ONLN + | ERTS_SCHDLR_SSPND_CHNG_WAITER), 0); schdlr_sspnd.online = no; if (no > online) { int ix; schdlr_sspnd.wait_curr_online = no; - if (ongoing_multi_scheduling_block()) - /* No schedulers to resume */; + if (ongoing_multi_scheduling_block()) { + for (ix = online; ix < no; ix++) + erts_sched_poke(ERTS_SCHED_SLEEP_INFO_IX(ix)); + } else if (erts_common_run_queue) { for (ix = online; ix < no; ix++) - erts_smp_atomic_set(&ERTS_SCHEDULER_IX(ix)->suspended, - 0); + scheduler_ix_resume_wake(ix); } else { if (plocks) { @@ -2685,6 +3157,7 @@ erts_set_schedulers_online(Process *p, erts_smp_runq_lock(rq); ERTS_RUNQ_RESET_SUSPEND_INFO(rq, 0x5); erts_smp_runq_unlock(rq); + scheduler_ix_resume_wake(ix); } /* * Spread evacuation paths among all online @@ -2699,7 +3172,6 @@ erts_set_schedulers_online(Process *p, erts_smp_mtx_unlock(&balance_info.update_mtx); erts_smp_mtx_lock(&schdlr_sspnd.mtx); } - erts_smp_cnd_broadcast(&schdlr_sspnd.cnd); res = ERTS_SCHDLR_SSPND_DONE; } else /* if (no < online) */ { @@ -2716,12 +3188,17 @@ erts_set_schedulers_online(Process *p, schdlr_sspnd.wait_curr_online = no+1; } - if (ongoing_multi_scheduling_block()) - erts_smp_cnd_broadcast(&schdlr_sspnd.cnd); - else if (erts_common_run_queue) { + if (ongoing_multi_scheduling_block()) { for (ix = no; ix < online; ix++) - erts_smp_atomic_set(&ERTS_SCHEDULER_IX(ix)->suspended, - 1); + erts_sched_poke(ERTS_SCHED_SLEEP_INFO_IX(ix)); + } + else if (erts_common_run_queue) { + for (ix = no; ix < online; ix++) { + ErtsSchedulerSleepInfo *ssi; + ssi = ERTS_SCHED_SLEEP_INFO_IX(ix); + erts_smp_atomic_bor(&ssi->flags, + ERTS_SSI_FLG_SUSPENDED); + } wake_all_schedulers(); } else { @@ -2748,7 +3225,10 @@ erts_set_schedulers_online(Process *p, erts_smp_atomic_set(&balance_info.used_runqs, no); erts_smp_mtx_unlock(&balance_info.update_mtx); erts_smp_mtx_lock(&schdlr_sspnd.mtx); - ERTS_FOREACH_OP_RUNQ(rq, wake_scheduler(rq, 0)); + for (ix = no; ix < online; ix++) { + ErtsRunQueue *rq = ERTS_RUNQ_IX(ix); + wake_scheduler(rq, 0, 1); + } } } @@ -2762,6 +3242,12 @@ erts_set_schedulers_online(Process *p, susp_sched_prep_block, susp_sched_resume_block, NULL); + ASSERT(res != ERTS_SCHDLR_SSPND_DONE + ? (ERTS_SCHDLR_SSPND_CHNG_WAITER + & erts_smp_atomic_read(&schdlr_sspnd.changing)) + : (ERTS_SCHDLR_SSPND_CHNG_WAITER + == erts_smp_atomic_read(&schdlr_sspnd.changing))); + erts_smp_atomic_band(&schdlr_sspnd.changing, ~ERTS_SCHDLR_SSPND_CHNG_WAITER); } } @@ -2776,11 +3262,12 @@ ErtsSchedSuspendResult erts_block_multi_scheduling(Process *p, ErtsProcLocks plocks, int on, int all) { int ix, res, have_unlocked_plocks = 0; + long changing; ErtsProcList *plp; erts_smp_mtx_lock(&schdlr_sspnd.mtx); - - if (schdlr_sspnd.changing) { + changing = erts_smp_atomic_read(&schdlr_sspnd.changing); + if (changing) { res = ERTS_SCHDLR_SSPND_YIELD_RESTART; /* Yield */ } else if (on) { /* ------ BLOCK ------ */ @@ -2794,19 +3281,22 @@ erts_block_multi_scheduling(Process *p, ErtsProcLocks plocks, int on, int all) res = ERTS_SCHDLR_SSPND_DONE_MSCHED_BLOCKED; } else { + int online = schdlr_sspnd.online; p->flags |= F_HAVE_BLCKD_MSCHED; if (plocks) { have_unlocked_plocks = 1; erts_smp_proc_unlock(p, plocks); } + ASSERT(0 == erts_smp_atomic_read(&schdlr_sspnd.msb.ongoing)); erts_smp_atomic_set(&schdlr_sspnd.msb.ongoing, 1); - if (schdlr_sspnd.online == 1) { + if (online == 1) { res = ERTS_SCHDLR_SSPND_DONE_MSCHED_BLOCKED; ASSERT(erts_smp_atomic_read(&schdlr_sspnd.active) == 1); ASSERT(p->scheduler_data->no == 1); } else { - schdlr_sspnd.changing = ERTS_SCHED_CHANGING_MULTI_SCHED; + ERTS_SCHDLR_SSPND_CHNG_SET((ERTS_SCHDLR_SSPND_CHNG_MSB + | ERTS_SCHDLR_SSPND_CHNG_WAITER), 0); if (p->scheduler_data->no == 1) { res = ERTS_SCHDLR_SSPND_DONE_MSCHED_BLOCKED; schdlr_sspnd.msb.wait_active = 1; @@ -2820,17 +3310,19 @@ erts_block_multi_scheduling(Process *p, ErtsProcLocks plocks, int on, int all) schdlr_sspnd.msb.wait_active = 2; } if (erts_common_run_queue) { - for (ix = 1; ix < schdlr_sspnd.online; ix++) - erts_smp_atomic_set(&ERTS_SCHEDULER_IX(ix)->suspended, 1); + for (ix = 1; ix < online; ix++) + erts_smp_atomic_bor(&ERTS_SCHED_SLEEP_INFO_IX(ix)->flags, + ERTS_SSI_FLG_SUSPENDED); wake_all_schedulers(); } else { erts_smp_mtx_unlock(&schdlr_sspnd.mtx); erts_smp_mtx_lock(&balance_info.update_mtx); erts_smp_atomic_set(&balance_info.used_runqs, 1); - for (ix = 0; ix < schdlr_sspnd.online; ix++) { + for (ix = 0; ix < online; ix++) { ErtsRunQueue *rq = ERTS_RUNQ_IX(ix); erts_smp_runq_lock(rq); + ASSERT(!(rq->flags & ERTS_RUNQ_FLG_SUSPENDED)); ERTS_RUNQ_RESET_MIGRATION_PATHS(rq, 0x7); erts_smp_runq_unlock(rq); } @@ -2855,6 +3347,13 @@ erts_block_multi_scheduling(Process *p, ErtsProcLocks plocks, int on, int all) susp_sched_prep_block, susp_sched_resume_block, NULL); + ASSERT(res != ERTS_SCHDLR_SSPND_DONE_MSCHED_BLOCKED + ? (ERTS_SCHDLR_SSPND_CHNG_WAITER + & erts_smp_atomic_read(&schdlr_sspnd.changing)) + : (ERTS_SCHDLR_SSPND_CHNG_WAITER + == erts_smp_atomic_read(&schdlr_sspnd.changing))); + erts_smp_atomic_band(&schdlr_sspnd.changing, + ~ERTS_SCHDLR_SSPND_CHNG_WAITER); } plp = proclist_create(p); plp->next = schdlr_sspnd.msb.procs; @@ -2898,7 +3397,7 @@ erts_block_multi_scheduling(Process *p, ErtsProcLocks plocks, int on, int all) if (schdlr_sspnd.msb.procs) res = ERTS_SCHDLR_SSPND_DONE_MSCHED_BLOCKED; else { - schdlr_sspnd.changing = ERTS_SCHED_CHANGING_MULTI_SCHED; + ERTS_SCHDLR_SSPND_CHNG_SET(ERTS_SCHDLR_SSPND_CHNG_MSB, 0); #ifdef DEBUG ERTS_FOREACH_RUNQ(rq, { @@ -2925,13 +3424,13 @@ erts_block_multi_scheduling(Process *p, ErtsProcLocks plocks, int on, int all) if (schdlr_sspnd.online == 1) { /* No schedulers to resume */ ASSERT(erts_smp_atomic_read(&schdlr_sspnd.active) == 1); - schdlr_sspnd.changing = 0; + ERTS_SCHDLR_SSPND_CHNG_SET(0, ERTS_SCHDLR_SSPND_CHNG_MSB); } else if (erts_common_run_queue) { for (ix = 1; ix < schdlr_sspnd.online; ix++) - erts_smp_atomic_set(&ERTS_SCHEDULER_IX(ix)->suspended, 0); + erts_smp_atomic_band(&ERTS_SCHED_SLEEP_INFO_IX(ix)->flags, + ~ERTS_SSI_FLG_SUSPENDED); wake_all_schedulers(); - erts_smp_cnd_broadcast(&schdlr_sspnd.cnd); } else { int online = schdlr_sspnd.online; @@ -2948,6 +3447,7 @@ erts_block_multi_scheduling(Process *p, ErtsProcLocks plocks, int on, int all) erts_smp_runq_lock(rq); ERTS_RUNQ_RESET_SUSPEND_INFO(rq, 0x4); erts_smp_runq_unlock(rq); + scheduler_ix_resume_wake(ix); } /* Spread evacuation paths among all online run queues */ @@ -2963,7 +3463,6 @@ erts_block_multi_scheduling(Process *p, ErtsProcLocks plocks, int on, int all) erts_smp_runq_unlock(ERTS_RUNQ_IX(0)); erts_smp_mtx_unlock(&balance_info.update_mtx); erts_smp_mtx_lock(&schdlr_sspnd.mtx); - erts_smp_cnd_broadcast(&schdlr_sspnd.cnd); } res = ERTS_SCHDLR_SSPND_DONE; } @@ -3035,18 +3534,34 @@ erts_multi_scheduling_blockers(Process *p) static void * sched_thread_func(void *vesdp) { +#ifdef ERTS_SMP + Uint no = ((ErtsSchedulerData *) vesdp)->no; +#endif #ifdef ERTS_ENABLE_LOCK_CHECK { char buf[31]; - Uint no = ((ErtsSchedulerData *) vesdp)->no; erts_snprintf(&buf[0], 31, "scheduler %bpu", no); erts_lc_set_thread_name(&buf[0]); } #endif - erts_alloc_reg_scheduler_id(((ErtsSchedulerData *) vesdp)->no); + erts_alloc_reg_scheduler_id(no); erts_tsd_set(sched_data_key, vesdp); #ifdef ERTS_SMP + + if (no <= erts_max_main_threads) { + erts_thr_set_main_status(1, (int) no); + if (erts_reader_groups) { + int rg = (int) no; + if (rg > erts_reader_groups) + rg = (((int) no) - 1) % erts_reader_groups + 1; + erts_smp_rwmtx_set_reader_group(rg); + } + } + erts_proc_lock_prepare_proc_lock_waiter(); + ERTS_SCHED_SLEEP_INFO_IX(no - 1)->event = erts_tse_fetch(); + + #endif erts_register_blockable_thread(); #ifdef HIPE @@ -3055,30 +3570,30 @@ sched_thread_func(void *vesdp) erts_thread_init_float(); erts_smp_mtx_lock(&schdlr_sspnd.mtx); - ASSERT(schdlr_sspnd.changing == ERTS_SCHED_CHANGING_ONLINE); + ASSERT(erts_smp_atomic_read(&schdlr_sspnd.changing) + & ERTS_SCHDLR_SSPND_CHNG_ONLN); - schdlr_sspnd.curr_online--; + if (--schdlr_sspnd.curr_online == schdlr_sspnd.wait_curr_online) { + erts_smp_atomic_band(&schdlr_sspnd.changing, + ~ERTS_SCHDLR_SSPND_CHNG_ONLN); + if (((ErtsSchedulerData *) vesdp)->no != 1) + erts_smp_cnd_signal(&schdlr_sspnd.cnd); + } - if (((ErtsSchedulerData *) vesdp)->no != 1) { - if (schdlr_sspnd.online == schdlr_sspnd.curr_online) { - schdlr_sspnd.changing = 0; - erts_smp_cnd_broadcast(&schdlr_sspnd.cnd); + if (((ErtsSchedulerData *) vesdp)->no == 1) { + if (schdlr_sspnd.curr_online != schdlr_sspnd.wait_curr_online) { + erts_smp_activity_begin(ERTS_ACTIVITY_WAIT, + susp_sched_prep_block, + susp_sched_resume_block, + NULL); + while (schdlr_sspnd.curr_online != schdlr_sspnd.wait_curr_online) + erts_smp_cnd_wait(&schdlr_sspnd.cnd, &schdlr_sspnd.mtx); + erts_smp_activity_end(ERTS_ACTIVITY_WAIT, + susp_sched_prep_block, + susp_sched_resume_block, + NULL); } - } - else if (schdlr_sspnd.curr_online == schdlr_sspnd.wait_curr_online) - schdlr_sspnd.changing = 0; - else { - erts_smp_activity_begin(ERTS_ACTIVITY_WAIT, - susp_sched_prep_block, - susp_sched_resume_block, - NULL); - while (schdlr_sspnd.curr_online != schdlr_sspnd.wait_curr_online) - erts_smp_cnd_wait(&schdlr_sspnd.cnd, &schdlr_sspnd.mtx); - erts_smp_activity_end(ERTS_ACTIVITY_WAIT, - susp_sched_prep_block, - susp_sched_resume_block, - NULL); - ASSERT(!schdlr_sspnd.changing); + ERTS_SCHDLR_SSPND_CHNG_SET(0, ERTS_SCHDLR_SSPND_CHNG_WAITER); } erts_smp_mtx_unlock(&schdlr_sspnd.mtx); @@ -3428,6 +3943,7 @@ processor_order_cmp(const void *vx, const void *vy) static void check_cpu_bind(ErtsSchedulerData *esdp) { + int rg = 0; int res; int cpu_id; erts_smp_runq_unlock(esdp->run_queue); @@ -3459,6 +3975,12 @@ check_cpu_bind(ErtsSchedulerData *esdp) erts_send_error_to_logger_nogl(dsbufp); } } + if (erts_reader_groups) { + if (esdp->cpu_id >= 0) + rg = reader_group_lookup(esdp->cpu_id); + else + rg = (((int) esdp->no) - 1) % erts_reader_groups + 1; + } erts_smp_runq_lock(esdp->run_queue); #ifdef ERTS_SMP if (erts_common_run_queue) @@ -3469,6 +3991,8 @@ check_cpu_bind(ErtsSchedulerData *esdp) #endif erts_smp_rwmtx_rwunlock(&erts_cpu_bind_rwmtx); + if (erts_reader_groups) + erts_smp_rwmtx_set_reader_group(rg); } static void @@ -3497,11 +4021,13 @@ signal_schedulers_bind_change(erts_cpu_topology_t *cpudata, int size) wake_all_schedulers(); } else { - ERTS_FOREACH_RUNQ(rq, - { + for (s_ix = 0; s_ix < erts_no_run_queues; s_ix++) { + ErtsRunQueue *rq = ERTS_RUNQ_IX(s_ix); + erts_smp_runq_lock(rq); rq->flags |= ERTS_RUNQ_FLG_CHK_CPU_BIND; - wake_scheduler(rq, 0); - }); + erts_smp_runq_unlock(rq); + wake_scheduler(rq, 0, 1); + }; } #else check_cpu_bind(erts_get_scheduler_data()); @@ -3541,6 +4067,490 @@ erts_init_scheduler_bind_type(char *how) return ERTS_INIT_SCHED_BIND_TYPE_SUCCESS; } +/* + * reader groups map + */ + +typedef struct { + int level[ERTS_TOPOLOGY_MAX_DEPTH+1]; +} erts_avail_cput; + +typedef struct { + int *map; + int size; + int groups; +} erts_reader_groups_map_test; + +typedef struct { + int id; + int sub_levels; + int reader_groups; +} erts_rg_count_t; + +typedef struct { + int logical; + int reader_group; +} erts_reader_groups_map_t; + +typedef struct { + erts_reader_groups_map_t *map; + int map_size; + int logical_processors; + int groups; +} erts_make_reader_groups_map_test; + +static int reader_groups_available_cpu_check; +static int reader_groups_logical_processors; +static int reader_groups_map_size; +static erts_reader_groups_map_t *reader_groups_map; + +#define ERTS_TOPOLOGY_RG ERTS_TOPOLOGY_MAX_DEPTH + +static void +make_reader_groups_map(erts_make_reader_groups_map_test *test); + +static Eterm +get_reader_groups_map(Process *c_p, + erts_reader_groups_map_t *map, + int map_size, + int logical_processors) +{ +#ifdef DEBUG + Eterm *endp; +#endif + Eterm res = NIL, tuple; + Eterm *hp; + int i; + + hp = HAlloc(c_p, logical_processors*(2+3)); +#ifdef DEBUG + endp = hp + logical_processors*(2+3); +#endif + for (i = map_size - 1; i >= 0; i--) { + if (map[i].logical >= 0) { + tuple = TUPLE2(hp, + make_small(map[i].logical), + make_small(map[i].reader_group)); + hp += 3; + res = CONS(hp, tuple, res); + hp += 2; + } + } + ASSERT(hp == endp); + return res; +} + +Eterm +erts_debug_reader_groups_map(Process *c_p, int groups) +{ + Eterm res; + erts_make_reader_groups_map_test test; + + test.groups = groups; + make_reader_groups_map(&test); + if (!test.map) + res = NIL; + else { + res = get_reader_groups_map(c_p, + test.map, + test.map_size, + test.logical_processors); + erts_free(ERTS_ALC_T_TMP, test.map); + } + return res; +} + + +Eterm +erts_get_reader_groups_map(Process *c_p) +{ + Eterm res; + erts_smp_rwmtx_rlock(&erts_cpu_bind_rwmtx); + res = get_reader_groups_map(c_p, + reader_groups_map, + reader_groups_map_size, + reader_groups_logical_processors); + erts_smp_rwmtx_runlock(&erts_cpu_bind_rwmtx); + return res; +} + +static void +make_available_cpu_topology(erts_avail_cput *no, + erts_avail_cput *avail, + erts_cpu_topology_t *cpudata, + int *size, + int test) +{ + int len = *size; + erts_cpu_topology_t last; + int a, i, j; + + no->level[ERTS_TOPOLOGY_NODE] = -1; + no->level[ERTS_TOPOLOGY_PROCESSOR] = -1; + no->level[ERTS_TOPOLOGY_PROCESSOR_NODE] = -1; + no->level[ERTS_TOPOLOGY_CORE] = -1; + no->level[ERTS_TOPOLOGY_THREAD] = -1; + no->level[ERTS_TOPOLOGY_LOGICAL] = -1; + + last.node = INT_MIN; + last.processor = INT_MIN; + last.processor_node = INT_MIN; + last.core = INT_MIN; + last.thread = INT_MIN; + last.logical = INT_MIN; + + a = 0; + + for (i = 0; i < len; i++) { + + if (!test && !erts_is_cpu_available(erts_cpuinfo, cpudata[i].logical)) + continue; + + if (last.node != cpudata[i].node) + goto node; + if (last.processor != cpudata[i].processor) + goto processor; + if (last.processor_node != cpudata[i].processor_node) + goto processor_node; + if (last.core != cpudata[i].core) + goto core; + ASSERT(last.thread != cpudata[i].thread); + goto thread; + + node: + no->level[ERTS_TOPOLOGY_NODE]++; + processor: + no->level[ERTS_TOPOLOGY_PROCESSOR]++; + processor_node: + no->level[ERTS_TOPOLOGY_PROCESSOR_NODE]++; + core: + no->level[ERTS_TOPOLOGY_CORE]++; + thread: + no->level[ERTS_TOPOLOGY_THREAD]++; + + no->level[ERTS_TOPOLOGY_LOGICAL]++; + + for (j = 0; j < ERTS_TOPOLOGY_LOGICAL; j++) + avail[a].level[j] = no->level[j]; + + avail[a].level[ERTS_TOPOLOGY_LOGICAL] = cpudata[i].logical; + avail[a].level[ERTS_TOPOLOGY_RG] = 0; + + ASSERT(last.logical != cpudata[a].logical); + + last = cpudata[i]; + a++; + } + + no->level[ERTS_TOPOLOGY_NODE]++; + no->level[ERTS_TOPOLOGY_PROCESSOR]++; + no->level[ERTS_TOPOLOGY_PROCESSOR_NODE]++; + no->level[ERTS_TOPOLOGY_CORE]++; + no->level[ERTS_TOPOLOGY_THREAD]++; + no->level[ERTS_TOPOLOGY_LOGICAL]++; + + *size = a; +} + +static int +reader_group_lookup(int logical) +{ + int start = logical % reader_groups_map_size; + int ix = start; + + do { + if (reader_groups_map[ix].logical == logical) { + ASSERT(reader_groups_map[ix].reader_group > 0); + return reader_groups_map[ix].reader_group; + } + ix++; + if (ix == reader_groups_map_size) + ix = 0; + } while (ix != start); + + erl_exit(ERTS_ABORT_EXIT, "Logical cpu id %d not found\n", logical); +} + +static void +reader_group_insert(erts_reader_groups_map_t *map, int map_size, + int logical, int reader_group) +{ + int start = logical % map_size; + int ix = start; + + do { + if (map[ix].logical < 0) { + map[ix].logical = logical; + map[ix].reader_group = reader_group; + return; + } + ix++; + if (ix == map_size) + ix = 0; + } while (ix != start); + + erl_exit(ERTS_ABORT_EXIT, "Reader groups map full\n"); +} + + +static int +sub_levels(erts_rg_count_t *rgc, int level, int aix, int avail_sz, erts_avail_cput *avail) +{ + int sub_level = level+1; + int last = -1; + rgc->sub_levels = 0; + + do { + if (last != avail[aix].level[sub_level]) { + rgc->sub_levels++; + last = avail[aix].level[sub_level]; + } + aix++; + } + while (aix < avail_sz && rgc->id == avail[aix].level[level]); + rgc->reader_groups = 0; + return aix; +} + +static int +write_reader_groups(int *rgp, erts_rg_count_t *rgcp, + int level, int a, + int avail_sz, erts_avail_cput *avail) +{ + int rg = *rgp; + int sub_level = level+1; + int sl_per_gr = rgcp->sub_levels / rgcp->reader_groups; + int xsl = rgcp->sub_levels % rgcp->reader_groups; + int sls = 0; + int last = -1; + int xsl_rg_lim = (rgcp->reader_groups - xsl) + rg + 1; + + ASSERT(level < 0 || avail[a].level[level] == rgcp->id) + + do { + if (last != avail[a].level[sub_level]) { + if (!sls) { + sls = sl_per_gr; + rg++; + if (rg >= xsl_rg_lim) + sls++; + } + last = avail[a].level[sub_level]; + sls--; + } + avail[a].level[ERTS_TOPOLOGY_RG] = rg; + a++; + } while (a < avail_sz && (level < 0 + || avail[a].level[level] == rgcp->id)); + + ASSERT(rgcp->reader_groups == rg - *rgp); + + *rgp = rg; + + return a; +} + +static int +rg_count_sub_levels_compare(const void *vx, const void *vy) +{ + erts_rg_count_t *x = (erts_rg_count_t *) vx; + erts_rg_count_t *y = (erts_rg_count_t *) vy; + if (x->sub_levels != y->sub_levels) + return y->sub_levels - x->sub_levels; + return x->id - y->id; +} + +static int +rg_count_id_compare(const void *vx, const void *vy) +{ + erts_rg_count_t *x = (erts_rg_count_t *) vx; + erts_rg_count_t *y = (erts_rg_count_t *) vy; + return x->id - y->id; +} + +static void +make_reader_groups_map(erts_make_reader_groups_map_test *test) +{ + int i, spread_level, avail_sz; + erts_avail_cput no, *avail; + erts_cpu_topology_t *cpudata; + erts_reader_groups_map_t *map; + int map_sz; + int groups = erts_reader_groups; + + if (test) { + test->map = NULL; + test->map_size = 0; + groups = test->groups; + } + + if (!groups) + return; + + if (!test) { + if (reader_groups_map) + erts_free(ERTS_ALC_T_RDR_GRPS_MAP, reader_groups_map); + + reader_groups_logical_processors = 0; + reader_groups_map_size = 0; + reader_groups_map = NULL; + } + + create_tmp_cpu_topology_copy(&cpudata, &avail_sz); + + if (!cpudata) + return; + + cpu_bind_order_sort(cpudata, + avail_sz, + ERTS_CPU_BIND_NO_SPREAD, + 1); + + avail = erts_alloc(ERTS_ALC_T_TMP, + sizeof(erts_avail_cput)*avail_sz); + + make_available_cpu_topology(&no, avail, cpudata, + &avail_sz, test != NULL); + + destroy_tmp_cpu_topology_copy(cpudata); + + map_sz = avail_sz*2+1; + + if (test) { + map = erts_alloc(ERTS_ALC_T_TMP, + (sizeof(erts_reader_groups_map_t) + * map_sz)); + test->map = map; + test->map_size = map_sz; + test->logical_processors = avail_sz; + } + else { + map = erts_alloc(ERTS_ALC_T_RDR_GRPS_MAP, + (sizeof(erts_reader_groups_map_t) + * map_sz)); + reader_groups_map = map; + reader_groups_logical_processors = avail_sz; + reader_groups_map_size = map_sz; + + } + + for (i = 0; i < map_sz; i++) { + map[i].logical = -1; + map[i].reader_group = 0; + } + + spread_level = ERTS_TOPOLOGY_CORE; + for (i = ERTS_TOPOLOGY_NODE; i < ERTS_TOPOLOGY_THREAD; i++) { + if (no.level[i] > groups) { + spread_level = i; + break; + } + } + + if (no.level[spread_level] <= groups) { + int a, rg, last = -1; + rg = 0; + ASSERT(spread_level == ERTS_TOPOLOGY_CORE); + for (a = 0; a < avail_sz; a++) { + if (last != avail[a].level[spread_level]) { + rg++; + last = avail[a].level[spread_level]; + } + reader_group_insert(map, + map_sz, + avail[a].level[ERTS_TOPOLOGY_LOGICAL], + rg); + } + } + else { /* groups < no.level[spread_level] */ + erts_rg_count_t *rg_count; + int a, rg, tl, toplevels; + + tl = spread_level-1; + + if (spread_level == ERTS_TOPOLOGY_NODE) + toplevels = 1; + else + toplevels = no.level[tl]; + + rg_count = erts_alloc(ERTS_ALC_T_TMP, + toplevels*sizeof(erts_rg_count_t)); + + if (toplevels == 1) { + rg_count[0].id = 0; + rg_count[0].sub_levels = no.level[spread_level]; + rg_count[0].reader_groups = groups; + } + else { + int rgs_per_tl, rgs; + rgs = groups; + rgs_per_tl = rgs / toplevels; + + a = 0; + for (i = 0; i < toplevels; i++) { + rg_count[i].id = avail[a].level[tl]; + a = sub_levels(&rg_count[i], tl, a, avail_sz, avail); + } + + qsort(rg_count, + toplevels, + sizeof(erts_rg_count_t), + rg_count_sub_levels_compare); + + for (i = 0; i < toplevels; i++) { + if (rg_count[i].sub_levels < rgs_per_tl) { + rg_count[i].reader_groups = rg_count[i].sub_levels; + rgs -= rg_count[i].sub_levels; + } + else { + rg_count[i].reader_groups = rgs_per_tl; + rgs -= rgs_per_tl; + } + } + + while (rgs > 0) { + for (i = 0; i < toplevels; i++) { + if (rg_count[i].sub_levels == rg_count[i].reader_groups) + break; + else { + rg_count[i].reader_groups++; + if (--rgs == 0) + break; + } + } + } + + qsort(rg_count, + toplevels, + sizeof(erts_rg_count_t), + rg_count_id_compare); + } + + a = i = rg = 0; + while (a < avail_sz) { + a = write_reader_groups(&rg, &rg_count[i], tl, + a, avail_sz, avail); + i++; + } + + ASSERT(groups == rg); + + for (a = 0; a < avail_sz; a++) + reader_group_insert(map, + map_sz, + avail[a].level[ERTS_TOPOLOGY_LOGICAL], + avail[a].level[ERTS_TOPOLOGY_RG]); + + erts_free(ERTS_ALC_T_TMP, rg_count); + } + + erts_free(ERTS_ALC_T_TMP, avail); +} + +/* + * CPU topology + */ + typedef struct { int *id; int used; @@ -4054,6 +5064,8 @@ erts_set_cpu_topology(Process *c_p, Eterm term) sizeof(erts_cpu_topology_t)*cpudata_size); } + make_reader_groups_map(NULL); + signal_schedulers_bind_change(cpudata, cpudata_size); done: @@ -4457,6 +5469,11 @@ early_cpu_bind_init(void) cpu_bind_order = ERTS_CPU_BIND_UNDEFINED; + reader_groups_available_cpu_check = 1; + reader_groups_logical_processors = 0; + reader_groups_map_size = 0; + reader_groups_map = NULL; + if (!erts_get_cpu_topology(erts_cpuinfo, system_cpudata) || ERTS_INIT_CPU_TOPOLOGY_OK != verify_topology(system_cpudata, system_cpudata_size)) { @@ -4492,6 +5509,8 @@ late_cpu_bind_init(void) : ERTS_CPU_BIND_NONE); } + make_reader_groups_map(NULL); + if (cpu_bind_order != ERTS_CPU_BIND_NONE) { erts_cpu_topology_t *cpudata; int cpudata_size; @@ -5357,7 +6376,7 @@ dequeue_process(ErtsRunQueue *runq, Process *p) } /* schedule a process */ -static ERTS_INLINE void +static ERTS_INLINE ErtsRunQueue * internal_add_to_runq(ErtsRunQueue *runq, Process *p) { Uint32 prev_status = p->status; @@ -5368,12 +6387,12 @@ internal_add_to_runq(ErtsRunQueue *runq, Process *p) ERTS_SMP_LC_ASSERT(erts_smp_lc_runq_is_locked(runq)); if (p->status_flags & ERTS_PROC_SFLG_INRUNQ) - return; + return NULL; else if (p->runq_flags & ERTS_PROC_RUNQ_FLG_RUNNING) { ASSERT(p->status != P_SUSPENDED); ERTS_DBG_CHK_PROCS_RUNQ_NOPROC(runq, p); p->status_flags |= ERTS_PROC_SFLG_PENDADD2SCHEDQ; - return; + return NULL; } ASSERT(!p->scheduler_data); #endif @@ -5412,20 +6431,23 @@ internal_add_to_runq(ErtsRunQueue *runq, Process *p) profile_runnable_proc(p, am_active); } - smp_notify_inc_runq(add_runq); - if (add_runq != runq) erts_smp_runq_unlock(add_runq); + + return add_runq; } void erts_add_to_runq(Process *p) { + ErtsRunQueue *notify_runq; ErtsRunQueue *runq = erts_get_runq_proc(p); erts_smp_runq_lock(runq); - internal_add_to_runq(runq, p); + notify_runq = internal_add_to_runq(runq, p); erts_smp_runq_unlock(runq); + smp_notify_inc_runq(notify_runq); + } /* Possibly remove a scheduled process we need to suspend */ @@ -5564,8 +6586,6 @@ erts_proc_migrate(Process *p, ErtsProcLocks *plcks, p->run_queue = to_rq; enqueue_process(to_rq, p); - smp_notify_inc_runq(to_rq); - return ERTS_MIGRATE_SUCCESS; } #endif /* ERTS_SMP */ @@ -5762,30 +6782,6 @@ erts_set_process_priority(Process *p, Eterm new_value) return old_value; } -#ifdef ERTS_SMP - -static ERTS_INLINE int -prepare_for_sys_schedule(void) -{ - while (!erts_port_task_have_outstanding_io_tasks() - && !erts_smp_atomic_xchg(&doing_sys_schedule, 1)) { - if (!erts_port_task_have_outstanding_io_tasks()) - return 1; - erts_smp_atomic_set(&doing_sys_schedule, 0); - } - return 0; -} - -#else - -static ERTS_INLINE int -prepare_for_sys_schedule(void) -{ - return !erts_port_task_have_outstanding_io_tasks(); -} - -#endif - /* note that P_RUNNING is only set so that we don't try to remove ** running processes from the schedule queue if they exit - a running ** process not being in the schedule queue!! @@ -5920,8 +6916,11 @@ Process *schedule(Process *p, int calls) p->status_flags &= ~ERTS_PROC_SFLG_RUNNING; if (p->status_flags & ERTS_PROC_SFLG_PENDADD2SCHEDQ) { + ErtsRunQueue *notify_runq; p->status_flags &= ~ERTS_PROC_SFLG_PENDADD2SCHEDQ; - internal_add_to_runq(rq, p); + notify_runq = internal_add_to_runq(rq, p); + if (notify_runq != rq) + smp_notify_inc_runq(notify_runq); } #endif @@ -5989,7 +6988,10 @@ Process *schedule(Process *p, int calls) | ERTS_RUNQ_FLG_CHK_CPU_BIND | ERTS_RUNQ_FLG_SUSPENDED)) { if ((rq->flags & ERTS_RUNQ_FLG_SUSPENDED) - || erts_smp_atomic_read(&esdp->suspended)) { + || (erts_smp_atomic_read(&esdp->ssi->flags) + & ERTS_SSI_FLG_SUSPENDED)) { + ASSERT(erts_smp_atomic_read(&esdp->ssi->flags) + & ERTS_SSI_FLG_SUSPENDED); suspend_scheduler(esdp); } if ((rq->flags & ERTS_RUNQ_FLG_CHK_CPU_BIND) @@ -5998,12 +7000,21 @@ Process *schedule(Process *p, int calls) } } -#ifdef ERTS_SMP_SCHEDULERS_NEED_TO_CHECK_CHILDREN - if (esdp->check_children) { - esdp->check_children = 0; - erts_smp_runq_unlock(rq); - erts_check_children(); - erts_smp_runq_lock(rq); +#if defined(ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK) \ + || defined(ERTS_SCHED_NEED_NONBLOCKABLE_AUX_WORK) + { + ErtsSchedulerSleepInfo *ssi = esdp->ssi; + long aux_work = erts_smp_atomic_read(&ssi->aux_work); + if (aux_work) { + erts_smp_runq_unlock(rq); +#ifdef ERTS_SCHED_NEED_BLOCKABLE_AUX_WORK + aux_work = blockable_aux_work(esdp, ssi, aux_work); +#endif +#ifdef ERTS_SCHED_NEED_NONBLOCKABLE_AUX_WORK + nonblockable_aux_work(esdp, ssi, aux_work); +#endif + erts_smp_runq_lock(rq); + } } #endif @@ -6035,7 +7046,10 @@ Process *schedule(Process *p, int calls) if (rq->flags & (ERTS_RUNQ_FLG_SHARED_RUNQ | ERTS_RUNQ_FLG_SUSPENDED)) { if ((rq->flags & ERTS_RUNQ_FLG_SUSPENDED) - || erts_smp_atomic_read(&esdp->suspended)) { + || (erts_smp_atomic_read(&esdp->ssi->flags) + & ERTS_SSI_FLG_SUSPENDED)) { + ASSERT(erts_smp_atomic_read(&esdp->ssi->flags) + & ERTS_SSI_FLG_SUSPENDED); non_empty_runq(rq); goto continue_check_activities_to_run; } @@ -6052,17 +7066,7 @@ Process *schedule(Process *p, int calls) } } - if (prepare_for_sys_schedule()) { - erts_smp_atomic_set(&function_calls, 0); - fcalls = 0; - sched_sys_wait(esdp->no, rq); - erts_smp_atomic_set(&doing_sys_schedule, 0); - } - else { - /* If all schedulers are waiting, one of them *should* - be waiting in erl_sys_schedule() */ - sched_cnd_wait(esdp->no, rq); - } + scheduler_wait(&fcalls, esdp, rq); non_empty_runq(rq); @@ -6124,7 +7128,7 @@ Process *schedule(Process *p, int calls) else { if (erts_common_run_queue) { if (erts_common_run_queue->waiting) - wake_one_scheduler(); + wake_scheduler(erts_common_run_queue, 0, 1); } else if (erts_smp_atomic_read(&no_empty_run_queues) != 0) { wake_scheduler_on_empty_runq(rq); @@ -6439,8 +7443,8 @@ erts_schedule_misc_op(void (*func)(void *), void *arg) else rq->misc.start = molp; rq->misc.end = molp; - smp_notify_inc_runq(rq); erts_smp_runq_unlock(rq); + smp_notify_inc_runq(rq); } static void @@ -6682,7 +7686,7 @@ erl_create_process(Process* parent, /* Parent of process (default group leader). Eterm args, /* Arguments for function (must be well-formed list). */ ErlSpawnOpts* so) /* Options for spawn. */ { - ErtsRunQueue *rq; + ErtsRunQueue *rq, *notify_runq; Process *p; Sint arity; /* Number of arguments. */ #ifndef HYBRID @@ -6995,10 +7999,12 @@ erl_create_process(Process* parent, /* Parent of process (default group leader). #endif p->status = P_WAITING; - internal_add_to_runq(rq, p); + notify_runq = internal_add_to_runq(rq, p); erts_smp_runq_unlock(rq); + smp_notify_inc_runq(notify_runq); + res = p->id; erts_smp_proc_unlock(p, ERTS_PROC_LOCKS_ALL); diff --git a/erts/emulator/beam/erl_process.h b/erts/emulator/beam/erl_process.h index 6ce74a917f..c3fef6d38e 100644 --- a/erts/emulator/beam/erl_process.h +++ b/erts/emulator/beam/erl_process.h @@ -89,6 +89,7 @@ extern int erts_sched_thread_suggested_stack_size; #define ERTS_SCHED_THREAD_MAX_STACK_SIZE 8192 /* Kilo words */ #ifdef ERTS_SMP +extern Uint erts_max_main_threads; #include "erl_bits.h" #endif @@ -219,6 +220,51 @@ typedef enum { 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 @@ -271,8 +317,9 @@ struct ErtsRunQueue_ { erts_smp_mtx_t mtx; erts_smp_cnd_t cnd; - erts_smp_atomic_t spin_waiter; - erts_smp_atomic_t spin_wake; +#ifdef ERTS_SMP + ErtsSchedulerSleepList sleepers; +#endif ErtsSchedulerData *scheduler; int waiting; /* < 0 in sys schedule; > 0 on cnd variable */ @@ -353,6 +400,7 @@ struct ErtsSchedulerData_ { 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 @@ -374,11 +422,6 @@ struct ErtsSchedulerData_ { #ifdef ERTS_SMP /* NOTE: These fields are modified under held mutexes by other threads */ -#ifdef ERTS_SMP_SCHEDULERS_NEED_TO_CHECK_CHILDREN - int check_children; /* run queue mutex */ - int blocked_check_children; /* schdlr_sspnd mutex */ -#endif - erts_smp_atomic_t suspended; /* Only used when common run queue */ erts_smp_atomic_t chk_cpu_bind; /* Only used when common run queue */ #endif }; @@ -1085,6 +1128,9 @@ void erts_handle_pending_exit(Process *, ErtsProcLocks); 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); @@ -1509,29 +1555,30 @@ extern int erts_disable_proc_not_running_opt; #define ERTS_MIN_PROCESSES 16 #endif -#ifdef ERTS_INCLUDE_SCHEDULER_INTERNALS -ERTS_GLB_INLINE void erts_smp_notify_inc_runq(ErtsRunQueue *runq); -void erts_smp_notify_inc_runq__(ErtsRunQueue *runq); -#endif /* ERTS_INCLUDE_SCHEDULER_INTERNALS */ +void erts_smp_notify_inc_runq(ErtsRunQueue *runq); -#if ERTS_GLB_INLINE_INCL_FUNC_DEF +#ifdef ERTS_SMP +void erts_sched_finish_poke(ErtsSchedulerSleepInfo *, long); +ERTS_GLB_INLINE void erts_sched_poke(ErtsSchedulerSleepInfo *ssi); -#ifdef ERTS_INCLUDE_SCHEDULER_INTERNALS +#if ERTS_GLB_INLINE_INCL_FUNC_DEF ERTS_GLB_INLINE void -erts_smp_notify_inc_runq(ErtsRunQueue *runq) +erts_sched_poke(ErtsSchedulerSleepInfo *ssi) { -#ifdef ERTS_SMP - ERTS_SMP_LC_ASSERT(erts_smp_lc_runq_is_locked(runq)); - if (runq->waiting) - erts_smp_notify_inc_runq__(runq); -#endif + 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 /* ERTS_INCLUDE_SCHEDULER_INTERNALS */ - #endif /* #if ERTS_GLB_INLINE_INCL_FUNC_DEF */ +#endif /* #ifdef ERTS_SMP */ + #include "erl_process_lock.h" #undef ERTS_INCLUDE_SCHEDULER_INTERNALS diff --git a/erts/emulator/beam/erl_process_lock.c b/erts/emulator/beam/erl_process_lock.c index 52440fb635..a4d12139e9 100644 --- a/erts/emulator/beam/erl_process_lock.c +++ b/erts/emulator/beam/erl_process_lock.c @@ -1,19 +1,19 @@ /* * %CopyrightBegin% - * - * Copyright Ericsson AB 2007-2009. All Rights Reserved. - * + * + * Copyright Ericsson AB 2007-2010. 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% */ @@ -71,9 +71,12 @@ const Process erts_proc_lock_busy; #ifdef ERTS_SMP -/*#define ERTS_PROC_LOCK_SPIN_ON_GATE*/ -#define ERTS_PROC_LOCK_SPIN_COUNT_MAX 16000 +#define ERTS_PROC_LOCK_SPIN_COUNT_MAX 2000 +#define ERTS_PROC_LOCK_SPIN_COUNT_SCHED_INC 32 #define ERTS_PROC_LOCK_SPIN_COUNT_BASE 1000 +#define ERTS_PROC_LOCK_AUX_SPIN_COUNT 50 + +#define ERTS_PROC_LOCK_SPIN_UNTIL_YIELD 25 #ifdef ERTS_PROC_LOCK_DEBUG #define ERTS_PROC_LOCK_HARD_DEBUG @@ -83,32 +86,19 @@ const Process erts_proc_lock_busy; static void check_queue(erts_proc_lock_t *lck); #endif - -typedef struct erts_proc_lock_waiter_t_ erts_proc_lock_waiter_t; -struct erts_proc_lock_waiter_t_ { - erts_proc_lock_waiter_t *next; - erts_proc_lock_waiter_t *prev; - ErtsProcLocks wait_locks; - erts_smp_gate_t gate; - erts_proc_lock_queues_t *queues; -}; +#if SIZEOF_INT < 4 +#error "The size of the 'uflgs' field of the erts_tse_t type is too small" +#endif struct erts_proc_lock_queues_t_ { erts_proc_lock_queues_t *next; - erts_proc_lock_waiter_t *queue[ERTS_PROC_LOCK_MAX_BIT+1]; -}; - -struct erts_proc_lock_thr_spec_data_t_ { - erts_proc_lock_queues_t *qs; - erts_proc_lock_waiter_t *wtr; + erts_tse_t *queue[ERTS_PROC_LOCK_MAX_BIT+1]; }; static erts_proc_lock_queues_t zeroqs = {0}; -static erts_smp_spinlock_t wtr_lock; -static erts_proc_lock_waiter_t *waiter_free_list; +static erts_smp_spinlock_t qs_lock; static erts_proc_lock_queues_t *queue_free_list; -static erts_tsd_key_t waiter_key; #ifdef ERTS_ENABLE_LOCK_CHECK static struct { @@ -122,35 +112,26 @@ static struct { erts_pix_lock_t erts_pix_locks[ERTS_NO_OF_PIX_LOCKS]; static int proc_lock_spin_count; -static int proc_lock_trans_spin_cost; +static int aux_thr_proc_lock_spin_count; -static void cleanup_waiter(void); +static void cleanup_tse(void); void erts_init_proc_lock(void) { int i; int cpus; - erts_smp_spinlock_init(&wtr_lock, "proc_lck_wtr_alloc"); + erts_smp_spinlock_init(&qs_lock, "proc_lck_qs_alloc"); for (i = 0; i < ERTS_NO_OF_PIX_LOCKS; i++) { -#if ERTS_PROC_LOCK_MUTEX_IMPL -#ifdef ERTS_ENABLE_LOCK_COUNT - erts_smp_mtx_init_x(&erts_pix_locks[i].u.mtx, "pix_lock", make_small(i)); -#else - erts_smp_mtx_init(&erts_pix_locks[i].u.mtx, "pix_lock"); -#endif -#else #ifdef ERTS_ENABLE_LOCK_COUNT - erts_smp_spinlock_init_x(&erts_pix_locks[i].u.spnlck, "pix_lock", make_small(i)); + erts_smp_spinlock_init_x(&erts_pix_locks[i].u.spnlck, + "pix_lock", make_small(i)); #else erts_smp_spinlock_init(&erts_pix_locks[i].u.spnlck, "pix_lock"); #endif -#endif } - waiter_free_list = NULL; queue_free_list = NULL; - erts_tsd_key_create(&waiter_key); - erts_thr_install_exit_handler(cleanup_waiter); + erts_thr_install_exit_handler(cleanup_tse); #ifdef ERTS_ENABLE_LOCK_CHECK lc_id.proc_lock_main = erts_lc_get_lock_order_id("proc_main"); lc_id.proc_lock_link = erts_lc_get_lock_order_id("proc_link"); @@ -158,86 +139,106 @@ erts_init_proc_lock(void) lc_id.proc_lock_status = erts_lc_get_lock_order_id("proc_status"); #endif cpus = erts_get_cpu_configured(erts_cpuinfo); - if (cpus > 1) - proc_lock_spin_count = (ERTS_PROC_LOCK_SPIN_COUNT_BASE - * ((int) erts_no_schedulers)); - else if (cpus == 1) - proc_lock_spin_count = 0; - else /* No of cpus unknown. Assume multi proc, but be conservative. */ + if (cpus > 1) { proc_lock_spin_count = ERTS_PROC_LOCK_SPIN_COUNT_BASE; - if (proc_lock_spin_count > ERTS_PROC_LOCK_SPIN_COUNT_MAX) - proc_lock_spin_count = ERTS_PROC_LOCK_SPIN_COUNT_MAX; - proc_lock_trans_spin_cost = proc_lock_spin_count/20; -} - -static ERTS_INLINE erts_proc_lock_waiter_t * -alloc_wtr(void) -{ - erts_proc_lock_waiter_t *wtr; - erts_smp_spin_lock(&wtr_lock); - wtr = waiter_free_list; - if (wtr) { - waiter_free_list = wtr->next; - ERTS_LC_ASSERT(queue_free_list); - wtr->queues = queue_free_list; - queue_free_list = wtr->queues->next; - erts_smp_spin_unlock(&wtr_lock); + proc_lock_spin_count += (ERTS_PROC_LOCK_SPIN_COUNT_SCHED_INC + * ((int) erts_no_schedulers)); + aux_thr_proc_lock_spin_count = ERTS_PROC_LOCK_AUX_SPIN_COUNT; } - else { - erts_smp_spin_unlock(&wtr_lock); - wtr = erts_alloc(ERTS_ALC_T_PROC_LCK_WTR, - sizeof(erts_proc_lock_waiter_t)); - erts_smp_gate_init(&wtr->gate); - wtr->wait_locks = (ErtsProcLocks) 0; - wtr->queues = erts_alloc(ERTS_ALC_T_PROC_LCK_QS, - sizeof(erts_proc_lock_queues_t)); - sys_memcpy((void *) wtr->queues, - (void *) &zeroqs, - sizeof(erts_proc_lock_queues_t)); + else if (cpus == 1) { + proc_lock_spin_count = 0; + aux_thr_proc_lock_spin_count = 0; } - return wtr; + else { /* No of cpus unknown. Assume multi proc, but be conservative. */ + proc_lock_spin_count = ERTS_PROC_LOCK_SPIN_COUNT_BASE/2; + aux_thr_proc_lock_spin_count = ERTS_PROC_LOCK_AUX_SPIN_COUNT/2; + } + if (proc_lock_spin_count > ERTS_PROC_LOCK_SPIN_COUNT_MAX) + proc_lock_spin_count = ERTS_PROC_LOCK_SPIN_COUNT_MAX; } #ifdef ERTS_ENABLE_LOCK_CHECK static void -check_unused_waiter(erts_proc_lock_waiter_t *wtr) +check_unused_tse(erts_tse_t *wtr) { int i; - ERTS_LC_ASSERT(wtr->wait_locks == 0); + erts_proc_lock_queues_t *queues = wtr->udata; + ERTS_LC_ASSERT(wtr->uflgs == 0); for (i = 0; i <= ERTS_PROC_LOCK_MAX_BIT; i++) - ERTS_LC_ASSERT(!wtr->queues->queue[i]); + ERTS_LC_ASSERT(!queues->queue[i]); } -#define CHECK_UNUSED_WAITER(W) check_unused_waiter((W)) +#define CHECK_UNUSED_TSE(W) check_unused_tse((W)) #else -#define CHECK_UNUSED_WAITER(W) +#define CHECK_UNUSED_TSE(W) #endif +static ERTS_INLINE erts_tse_t * +tse_fetch(erts_pix_lock_t *pix_lock) +{ + erts_tse_t *tse = erts_tse_fetch(); + if (!tse->udata) { + erts_proc_lock_queues_t *qs; +#if ERTS_PROC_LOCK_SPINLOCK_IMPL && !ERTS_PROC_LOCK_ATOMIC_IMPL + if (pix_lock) + erts_pix_unlock(pix_lock); +#endif + erts_smp_spin_lock(&qs_lock); + qs = queue_free_list; + if (qs) { + queue_free_list = queue_free_list->next; + erts_smp_spin_unlock(&qs_lock); + } + else { + erts_smp_spin_unlock(&qs_lock); + qs = erts_alloc(ERTS_ALC_T_PROC_LCK_QS, + sizeof(erts_proc_lock_queues_t)); + sys_memcpy((void *) qs, + (void *) &zeroqs, + sizeof(erts_proc_lock_queues_t)); + } + tse->udata = qs; +#if ERTS_PROC_LOCK_SPINLOCK_IMPL && !ERTS_PROC_LOCK_ATOMIC_IMPL + if (pix_lock) + erts_pix_lock(pix_lock); +#endif + } + tse->uflgs = 0; + return tse; +} static ERTS_INLINE void -free_wtr(erts_proc_lock_waiter_t *wtr) +tse_return(erts_tse_t *tse, int force_free_q) { - CHECK_UNUSED_WAITER(wtr); - erts_smp_spin_lock(&wtr_lock); - wtr->next = waiter_free_list; - waiter_free_list = wtr; - wtr->queues->next = queue_free_list; - queue_free_list = wtr->queues; - erts_smp_spin_unlock(&wtr_lock); + CHECK_UNUSED_TSE(tse); + if (force_free_q || erts_tse_is_tmp(tse)) { + erts_proc_lock_queues_t *qs = tse->udata; + ASSERT(qs); + erts_smp_spin_lock(&qs_lock); + qs->next = queue_free_list; + queue_free_list = qs; + erts_smp_spin_unlock(&qs_lock); + tse->udata = NULL; + } + erts_tse_return(tse); } void erts_proc_lock_prepare_proc_lock_waiter(void) { - erts_tsd_set(waiter_key, (void *) alloc_wtr()); + tse_return(tse_fetch(NULL), 0); } static void -cleanup_waiter(void) +cleanup_tse(void) { - erts_proc_lock_waiter_t *wtr = erts_tsd_get(waiter_key); - if (wtr) - free_wtr(wtr); + erts_tse_t *tse = erts_tse_fetch(); + if (tse) { + if (tse->udata) + tse_return(tse, 1); + else + erts_tse_return(tse); + } } @@ -250,7 +251,7 @@ cleanup_waiter(void) static ERTS_INLINE void enqueue_waiter(erts_proc_lock_queues_t *qs, int ix, - erts_proc_lock_waiter_t *wtr) + erts_tse_t *wtr) { if (!qs->queue[ix]) { qs->queue[ix] = wtr; @@ -266,10 +267,10 @@ enqueue_waiter(erts_proc_lock_queues_t *qs, } } -static erts_proc_lock_waiter_t * +static erts_tse_t * dequeue_waiter(erts_proc_lock_queues_t *qs, int ix) { - erts_proc_lock_waiter_t *wtr = qs->queue[ix]; + erts_tse_t *wtr = qs->queue[ix]; ERTS_LC_ASSERT(qs->queue[ix]); if (wtr->next == wtr) { ERTS_LC_ASSERT(qs->queue[ix]->prev == wtr); @@ -295,10 +296,10 @@ dequeue_waiter(erts_proc_lock_queues_t *qs, int ix) * lock. */ static ERTS_INLINE void -try_aquire(erts_proc_lock_t *lck, erts_proc_lock_waiter_t *wtr) +try_aquire(erts_proc_lock_t *lck, erts_tse_t *wtr) { ErtsProcLocks got_locks = (ErtsProcLocks) 0; - ErtsProcLocks locks = wtr->wait_locks; + ErtsProcLocks locks = wtr->uflgs; int lock_no; ERTS_LC_ASSERT(lck->queues); @@ -334,7 +335,7 @@ try_aquire(erts_proc_lock_t *lck, erts_proc_lock_waiter_t *wtr) } } - wtr->wait_locks &= ~got_locks; + wtr->uflgs &= ~got_locks; } /* @@ -350,8 +351,8 @@ transfer_locks(Process *p, int unlock) { int transferred = 0; - erts_proc_lock_waiter_t *wake = NULL; - erts_proc_lock_waiter_t *wtr; + erts_tse_t *wake = NULL; + erts_tse_t *wtr; ErtsProcLocks unset_waiter = 0; ErtsProcLocks tlocks = trnsfr_lcks; int lock_no; @@ -377,11 +378,11 @@ transfer_locks(Process *p, ERTS_LC_ASSERT(wtr); if (!qs->queue[lock_no]) unset_waiter |= lock; - ERTS_LC_ASSERT(wtr->wait_locks & lock); - wtr->wait_locks &= ~lock; - if (wtr->wait_locks) + ERTS_LC_ASSERT(wtr->uflgs & lock); + wtr->uflgs &= ~lock; + if (wtr->uflgs) try_aquire(&p->lock, wtr); - if (!wtr->wait_locks) { + if (!wtr->uflgs) { /* * The other thread got all locks it needs; * need to wake it up. @@ -412,9 +413,10 @@ transfer_locks(Process *p, erts_pix_unlock(pix_lock); do { - erts_proc_lock_waiter_t *tmp = wake; + erts_tse_t *tmp = wake; wake = wake->next; - erts_smp_gate_let_through(&tmp->gate, 1); + erts_atomic_set(&tmp->uaflgs, 0); + erts_tse_set(tmp); } while (wake); if (!unlock) @@ -462,26 +464,16 @@ wait_for_locks(Process *p, ErtsProcLocks olflgs) { erts_pix_lock_t *pix_lock = pixlck ? pixlck : ERTS_PID2PIXLOCK(p->id); - int tsd; - erts_proc_lock_waiter_t *wtr; + erts_tse_t *wtr; + erts_proc_lock_queues_t *qs; /* Acquire a waiter object on which this thread can wait. */ - wtr = erts_tsd_get(waiter_key); - if (wtr) - tsd = 1; - else { -#if ERTS_PROC_LOCK_SPINLOCK_IMPL && !ERTS_PROC_LOCK_ATOMIC_IMPL - erts_pix_unlock(pix_lock); -#endif - wtr = alloc_wtr(); - tsd = 0; -#if ERTS_PROC_LOCK_SPINLOCK_IMPL && !ERTS_PROC_LOCK_ATOMIC_IMPL - erts_pix_lock(pix_lock); -#endif - } + wtr = tse_fetch(pix_lock); /* Record which locks this waiter needs. */ - wtr->wait_locks = need_locks; + wtr->uflgs = need_locks; + + ASSERT((wtr->uflgs & ~ERTS_PROC_LOCKS_ALL) == 0); #if ERTS_PROC_LOCK_ATOMIC_IMPL erts_pix_lock(pix_lock); @@ -489,14 +481,16 @@ wait_for_locks(Process *p, ERTS_LC_ASSERT(erts_lc_pix_lock_is_locked(pix_lock)); + qs = wtr->udata; + ASSERT(qs); /* Provide the process with waiter queues, if it doesn't have one. */ if (!p->lock.queues) { - wtr->queues->next = NULL; - p->lock.queues = wtr->queues; + qs->next = NULL; + p->lock.queues = qs; } else { - wtr->queues->next = p->lock.queues->next; - p->lock.queues->next = wtr->queues; + qs->next = p->lock.queues->next; + p->lock.queues->next = qs; } #ifdef ERTS_PROC_LOCK_HARD_DEBUG @@ -506,46 +500,59 @@ wait_for_locks(Process *p, /* Try to aquire locks one at a time in lock order and set wait flag */ try_aquire(&p->lock, wtr); + ASSERT((wtr->uflgs & ~ERTS_PROC_LOCKS_ALL) == 0); + #ifdef ERTS_PROC_LOCK_HARD_DEBUG check_queue(&p->lock); #endif - if (wtr->wait_locks) { /* We didn't get them all; need to wait... */ - /* Got to wait for locks... */ + if (wtr->uflgs) { + /* We didn't get them all; need to wait... */ + + ASSERT((wtr->uflgs & ~ERTS_PROC_LOCKS_ALL) == 0); + + erts_atomic_set(&wtr->uaflgs, 1); erts_pix_unlock(pix_lock); - /* - * Wait for needed locks. When we return all needed locks have - * have been acquired by other threads and transfered to us. - */ -#ifdef ERTS_PROC_LOCK_SPIN_ON_GATE - erts_smp_gate_swait(&wtr->gate, proc_lock_spin_count); -#else - erts_smp_gate_wait(&wtr->gate); -#endif + while (1) { + int res; + erts_tse_reset(wtr); + + if (erts_atomic_read(&wtr->uaflgs) == 0) + break; + + /* + * Wait for needed locks. When we are woken all needed locks have + * have been acquired by other threads and transfered to us. + * However, we need to be prepared for spurious wakeups. + */ + do { + res = erts_tse_wait(wtr); /* might return EINTR */ + } while (res != 0); + } erts_pix_lock(pix_lock); + + ASSERT(wtr->uflgs == 0); } /* Recover some queues to store in the waiter. */ ERTS_LC_ASSERT(p->lock.queues); if (p->lock.queues->next) { - wtr->queues = p->lock.queues->next; - p->lock.queues->next = wtr->queues->next; + qs = p->lock.queues->next; + p->lock.queues->next = qs->next; } else { - wtr->queues = p->lock.queues; + qs = p->lock.queues; p->lock.queues = NULL; } + wtr->udata = qs; erts_pix_unlock(pix_lock); ERTS_LC_ASSERT(locks == (ERTS_PROC_LOCK_FLGS_READ_(&p->lock) & locks)); - if (tsd) - CHECK_UNUSED_WAITER(wtr); - else - free_wtr(wtr); + tse_return(wtr, 0); } /* @@ -563,52 +570,57 @@ erts_proc_lock_failed(Process *p, ErtsProcLocks locks, ErtsProcLocks old_lflgs) { -#ifdef ERTS_PROC_LOCK_SPIN_ON_GATE - int spin_count = 0; -#else - int spin_count = proc_lock_spin_count; -#endif - + int until_yield = ERTS_PROC_LOCK_SPIN_UNTIL_YIELD; + int thr_spin_count; + int spin_count; ErtsProcLocks need_locks = locks; ErtsProcLocks olflgs = old_lflgs; - while (need_locks != 0) - { - ErtsProcLocks can_grab = in_order_locks(olflgs, need_locks); + if (erts_thr_get_main_status()) + thr_spin_count = proc_lock_spin_count; + else + thr_spin_count = aux_thr_proc_lock_spin_count; + + spin_count = thr_spin_count; + + while (need_locks != 0) { + ErtsProcLocks can_grab; + + can_grab = in_order_locks(olflgs, need_locks); - if (can_grab == 0) - { + if (can_grab == 0) { /* Someone already has the lowest-numbered lock we want. */ - if (spin_count-- <= 0) - { + if (spin_count-- <= 0) { /* Too many retries, give up and sleep for the lock. */ wait_for_locks(p, pixlck, locks, need_locks, olflgs); return; } + ERTS_SPIN_BODY; + + if (--until_yield == 0) { + until_yield = ERTS_PROC_LOCK_SPIN_UNTIL_YIELD; + erts_thr_yield(); + } + olflgs = ERTS_PROC_LOCK_FLGS_READ_(&p->lock); } - else - { + else { /* Try to grab all of the grabbable locks at once with cmpxchg. */ ErtsProcLocks grabbed = olflgs | can_grab; ErtsProcLocks nflgs = - ERTS_PROC_LOCK_FLGS_CMPXCHG_(&p->lock, grabbed, olflgs); + ERTS_PROC_LOCK_FLGS_CMPXCHG_ACQB_(&p->lock, grabbed, olflgs); - if (nflgs == olflgs) - { + if (nflgs == olflgs) { /* Success! We grabbed the 'can_grab' locks. */ olflgs = grabbed; need_locks &= ~can_grab; -#ifndef ERTS_PROC_LOCK_SPIN_ON_GATE /* Since we made progress, reset the spin count. */ - spin_count = proc_lock_spin_count; -#endif + spin_count = thr_spin_count; } - else - { + else { /* Compare-and-exchange failed, try again. */ olflgs = nflgs; } @@ -1407,7 +1419,7 @@ check_queue(erts_proc_lock_t *lck) wtr = (((ErtsProcLocks) 1) << lock_no) << ERTS_PROC_LOCK_WAITER_SHIFT; if (lflgs & wtr) { int n; - erts_proc_lock_waiter_t *wtr; + erts_tse_t *wtr; ERTS_LC_ASSERT(lck->queues && lck->queues->queue[lock_no]); wtr = lck->queues->queue[lock_no]; n = 0; diff --git a/erts/emulator/beam/erl_process_lock.h b/erts/emulator/beam/erl_process_lock.h index d71e5a0a6e..7cfc9893fa 100644 --- a/erts/emulator/beam/erl_process_lock.h +++ b/erts/emulator/beam/erl_process_lock.h @@ -255,11 +255,7 @@ void erts_proc_lc_unrequire_lock(Process *p, ErtsProcLocks locks); typedef struct { union { -#if ERTS_PROC_LOCK_MUTEX_IMPL - erts_smp_mtx_t mtx; -#else erts_smp_spinlock_t spnlck; -#endif char buf[64]; /* Try to get locks in different cache lines */ } u; } erts_pix_lock_t; @@ -277,9 +273,12 @@ typedef struct { ((ErtsProcLocks) erts_smp_atomic_band(&(L)->flags, (long) (MSK))) #define ERTS_PROC_LOCK_FLGS_BOR_(L, MSK) \ ((ErtsProcLocks) erts_smp_atomic_bor(&(L)->flags, (long) (MSK))) -#define ERTS_PROC_LOCK_FLGS_CMPXCHG_(L, NEW, EXPECTED) \ - ((ErtsProcLocks) erts_smp_atomic_cmpxchg(&(L)->flags, \ - (long) (NEW), (long) (EXPECTED))) +#define ERTS_PROC_LOCK_FLGS_CMPXCHG_ACQB_(L, NEW, EXPECTED) \ + ((ErtsProcLocks) erts_smp_atomic_cmpxchg_acqb(&(L)->flags, \ + (long) (NEW), (long) (EXPECTED))) +#define ERTS_PROC_LOCK_FLGS_CMPXCHG_RELB_(L, NEW, EXPECTED) \ + ((ErtsProcLocks) erts_smp_atomic_cmpxchg_relb(&(L)->flags, \ + (long) (NEW), (long) (EXPECTED))) #define ERTS_PROC_LOCK_FLGS_READ_(L) \ ((ErtsProcLocks) erts_smp_atomic_read(&(L)->flags)) @@ -289,6 +288,9 @@ ERTS_GLB_INLINE ErtsProcLocks erts_proc_lock_flags_band(erts_proc_lock_t *, ErtsProcLocks); ERTS_GLB_INLINE ErtsProcLocks erts_proc_lock_flags_bor(erts_proc_lock_t *, ErtsProcLocks); +ERTS_GLB_INLINE ErtsProcLocks erts_proc_lock_flags_cmpxchg(erts_proc_lock_t *, + ErtsProcLocks, + ErtsProcLocks); #if ERTS_GLB_INLINE_INCL_FUNC_DEF @@ -322,7 +324,9 @@ erts_proc_lock_flags_cmpxchg(erts_proc_lock_t *lck, ErtsProcLocks new, #define ERTS_PROC_LOCK_FLGS_BAND_(L, MSK) erts_proc_lock_flags_band((L), (MSK)) #define ERTS_PROC_LOCK_FLGS_BOR_(L, MSK) erts_proc_lock_flags_bor((L), (MSK)) -#define ERTS_PROC_LOCK_FLGS_CMPXCHG_(L, NEW, EXPECTED) \ +#define ERTS_PROC_LOCK_FLGS_CMPXCHG_ACQB_(L, NEW, EXPECTED) \ + erts_proc_lock_flags_cmpxchg((L), (NEW), (EXPECTED)) +#define ERTS_PROC_LOCK_FLGS_CMPXCHG_RELB_(L, NEW, EXPECTED) \ erts_proc_lock_flags_cmpxchg((L), (NEW), (EXPECTED)) #define ERTS_PROC_LOCK_FLGS_READ_(L) ((L)->flags) @@ -348,9 +352,9 @@ ERTS_GLB_INLINE ErtsProcLocks erts_smp_proc_raw_trylock__(Process *p, ErtsProcLocks locks); #ifdef ERTS_ENABLE_LOCK_COUNT ERTS_GLB_INLINE void erts_smp_proc_lock_x__(Process *, - erts_pix_lock_t *, - ErtsProcLocks, - char *file, unsigned int line); + erts_pix_lock_t *, + ErtsProcLocks, + char *file, unsigned int line); #else ERTS_GLB_INLINE void erts_smp_proc_lock__(Process *, erts_pix_lock_t *, @@ -372,30 +376,18 @@ ERTS_GLB_INLINE void erts_proc_lock_op_debug(Process *, ErtsProcLocks, int); ERTS_GLB_INLINE void erts_pix_lock(erts_pix_lock_t *pixlck) { ERTS_LC_ASSERT(pixlck); -#if ERTS_PROC_LOCK_MUTEX_IMPL - erts_smp_mtx_lock(&pixlck->u.mtx); -#else erts_smp_spin_lock(&pixlck->u.spnlck); -#endif } ERTS_GLB_INLINE void erts_pix_unlock(erts_pix_lock_t *pixlck) { ERTS_LC_ASSERT(pixlck); -#if ERTS_PROC_LOCK_MUTEX_IMPL - erts_smp_mtx_unlock(&pixlck->u.mtx); -#else erts_smp_spin_unlock(&pixlck->u.spnlck); -#endif } ERTS_GLB_INLINE int erts_lc_pix_lock_is_locked(erts_pix_lock_t *pixlck) { -#if ERTS_PROC_LOCK_MUTEX_IMPL - return erts_smp_lc_mtx_is_locked(&pixlck->u.mtx); -#else return erts_smp_lc_spinlock_is_locked(&pixlck->u.spnlck); -#endif } /* @@ -417,9 +409,9 @@ erts_smp_proc_raw_trylock__(Process *p, ErtsProcLocks locks) ErtsProcLocks expct_lflgs = 0; while (1) { - ErtsProcLocks lflgs = ERTS_PROC_LOCK_FLGS_CMPXCHG_(&p->lock, - expct_lflgs | locks, - expct_lflgs); + ErtsProcLocks lflgs = ERTS_PROC_LOCK_FLGS_CMPXCHG_ACQB_(&p->lock, + expct_lflgs | locks, + expct_lflgs); if (ERTS_LIKELY(lflgs == expct_lflgs)) { /* We successfully grabbed all locks. */ return 0; @@ -535,7 +527,7 @@ erts_smp_proc_unlock__(Process *p, if (want_lflgs != old_lflgs) { ErtsProcLocks new_lflgs = - ERTS_PROC_LOCK_FLGS_CMPXCHG_(&p->lock, want_lflgs, old_lflgs); + ERTS_PROC_LOCK_FLGS_CMPXCHG_RELB_(&p->lock, want_lflgs, old_lflgs); if (new_lflgs != old_lflgs) { /* cmpxchg failed, try again. */ diff --git a/erts/emulator/beam/erl_smp.h b/erts/emulator/beam/erl_smp.h index 03d2a586e3..b41fa70476 100644 --- a/erts/emulator/beam/erl_smp.h +++ b/erts/emulator/beam/erl_smp.h @@ -1,19 +1,19 @@ /* * %CopyrightBegin% - * - * Copyright Ericsson AB 2005-2009. All Rights Reserved. - * + * + * Copyright Ericsson AB 2005-2010. 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% */ /* @@ -43,9 +43,17 @@ typedef erts_thr_init_data_t erts_smp_thr_init_data_t; typedef erts_tid_t erts_smp_tid_t; typedef erts_mtx_t erts_smp_mtx_t; typedef erts_cnd_t erts_smp_cnd_t; +#define ERTS_SMP_RWMTX_OPT_DEFAULT_INITER ERTS_RWMTX_OPT_DEFAULT_INITER +#define ERTS_SMP_RWMTX_TYPE_NORMAL ERTS_RWMTX_TYPE_NORMAL +#define ERTS_SMP_RWMTX_TYPE_FREQUENT_READ ERTS_RWMTX_TYPE_FREQUENT_READ +#define ERTS_SMP_RWMTX_TYPE_EXTREMELY_FREQUENT_READ \ + ERTS_RWMTX_TYPE_EXTREMELY_FREQUENT_READ +#define ERTS_SMP_RWMTX_LONG_LIVED ERTS_RWMTX_LONG_LIVED +#define ERTS_SMP_RWMTX_SHORT_LIVED ERTS_RWMTX_SHORT_LIVED +#define ERTS_SMP_RWMTX_UNKNOWN_LIVED ERTS_RWMTX_UNKNOWN_LIVED +typedef erts_rwmtx_opt_t erts_smp_rwmtx_opt_t; typedef erts_rwmtx_t erts_smp_rwmtx_t; typedef erts_tsd_key_t erts_smp_tsd_key_t; -typedef erts_gate_t erts_smp_gate_t; typedef ethr_atomic_t erts_smp_atomic_t; typedef erts_spinlock_t erts_smp_spinlock_t; typedef erts_rwlock_t erts_smp_rwlock_t; @@ -54,15 +62,27 @@ void erts_thr_fatal_error(int, char *); /* implemented in erl_init.c */ #else /* #ifdef ERTS_SMP */ -#define ERTS_SMP_THR_OPTS_DEFAULT_INITER 0 +#define ERTS_SMP_THR_OPTS_DEFAULT_INITER {0} typedef int erts_smp_thr_opts_t; typedef int erts_smp_thr_init_data_t; typedef int erts_smp_tid_t; typedef int erts_smp_mtx_t; typedef int erts_smp_cnd_t; +#define ERTS_SMP_RWMTX_OPT_DEFAULT_INITER {0} +#define ERTS_SMP_RWMTX_TYPE_NORMAL 0 +#define ERTS_SMP_RWMTX_TYPE_FREQUENT_READ 0 +#define ERTS_SMP_RWMTX_TYPE_EXTREMELY_FREQUENT_READ 0 +#define ERTS_SMP_RWMTX_LONG_LIVED 0 +#define ERTS_SMP_RWMTX_SHORT_LIVED 0 +#define ERTS_SMP_RWMTX_UNKNOWN_LIVED 0 +typedef struct { + char type; + char lived; + int main_spincount; + int aux_spincount; +} erts_smp_rwmtx_opt_t; typedef int erts_smp_rwmtx_t; typedef int erts_smp_tsd_key_t; -typedef int erts_smp_gate_t; typedef long erts_smp_atomic_t; #if __GNUC__ > 2 typedef struct { } erts_smp_spinlock_t; @@ -103,8 +123,6 @@ ERTS_GLB_INLINE void erts_smp_mtx_init_locked_x(erts_smp_mtx_t *mtx, ERTS_GLB_INLINE void erts_smp_mtx_init(erts_smp_mtx_t *mtx, char *name); ERTS_GLB_INLINE void erts_smp_mtx_init_locked(erts_smp_mtx_t *mtx, char *name); ERTS_GLB_INLINE void erts_smp_mtx_destroy(erts_smp_mtx_t *mtx); -ERTS_GLB_INLINE void erts_smp_mtx_set_forksafe(erts_smp_mtx_t *mtx); -ERTS_GLB_INLINE void erts_smp_mtx_unset_forksafe(erts_smp_mtx_t *mtx); ERTS_GLB_INLINE int erts_smp_mtx_trylock(erts_smp_mtx_t *mtx); #ifdef ERTS_ENABLE_LOCK_COUNT ERTS_GLB_INLINE void erts_smp_mtx_lock_x(erts_smp_mtx_t *mtx, char *file, int line); @@ -119,9 +137,17 @@ ERTS_GLB_INLINE void erts_smp_cnd_wait(erts_smp_cnd_t *cnd, erts_smp_mtx_t *mtx); ERTS_GLB_INLINE void erts_smp_cnd_signal(erts_smp_cnd_t *cnd); ERTS_GLB_INLINE void erts_smp_cnd_broadcast(erts_smp_cnd_t *cnd); +ERTS_GLB_INLINE void erts_smp_rwmtx_set_reader_group(int no); +ERTS_GLB_INLINE void erts_smp_rwmtx_init_opt_x(erts_smp_rwmtx_t *rwmtx, + erts_smp_rwmtx_opt_t *opt, + char *name, + Eterm extra); ERTS_GLB_INLINE void erts_smp_rwmtx_init_x(erts_smp_rwmtx_t *rwmtx, char *name, Eterm extra); +ERTS_GLB_INLINE void erts_smp_rwmtx_init_opt(erts_smp_rwmtx_t *rwmtx, + erts_smp_rwmtx_opt_t *opt, + char *name); ERTS_GLB_INLINE void erts_smp_rwmtx_init(erts_smp_rwmtx_t *rwmtx, char *name); ERTS_GLB_INLINE void erts_smp_rwmtx_destroy(erts_smp_rwmtx_t *rwmtx); @@ -155,6 +181,16 @@ ERTS_GLB_INLINE long erts_smp_atomic_cmpxchg(erts_smp_atomic_t *xchgp, long expected); ERTS_GLB_INLINE long erts_smp_atomic_bor(erts_smp_atomic_t *var, long mask); ERTS_GLB_INLINE long erts_smp_atomic_band(erts_smp_atomic_t *var, long mask); +ERTS_GLB_INLINE long erts_smp_atomic_read_acqb(erts_smp_atomic_t *var); +ERTS_GLB_INLINE void erts_smp_atomic_set_relb(erts_smp_atomic_t *var, long i); +ERTS_GLB_INLINE void erts_smp_atomic_dec_relb(erts_smp_atomic_t *decp); +ERTS_GLB_INLINE long erts_smp_atomic_dectest_relb(erts_smp_atomic_t *decp); +ERTS_GLB_INLINE long erts_smp_atomic_cmpxchg_acqb(erts_smp_atomic_t *xchgp, + long new, + long exp); +ERTS_GLB_INLINE long erts_smp_atomic_cmpxchg_relb(erts_smp_atomic_t *xchgp, + long new, + long exp); ERTS_GLB_INLINE void erts_smp_spinlock_init_x(erts_smp_spinlock_t *lock, char *name, Eterm extra); @@ -190,12 +226,6 @@ ERTS_GLB_INLINE void erts_smp_tsd_key_create(erts_smp_tsd_key_t *keyp); ERTS_GLB_INLINE void erts_smp_tsd_key_delete(erts_smp_tsd_key_t key); ERTS_GLB_INLINE void erts_smp_tsd_set(erts_smp_tsd_key_t key, void *value); ERTS_GLB_INLINE void * erts_smp_tsd_get(erts_smp_tsd_key_t key); -ERTS_GLB_INLINE void erts_smp_gate_init(erts_smp_gate_t *gp); -ERTS_GLB_INLINE void erts_smp_gate_destroy(erts_smp_gate_t *gp); -ERTS_GLB_INLINE void erts_smp_gate_close(erts_smp_gate_t *gp); -ERTS_GLB_INLINE void erts_smp_gate_let_through(erts_smp_gate_t *gp, unsigned no); -ERTS_GLB_INLINE void erts_smp_gate_wait(erts_smp_gate_t *gp); -ERTS_GLB_INLINE void erts_smp_gate_swait(erts_smp_gate_t *gp, int spincount); #ifdef ERTS_THR_HAVE_SIG_FUNCS #define ERTS_SMP_THR_HAVE_SIG_FUNCS 1 @@ -331,22 +361,6 @@ erts_smp_mtx_destroy(erts_smp_mtx_t *mtx) #endif } -ERTS_GLB_INLINE void -erts_smp_mtx_set_forksafe(erts_smp_mtx_t *mtx) -{ -#ifdef ERTS_SMP - erts_mtx_set_forksafe(mtx); -#endif -} - -ERTS_GLB_INLINE void -erts_smp_mtx_unset_forksafe(erts_smp_mtx_t *mtx) -{ -#ifdef ERTS_SMP - erts_mtx_unset_forksafe(mtx); -#endif -} - ERTS_GLB_INLINE int erts_smp_mtx_trylock(erts_smp_mtx_t *mtx) { @@ -433,6 +447,25 @@ erts_smp_cnd_broadcast(erts_smp_cnd_t *cnd) } ERTS_GLB_INLINE void +erts_smp_rwmtx_set_reader_group(int no) +{ +#ifdef ERTS_SMP + erts_rwmtx_set_reader_group(no); +#endif +} + +ERTS_GLB_INLINE void +erts_smp_rwmtx_init_opt_x(erts_smp_rwmtx_t *rwmtx, + erts_smp_rwmtx_opt_t *opt, + char *name, + Eterm extra) +{ +#ifdef ERTS_SMP + erts_rwmtx_init_opt_x(rwmtx, opt, name, extra); +#endif +} + +ERTS_GLB_INLINE void erts_smp_rwmtx_init_x(erts_smp_rwmtx_t *rwmtx, char *name, Eterm extra) { #ifdef ERTS_SMP @@ -441,6 +474,16 @@ erts_smp_rwmtx_init_x(erts_smp_rwmtx_t *rwmtx, char *name, Eterm extra) } ERTS_GLB_INLINE void +erts_smp_rwmtx_init_opt(erts_smp_rwmtx_t *rwmtx, + erts_smp_rwmtx_opt_t *opt, + char *name) +{ +#ifdef ERTS_SMP + erts_rwmtx_init_opt(rwmtx, opt, name); +#endif +} + +ERTS_GLB_INLINE void erts_smp_rwmtx_init(erts_smp_rwmtx_t *rwmtx, char *name) { #ifdef ERTS_SMP @@ -709,6 +752,73 @@ erts_smp_atomic_band(erts_smp_atomic_t *var, long mask) #endif } +ERTS_GLB_INLINE long +erts_smp_atomic_read_acqb(erts_smp_atomic_t *var) +{ +#ifdef ERTS_SMP + return erts_atomic_read_acqb(var); +#else + return *var; +#endif +} + +ERTS_GLB_INLINE void +erts_smp_atomic_set_relb(erts_smp_atomic_t *var, long i) +{ +#ifdef ERTS_SMP + erts_atomic_set_relb(var, i); +#else + *var = i; +#endif +} + +ERTS_GLB_INLINE void erts_smp_atomic_dec_relb(erts_smp_atomic_t *decp) +{ +#ifdef ERTS_SMP + erts_atomic_dec_relb(decp); +#else + --(*decp); +#endif +} + +ERTS_GLB_INLINE long +erts_smp_atomic_dectest_relb(erts_smp_atomic_t *decp) +{ +#ifdef ERTS_SMP + return erts_atomic_dectest_relb(decp); +#else + return --(*decp); +#endif +} + +ERTS_GLB_INLINE long erts_smp_atomic_cmpxchg_acqb(erts_smp_atomic_t *xchgp, + long new, + long exp) +{ +#ifdef ERTS_SMP + return erts_atomic_cmpxchg_acqb(xchgp, new, exp); +#else + long old = *xchgp; + if (old == exp) + *xchgp = new; + return old; +#endif +} + +ERTS_GLB_INLINE long erts_smp_atomic_cmpxchg_relb(erts_smp_atomic_t *xchgp, + long new, + long exp) +{ +#ifdef ERTS_SMP + return erts_atomic_cmpxchg_relb(xchgp, new, exp); +#else + long old = *xchgp; + if (old == exp) + *xchgp = new; + return old; +#endif +} + ERTS_GLB_INLINE void erts_smp_spinlock_init_x(erts_smp_spinlock_t *lock, char *name, Eterm extra) { @@ -919,54 +1029,6 @@ erts_smp_tsd_get(erts_smp_tsd_key_t key) #endif } -ERTS_GLB_INLINE void -erts_smp_gate_init(erts_smp_gate_t *gp) -{ -#ifdef ERTS_SMP - erts_gate_init((erts_gate_t *) gp); -#endif -} - -ERTS_GLB_INLINE void -erts_smp_gate_destroy(erts_smp_gate_t *gp) -{ -#ifdef ERTS_SMP - erts_gate_destroy((erts_gate_t *) gp); -#endif -} - -ERTS_GLB_INLINE void -erts_smp_gate_close(erts_smp_gate_t *gp) -{ -#ifdef ERTS_SMP - erts_gate_close((erts_gate_t *) gp); -#endif -} - -ERTS_GLB_INLINE void -erts_smp_gate_let_through(erts_smp_gate_t *gp, unsigned no) -{ -#ifdef ERTS_SMP - erts_gate_let_through((erts_gate_t *) gp, no); -#endif -} - -ERTS_GLB_INLINE void -erts_smp_gate_wait(erts_smp_gate_t *gp) -{ -#ifdef ERTS_SMP - erts_gate_wait((erts_gate_t *) gp); -#endif -} - -ERTS_GLB_INLINE void -erts_smp_gate_swait(erts_smp_gate_t *gp, int spincount) -{ -#ifdef ERTS_SMP - erts_gate_swait((erts_gate_t *) gp, spincount); -#endif -} - #ifdef ERTS_THR_HAVE_SIG_FUNCS #define ERTS_SMP_THR_HAVE_SIG_FUNCS 1 diff --git a/erts/emulator/beam/erl_threads.h b/erts/emulator/beam/erl_threads.h index 35b338c6eb..0b7269262e 100644 --- a/erts/emulator/beam/erl_threads.h +++ b/erts/emulator/beam/erl_threads.h @@ -25,6 +25,11 @@ #ifndef ERL_THREAD_H__ #define ERL_THREAD_H__ +#define ERTS_SPIN_BODY ETHR_SPIN_BODY + +#define ERTS_MAX_READER_GROUPS 8 +extern int erts_reader_groups; + #include "sys.h" #ifdef USE_THREADS @@ -48,6 +53,7 @@ #define ERTS_THR_OPTS_DEFAULT_INITER ETHR_THR_OPTS_DEFAULT_INITER typedef ethr_thr_opts erts_thr_opts_t; typedef ethr_init_data erts_thr_init_data_t; +typedef ethr_late_init_data erts_thr_late_init_data_t; typedef ethr_tid erts_tid_t; /* mutex */ @@ -73,8 +79,19 @@ typedef struct { erts_lcnt_lock_t lcnt; #endif } erts_rwmtx_t; + +#define ERTS_RWMTX_OPT_DEFAULT_INITER ETHR_RWMUTEX_OPT_DEFAULT_INITER +#define ERTS_RWMTX_TYPE_NORMAL ETHR_RWMUTEX_TYPE_NORMAL +#define ERTS_RWMTX_TYPE_FREQUENT_READ ETHR_RWMUTEX_TYPE_FREQUENT_READ +#define ERTS_RWMTX_TYPE_EXTREMELY_FREQUENT_READ \ + ETHR_RWMUTEX_TYPE_EXTREMELY_FREQUENT_READ +#define ERTS_RWMTX_LONG_LIVED ETHR_RWMUTEX_LONG_LIVED +#define ERTS_RWMTX_SHORT_LIVED ETHR_RWMUTEX_SHORT_LIVED +#define ERTS_RWMTX_UNKNOWN_LIVED ETHR_RWMUTEX_UNKNOWN_LIVED +typedef ethr_rwmutex_opt erts_rwmtx_opt_t; + typedef ethr_tsd_key erts_tsd_key_t; -typedef ethr_gate erts_gate_t; +typedef ethr_ts_event erts_tse_t; typedef ethr_atomic_t erts_atomic_t; /* spinlock */ @@ -103,25 +120,14 @@ typedef ethr_timeval erts_thr_timeval_t; __decl_noreturn void __noreturn erts_thr_fatal_error(int, char *); /* implemented in erl_init.c */ -#ifdef ERTS_ENABLE_LOCK_CHECK -#define ERTS_REC_MTX_INITER \ - {ETHR_REC_MUTEX_INITER, \ - ERTS_LC_LOCK_INIT(-1,THE_NON_VALUE,ERTS_LC_FLG_LT_MUTEX)} -#define ERTS_MTX_INITER \ - {ETHR_MUTEX_INITER, \ - ERTS_LC_LOCK_INIT(-1, THE_NON_VALUE, ERTS_LC_FLG_LT_MUTEX)} -#else -#define ERTS_REC_MTX_INITER {ETHR_REC_MUTEX_INITER} -#define ERTS_MTX_INITER {ETHR_MUTEX_INITER} -#endif -#define ERTS_CND_INITER ETHR_COND_INITER #define ERTS_THR_INIT_DATA_DEF_INITER ETHR_INIT_DATA_DEFAULT_INITER +#define ERTS_THR_LATE_INIT_DATA_DEF_INITER \ + ETHR_LATE_INIT_DATA_DEFAULT_INITER #ifdef ETHR_HAVE_ETHR_REC_MUTEX_INIT # define ERTS_HAVE_REC_MTX_INIT ETHR_HAVE_ETHR_REC_MUTEX_INIT #endif - #else /* #ifdef USE_THREADS */ #define ERTS_THR_MEMORY_BARRIER @@ -129,12 +135,26 @@ __decl_noreturn void __noreturn erts_thr_fatal_error(int, char *); #define ERTS_THR_OPTS_DEFAULT_INITER 0 typedef int erts_thr_opts_t; typedef int erts_thr_init_data_t; +typedef int erts_thr_late_init_data_t; typedef int erts_tid_t; typedef int erts_mtx_t; typedef int erts_cnd_t; +#define ERTS_RWMTX_OPT_DEFAULT_INITER {0} +#define ERTS_RWMTX_TYPE_NORMAL 0 +#define ERTS_RWMTX_TYPE_FREQUENT_READ 0 +#define ERTS_RWMTX_TYPE_EXTREMELY_FREQUENT_READ 0 +#define ERTS_RWMTX_LONG_LIVED 0 +#define ERTS_RWMTX_SHORT_LIVED 0 +#define ERTS_RWMTX_UNKNOWN_LIVED 0 +typedef struct { + char type; + char lived; + int main_spincount; + int aux_spincount; +} erts_rwmtx_opt_t; typedef int erts_rwmtx_t; typedef int erts_tsd_key_t; -typedef int erts_gate_t; +typedef int erts_tse_t; typedef long erts_atomic_t; #if __GNUC__ > 2 typedef struct { } erts_spinlock_t; @@ -148,7 +168,6 @@ typedef struct { long tv_nsec; } erts_thr_timeval_t; -#define ERTS_REC_MTX_INITER 0 #define ERTS_MTX_INITER 0 #define ERTS_CND_INITER 0 #define ERTS_THR_INIT_DATA_DEF_INITER 0 @@ -158,6 +177,7 @@ typedef struct { #endif /* #ifdef USE_THREADS */ ERTS_GLB_INLINE void erts_thr_init(erts_thr_init_data_t *id); +ERTS_GLB_INLINE void erts_thr_late_init(erts_thr_late_init_data_t *id); ERTS_GLB_INLINE void erts_thr_create(erts_tid_t *tid, void * (*func)(void *), void *arg, erts_thr_opts_t *opts); ERTS_GLB_INLINE void erts_thr_join(erts_tid_t tid, void **thr_res); @@ -166,9 +186,6 @@ ERTS_GLB_INLINE void erts_thr_exit(void *res); ERTS_GLB_INLINE void erts_thr_install_exit_handler(void (*exit_handler)(void)); ERTS_GLB_INLINE erts_tid_t erts_thr_self(void); ERTS_GLB_INLINE int erts_equal_tids(erts_tid_t x, erts_tid_t y); -#ifdef ERTS_HAVE_REC_MTX_INIT -ERTS_GLB_INLINE void erts_rec_mtx_init(erts_mtx_t *mtx); -#endif ERTS_GLB_INLINE void erts_mtx_init_x(erts_mtx_t *mtx, char *name, Eterm extra); ERTS_GLB_INLINE void erts_mtx_init_x_opt(erts_mtx_t *mtx, char *name, Eterm extra, Uint16 opt); ERTS_GLB_INLINE void erts_mtx_init_locked_x(erts_mtx_t *mtx, @@ -177,8 +194,6 @@ ERTS_GLB_INLINE void erts_mtx_init_locked_x(erts_mtx_t *mtx, ERTS_GLB_INLINE void erts_mtx_init(erts_mtx_t *mtx, char *name); ERTS_GLB_INLINE void erts_mtx_init_locked(erts_mtx_t *mtx, char *name); ERTS_GLB_INLINE void erts_mtx_destroy(erts_mtx_t *mtx); -ERTS_GLB_INLINE void erts_mtx_set_forksafe(erts_mtx_t *mtx); -ERTS_GLB_INLINE void erts_mtx_unset_forksafe(erts_mtx_t *mtx); ERTS_GLB_INLINE int erts_mtx_trylock(erts_mtx_t *mtx); #ifdef ERTS_ENABLE_LOCK_COUNT ERTS_GLB_INLINE void erts_mtx_lock_x(erts_mtx_t *mtx, char *file, unsigned int line); @@ -192,9 +207,17 @@ ERTS_GLB_INLINE void erts_cnd_destroy(erts_cnd_t *cnd); ERTS_GLB_INLINE void erts_cnd_wait(erts_cnd_t *cnd, erts_mtx_t *mtx); ERTS_GLB_INLINE void erts_cnd_signal(erts_cnd_t *cnd); ERTS_GLB_INLINE void erts_cnd_broadcast(erts_cnd_t *cnd); +ERTS_GLB_INLINE void erts_rwmtx_set_reader_group(int no); +ERTS_GLB_INLINE void erts_rwmtx_init_opt_x(erts_rwmtx_t *rwmtx, + erts_rwmtx_opt_t *opt, + char *name, + Eterm extra); ERTS_GLB_INLINE void erts_rwmtx_init_x(erts_rwmtx_t *rwmtx, char *name, Eterm extra); +ERTS_GLB_INLINE void erts_rwmtx_init_opt(erts_rwmtx_t *rwmtx, + erts_rwmtx_opt_t *opt, + char *name); ERTS_GLB_INLINE void erts_rwmtx_init(erts_rwmtx_t *rwmtx, char *name); ERTS_GLB_INLINE void erts_rwmtx_destroy(erts_rwmtx_t *rwmtx); @@ -228,6 +251,20 @@ ERTS_GLB_INLINE long erts_atomic_cmpxchg(erts_atomic_t *xchgp, long expected); ERTS_GLB_INLINE long erts_atomic_bor(erts_atomic_t *var, long mask); ERTS_GLB_INLINE long erts_atomic_band(erts_atomic_t *var, long mask); +ERTS_GLB_INLINE long erts_atomic_read_acqb(erts_atomic_t *var); +ERTS_GLB_INLINE void erts_atomic_set_relb(erts_atomic_t *var, long i); +ERTS_GLB_INLINE void erts_atomic_dec_relb(erts_atomic_t *decp); +ERTS_GLB_INLINE long erts_atomic_dectest_relb(erts_atomic_t *decp); +ERTS_GLB_INLINE long erts_atomic_cmpxchg_acqb(erts_atomic_t *xchgp, + long new, + long exp); +ERTS_GLB_INLINE long erts_atomic_cmpxchg_relb(erts_atomic_t *xchgp, + long new, + long exp); +ERTS_GLB_INLINE void erts_spinlock_init_x_opt(erts_spinlock_t *lock, + char *name, + Eterm extra, + Uint16 opt); ERTS_GLB_INLINE void erts_spinlock_init_x(erts_spinlock_t *lock, char *name, Eterm extra); @@ -263,12 +300,16 @@ ERTS_GLB_INLINE void erts_tsd_key_create(erts_tsd_key_t *keyp); ERTS_GLB_INLINE void erts_tsd_key_delete(erts_tsd_key_t key); ERTS_GLB_INLINE void erts_tsd_set(erts_tsd_key_t key, void *value); ERTS_GLB_INLINE void * erts_tsd_get(erts_tsd_key_t key); -ERTS_GLB_INLINE void erts_gate_init(erts_gate_t *gp); -ERTS_GLB_INLINE void erts_gate_destroy(erts_gate_t *gp); -ERTS_GLB_INLINE void erts_gate_close(erts_gate_t *gp); -ERTS_GLB_INLINE void erts_gate_let_through(erts_gate_t *gp, unsigned no); -ERTS_GLB_INLINE void erts_gate_wait(erts_gate_t *gp); -ERTS_GLB_INLINE void erts_gate_swait(erts_gate_t *gp, int spincount); +ERTS_GLB_INLINE erts_tse_t *erts_tse_fetch(void); +ERTS_GLB_INLINE void erts_tse_return(erts_tse_t *ep); +ERTS_GLB_INLINE void erts_tse_set(erts_tse_t *ep); +ERTS_GLB_INLINE void erts_tse_reset(erts_tse_t *ep); +ERTS_GLB_INLINE int erts_tse_wait(erts_tse_t *ep); +ERTS_GLB_INLINE int erts_tse_swait(erts_tse_t *ep, int spincount); +ERTS_GLB_INLINE int erts_tse_is_tmp(erts_tse_t *ep); +ERTS_GLB_INLINE void erts_thr_set_main_status(int, int); +ERTS_GLB_INLINE int erts_thr_get_main_status(void); +ERTS_GLB_INLINE void erts_thr_yield(void); #ifdef ETHR_HAVE_ETHR_SIG_FUNCS #define ERTS_THR_HAVE_SIG_FUNCS 1 @@ -290,6 +331,16 @@ erts_thr_init(erts_thr_init_data_t *id) } ERTS_GLB_INLINE void +erts_thr_late_init(erts_thr_late_init_data_t *id) +{ +#ifdef USE_THREADS + int res = ethr_late_init(id); + if (res) + erts_thr_fatal_error(res, "complete initialization of thread library"); +#endif +} + +ERTS_GLB_INLINE void erts_thr_create(erts_tid_t *tid, void * (*func)(void *), void *arg, erts_thr_opts_t *opts) { @@ -362,20 +413,6 @@ erts_equal_tids(erts_tid_t x, erts_tid_t y) #endif } - -#ifdef ERTS_HAVE_REC_MTX_INIT -ERTS_GLB_INLINE void -erts_rec_mtx_init(erts_mtx_t *mtx) -{ -#ifdef USE_THREADS - int res = ethr_rec_mutex_init(&mtx->mtx); - if (res) - erts_thr_fatal_error(res, "initialize recursive mutex"); -#endif -} -#endif - - ERTS_GLB_INLINE void erts_mtx_init_x(erts_mtx_t *mtx, char *name, Eterm extra) { @@ -422,9 +459,7 @@ erts_mtx_init_locked_x(erts_mtx_t *mtx, char *name, Eterm extra) #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_init_lock_x(&mtx->lcnt, name, ERTS_LCNT_LT_MUTEX, extra); #endif - res = ethr_mutex_lock(&mtx->mtx); - if (res) - erts_thr_fatal_error(res, "lock mutex"); + ethr_mutex_lock(&mtx->mtx); #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_trylock(1, &mtx->lc); #endif @@ -463,9 +498,7 @@ erts_mtx_init_locked(erts_mtx_t *mtx, char *name) #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_init_lock(&mtx->lcnt, name, ERTS_LCNT_LT_MUTEX); #endif - res = ethr_mutex_lock(&mtx->mtx); - if (res) - erts_thr_fatal_error(res, "lock mutex"); + ethr_mutex_lock(&mtx->mtx); #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_trylock(1, &mtx->lc); #endif @@ -492,26 +525,6 @@ erts_mtx_destroy(erts_mtx_t *mtx) #endif } -ERTS_GLB_INLINE void -erts_mtx_set_forksafe(erts_mtx_t *mtx) -{ -#ifdef USE_THREADS - int res = ethr_mutex_set_forksafe(&mtx->mtx); - if (res != 0 && res != ENOTSUP) - erts_thr_fatal_error(res, "set mutex forksafe"); -#endif -} - -ERTS_GLB_INLINE void -erts_mtx_unset_forksafe(erts_mtx_t *mtx) -{ -#ifdef USE_THREADS - int res = ethr_mutex_unset_forksafe(&mtx->mtx); - if (res != 0 && res != ENOTSUP) - erts_thr_fatal_error(res, "unset mutex forksafe"); -#endif -} - ERTS_GLB_INLINE int erts_mtx_trylock(erts_mtx_t *mtx) { @@ -531,11 +544,7 @@ erts_mtx_trylock(erts_mtx_t *mtx) #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_trylock(&mtx->lcnt, res); -#endif - - if (res != 0 && res != EBUSY) - erts_thr_fatal_error(res, "try lock mutex"); - +#endif return res; #else return 0; @@ -551,19 +560,16 @@ erts_mtx_lock(erts_mtx_t *mtx) #endif { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_lock(&mtx->lc); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock(&mtx->lcnt); #endif - res = ethr_mutex_lock(&mtx->mtx); + ethr_mutex_lock(&mtx->mtx); #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock_post_x(&mtx->lcnt, file, line); #endif - if (res) - erts_thr_fatal_error(res, "lock mutex"); #endif } @@ -571,16 +577,13 @@ ERTS_GLB_INLINE void erts_mtx_unlock(erts_mtx_t *mtx) { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_unlock(&mtx->lc); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_unlock(&mtx->lcnt); #endif - res = ethr_mutex_unlock(&mtx->mtx); - if (res) - erts_thr_fatal_error(res, "unlock mutex"); + ethr_mutex_unlock(&mtx->mtx); #endif } @@ -648,9 +651,7 @@ ERTS_GLB_INLINE void erts_cnd_signal(erts_cnd_t *cnd) { #ifdef USE_THREADS - int res = ethr_cond_signal(cnd); - if (res) - erts_thr_fatal_error(res, "signal on condition variable"); + ethr_cond_signal(cnd); #endif } @@ -659,19 +660,34 @@ ERTS_GLB_INLINE void erts_cnd_broadcast(erts_cnd_t *cnd) { #ifdef USE_THREADS - int res = ethr_cond_broadcast(cnd); - if (res) - erts_thr_fatal_error(res, "broadcast on condition variable"); + ethr_cond_broadcast(cnd); #endif } /* rwmutex */ ERTS_GLB_INLINE void -erts_rwmtx_init_x(erts_rwmtx_t *rwmtx, char *name, Eterm extra) +erts_rwmtx_set_reader_group(int no) { #ifdef USE_THREADS - int res = ethr_rwmutex_init(&rwmtx->rwmtx); + int res; +#ifdef ERTS_ENABLE_LOCK_CHECK + erts_lc_check_no_locked_of_type(ERTS_LC_FLG_LT_RWMUTEX); +#endif + res = ethr_rwmutex_set_reader_group(no); + if (res != 0) + erts_thr_fatal_error(res, "set reader group"); +#endif +} + +ERTS_GLB_INLINE void +erts_rwmtx_init_opt_x(erts_rwmtx_t *rwmtx, + erts_rwmtx_opt_t *opt, + char *name, + Eterm extra) +{ +#ifdef USE_THREADS + int res = ethr_rwmutex_init_opt(&rwmtx->rwmtx, opt); if (res != 0) erts_thr_fatal_error(res, "initialize rwmutex"); #ifdef ERTS_ENABLE_LOCK_CHECK @@ -684,10 +700,20 @@ erts_rwmtx_init_x(erts_rwmtx_t *rwmtx, char *name, Eterm extra) } ERTS_GLB_INLINE void -erts_rwmtx_init(erts_rwmtx_t *rwmtx, char *name) +erts_rwmtx_init_x(erts_rwmtx_t *rwmtx, + char *name, + Eterm extra) +{ + erts_rwmtx_init_opt_x(rwmtx, NULL, name, extra); +} + +ERTS_GLB_INLINE void +erts_rwmtx_init_opt(erts_rwmtx_t *rwmtx, + erts_rwmtx_opt_t *opt, + char *name) { #ifdef USE_THREADS - int res = ethr_rwmutex_init(&rwmtx->rwmtx); + int res = ethr_rwmutex_init_opt(&rwmtx->rwmtx, opt); if (res != 0) erts_thr_fatal_error(res, "initialize rwmutex"); #ifdef ERTS_ENABLE_LOCK_CHECK @@ -700,6 +726,12 @@ erts_rwmtx_init(erts_rwmtx_t *rwmtx, char *name) } ERTS_GLB_INLINE void +erts_rwmtx_init(erts_rwmtx_t *rwmtx, char *name) +{ + erts_rwmtx_init_opt(rwmtx, NULL, name); +} + +ERTS_GLB_INLINE void erts_rwmtx_destroy(erts_rwmtx_t *rwmtx) { #ifdef USE_THREADS @@ -736,9 +768,6 @@ erts_rwmtx_tryrlock(erts_rwmtx_t *rwmtx) #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_trylock_opt(&rwmtx->lcnt, res, ERTS_LCNT_LO_READ); #endif - - if (res != 0 && res != EBUSY) - erts_thr_fatal_error(res, "try read lock rwmutex"); return res; #else @@ -754,19 +783,16 @@ erts_rwmtx_rlock(erts_rwmtx_t *rwmtx) #endif { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_lock_flg(&rwmtx->lc, ERTS_LC_FLG_LO_READ); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock_opt(&rwmtx->lcnt, ERTS_LCNT_LO_READ); #endif - res = ethr_rwmutex_rlock(&rwmtx->rwmtx); + ethr_rwmutex_rlock(&rwmtx->rwmtx); #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock_post_x(&rwmtx->lcnt, file, line); #endif - if (res != 0) - erts_thr_fatal_error(res, "read lock rwmutex"); #endif } @@ -774,16 +800,13 @@ ERTS_GLB_INLINE void erts_rwmtx_runlock(erts_rwmtx_t *rwmtx) { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_unlock_flg(&rwmtx->lc, ERTS_LC_FLG_LO_READ); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_unlock_opt(&rwmtx->lcnt, ERTS_LCNT_LO_READ); #endif - res = ethr_rwmutex_runlock(&rwmtx->rwmtx); - if (res != 0) - erts_thr_fatal_error(res, "read unlock rwmutex"); + ethr_rwmutex_runlock(&rwmtx->rwmtx); #endif } @@ -808,9 +831,6 @@ erts_rwmtx_tryrwlock(erts_rwmtx_t *rwmtx) #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_trylock_opt(&rwmtx->lcnt, res, ERTS_LCNT_LO_READ_WRITE); #endif - - if (res != 0 && res != EBUSY) - erts_thr_fatal_error(res, "try write lock rwmutex"); return res; #else @@ -826,19 +846,16 @@ erts_rwmtx_rwlock(erts_rwmtx_t *rwmtx) #endif { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_lock_flg(&rwmtx->lc, ERTS_LC_FLG_LO_READ_WRITE); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock_opt(&rwmtx->lcnt, ERTS_LCNT_LO_READ_WRITE); #endif - res = ethr_rwmutex_rwlock(&rwmtx->rwmtx); + ethr_rwmutex_rwlock(&rwmtx->rwmtx); #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock_post_x(&rwmtx->lcnt, file, line); #endif - if (res != 0) - erts_thr_fatal_error(res, "write lock rwmutex"); #endif } @@ -846,16 +863,13 @@ ERTS_GLB_INLINE void erts_rwmtx_rwunlock(erts_rwmtx_t *rwmtx) { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_unlock_flg(&rwmtx->lc, ERTS_LC_FLG_LO_READ_WRITE); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_unlock_opt(&rwmtx->lcnt, ERTS_LCNT_LO_READ_WRITE); #endif - res = ethr_rwmutex_rwunlock(&rwmtx->rwmtx); - if (res != 0) - erts_thr_fatal_error(res, "write unlock rwmutex"); + ethr_rwmutex_rwunlock(&rwmtx->rwmtx); #endif } @@ -917,9 +931,7 @@ ERTS_GLB_INLINE void erts_atomic_init(erts_atomic_t *var, long i) { #ifdef USE_THREADS - int res = ethr_atomic_init(var, i); - if (res) - erts_thr_fatal_error(res, "perform atomic init"); + ethr_atomic_init(var, i); #else *var = i; #endif @@ -929,9 +941,7 @@ ERTS_GLB_INLINE void erts_atomic_set(erts_atomic_t *var, long i) { #ifdef USE_THREADS - int res = ethr_atomic_set(var, i); - if (res) - erts_thr_fatal_error(res, "perform atomic set"); + ethr_atomic_set(var, i); #else *var = i; #endif @@ -941,11 +951,7 @@ ERTS_GLB_INLINE long erts_atomic_read(erts_atomic_t *var) { #ifdef USE_THREADS - long i; - int res = ethr_atomic_read(var, &i); - if (res) - erts_thr_fatal_error(res, "perform atomic read"); - return i; + return ethr_atomic_read(var); #else return *var; #endif @@ -955,11 +961,7 @@ ERTS_GLB_INLINE long erts_atomic_inctest(erts_atomic_t *incp) { #ifdef USE_THREADS - long test; - int res = ethr_atomic_inctest(incp, &test); - if (res) - erts_thr_fatal_error(res, "perform atomic increment and test"); - return test; + return ethr_atomic_inc_read(incp); #else return ++(*incp); #endif @@ -969,11 +971,7 @@ ERTS_GLB_INLINE long erts_atomic_dectest(erts_atomic_t *decp) { #ifdef USE_THREADS - long test; - int res = ethr_atomic_dectest(decp, &test); - if (res) - erts_thr_fatal_error(res, "perform atomic decrement and test"); - return test; + return ethr_atomic_dec_read(decp); #else return --(*decp); #endif @@ -983,9 +981,7 @@ ERTS_GLB_INLINE void erts_atomic_inc(erts_atomic_t *incp) { #ifdef USE_THREADS - int res = ethr_atomic_inc(incp); - if (res) - erts_thr_fatal_error(res, "perform atomic increment"); + ethr_atomic_inc(incp); #else ++(*incp); #endif @@ -995,9 +991,7 @@ ERTS_GLB_INLINE void erts_atomic_dec(erts_atomic_t *decp) { #ifdef USE_THREADS - int res = ethr_atomic_dec(decp); - if (res) - erts_thr_fatal_error(res, "perform atomic decrement"); + ethr_atomic_dec(decp); #else --(*decp); #endif @@ -1007,11 +1001,7 @@ ERTS_GLB_INLINE long erts_atomic_addtest(erts_atomic_t *addp, long i) { #ifdef USE_THREADS - long test; - int res = ethr_atomic_addtest(addp, i, &test); - if (res) - erts_thr_fatal_error(res, "perform atomic addition and test"); - return test; + return ethr_atomic_add_read(addp, i); #else return *addp += i; #endif @@ -1021,9 +1011,7 @@ ERTS_GLB_INLINE void erts_atomic_add(erts_atomic_t *addp, long i) { #ifdef USE_THREADS - int res = ethr_atomic_add(addp, i); - if (res) - erts_thr_fatal_error(res, "perform atomic addition"); + ethr_atomic_add(addp, i); #else *addp += i; #endif @@ -1034,9 +1022,7 @@ erts_atomic_xchg(erts_atomic_t *xchgp, long new) { long old; #ifdef USE_THREADS - int res = ethr_atomic_xchg(xchgp, new, &old); - if (res) - erts_thr_fatal_error(res, "perform atomic exchange"); + return ethr_atomic_xchg(xchgp, new); #else old = *xchgp; *xchgp = new; @@ -1048,11 +1034,7 @@ ERTS_GLB_INLINE long erts_atomic_cmpxchg(erts_atomic_t *xchgp, long new, long expected) { #ifdef USE_THREADS - long old; - int res = ethr_atomic_cmpxchg(xchgp, new, expected, &old); - if (ERTS_UNLIKELY(res != 0)) - erts_thr_fatal_error(res, "perform atomic exchange"); - return old; + return ethr_atomic_cmpxchg(xchgp, new, expected); #else long old = *xchgp; if (old == expected) @@ -1064,31 +1046,95 @@ erts_atomic_cmpxchg(erts_atomic_t *xchgp, long new, long expected) ERTS_GLB_INLINE long erts_atomic_bor(erts_atomic_t *var, long mask) { - long old; #ifdef USE_THREADS - int res = ethr_atomic_or_old(var, mask, &old); - if (res != 0) - erts_thr_fatal_error(res, "perform atomic bitwise or"); + return ethr_atomic_read_bor(var, mask); #else + long old; old = *var; *var |= mask; -#endif return old; +#endif } ERTS_GLB_INLINE long erts_atomic_band(erts_atomic_t *var, long mask) { - long old; #ifdef USE_THREADS - int res = ethr_atomic_and_old(var, mask, &old); - if (res != 0) - erts_thr_fatal_error(res, "perform atomic bitwise and"); + return ethr_atomic_read_band(var, mask); #else + long old; old = *var; *var &= mask; + return old; +#endif +} + +ERTS_GLB_INLINE long +erts_atomic_read_acqb(erts_atomic_t *var) +{ +#ifdef USE_THREADS + return ethr_atomic_read_acqb(var); +#else + return *var; +#endif +} + +ERTS_GLB_INLINE void +erts_atomic_set_relb(erts_atomic_t *var, long i) +{ +#ifdef USE_THREADS + ethr_atomic_set_relb(var, i); +#else + *var = i; +#endif +} + +ERTS_GLB_INLINE void +erts_atomic_dec_relb(erts_atomic_t *decp) +{ +#ifdef USE_THREADS + ethr_atomic_dec_relb(decp); +#else + --(*decp); +#endif +} + +ERTS_GLB_INLINE long +erts_atomic_dectest_relb(erts_atomic_t *decp) +{ +#ifdef USE_THREADS + return ethr_atomic_dec_read_relb(decp); +#else + return --(*decp); #endif +} + +ERTS_GLB_INLINE long erts_atomic_cmpxchg_acqb(erts_atomic_t *xchgp, + long new, + long exp) +{ +#ifdef USE_THREADS + return ethr_atomic_cmpxchg_acqb(xchgp, new, exp); +#else + long old = *xchgp; + if (old == exp) + *xchgp = new; + return old; +#endif +} + +ERTS_GLB_INLINE long erts_atomic_cmpxchg_relb(erts_atomic_t *xchgp, + long new, + long exp) +{ +#ifdef USE_THREADS + return ethr_atomic_cmpxchg_relb(xchgp, new, exp); +#else + long old = *xchgp; + if (old == exp) + *xchgp = new; return old; +#endif } /* spinlock */ @@ -1112,6 +1158,26 @@ erts_spinlock_init_x(erts_spinlock_t *lock, char *name, Eterm extra) } ERTS_GLB_INLINE void +erts_spinlock_init_x_opt(erts_spinlock_t *lock, char *name, Eterm extra, + Uint16 opt) +{ +#ifdef USE_THREADS + int res = ethr_spinlock_init(&lock->slck); + if (res) + erts_thr_fatal_error(res, "init spinlock"); +#ifdef ERTS_ENABLE_LOCK_CHECK + erts_lc_init_lock_x(&lock->lc, name, ERTS_LC_FLG_LT_SPINLOCK, extra); +#endif +#ifdef ERTS_ENABLE_LOCK_COUNT + erts_lcnt_init_lock_x(&lock->lcnt, name, ERTS_LCNT_LT_SPINLOCK|opt, extra); +#endif +#else + (void)lock; +#endif +} + + +ERTS_GLB_INLINE void erts_spinlock_init(erts_spinlock_t *lock, char *name) { #ifdef USE_THREADS @@ -1152,16 +1218,13 @@ ERTS_GLB_INLINE void erts_spin_unlock(erts_spinlock_t *lock) { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_unlock(&lock->lc); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_unlock(&lock->lcnt); #endif - res = ethr_spin_unlock(&lock->slck); - if (res) - erts_thr_fatal_error(res, "release spin lock"); + ethr_spin_unlock(&lock->slck); #else (void)lock; #endif @@ -1175,19 +1238,16 @@ erts_spin_lock(erts_spinlock_t *lock) #endif { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_lock(&lock->lc); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock(&lock->lcnt); #endif - res = ethr_spin_lock(&lock->slck); + ethr_spin_lock(&lock->slck); #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock_post_x(&lock->lcnt, file, line); #endif - if (res) - erts_thr_fatal_error(res, "take spin lock"); #else (void)lock; #endif @@ -1268,16 +1328,13 @@ ERTS_GLB_INLINE void erts_read_unlock(erts_rwlock_t *lock) { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_unlock_flg(&lock->lc, ERTS_LC_FLG_LO_READ); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_unlock_opt(&lock->lcnt, ERTS_LCNT_LO_READ); #endif - res = ethr_read_unlock(&lock->rwlck); - if (res) - erts_thr_fatal_error(res, "release read lock"); + ethr_read_unlock(&lock->rwlck); #else (void)lock; #endif @@ -1291,19 +1348,16 @@ erts_read_lock(erts_rwlock_t *lock) #endif { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_lock_flg(&lock->lc, ERTS_LC_FLG_LO_READ); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock_opt(&lock->lcnt, ERTS_LCNT_LO_READ); #endif - res = ethr_read_lock(&lock->rwlck); + ethr_read_lock(&lock->rwlck); #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock_post_x(&lock->lcnt, file, line); #endif - if (res) - erts_thr_fatal_error(res, "take read lock"); #else (void)lock; #endif @@ -1313,16 +1367,13 @@ ERTS_GLB_INLINE void erts_write_unlock(erts_rwlock_t *lock) { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_unlock_flg(&lock->lc, ERTS_LC_FLG_LO_READ_WRITE); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_unlock_opt(&lock->lcnt, ERTS_LCNT_LO_READ_WRITE); #endif - res = ethr_write_unlock(&lock->rwlck); - if (res) - erts_thr_fatal_error(res, "release write lock"); + ethr_write_unlock(&lock->rwlck); #else (void)lock; #endif @@ -1336,19 +1387,16 @@ erts_write_lock(erts_rwlock_t *lock) #endif { #ifdef USE_THREADS - int res; #ifdef ERTS_ENABLE_LOCK_CHECK erts_lc_lock_flg(&lock->lc, ERTS_LC_FLG_LO_READ_WRITE); #endif #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock_opt(&lock->lcnt, ERTS_LCNT_LO_READ_WRITE); #endif - res = ethr_write_lock(&lock->rwlck); + ethr_write_lock(&lock->rwlck); #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_lock_post_x(&lock->lcnt, file, line); #endif - if (res) - erts_thr_fatal_error(res, "take write lock"); #else (void)lock; #endif @@ -1432,66 +1480,95 @@ erts_tsd_get(erts_tsd_key_t key) #endif } -ERTS_GLB_INLINE void -erts_gate_init(erts_gate_t *gp) +ERTS_GLB_INLINE erts_tse_t *erts_tse_fetch(void) { #ifdef USE_THREADS - int res = ethr_gate_init((ethr_gate *) gp); - if (res != 0) - erts_thr_fatal_error(res, "initialize gate"); + return (erts_tse_t *) ethr_get_ts_event(); +#else + return (erts_tse_t *) NULL; #endif } -ERTS_GLB_INLINE void -erts_gate_destroy(erts_gate_t *gp) +ERTS_GLB_INLINE void erts_tse_return(erts_tse_t *ep) { #ifdef USE_THREADS - int res = ethr_gate_destroy((ethr_gate *) gp); - if (res != 0) - erts_thr_fatal_error(res, "destroy gate"); + ethr_leave_ts_event(ep); #endif } -ERTS_GLB_INLINE void -erts_gate_close(erts_gate_t *gp) +ERTS_GLB_INLINE void erts_tse_set(erts_tse_t *ep) { #ifdef USE_THREADS - int res = ethr_gate_close((ethr_gate *) gp); - if (res != 0) - erts_thr_fatal_error(res, "close gate"); + ethr_event_set(&((ethr_ts_event *) ep)->event); #endif } -ERTS_GLB_INLINE void -erts_gate_let_through(erts_gate_t *gp, unsigned no) +ERTS_GLB_INLINE void erts_tse_reset(erts_tse_t *ep) { #ifdef USE_THREADS - int res = ethr_gate_let_through((ethr_gate *) gp, no); - if (res != 0) - erts_thr_fatal_error(res, "let through gate"); + ethr_event_reset(&((ethr_ts_event *) ep)->event); #endif } -ERTS_GLB_INLINE void -erts_gate_wait(erts_gate_t *gp) +ERTS_GLB_INLINE int erts_tse_wait(erts_tse_t *ep) +{ +#ifdef USE_THREADS + return ethr_event_wait(&((ethr_ts_event *) ep)->event); +#else + return ENOTSUP; +#endif +} + +ERTS_GLB_INLINE int erts_tse_swait(erts_tse_t *ep, int spincount) +{ +#ifdef USE_THREADS + return ethr_event_swait(&((ethr_ts_event *) ep)->event, spincount); +#else + return ENOTSUP; +#endif +} + +ERTS_GLB_INLINE int erts_tse_is_tmp(erts_tse_t *ep) { #ifdef USE_THREADS - int res = ethr_gate_wait((ethr_gate *) gp); + return (ep->iflgs & ETHR_TS_EV_TMP) == ETHR_TS_EV_TMP; +#else + return 0; +#endif +} + +ERTS_GLB_INLINE void erts_thr_set_main_status(int on, int no) +{ +#ifdef USE_THREADS + int res = ethr_set_main_thr_status(on, no); if (res != 0) - erts_thr_fatal_error(res, "wait on gate"); + erts_thr_fatal_error(res, "set thread main status"); #endif } -ERTS_GLB_INLINE void -erts_gate_swait(erts_gate_t *gp, int spincount) +ERTS_GLB_INLINE int erts_thr_get_main_status(void) { #ifdef USE_THREADS - int res = ethr_gate_swait((ethr_gate *) gp, spincount); + int main_status; + int res = ethr_get_main_thr_status(&main_status); if (res != 0) - erts_thr_fatal_error(res, "swait on gate"); + erts_thr_fatal_error(res, "get thread main status"); + return main_status; +#else + return 1; #endif } +ERTS_GLB_INLINE void erts_thr_yield(void) +{ +#ifdef USE_THREADS + int res = ETHR_YIELD(); + if (res != 0) + erts_thr_fatal_error(res, "yield"); +#endif +} + + #ifdef ETHR_HAVE_ETHR_SIG_FUNCS ERTS_GLB_INLINE void diff --git a/erts/emulator/beam/export.c b/erts/emulator/beam/export.c index 66b05c0e9d..5e81a2d624 100644 --- a/erts/emulator/beam/export.c +++ b/erts/emulator/beam/export.c @@ -43,8 +43,6 @@ static erts_smp_rwmtx_t export_table_lock; /* Locks the secondary export table. #define export_read_unlock() erts_smp_rwmtx_runlock(&export_table_lock) #define export_write_lock() erts_smp_rwmtx_rwlock(&export_table_lock) #define export_write_unlock() erts_smp_rwmtx_rwunlock(&export_table_lock) -#define export_init_lock() erts_smp_rwmtx_init(&export_table_lock, \ - "export_tab") extern BeamInstr* em_call_error_handler; extern BeamInstr* em_call_traced_function; @@ -111,8 +109,12 @@ void init_export_table(void) { HashFunctions f; + erts_smp_rwmtx_opt_t rwmtx_opt = ERTS_SMP_THR_OPTS_DEFAULT_INITER; + rwmtx_opt.type = ERTS_SMP_RWMTX_TYPE_FREQUENT_READ; + rwmtx_opt.lived = ERTS_SMP_RWMTX_LONG_LIVED; + + erts_smp_rwmtx_init_opt(&export_table_lock, &rwmtx_opt, "export_tab"); - export_init_lock(); f.hash = (H_FUN) export_hash; f.cmp = (HCMP_FUN) export_cmp; f.alloc = (HALLOC_FUN) export_alloc; diff --git a/erts/emulator/beam/register.c b/erts/emulator/beam/register.c index 900ebcbbf7..c9bb7bbe91 100644 --- a/erts/emulator/beam/register.c +++ b/erts/emulator/beam/register.c @@ -39,8 +39,6 @@ static Hash process_reg; static erts_smp_rwmtx_t regtab_rwmtx; -#define reg_lock_init() erts_smp_rwmtx_init(®tab_rwmtx, \ - "reg_tab") #define reg_try_read_lock() erts_smp_rwmtx_tryrlock(®tab_rwmtx) #define reg_try_write_lock() erts_smp_rwmtx_tryrwlock(®tab_rwmtx) #define reg_read_lock() erts_smp_rwmtx_rlock(®tab_rwmtx) @@ -147,8 +145,11 @@ static void reg_free(RegProc *obj) void init_register_table(void) { HashFunctions f; + erts_smp_rwmtx_opt_t rwmtx_opt = ERTS_SMP_THR_OPTS_DEFAULT_INITER; + rwmtx_opt.type = ERTS_SMP_RWMTX_TYPE_FREQUENT_READ; + rwmtx_opt.lived = ERTS_SMP_RWMTX_LONG_LIVED; - reg_lock_init(); + erts_smp_rwmtx_init_opt(®tab_rwmtx, &rwmtx_opt, "reg_tab"); f.hash = (H_FUN) reg_hash; f.cmp = (HCMP_FUN) reg_cmp; diff --git a/erts/emulator/sys/unix/sys.c b/erts/emulator/sys/unix/sys.c index 737ffd9f94..400ef6c0ce 100644 --- a/erts/emulator/sys/unix/sys.c +++ b/erts/emulator/sys/unix/sys.c @@ -384,18 +384,6 @@ MALLOC_USE_HASH(1); #endif #ifdef USE_THREADS -static void *ethr_internal_alloc(size_t size) -{ - return erts_alloc_fnf(ERTS_ALC_T_ETHR_INTERNAL, (Uint) size); -} -static void *ethr_internal_realloc(void *ptr, size_t size) -{ - return erts_realloc_fnf(ERTS_ALC_T_ETHR_INTERNAL, ptr, (Uint) size); -} -static void ethr_internal_free(void *ptr) -{ - erts_free(ERTS_ALC_T_ETHR_INTERNAL, ptr); -} #ifdef ERTS_THR_HAVE_SIG_FUNCS /* @@ -488,9 +476,6 @@ erts_sys_pre_init(void) #ifdef USE_THREADS { erts_thr_init_data_t eid = ERTS_THR_INIT_DATA_DEF_INITER; - eid.alloc = ethr_internal_alloc; - eid.realloc = ethr_internal_realloc; - eid.free = ethr_internal_free; eid.thread_create_child_func = thr_create_prepare_child; /* Before creation in parent */ @@ -538,13 +523,14 @@ erts_sys_pre_init(void) #endif #endif /* USE_THREADS */ erts_smp_atomic_init(&sys_misc_mem_sz, 0); - erts_smp_rwmtx_init(&environ_rwmtx, "environ"); } void erl_sys_init(void) { + erts_smp_rwmtx_init(&environ_rwmtx, "environ"); #if !DISABLE_VFORK + { int res; char bindir[MAXPATHLEN]; size_t bindirsz = sizeof(bindir); @@ -574,6 +560,7 @@ erl_sys_init(void) bindir, DIR_SEPARATOR_CHAR, CHILD_SETUP_PROG_NAME); + } #endif #ifdef USE_SETLINEBUF diff --git a/erts/emulator/sys/win32/sys.c b/erts/emulator/sys/win32/sys.c index bd02cf85eb..54f71b202d 100644 --- a/erts/emulator/sys/win32/sys.c +++ b/erts/emulator/sys/win32/sys.c @@ -2974,19 +2974,6 @@ check_supported_os_version(void) } #ifdef USE_THREADS -static void *ethr_internal_alloc(size_t size) -{ - return erts_alloc_fnf(ERTS_ALC_T_ETHR_INTERNAL, (Uint) size); -} -static void *ethr_internal_realloc(void *ptr, size_t size) -{ - return erts_realloc_fnf(ERTS_ALC_T_ETHR_INTERNAL, ptr, (Uint) size); -} -static void ethr_internal_free(void *ptr) -{ - erts_free(ERTS_ALC_T_ETHR_INTERNAL, ptr); -} - #ifdef ERTS_ENABLE_LOCK_COUNT static void thr_create_prepare_child(void *vtcdp) @@ -3005,14 +2992,9 @@ erts_sys_pre_init(void) #ifdef USE_THREADS { erts_thr_init_data_t eid = ERTS_THR_INIT_DATA_DEF_INITER; - eid.alloc = ethr_internal_alloc; - eid.realloc = ethr_internal_realloc; - eid.free = ethr_internal_free; - #ifdef ERTS_ENABLE_LOCK_COUNT eid.thread_create_child_func = thr_create_prepare_child; #endif - erts_thr_init(&eid); #ifdef ERTS_ENABLE_LOCK_COUNT erts_lcnt_init(); diff --git a/erts/emulator/test/scheduler_SUITE.erl b/erts/emulator/test/scheduler_SUITE.erl index ab727b7cb1..4f4458802c 100644 --- a/erts/emulator/test/scheduler_SUITE.erl +++ b/erts/emulator/test/scheduler_SUITE.erl @@ -33,7 +33,7 @@ -include("test_server.hrl"). %-compile(export_all). --export([all/1, init_per_testcase/2, fin_per_testcase/2]). +-export([all/1, init_per_testcase/2, fin_per_testcase/2, end_per_suite/1]). -export([equal/1, few_low/1, @@ -49,7 +49,8 @@ cpu_topology/1, sct_cmd/1, sbt_cmd/1, - scheduler_suspend/1]). + scheduler_suspend/1, + reader_groups/1]). -define(DEFAULT_TIMEOUT, ?t:minutes(10)). @@ -67,7 +68,8 @@ all(suite) -> equal_with_high_max, bound_process, scheduler_bind, - scheduler_suspend]. + scheduler_suspend, + reader_groups]. init_per_testcase(Case, Config) when is_list(Config) -> Dog = ?t:timetrap(?DEFAULT_TIMEOUT), @@ -81,6 +83,10 @@ fin_per_testcase(_Case, Config) when is_list(Config) -> ?t:timetrap_cancel(Dog), ok. +end_per_suite(Config) -> + catch erts_debug:set_internal_state(available_internal_state, false), + Config. + -define(ERTS_RUNQ_CHECK_BALANCE_REDS_PER_SCHED, (2000*2000)). -define(DEFAULT_TEST_REDS_PER_SCHED, 200000000). @@ -965,12 +971,361 @@ sst3_loop(S, N) -> erlang:system_flag(schedulers_online, 1), erlang:system_flag(schedulers_online, S), sst3_loop(S, N-1). + +reader_groups(Config) when is_list(Config) -> + %% White box testing. These results are correct, but other results + %% could be too... + + %% The actual tilepro64 topology + CPUT0 = [{processor,[{node,[{core,{logical,0}}, + {core,{logical,1}}, + {core,{logical,2}}, + {core,{logical,8}}, + {core,{logical,9}}, + {core,{logical,10}}, + {core,{logical,11}}, + {core,{logical,16}}, + {core,{logical,17}}, + {core,{logical,18}}, + {core,{logical,19}}, + {core,{logical,24}}, + {core,{logical,25}}, + {core,{logical,27}}, + {core,{logical,29}}]}, + {node,[{core,{logical,3}}, + {core,{logical,4}}, + {core,{logical,5}}, + {core,{logical,6}}, + {core,{logical,7}}, + {core,{logical,12}}, + {core,{logical,13}}, + {core,{logical,14}}, + {core,{logical,15}}, + {core,{logical,20}}, + {core,{logical,21}}, + {core,{logical,22}}, + {core,{logical,23}}, + {core,{logical,28}}, + {core,{logical,30}}]}, + {node,[{core,{logical,31}}, + {core,{logical,36}}, + {core,{logical,37}}, + {core,{logical,38}}, + {core,{logical,44}}, + {core,{logical,45}}, + {core,{logical,46}}, + {core,{logical,47}}, + {core,{logical,51}}, + {core,{logical,52}}, + {core,{logical,53}}, + {core,{logical,54}}, + {core,{logical,55}}, + {core,{logical,60}}, + {core,{logical,61}}]}, + {node,[{core,{logical,26}}, + {core,{logical,32}}, + {core,{logical,33}}, + {core,{logical,34}}, + {core,{logical,35}}, + {core,{logical,39}}, + {core,{logical,40}}, + {core,{logical,41}}, + {core,{logical,42}}, + {core,{logical,43}}, + {core,{logical,48}}, + {core,{logical,49}}, + {core,{logical,50}}, + {core,{logical,58}}]}]}], + + ?line [{0,1},{1,1},{2,1},{3,3},{4,3},{5,3},{6,3},{7,3},{8,1},{9,1},{10,1}, + {11,1},{12,3},{13,3},{14,4},{15,4},{16,2},{17,2},{18,2},{19,2}, + {20,4},{21,4},{22,4},{23,4},{24,2},{25,2},{26,7},{27,2},{28,4}, + {29,2},{30,4},{31,5},{32,7},{33,7},{34,7},{35,7},{36,5},{37,5}, + {38,5},{39,7},{40,7},{41,8},{42,8},{43,8},{44,5},{45,5},{46,5}, + {47,6},{48,8},{49,8},{50,8},{51,6},{52,6},{53,6},{54,6},{55,6}, + {58,8},{60,6},{61,6}] + = reader_groups_map(CPUT0, 8), + + CPUT1 = [n([p([c([t(l(0)),t(l(1)),t(l(2)),t(l(3))]), + c([t(l(4)),t(l(5)),t(l(6)),t(l(7))]), + c([t(l(8)),t(l(9)),t(l(10)),t(l(11))]), + c([t(l(12)),t(l(13)),t(l(14)),t(l(15))])]), + p([c([t(l(16)),t(l(17)),t(l(18)),t(l(19))]), + c([t(l(20)),t(l(21)),t(l(22)),t(l(23))]), + c([t(l(24)),t(l(25)),t(l(26)),t(l(27))]), + c([t(l(28)),t(l(29)),t(l(30)),t(l(31))])])]), + n([p([c([t(l(32)),t(l(33)),t(l(34)),t(l(35))]), + c([t(l(36)),t(l(37)),t(l(38)),t(l(39))]), + c([t(l(40)),t(l(41)),t(l(42)),t(l(43))]), + c([t(l(44)),t(l(45)),t(l(46)),t(l(47))])]), + p([c([t(l(48)),t(l(49)),t(l(50)),t(l(51))]), + c([t(l(52)),t(l(53)),t(l(54)),t(l(55))]), + c([t(l(56)),t(l(57)),t(l(58)),t(l(59))]), + c([t(l(60)),t(l(61)),t(l(62)),t(l(63))])])]), + n([p([c([t(l(64)),t(l(65)),t(l(66)),t(l(67))]), + c([t(l(68)),t(l(69)),t(l(70)),t(l(71))]), + c([t(l(72)),t(l(73)),t(l(74)),t(l(75))]), + c([t(l(76)),t(l(77)),t(l(78)),t(l(79))])]), + p([c([t(l(80)),t(l(81)),t(l(82)),t(l(83))]), + c([t(l(84)),t(l(85)),t(l(86)),t(l(87))]), + c([t(l(88)),t(l(89)),t(l(90)),t(l(91))]), + c([t(l(92)),t(l(93)),t(l(94)),t(l(95))])])]), + n([p([c([t(l(96)),t(l(97)),t(l(98)),t(l(99))]), + c([t(l(100)),t(l(101)),t(l(102)),t(l(103))]), + c([t(l(104)),t(l(105)),t(l(106)),t(l(107))]), + c([t(l(108)),t(l(109)),t(l(110)),t(l(111))])]), + p([c([t(l(112)),t(l(113)),t(l(114)),t(l(115))]), + c([t(l(116)),t(l(117)),t(l(118)),t(l(119))]), + c([t(l(120)),t(l(121)),t(l(122)),t(l(123))]), + c([t(l(124)),t(l(125)),t(l(126)),t(l(127))])])])], + + ?line [{0,1},{1,1},{2,1},{3,1},{4,2},{5,2},{6,2},{7,2},{8,3},{9,3}, + {10,3},{11,3},{12,4},{13,4},{14,4},{15,4},{16,5},{17,5},{18,5}, + {19,5},{20,6},{21,6},{22,6},{23,6},{24,7},{25,7},{26,7},{27,7}, + {28,8},{29,8},{30,8},{31,8},{32,9},{33,9},{34,9},{35,9},{36,10}, + {37,10},{38,10},{39,10},{40,11},{41,11},{42,11},{43,11},{44,12}, + {45,12},{46,12},{47,12},{48,13},{49,13},{50,13},{51,13},{52,14}, + {53,14},{54,14},{55,14},{56,15},{57,15},{58,15},{59,15},{60,16}, + {61,16},{62,16},{63,16},{64,17},{65,17},{66,17},{67,17},{68,18}, + {69,18},{70,18},{71,18},{72,19},{73,19},{74,19},{75,19},{76,20}, + {77,20},{78,20},{79,20},{80,21},{81,21},{82,21},{83,21},{84,22}, + {85,22},{86,22},{87,22},{88,23},{89,23},{90,23},{91,23},{92,24}, + {93,24},{94,24},{95,24},{96,25},{97,25},{98,25},{99,25},{100,26}, + {101,26},{102,26},{103,26},{104,27},{105,27},{106,27},{107,27}, + {108,28},{109,28},{110,28},{111,28},{112,29},{113,29},{114,29}, + {115,29},{116,30},{117,30},{118,30},{119,30},{120,31},{121,31}, + {122,31},{123,31},{124,32},{125,32},{126,32},{127,32}] + = reader_groups_map(CPUT1, 128), + + ?line [{0,1},{1,1},{2,1},{3,1},{4,1},{5,1},{6,1},{7,1},{8,1},{9,1},{10,1}, + {11,1},{12,1},{13,1},{14,1},{15,1},{16,1},{17,1},{18,1},{19,1}, + {20,1},{21,1},{22,1},{23,1},{24,1},{25,1},{26,1},{27,1},{28,1}, + {29,1},{30,1},{31,1},{32,1},{33,1},{34,1},{35,1},{36,1},{37,1}, + {38,1},{39,1},{40,1},{41,1},{42,1},{43,1},{44,1},{45,1},{46,1}, + {47,1},{48,1},{49,1},{50,1},{51,1},{52,1},{53,1},{54,1},{55,1}, + {56,1},{57,1},{58,1},{59,1},{60,1},{61,1},{62,1},{63,1},{64,2}, + {65,2},{66,2},{67,2},{68,2},{69,2},{70,2},{71,2},{72,2},{73,2}, + {74,2},{75,2},{76,2},{77,2},{78,2},{79,2},{80,2},{81,2},{82,2}, + {83,2},{84,2},{85,2},{86,2},{87,2},{88,2},{89,2},{90,2},{91,2}, + {92,2},{93,2},{94,2},{95,2},{96,2},{97,2},{98,2},{99,2},{100,2}, + {101,2},{102,2},{103,2},{104,2},{105,2},{106,2},{107,2},{108,2}, + {109,2},{110,2},{111,2},{112,2},{113,2},{114,2},{115,2},{116,2}, + {117,2},{118,2},{119,2},{120,2},{121,2},{122,2},{123,2},{124,2}, + {125,2},{126,2},{127,2}] + = reader_groups_map(CPUT1, 2), + + ?line [{0,1},{1,1},{2,1},{3,1},{4,2},{5,2},{6,2},{7,2},{8,3},{9,3},{10,3}, + {11,3},{12,3},{13,3},{14,3},{15,3},{16,4},{17,4},{18,4},{19,4}, + {20,4},{21,4},{22,4},{23,4},{24,5},{25,5},{26,5},{27,5},{28,5}, + {29,5},{30,5},{31,5},{32,6},{33,6},{34,6},{35,6},{36,6},{37,6}, + {38,6},{39,6},{40,7},{41,7},{42,7},{43,7},{44,7},{45,7},{46,7}, + {47,7},{48,8},{49,8},{50,8},{51,8},{52,8},{53,8},{54,8},{55,8}, + {56,9},{57,9},{58,9},{59,9},{60,9},{61,9},{62,9},{63,9},{64,10}, + {65,10},{66,10},{67,10},{68,10},{69,10},{70,10},{71,10},{72,11}, + {73,11},{74,11},{75,11},{76,11},{77,11},{78,11},{79,11},{80,12}, + {81,12},{82,12},{83,12},{84,12},{85,12},{86,12},{87,12},{88,13}, + {89,13},{90,13},{91,13},{92,13},{93,13},{94,13},{95,13},{96,14}, + {97,14},{98,14},{99,14},{100,14},{101,14},{102,14},{103,14}, + {104,15},{105,15},{106,15},{107,15},{108,15},{109,15},{110,15}, + {111,15},{112,16},{113,16},{114,16},{115,16},{116,16},{117,16}, + {118,16},{119,16},{120,17},{121,17},{122,17},{123,17},{124,17}, + {125,17},{126,17},{127,17}] + = reader_groups_map(CPUT1, 17), + + ?line [{0,1},{1,1},{2,1},{3,1},{4,1},{5,1},{6,1},{7,1},{8,1},{9,1},{10,1}, + {11,1},{12,1},{13,1},{14,1},{15,1},{16,2},{17,2},{18,2},{19,2}, + {20,2},{21,2},{22,2},{23,2},{24,2},{25,2},{26,2},{27,2},{28,2}, + {29,2},{30,2},{31,2},{32,3},{33,3},{34,3},{35,3},{36,3},{37,3}, + {38,3},{39,3},{40,3},{41,3},{42,3},{43,3},{44,3},{45,3},{46,3}, + {47,3},{48,4},{49,4},{50,4},{51,4},{52,4},{53,4},{54,4},{55,4}, + {56,4},{57,4},{58,4},{59,4},{60,4},{61,4},{62,4},{63,4},{64,5}, + {65,5},{66,5},{67,5},{68,5},{69,5},{70,5},{71,5},{72,5},{73,5}, + {74,5},{75,5},{76,5},{77,5},{78,5},{79,5},{80,6},{81,6},{82,6}, + {83,6},{84,6},{85,6},{86,6},{87,6},{88,6},{89,6},{90,6},{91,6}, + {92,6},{93,6},{94,6},{95,6},{96,7},{97,7},{98,7},{99,7},{100,7}, + {101,7},{102,7},{103,7},{104,7},{105,7},{106,7},{107,7},{108,7}, + {109,7},{110,7},{111,7},{112,7},{113,7},{114,7},{115,7},{116,7}, + {117,7},{118,7},{119,7},{120,7},{121,7},{122,7},{123,7},{124,7}, + {125,7},{126,7},{127,7}] + = reader_groups_map(CPUT1, 7), + + ?line CPUT2 = [p([c(l(0)),c(l(1)),c(l(2)),c(l(3)),c(l(4))]), + p([t(l(5)),t(l(6)),t(l(7)),t(l(8)),t(l(9))]), + p([t(l(10))]), + p([c(l(11)),c(l(12)),c(l(13))]), + p([c(l(14)),c(l(15))])], + + ?line [{0,1},{1,1},{2,1},{3,1},{4,1}, + {5,2},{6,2},{7,2},{8,2},{9,2}, + {10,3}, + {11,4},{12,4},{13,4}, + {14,5},{15,5}] = reader_groups_map(CPUT2, 5), + + + ?line [{0,1},{1,1},{2,2},{3,2},{4,2}, + {5,3},{6,3},{7,3},{8,3},{9,3}, + {10,4}, + {11,5},{12,5},{13,5}, + {14,6},{15,6}] = reader_groups_map(CPUT2, 6), + + ?line [{0,1},{1,1},{2,2},{3,2},{4,2}, + {5,3},{6,3},{7,3},{8,3},{9,3}, + {10,4}, + {11,5},{12,6},{13,6}, + {14,7},{15,7}] = reader_groups_map(CPUT2, 7), + + ?line [{0,1},{1,1},{2,2},{3,2},{4,2}, + {5,3},{6,3},{7,3},{8,3},{9,3}, + {10,4}, + {11,5},{12,6},{13,6}, + {14,7},{15,8}] = reader_groups_map(CPUT2, 8), + + ?line [{0,1},{1,2},{2,2},{3,3},{4,3}, + {5,4},{6,4},{7,4},{8,4},{9,4}, + {10,5}, + {11,6},{12,7},{13,7}, + {14,8},{15,9}] = reader_groups_map(CPUT2, 9), + + ?line [{0,1},{1,2},{2,2},{3,3},{4,3}, + {5,4},{6,4},{7,4},{8,4},{9,4}, + {10,5}, + {11,6},{12,7},{13,8}, + {14,9},{15,10}] = reader_groups_map(CPUT2, 10), + + ?line [{0,1},{1,2},{2,3},{3,4},{4,4}, + {5,5},{6,5},{7,5},{8,5},{9,5}, + {10,6}, + {11,7},{12,8},{13,9}, + {14,10},{15,11}] = reader_groups_map(CPUT2, 11), + + ?line [{0,1},{1,2},{2,3},{3,4},{4,5}, + {5,6},{6,6},{7,6},{8,6},{9,6}, + {10,7}, + {11,8},{12,9},{13,10}, + {14,11},{15,12}] = reader_groups_map(CPUT2, 100), + + CPUT3 = [p([t(l(5)),t(l(6)),t(l(7)),t(l(8)),t(l(9))]), + p([t(l(10))]), + p([c(l(11)),c(l(12)),c(l(13))]), + p([c(l(14)),c(l(15))]), + p([c(l(0)),c(l(1)),c(l(2)),c(l(3)),c(l(4))])], + + ?line [{0,5},{1,5},{2,6},{3,6},{4,6}, + {5,1},{6,1},{7,1},{8,1},{9,1}, + {10,2},{11,3},{12,3},{13,3}, + {14,4},{15,4}] = reader_groups_map(CPUT3, 6), + + CPUT4 = [p([t(l(0)),t(l(1)),t(l(2)),t(l(3)),t(l(4))]), + p([t(l(5))]), + p([c(l(6)),c(l(7)),c(l(8))]), + p([c(l(9)),c(l(10))]), + p([c(l(11)),c(l(12)),c(l(13)),c(l(14)),c(l(15))])], + + ?line [{0,1},{1,1},{2,1},{3,1},{4,1}, + {5,2}, + {6,3},{7,3},{8,3}, + {9,4},{10,4}, + {11,5},{12,5},{13,6},{14,6},{15,6}] = reader_groups_map(CPUT4, 6), + + ?line [{0,1},{1,1},{2,1},{3,1},{4,1}, + {5,2}, + {6,3},{7,4},{8,4}, + {9,5},{10,5}, + {11,6},{12,6},{13,7},{14,7},{15,7}] = reader_groups_map(CPUT4, 7), + + ?line [{0,1},{65535,2}] = reader_groups_map([c(l(0)),c(l(65535))], 10), + + ?line ok. +reader_groups_map(CPUT, Groups) -> + Old = erlang:system_info({cpu_topology, defined}), + erlang:system_flag(cpu_topology, CPUT), + enable_internal_state(), + Res = erts_debug:get_internal_state({reader_groups_map, Groups}), + erlang:system_flag(cpu_topology, Old), + lists:sort(Res). + %% %% Utils %% +tilera_cpu_topology() -> + [{processor,[{node,[{core,{logical,0}}, + {core,{logical,1}}, + {core,{logical,2}}, + {core,{logical,8}}, + {core,{logical,9}}, + {core,{logical,10}}, + {core,{logical,11}}, + {core,{logical,16}}, + {core,{logical,17}}, + {core,{logical,18}}, + {core,{logical,19}}, + {core,{logical,24}}, + {core,{logical,25}}, + {core,{logical,27}}, + {core,{logical,29}}]}, + {node,[{core,{logical,3}}, + {core,{logical,4}}, + {core,{logical,5}}, + {core,{logical,6}}, + {core,{logical,7}}, + {core,{logical,12}}, + {core,{logical,13}}, + {core,{logical,14}}, + {core,{logical,15}}, + {core,{logical,20}}, + {core,{logical,21}}, + {core,{logical,22}}, + {core,{logical,23}}, + {core,{logical,28}}, + {core,{logical,30}}]}, + {node,[{core,{logical,31}}, + {core,{logical,36}}, + {core,{logical,37}}, + {core,{logical,38}}, + {core,{logical,44}}, + {core,{logical,45}}, + {core,{logical,46}}, + {core,{logical,47}}, + {core,{logical,51}}, + {core,{logical,52}}, + {core,{logical,53}}, + {core,{logical,54}}, + {core,{logical,55}}, + {core,{logical,60}}, + {core,{logical,61}}]}, + {node,[{core,{logical,26}}, + {core,{logical,32}}, + {core,{logical,33}}, + {core,{logical,34}}, + {core,{logical,35}}, + {core,{logical,39}}, + {core,{logical,40}}, + {core,{logical,41}}, + {core,{logical,42}}, + {core,{logical,43}}, + {core,{logical,48}}, + {core,{logical,49}}, + {core,{logical,50}}, + {core,{logical,58}}]}]}]. + +l(Id) -> + {logical, Id}. + +t(X) -> + {thread, X}. + +c(X) -> + {core, X}. + +p(X) -> + {processor, X}. + +n(X) -> + {node, X}. + mcall(Node, Funs) -> Parent = self(), Refs = lists:map(fun (Fun) -> |