ed/raddebugger
1
0
Fork 0
mirror of https://github.com/Ed94/raddebugger.git synced 2026-07-14 21:41:25 -07:00
Code Issues Packages Projects Releases Wiki Activity

switch match tasks to atomically update match nodes, rather than push results into a queue

Browse Source
This commit is contained in:
Ryan Fleury
2024-11-12 13:00:02 -08:00
parent 0ec6b8e8a8
commit 9808157f45
3 changed files with 33 additions and 165 deletions
Download Patch File Download Diff File Expand all files Collapse all files
src/dbgi/dbgi.c
+25 -155
View File
@@ -1509,14 +1509,12 @@ di_match_store_alloc(void)
store->params_rw_mutex = os_rw_mutex_alloc();
store->match_name_slots_count = 4096;
store->match_name_slots = push_array(arena, DI_MatchNameSlot, store->match_name_slots_count);
store->match_rw_mutex = os_rw_mutex_alloc();
store->match_cv = os_condition_variable_alloc();
store->u2m_ring_cv = os_condition_variable_alloc();
store->u2m_ring_mutex = os_mutex_alloc();
store->u2m_ring_size = KB(2);
store->u2m_ring_base = push_array_no_zero(arena, U8, store->u2m_ring_size);
store->m2u_ring_cv = os_condition_variable_alloc();
store->m2u_ring_mutex = os_mutex_alloc();
store->m2u_ring_size = KB(2);
store->m2u_ring_base = push_array_no_zero(arena, U8, store->m2u_ring_size);
return store;
}
@@ -1550,6 +1548,7 @@ di_match_store_begin(DI_MatchStore *store, DI_KeyArray keys)
prev = node->prev;
if(node->last_gen_touched+64 < store->gen)
{
node->alloc_gen += 1;
U64 slot_idx = node->hash%store->match_name_slots_count;
DI_MatchNameSlot *slot = &store->match_name_slots[slot_idx];
DLLRemove_NP(store->first_lru_match_name, store->last_lru_match_name, node, lru_next, lru_prev);
@@ -1564,67 +1563,6 @@ di_match_store_begin(DI_MatchStore *store, DI_KeyArray keys)
}
}
// rjf: consume & store pending results
{
U64 endt_us = os_now_microseconds() + 1000000;
for(B32 done = 0; !done;)
{
Temp scratch = scratch_begin(0, 0);
// rjf: get next result
String8 result_name = {0};
U64 result_params_hash = 0;
U64 result_dbgi_idx = 0;
RDI_SectionKind result_section_kind = RDI_SectionKind_NULL;
U32 result_idx = 0;
OS_MutexScope(store->m2u_ring_mutex)
{
U64 unconsumed_size = store->m2u_ring_write_pos - store->m2u_ring_read_pos;
if(unconsumed_size >= sizeof(U64)*4)
{
store->m2u_ring_read_pos += ring_read_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, &result_name.size);
result_name.str = push_array(scratch.arena, U8, result_name.size);
store->m2u_ring_read_pos += ring_read(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, result_name.str, result_name.size);
store->m2u_ring_read_pos += 7;
store->m2u_ring_read_pos -= store->m2u_ring_read_pos%8;
store->m2u_ring_read_pos += ring_read_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, &result_params_hash);
store->m2u_ring_read_pos += ring_read_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, &result_dbgi_idx);
store->m2u_ring_read_pos += ring_read_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, &result_section_kind);
store->m2u_ring_read_pos += ring_read_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, &result_idx);
}
}
os_condition_variable_broadcast(store->m2u_ring_cv);
// rjf: store result
if(result_name.size == 0)
{
done = 1;
}
else
{
U64 result_hash = di_hash_from_string(result_name, 0);
U64 result_slot_idx = result_hash%store->match_name_slots_count;
DI_MatchNameSlot *result_slot = &store->match_name_slots[result_slot_idx];
for(DI_MatchNameNode *n = result_slot->first; n != 0; n = n->next)
{
if(n->hash == result_hash && str8_match(result_name, n->name, 0))
{
n->cmp_params_hash = result_params_hash;
n->section_kind = result_section_kind;
break;
}
}
}
// rjf: timeout
if(os_now_microseconds() >= endt_us)
{
done = 1;
}
scratch_end(scratch);
}
}
ProfEnd();
}
@@ -1653,9 +1591,11 @@ di_match_store_section_kind_from_name(DI_MatchStore *store, String8 name, U64 en
if(node == 0)
{
node = store->first_free_match_name;
U64 alloc_gen = 0;
if(node)
{
SLLStackPop(store->first_free_match_name);
alloc_gen = node->alloc_gen;
}
else
{
@@ -1663,6 +1603,7 @@ di_match_store_section_kind_from_name(DI_MatchStore *store, String8 name, U64 en
}
MemoryZeroStruct(node);
node->hash = hash;
node->alloc_gen = alloc_gen + 1;
DLLPushBack(slot->first, slot->last, node);
node->first_gen_touched = store->gen;
DLLInsert_NP(store->first_lru_match_name, store->last_lru_match_name, (DI_MatchNameNode *)0, node, lru_next, lru_prev);
@@ -1675,7 +1616,7 @@ di_match_store_section_kind_from_name(DI_MatchStore *store, String8 name, U64 en
DLLRemove_NP(store->first_lru_match_name, store->last_lru_match_name, node, lru_next, lru_prev);
DLLInsert_NP(store->first_lru_match_name, store->last_lru_match_name, (DI_MatchNameNode *)0, node, lru_next, lru_prev);
// rjf: if this node is new, request it for the given parameters
// rjf: if this node is new w.r.t. the store's current parameters, request it
if(node->req_params_hash != store->params_hash)
{
B32 sent = 0;
@@ -1685,6 +1626,8 @@ di_match_store_section_kind_from_name(DI_MatchStore *store, String8 name, U64 en
U64 available_size = store->u2m_ring_size - unconsumed_size;
if(available_size >= sizeof(U64) + name.size)
{
store->u2m_ring_write_pos += ring_write_struct(store->u2m_ring_base, store->u2m_ring_size, store->u2m_ring_write_pos, &node);
store->u2m_ring_write_pos += ring_write_struct(store->u2m_ring_base, store->u2m_ring_size, store->u2m_ring_write_pos, &node->alloc_gen);
store->u2m_ring_write_pos += ring_write_struct(store->u2m_ring_base, store->u2m_ring_size, store->u2m_ring_write_pos, &name.size);
store->u2m_ring_write_pos += ring_write(store->u2m_ring_base, store->u2m_ring_size, store->u2m_ring_write_pos, name.str, name.size);
store->u2m_ring_write_pos += 7;
@@ -1706,70 +1649,20 @@ di_match_store_section_kind_from_name(DI_MatchStore *store, String8 name, U64 en
}
}
// rjf: if this node's state is stale, consume results from match work & store them
if(node->req_params_hash != node->cmp_params_hash)
// rjf: if this node's state is stale, wait for it if we need to
if(os_now_microseconds() < endt_us && node->req_params_hash != ins_atomic_u64_eval(&node->cmp_params_hash))
{
for(B32 done = 0; !done;)
OS_MutexScopeR(store->match_rw_mutex) for(;;)
{
Temp scratch = scratch_begin(0, 0);
// rjf: get next result
String8 result_name = {0};
U64 result_params_hash = 0;
U64 result_dbgi_idx = 0;
RDI_SectionKind result_section_kind = RDI_SectionKind_NULL;
U32 result_idx = 0;
OS_MutexScope(store->m2u_ring_mutex) for(;;)
if(node->req_params_hash == ins_atomic_u64_eval(&node->cmp_params_hash))
{
U64 unconsumed_size = store->m2u_ring_write_pos - store->m2u_ring_read_pos;
if(unconsumed_size >= sizeof(U64)*4)
{
store->m2u_ring_read_pos += ring_read_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, &result_name.size);
result_name.str = push_array(scratch.arena, U8, result_name.size);
store->m2u_ring_read_pos += ring_read(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, result_name.str, result_name.size);
store->m2u_ring_read_pos += 7;
store->m2u_ring_read_pos -= store->m2u_ring_read_pos%8;
store->m2u_ring_read_pos += ring_read_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, &result_params_hash);
store->m2u_ring_read_pos += ring_read_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, &result_dbgi_idx);
store->m2u_ring_read_pos += ring_read_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, &result_section_kind);
store->m2u_ring_read_pos += ring_read_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_read_pos, &result_idx);
break;
}
if(os_now_microseconds() >= endt_us)
{
break;
}
os_condition_variable_wait(store->m2u_ring_cv, store->m2u_ring_mutex, endt_us);
break;
}
os_condition_variable_broadcast(store->m2u_ring_cv);
// rjf: store result
U64 result_hash = di_hash_from_string(result_name, 0);
U64 result_slot_idx = result_hash%store->match_name_slots_count;
DI_MatchNameSlot *result_slot = &store->match_name_slots[result_slot_idx];
for(DI_MatchNameNode *n = result_slot->first; n != 0; n = n->next)
{
if(n->hash == result_hash && str8_match(result_name, n->name, 0))
{
n->cmp_params_hash = result_params_hash;
n->section_kind = result_section_kind;
break;
}
}
// rjf: we're done if we got the hash we were looking for
if(result_hash == hash && result_params_hash == store->params_hash)
{
done = 1;
}
// rjf: we're done if we're out of time
if(os_now_microseconds() >= endt_us)
{
done = 1;
break;
}
scratch_end(scratch);
os_condition_variable_wait_rw_r(store->match_cv, store->match_rw_mutex, endt_us);
}
}
@@ -1785,13 +1678,17 @@ ASYNC_WORK_DEF(di_match_work)
Temp scratch = scratch_begin(0, 0);
DI_MatchStore *store = (DI_MatchStore *)input;
{
//- rjf: get next name
//- rjf: get next request
DI_MatchNameNode *node = 0;
U64 alloc_gen = 0;
String8 name = {0};
ProfScope("get next name") OS_MutexScope(store->u2m_ring_mutex) for(;;)
{
U64 unconsumed_size = store->u2m_ring_write_pos - store->u2m_ring_read_pos;
if(unconsumed_size >= sizeof(U64))
{
store->u2m_ring_read_pos += ring_read_struct(store->u2m_ring_base, store->u2m_ring_size, store->u2m_ring_read_pos, &node);
store->u2m_ring_read_pos += ring_read_struct(store->u2m_ring_base, store->u2m_ring_size, store->u2m_ring_read_pos, &alloc_gen);
store->u2m_ring_read_pos += ring_read_struct(store->u2m_ring_base, store->u2m_ring_size, store->u2m_ring_read_pos, &name.size);
name.str = push_array(scratch.arena, U8, name.size);
store->u2m_ring_read_pos += ring_read(store->u2m_ring_base, store->u2m_ring_size, store->u2m_ring_read_pos, name.str, name.size);
@@ -1839,39 +1736,12 @@ ASYNC_WORK_DEF(di_match_work)
RDI_NameMap *name_map = rdi_element_from_name_idx(rdi, NameMaps, name_map_kinds[name_map_kind_idx]);
RDI_ParsedNameMap parsed_name_map = {0};
rdi_parsed_from_name_map(rdi, name_map, &parsed_name_map);
RDI_NameMapNode *node = rdi_name_map_lookup(rdi, &parsed_name_map, name.str, name.size);
RDI_NameMapNode *map_node = rdi_name_map_lookup(rdi, &parsed_name_map, name.str, name.size);
U32 num = 0;
U32 *run = rdi_matches_from_map_node(rdi, node, &num);
for(U32 run_idx = 0; run_idx < num; run_idx += 1)
U32 *run = rdi_matches_from_map_node(rdi, map_node, &num);
if(num != 0)
{
ProfScope("write result") OS_MutexScope(store->m2u_ring_mutex) for(;;)
{
U64 unconsumed_size = store->m2u_ring_write_pos - store->m2u_ring_read_pos;
U64 available_size = store->m2u_ring_size - unconsumed_size;
U64 needed_size = 0;
needed_size += sizeof(U64);
needed_size += name.size;
needed_size += 7;
needed_size -= needed_size%8;
needed_size += sizeof(U64);
needed_size += sizeof(U64);
needed_size += sizeof(RDI_SectionKind);
needed_size += sizeof(U32);
if(available_size >= needed_size)
{
store->m2u_ring_write_pos += ring_write_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_write_pos, &name.size);
store->m2u_ring_write_pos += ring_write(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_write_pos, name.str, name.size);
store->m2u_ring_write_pos += 7;
store->m2u_ring_write_pos -= store->m2u_ring_write_pos%8;
store->m2u_ring_write_pos += ring_write_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_write_pos, &params_hash);
store->m2u_ring_write_pos += ring_write_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_write_pos, &dbgi_idx);
store->m2u_ring_write_pos += ring_write_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_write_pos, &name_map_section_kinds[name_map_kind_idx]);
store->m2u_ring_write_pos += ring_write_struct(store->m2u_ring_base, store->m2u_ring_size, store->m2u_ring_write_pos, &run[run_idx]);
break;
}
os_condition_variable_wait(store->m2u_ring_cv, store->m2u_ring_mutex, max_U64);
}
os_condition_variable_broadcast(store->m2u_ring_cv);
ins_atomic_u32_eval_assign(&node->section_kind, name_map_section_kinds[name_map_kind_idx]);
}
}
di_scope_close(di_scope);
src/dbgi/dbgi.h
+7 -9
View File
@@ -269,16 +269,20 @@ struct DI_Match
typedef struct DI_MatchNameNode DI_MatchNameNode;
struct DI_MatchNameNode
{
// rjf: synchronously written by usage code
DI_MatchNameNode *next;
DI_MatchNameNode *prev;
DI_MatchNameNode *lru_next;
DI_MatchNameNode *lru_prev;
U64 alloc_gen;
U64 first_gen_touched;
U64 last_gen_touched;
U64 req_params_hash;
U64 cmp_params_hash;
String8 name;
U64 hash;
// rjf: atomically written by match work
U64 cmp_params_hash;
RDI_SectionKind section_kind;
// DI_Match *first_match;
// DI_Match *last_match;
@@ -312,6 +316,8 @@ struct DI_MatchStore
DI_MatchNameNode *first_lru_match_name;
DI_MatchNameNode *last_lru_match_name;
U64 active_match_name_nodes_count;
OS_Handle match_rw_mutex;
OS_Handle match_cv;
// rjf: user -> match work ring buffer
OS_Handle u2m_ring_cv;
@@ -320,14 +326,6 @@ struct DI_MatchStore
U8 *u2m_ring_base;
U64 u2m_ring_write_pos;
U64 u2m_ring_read_pos;
// rjf: match work -> user ring buffer
OS_Handle m2u_ring_cv;
OS_Handle m2u_ring_mutex;
U64 m2u_ring_size;
U8 *m2u_ring_base;
U64 m2u_ring_write_pos;
U64 m2u_ring_read_pos;
};
////////////////////////////////
src/raddbg/raddbg_core.c
+1 -1
View File
@@ -8894,7 +8894,7 @@ EV_VIEW_RULE_EXPR_EXPAND_RANGE_INFO_FUNCTION_DEF(scheduler_process)
expr->space = rd_eval_space_from_ctrl_entity(thread, RD_EvalSpaceKind_MetaCtrlEntity);
expr->mode = E_Mode_Offset;
expr->type_key = e_type_key_cons_base(type(CTRL_ThreadMetaEval));;
info.row_strings[row_expr_idx] = thread->string;
info.row_strings[row_expr_idx] = push_str8f(arena, "thread:%S", thread->string);
info.row_exprs[row_expr_idx] = expr;
info.row_members[row_expr_idx] = &e_member_nil;
}