Files
Odin/core/runtime/dynamic_map_internal.odin
gingerBill f98c4d6837 Improve the Allocator interface to support returning Allocator_Error to allow for safer calls
Virtually all code (except for user-written custom allocators) should work as normal. Extra features will need to be added to make the current procedures support the `Allocator_Error` return value (akin to #optional_ok)
2021-04-19 12:31:31 +01:00

384 lines
12 KiB
Odin

package runtime
import "intrinsics"
_ :: intrinsics;
INITIAL_MAP_CAP :: 16;
// Temporary data structure for comparing hashes and keys
Map_Hash :: struct {
hash: uintptr,
key_ptr: rawptr, // address of Map_Entry_Header.key
}
__get_map_hash :: proc "contextless" (k: ^$K) -> (map_hash: Map_Hash) {
hasher := intrinsics.type_hasher_proc(K);
map_hash.key_ptr = k;
map_hash.hash = hasher(k, 0);
return;
}
__get_map_hash_from_entry :: proc "contextless" (h: Map_Header, entry: ^Map_Entry_Header) -> (hash: Map_Hash) {
hash.hash = entry.hash;
hash.key_ptr = rawptr(uintptr(entry) + h.key_offset);
return;
}
Map_Find_Result :: struct {
hash_index: int,
entry_prev: int,
entry_index: int,
}
Map_Entry_Header :: struct {
hash: uintptr,
next: int,
/*
key: Key_Value,
value: Value_Type,
*/
}
Map_Header :: struct {
m: ^Raw_Map,
equal: Equal_Proc,
entry_size: int,
entry_align: int,
key_offset: uintptr,
key_size: int,
value_offset: uintptr,
value_size: int,
}
INITIAL_HASH_SEED :: 0xcbf29ce484222325;
_fnv64a :: proc "contextless" (data: []byte, seed: u64 = INITIAL_HASH_SEED) -> u64 {
h: u64 = seed;
for b in data {
h = (h ~ u64(b)) * 0x100000001b3;
}
return h;
}
default_hash :: #force_inline proc "contextless" (data: []byte) -> uintptr {
return uintptr(_fnv64a(data));
}
default_hash_string :: #force_inline proc "contextless" (s: string) -> uintptr {
return default_hash(transmute([]byte)(s));
}
default_hash_ptr :: #force_inline proc "contextless" (data: rawptr, size: int) -> uintptr {
s := Raw_Slice{data, size};
return default_hash(transmute([]byte)(s));
}
@(private)
_default_hasher_const :: #force_inline proc "contextless" (data: rawptr, seed: uintptr, $N: uint) -> uintptr where N <= 16 {
h := u64(seed) + 0xcbf29ce484222325;
p := uintptr(data);
#unroll for _ in 0..<N {
b := u64((^byte)(p)^);
h = (h ~ b) * 0x100000001b3;
p += 1;
}
return uintptr(h);
}
default_hasher_n :: #force_inline proc "contextless" (data: rawptr, seed: uintptr, N: int) -> uintptr {
h := u64(seed) + 0xcbf29ce484222325;
p := uintptr(data);
for _ in 0..<N {
b := u64((^byte)(p)^);
h = (h ~ b) * 0x100000001b3;
p += 1;
}
return uintptr(h);
}
// NOTE(bill): There are loads of predefined ones to improve optimizations for small types
default_hasher1 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 1); }
default_hasher2 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 2); }
default_hasher3 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 3); }
default_hasher4 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 4); }
default_hasher5 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 5); }
default_hasher6 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 6); }
default_hasher7 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 7); }
default_hasher8 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 8); }
default_hasher9 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 9); }
default_hasher10 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 10); }
default_hasher11 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 11); }
default_hasher12 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 12); }
default_hasher13 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 13); }
default_hasher14 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 14); }
default_hasher15 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 15); }
default_hasher16 :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr { return #force_inline _default_hasher_const(data, seed, 16); }
default_hasher_string :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr {
h := u64(seed) + 0xcbf29ce484222325;
str := (^[]byte)(data)^;
for b in str {
h = (h ~ u64(b)) * 0x100000001b3;
}
return uintptr(h);
}
default_hasher_cstring :: proc "contextless" (data: rawptr, seed: uintptr) -> uintptr {
h := u64(seed) + 0xcbf29ce484222325;
ptr := (^uintptr)(data)^;
for (^byte)(ptr)^ != 0 {
b := (^byte)(ptr)^;
h = (h ~ u64(b)) * 0x100000001b3;
ptr += 1;
}
return uintptr(h);
}
__get_map_header :: proc "contextless" (m: ^$T/map[$K]$V) -> Map_Header {
header := Map_Header{m = (^Raw_Map)(m)};
Entry :: struct {
hash: uintptr,
next: int,
key: K,
value: V,
};
header.equal = intrinsics.type_equal_proc(K);
header.entry_size = size_of(Entry);
header.entry_align = align_of(Entry);
header.key_offset = offset_of(Entry, key);
header.key_size = size_of(K);
header.value_offset = offset_of(Entry, value);
header.value_size = size_of(V);
return header;
}
__slice_resize :: proc(array_: ^$T/[]$E, new_count: int, allocator: Allocator, loc := #caller_location) -> bool {
array := (^Raw_Slice)(array_);
if new_count < array.len {
return true;
}
assert(allocator.procedure != nil);
old_size := array.len*size_of(T);
new_size := new_count*size_of(T);
new_data, err := mem_resize(array.data, old_size, new_size, align_of(T), allocator, loc);
if new_data == nil || err != nil {
return false;
}
array.data = new_data;
array.len = new_count;
return true;
}
__dynamic_map_reserve :: proc(using header: Map_Header, cap: int, loc := #caller_location) {
__dynamic_array_reserve(&m.entries, entry_size, entry_align, cap, loc);
old_len := len(m.hashes);
__slice_resize(&m.hashes, cap, m.entries.allocator, loc);
for i in old_len..<len(m.hashes) {
m.hashes[i] = -1;
}
}
__dynamic_map_rehash :: proc(using header: Map_Header, new_count: int, loc := #caller_location) #no_bounds_check {
new_header: Map_Header = header;
nm := Raw_Map{};
nm.entries.allocator = m.entries.allocator;
new_header.m = &nm;
c := context;
if m.entries.allocator.procedure != nil {
c.allocator = m.entries.allocator;
}
context = c;
new_count := new_count;
new_count = max(new_count, 2*m.entries.len);
__dynamic_array_reserve(&nm.entries, entry_size, entry_align, m.entries.len, loc);
__slice_resize(&nm.hashes, new_count, m.entries.allocator, loc);
for i in 0 ..< new_count {
nm.hashes[i] = -1;
}
for i in 0 ..< m.entries.len {
if len(nm.hashes) == 0 {
__dynamic_map_grow(new_header, loc);
}
entry_header := __dynamic_map_get_entry(header, i);
entry_hash := __get_map_hash_from_entry(header, entry_header);
fr := __dynamic_map_find(new_header, entry_hash);
j := __dynamic_map_add_entry(new_header, entry_hash, loc);
if fr.entry_prev < 0 {
nm.hashes[fr.hash_index] = j;
} else {
e := __dynamic_map_get_entry(new_header, fr.entry_prev);
e.next = j;
}
e := __dynamic_map_get_entry(new_header, j);
__dynamic_map_copy_entry(header, e, entry_header);
e.next = fr.entry_index;
if __dynamic_map_full(new_header) {
__dynamic_map_grow(new_header, loc);
}
}
delete(m.hashes, m.entries.allocator, loc);
free(m.entries.data, m.entries.allocator, loc);
header.m^ = nm;
}
__dynamic_map_get :: proc(h: Map_Header, hash: Map_Hash) -> rawptr {
index := __dynamic_map_find(h, hash).entry_index;
if index >= 0 {
data := uintptr(__dynamic_map_get_entry(h, index));
return rawptr(data + h.value_offset);
}
return nil;
}
__dynamic_map_set :: proc(h: Map_Header, hash: Map_Hash, value: rawptr, loc := #caller_location) #no_bounds_check {
index: int;
assert(value != nil);
if len(h.m.hashes) == 0 {
__dynamic_map_reserve(h, INITIAL_MAP_CAP, loc);
__dynamic_map_grow(h, loc);
}
fr := __dynamic_map_find(h, hash);
if fr.entry_index >= 0 {
index = fr.entry_index;
} else {
index = __dynamic_map_add_entry(h, hash, loc);
if fr.entry_prev >= 0 {
entry := __dynamic_map_get_entry(h, fr.entry_prev);
entry.next = index;
} else {
h.m.hashes[fr.hash_index] = index;
}
}
e := __dynamic_map_get_entry(h, index);
e.hash = hash.hash;
key := rawptr(uintptr(e) + h.key_offset);
mem_copy(key, hash.key_ptr, h.key_size);
val := rawptr(uintptr(e) + h.value_offset);
mem_copy(val, value, h.value_size);
if __dynamic_map_full(h) {
__dynamic_map_grow(h, loc);
}
}
__dynamic_map_grow :: proc(using h: Map_Header, loc := #caller_location) {
// TODO(bill): Determine an efficient growing rate
new_count := max(4*m.entries.cap + 7, INITIAL_MAP_CAP);
__dynamic_map_rehash(h, new_count, loc);
}
__dynamic_map_full :: #force_inline proc "contextless" (using h: Map_Header) -> bool {
return int(0.75 * f64(len(m.hashes))) <= m.entries.cap;
}
__dynamic_map_hash_equal :: proc "contextless" (h: Map_Header, a, b: Map_Hash) -> bool {
if a.hash == b.hash {
return h.equal(a.key_ptr, b.key_ptr);
}
return false;
}
__dynamic_map_find :: proc(using h: Map_Header, hash: Map_Hash) -> Map_Find_Result #no_bounds_check {
fr := Map_Find_Result{-1, -1, -1};
if n := uintptr(len(m.hashes)); n > 0 {
fr.hash_index = int(hash.hash % n);
fr.entry_index = m.hashes[fr.hash_index];
for fr.entry_index >= 0 {
entry := __dynamic_map_get_entry(h, fr.entry_index);
entry_hash := __get_map_hash_from_entry(h, entry);
if __dynamic_map_hash_equal(h, entry_hash, hash) {
return fr;
}
fr.entry_prev = fr.entry_index;
fr.entry_index = entry.next;
}
}
return fr;
}
__dynamic_map_add_entry :: proc(using h: Map_Header, hash: Map_Hash, loc := #caller_location) -> int {
prev := m.entries.len;
c := __dynamic_array_append_nothing(&m.entries, entry_size, entry_align, loc);
if c != prev {
end := __dynamic_map_get_entry(h, c-1);
end.hash = hash.hash;
mem_copy(rawptr(uintptr(end) + key_offset), hash.key_ptr, key_size);
end.next = -1;
}
return prev;
}
__dynamic_map_delete_key :: proc(using h: Map_Header, hash: Map_Hash) {
fr := __dynamic_map_find(h, hash);
if fr.entry_index >= 0 {
__dynamic_map_erase(h, fr);
}
}
__dynamic_map_get_entry :: proc(using h: Map_Header, index: int) -> ^Map_Entry_Header {
assert(0 <= index && index < m.entries.len);
return (^Map_Entry_Header)(uintptr(m.entries.data) + uintptr(index*entry_size));
}
__dynamic_map_copy_entry :: proc "contextless" (h: Map_Header, new, old: ^Map_Entry_Header) {
mem_copy(new, old, h.entry_size);
}
__dynamic_map_erase :: proc(using h: Map_Header, fr: Map_Find_Result) #no_bounds_check {
if fr.entry_prev < 0 {
m.hashes[fr.hash_index] = __dynamic_map_get_entry(h, fr.entry_index).next;
} else {
prev := __dynamic_map_get_entry(h, fr.entry_prev);
curr := __dynamic_map_get_entry(h, fr.entry_index);
prev.next = curr.next;
}
if (fr.entry_index == m.entries.len-1) {
// NOTE(bill): No need to do anything else, just pop
} else {
old := __dynamic_map_get_entry(h, fr.entry_index);
end := __dynamic_map_get_entry(h, m.entries.len-1);
__dynamic_map_copy_entry(h, old, end);
old_hash := __get_map_hash_from_entry(h, old);
if last := __dynamic_map_find(h, old_hash); last.entry_prev >= 0 {
last_entry := __dynamic_map_get_entry(h, last.entry_prev);
last_entry.next = fr.entry_index;
} else {
m.hashes[last.hash_index] = fr.entry_index;
}
}
m.entries.len -= 1;
}