0
0
mirror of https://github.com/python/cpython.git synced 2024-11-22 05:26:10 +01:00
cpython/Python/hashtable.c
Victor Stinner 13a00078b8
gh-108634: Py_TRACE_REFS uses a hash table (#108663)
Python built with "configure --with-trace-refs" (tracing references)
is now ABI compatible with Python release build and debug build.
Moreover, it now also supports the Limited API.

Change Py_TRACE_REFS build:

* Remove _PyObject_EXTRA_INIT macro.
* The PyObject structure no longer has two extra members (_ob_prev
  and _ob_next).
* Use a hash table (_Py_hashtable_t) to trace references (all
  objects): PyInterpreterState.object_state.refchain.
* Py_TRACE_REFS build is now ABI compatible with release build and
  debug build.
* Limited C API extensions can now be built with Py_TRACE_REFS:
  xxlimited, xxlimited_35, _testclinic_limited.
* No longer rename PyModule_Create2() and PyModule_FromDefAndSpec2()
  functions to PyModule_Create2TraceRefs() and
  PyModule_FromDefAndSpec2TraceRefs().
* _Py_PrintReferenceAddresses() is now called before
  finalize_interp_delete() which deletes the refchain hash table.
* test_tracemalloc find_trace() now also filters by size to ignore
  the memory allocated by _PyRefchain_Trace().

Test changes for Py_TRACE_REFS:

* Add test.support.Py_TRACE_REFS constant.
* Add test_sys.test_getobjects() to test sys.getobjects() function.
* test_exceptions skips test_recursion_normalizing_with_no_memory()
  and test_memory_error_in_PyErr_PrintEx() if Python is built with
  Py_TRACE_REFS.
* test_repl skips test_no_memory().
* test_capi skisp test_set_nomemory().
2023-08-31 18:33:34 +02:00

425 lines
11 KiB
C

/* The implementation of the hash table (_Py_hashtable_t) is based on the
cfuhash project:
http://sourceforge.net/projects/libcfu/
Copyright of cfuhash:
----------------------------------
Creation date: 2005-06-24 21:22:40
Authors: Don
Change log:
Copyright (c) 2005 Don Owens
All rights reserved.
This code is released under the BSD license:
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions
are met:
* Redistributions of source code must retain the above copyright
notice, this list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above
copyright notice, this list of conditions and the following
disclaimer in the documentation and/or other materials provided
with the distribution.
* Neither the name of the author nor the names of its
contributors may be used to endorse or promote products derived
from this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
"AS IS" AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
FOR A PARTICULAR PURPOSE ARE DISCLAIMED. IN NO EVENT SHALL THE
COPYRIGHT OWNER OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES
(INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT,
STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE)
ARISING IN ANY WAY OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED
OF THE POSSIBILITY OF SUCH DAMAGE.
----------------------------------
*/
#include "Python.h"
#include "pycore_hashtable.h"
#include "pycore_pyhash.h" // _Py_HashPointerRaw()
#define HASHTABLE_MIN_SIZE 16
#define HASHTABLE_HIGH 0.50
#define HASHTABLE_LOW 0.10
#define HASHTABLE_REHASH_FACTOR 2.0 / (HASHTABLE_LOW + HASHTABLE_HIGH)
#define BUCKETS_HEAD(SLIST) \
((_Py_hashtable_entry_t *)_Py_SLIST_HEAD(&(SLIST)))
#define TABLE_HEAD(HT, BUCKET) \
((_Py_hashtable_entry_t *)_Py_SLIST_HEAD(&(HT)->buckets[BUCKET]))
#define ENTRY_NEXT(ENTRY) \
((_Py_hashtable_entry_t *)_Py_SLIST_ITEM_NEXT(ENTRY))
/* Forward declaration */
static int hashtable_rehash(_Py_hashtable_t *ht);
static void
_Py_slist_init(_Py_slist_t *list)
{
list->head = NULL;
}
static void
_Py_slist_prepend(_Py_slist_t *list, _Py_slist_item_t *item)
{
item->next = list->head;
list->head = item;
}
static void
_Py_slist_remove(_Py_slist_t *list, _Py_slist_item_t *previous,
_Py_slist_item_t *item)
{
if (previous != NULL)
previous->next = item->next;
else
list->head = item->next;
}
Py_uhash_t
_Py_hashtable_hash_ptr(const void *key)
{
return (Py_uhash_t)_Py_HashPointerRaw(key);
}
int
_Py_hashtable_compare_direct(const void *key1, const void *key2)
{
return (key1 == key2);
}
/* makes sure the real size of the buckets array is a power of 2 */
static size_t
round_size(size_t s)
{
size_t i;
if (s < HASHTABLE_MIN_SIZE)
return HASHTABLE_MIN_SIZE;
i = 1;
while (i < s)
i <<= 1;
return i;
}
size_t
_Py_hashtable_size(const _Py_hashtable_t *ht)
{
size_t size = sizeof(_Py_hashtable_t);
/* buckets */
size += ht->nbuckets * sizeof(_Py_hashtable_entry_t *);
/* entries */
size += ht->nentries * sizeof(_Py_hashtable_entry_t);
return size;
}
size_t
_Py_hashtable_len(const _Py_hashtable_t *ht)
{
return ht->nentries;
}
_Py_hashtable_entry_t *
_Py_hashtable_get_entry_generic(_Py_hashtable_t *ht, const void *key)
{
Py_uhash_t key_hash = ht->hash_func(key);
size_t index = key_hash & (ht->nbuckets - 1);
_Py_hashtable_entry_t *entry = TABLE_HEAD(ht, index);
while (1) {
if (entry == NULL) {
return NULL;
}
if (entry->key_hash == key_hash && ht->compare_func(key, entry->key)) {
break;
}
entry = ENTRY_NEXT(entry);
}
return entry;
}
// Specialized for:
// hash_func == _Py_hashtable_hash_ptr
// compare_func == _Py_hashtable_compare_direct
static _Py_hashtable_entry_t *
_Py_hashtable_get_entry_ptr(_Py_hashtable_t *ht, const void *key)
{
Py_uhash_t key_hash = _Py_hashtable_hash_ptr(key);
size_t index = key_hash & (ht->nbuckets - 1);
_Py_hashtable_entry_t *entry = TABLE_HEAD(ht, index);
while (1) {
if (entry == NULL) {
return NULL;
}
// Compare directly keys (ignore entry->key_hash)
if (entry->key == key) {
break;
}
entry = ENTRY_NEXT(entry);
}
return entry;
}
void*
_Py_hashtable_steal(_Py_hashtable_t *ht, const void *key)
{
Py_uhash_t key_hash = ht->hash_func(key);
size_t index = key_hash & (ht->nbuckets - 1);
_Py_hashtable_entry_t *entry = TABLE_HEAD(ht, index);
_Py_hashtable_entry_t *previous = NULL;
while (1) {
if (entry == NULL) {
// not found
return NULL;
}
if (entry->key_hash == key_hash && ht->compare_func(key, entry->key)) {
break;
}
previous = entry;
entry = ENTRY_NEXT(entry);
}
_Py_slist_remove(&ht->buckets[index], (_Py_slist_item_t *)previous,
(_Py_slist_item_t *)entry);
ht->nentries--;
void *value = entry->value;
ht->alloc.free(entry);
if ((float)ht->nentries / (float)ht->nbuckets < HASHTABLE_LOW) {
// Ignore failure: error cannot be reported to the caller
hashtable_rehash(ht);
}
return value;
}
int
_Py_hashtable_set(_Py_hashtable_t *ht, const void *key, void *value)
{
_Py_hashtable_entry_t *entry;
#ifndef NDEBUG
/* Don't write the assertion on a single line because it is interesting
to know the duplicated entry if the assertion failed. The entry can
be read using a debugger. */
entry = ht->get_entry_func(ht, key);
assert(entry == NULL);
#endif
entry = ht->alloc.malloc(sizeof(_Py_hashtable_entry_t));
if (entry == NULL) {
/* memory allocation failed */
return -1;
}
entry->key_hash = ht->hash_func(key);
entry->key = (void *)key;
entry->value = value;
ht->nentries++;
if ((float)ht->nentries / (float)ht->nbuckets > HASHTABLE_HIGH) {
if (hashtable_rehash(ht) < 0) {
ht->nentries--;
ht->alloc.free(entry);
return -1;
}
}
size_t index = entry->key_hash & (ht->nbuckets - 1);
_Py_slist_prepend(&ht->buckets[index], (_Py_slist_item_t*)entry);
return 0;
}
void*
_Py_hashtable_get(_Py_hashtable_t *ht, const void *key)
{
_Py_hashtable_entry_t *entry = ht->get_entry_func(ht, key);
if (entry != NULL) {
return entry->value;
}
else {
return NULL;
}
}
int
_Py_hashtable_foreach(_Py_hashtable_t *ht,
_Py_hashtable_foreach_func func,
void *user_data)
{
for (size_t hv = 0; hv < ht->nbuckets; hv++) {
_Py_hashtable_entry_t *entry = TABLE_HEAD(ht, hv);
while (entry != NULL) {
int res = func(ht, entry->key, entry->value, user_data);
if (res) {
return res;
}
entry = ENTRY_NEXT(entry);
}
}
return 0;
}
static int
hashtable_rehash(_Py_hashtable_t *ht)
{
size_t new_size = round_size((size_t)(ht->nentries * HASHTABLE_REHASH_FACTOR));
if (new_size == ht->nbuckets) {
return 0;
}
size_t buckets_size = new_size * sizeof(ht->buckets[0]);
_Py_slist_t *new_buckets = ht->alloc.malloc(buckets_size);
if (new_buckets == NULL) {
/* memory allocation failed */
return -1;
}
memset(new_buckets, 0, buckets_size);
for (size_t bucket = 0; bucket < ht->nbuckets; bucket++) {
_Py_hashtable_entry_t *entry = BUCKETS_HEAD(ht->buckets[bucket]);
while (entry != NULL) {
assert(ht->hash_func(entry->key) == entry->key_hash);
_Py_hashtable_entry_t *next = ENTRY_NEXT(entry);
size_t entry_index = entry->key_hash & (new_size - 1);
_Py_slist_prepend(&new_buckets[entry_index], (_Py_slist_item_t*)entry);
entry = next;
}
}
ht->alloc.free(ht->buckets);
ht->nbuckets = new_size;
ht->buckets = new_buckets;
return 0;
}
_Py_hashtable_t *
_Py_hashtable_new_full(_Py_hashtable_hash_func hash_func,
_Py_hashtable_compare_func compare_func,
_Py_hashtable_destroy_func key_destroy_func,
_Py_hashtable_destroy_func value_destroy_func,
_Py_hashtable_allocator_t *allocator)
{
_Py_hashtable_allocator_t alloc;
if (allocator == NULL) {
alloc.malloc = PyMem_Malloc;
alloc.free = PyMem_Free;
}
else {
alloc = *allocator;
}
_Py_hashtable_t *ht = (_Py_hashtable_t *)alloc.malloc(sizeof(_Py_hashtable_t));
if (ht == NULL) {
return ht;
}
ht->nbuckets = HASHTABLE_MIN_SIZE;
ht->nentries = 0;
size_t buckets_size = ht->nbuckets * sizeof(ht->buckets[0]);
ht->buckets = alloc.malloc(buckets_size);
if (ht->buckets == NULL) {
alloc.free(ht);
return NULL;
}
memset(ht->buckets, 0, buckets_size);
ht->get_entry_func = _Py_hashtable_get_entry_generic;
ht->hash_func = hash_func;
ht->compare_func = compare_func;
ht->key_destroy_func = key_destroy_func;
ht->value_destroy_func = value_destroy_func;
ht->alloc = alloc;
if (ht->hash_func == _Py_hashtable_hash_ptr
&& ht->compare_func == _Py_hashtable_compare_direct)
{
ht->get_entry_func = _Py_hashtable_get_entry_ptr;
}
return ht;
}
_Py_hashtable_t *
_Py_hashtable_new(_Py_hashtable_hash_func hash_func,
_Py_hashtable_compare_func compare_func)
{
return _Py_hashtable_new_full(hash_func, compare_func,
NULL, NULL, NULL);
}
static void
_Py_hashtable_destroy_entry(_Py_hashtable_t *ht, _Py_hashtable_entry_t *entry)
{
if (ht->key_destroy_func) {
ht->key_destroy_func(entry->key);
}
if (ht->value_destroy_func) {
ht->value_destroy_func(entry->value);
}
ht->alloc.free(entry);
}
void
_Py_hashtable_clear(_Py_hashtable_t *ht)
{
for (size_t i=0; i < ht->nbuckets; i++) {
_Py_hashtable_entry_t *entry = TABLE_HEAD(ht, i);
while (entry != NULL) {
_Py_hashtable_entry_t *next = ENTRY_NEXT(entry);
_Py_hashtable_destroy_entry(ht, entry);
entry = next;
}
_Py_slist_init(&ht->buckets[i]);
}
ht->nentries = 0;
// Ignore failure: clear function is not expected to fail
// because of a memory allocation failure.
(void)hashtable_rehash(ht);
}
void
_Py_hashtable_destroy(_Py_hashtable_t *ht)
{
for (size_t i = 0; i < ht->nbuckets; i++) {
_Py_hashtable_entry_t *entry = TABLE_HEAD(ht, i);
while (entry) {
_Py_hashtable_entry_t *entry_next = ENTRY_NEXT(entry);
_Py_hashtable_destroy_entry(ht, entry);
entry = entry_next;
}
}
ht->alloc.free(ht->buckets);
ht->alloc.free(ht);
}