binutils-gdb/libctf/ctf-hash.c

850 lines
21 KiB
C
Raw Normal View History

/* Interface to hashtable implementations.
Copyright (C) 2006-2021 Free Software Foundation, Inc.
This file is part of libctf.
libctf is free software; you can redistribute it and/or modify it under
the terms of the GNU General Public License as published by the Free
Software Foundation; either version 3, or (at your option) any later
version.
This program is distributed in the hope that it will be useful, but
WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
See the GNU General Public License for more details.
You should have received a copy of the GNU General Public License
along with this program; see the file COPYING. If not see
<http://www.gnu.org/licenses/>. */
#include <ctf-impl.h>
#include <string.h>
#include "libiberty.h"
#include "hashtab.h"
libctf, hash: introduce the ctf_dynset There are many places in the deduplicator which use hashtables as tiny sets: keys with no value (and usually, but not always, no freeing function) often with only one or a few members. For each of these, even after the last change to not store the freeing functions, we are storing a little malloced block for each item just to track the key/value pair, and a little malloced block for the hash table itself just to track the freeing function because we can't use libiberty hashtab's freeing function because we are using that to free the little malloced per-item block. If we only have a key, we don't need any of that: we can ditch the per-malloced block because we don't have a value, and we can ditch the per-hashtab structure because we don't need to independently track the freeing functions since libiberty hashtab is doing it for us. That means we don't need an owner field in the (now nonexistent) item block either. Roughly speaking, this datatype saves about 25% in time and 20% in peak memory usage for normal links, even fairly big ones. So this might seem redundant, but it's really worth it. Instead of a _lookup function, a dynset has two distinct functions: ctf_dynset_exists, which returns true or false and an optional pointer to the set member, and ctf_dynhash_lookup_any, which is used if all members of the set are expected to be equivalent and we just want *any* member and we don't care which one. There is no iterator in this set of functions, not because we don't iterate over dynset members -- we do, a lot -- but because the iterator here is a member of an entirely new family of much more convenient iteration functions, introduced in the next commit. libctf/ * ctf-hash.c (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (DYNSET_EMPTY_ENTRY_REPLACEMENT): New. (DYNSET_DELETED_ENTRY_REPLACEMENT): New. (key_to_internal): New. (internal_to_key): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. (ctf_hash_insert_type): Coding style. (ctf_hash_define_type): Likewise. * ctf-impl.h (ctf_dynset_t): New. (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. * ctf-inlines.h (ctf_dynset_cinsert): New.
2020-06-03 05:26:38 +08:00
/* We have three hashtable implementations:
- ctf_hash_* is an interface to a fixed-size hash from const char * ->
ctf_id_t with number of elements specified at creation time, that should
support addition of items but need not support removal.
- ctf_dynhash_* is an interface to a dynamically-expanding hash with
unknown size that should support addition of large numbers of items, and
removal as well, and is used only at type-insertion time and during
linking.
- ctf_dynset_* is an interface to a dynamically-expanding hash that contains
only keys: no values.
These can be implemented by the same underlying hashmap if you wish. */
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
/* The helem is used for general key/value mappings in both the ctf_hash and
ctf_dynhash: the owner may not have space allocated for it, and will be
garbage (not NULL!) in that case. */
typedef struct ctf_helem
{
void *key; /* Either a pointer, or a coerced ctf_id_t. */
void *value; /* The value (possibly a coerced int). */
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
ctf_dynhash_t *owner; /* The hash that owns us. */
} ctf_helem_t;
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
/* Equally, the key_free and value_free may not exist. */
struct ctf_dynhash
{
struct htab *htab;
ctf_hash_free_fun key_free;
ctf_hash_free_fun value_free;
};
libctf, hash: introduce the ctf_dynset There are many places in the deduplicator which use hashtables as tiny sets: keys with no value (and usually, but not always, no freeing function) often with only one or a few members. For each of these, even after the last change to not store the freeing functions, we are storing a little malloced block for each item just to track the key/value pair, and a little malloced block for the hash table itself just to track the freeing function because we can't use libiberty hashtab's freeing function because we are using that to free the little malloced per-item block. If we only have a key, we don't need any of that: we can ditch the per-malloced block because we don't have a value, and we can ditch the per-hashtab structure because we don't need to independently track the freeing functions since libiberty hashtab is doing it for us. That means we don't need an owner field in the (now nonexistent) item block either. Roughly speaking, this datatype saves about 25% in time and 20% in peak memory usage for normal links, even fairly big ones. So this might seem redundant, but it's really worth it. Instead of a _lookup function, a dynset has two distinct functions: ctf_dynset_exists, which returns true or false and an optional pointer to the set member, and ctf_dynhash_lookup_any, which is used if all members of the set are expected to be equivalent and we just want *any* member and we don't care which one. There is no iterator in this set of functions, not because we don't iterate over dynset members -- we do, a lot -- but because the iterator here is a member of an entirely new family of much more convenient iteration functions, introduced in the next commit. libctf/ * ctf-hash.c (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (DYNSET_EMPTY_ENTRY_REPLACEMENT): New. (DYNSET_DELETED_ENTRY_REPLACEMENT): New. (key_to_internal): New. (internal_to_key): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. (ctf_hash_insert_type): Coding style. (ctf_hash_define_type): Likewise. * ctf-impl.h (ctf_dynset_t): New. (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. * ctf-inlines.h (ctf_dynset_cinsert): New.
2020-06-03 05:26:38 +08:00
/* Hash and eq functions for the dynhash and hash. */
unsigned int
ctf_hash_integer (const void *ptr)
{
ctf_helem_t *hep = (ctf_helem_t *) ptr;
return htab_hash_pointer (hep->key);
}
int
ctf_hash_eq_integer (const void *a, const void *b)
{
ctf_helem_t *hep_a = (ctf_helem_t *) a;
ctf_helem_t *hep_b = (ctf_helem_t *) b;
return htab_eq_pointer (hep_a->key, hep_b->key);
}
unsigned int
ctf_hash_string (const void *ptr)
{
ctf_helem_t *hep = (ctf_helem_t *) ptr;
return htab_hash_string (hep->key);
}
int
ctf_hash_eq_string (const void *a, const void *b)
{
ctf_helem_t *hep_a = (ctf_helem_t *) a;
ctf_helem_t *hep_b = (ctf_helem_t *) b;
return !strcmp((const char *) hep_a->key, (const char *) hep_b->key);
}
/* Hash a type_key. */
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-14 04:31:26 +08:00
unsigned int
ctf_hash_type_key (const void *ptr)
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-14 04:31:26 +08:00
{
ctf_helem_t *hep = (ctf_helem_t *) ptr;
ctf_link_type_key_t *k = (ctf_link_type_key_t *) hep->key;
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-14 04:31:26 +08:00
return htab_hash_pointer (k->cltk_fp) + 59
* htab_hash_pointer ((void *) (uintptr_t) k->cltk_idx);
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-14 04:31:26 +08:00
}
int
ctf_hash_eq_type_key (const void *a, const void *b)
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-14 04:31:26 +08:00
{
ctf_helem_t *hep_a = (ctf_helem_t *) a;
ctf_helem_t *hep_b = (ctf_helem_t *) b;
ctf_link_type_key_t *key_a = (ctf_link_type_key_t *) hep_a->key;
ctf_link_type_key_t *key_b = (ctf_link_type_key_t *) hep_b->key;
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-14 04:31:26 +08:00
return (key_a->cltk_fp == key_b->cltk_fp)
&& (key_a->cltk_idx == key_b->cltk_idx);
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-14 04:31:26 +08:00
}
/* Hash a type_id_key. */
unsigned int
ctf_hash_type_id_key (const void *ptr)
{
ctf_helem_t *hep = (ctf_helem_t *) ptr;
ctf_type_id_key_t *k = (ctf_type_id_key_t *) hep->key;
return htab_hash_pointer ((void *) (uintptr_t) k->ctii_input_num)
+ 59 * htab_hash_pointer ((void *) (uintptr_t) k->ctii_type);
}
int
ctf_hash_eq_type_id_key (const void *a, const void *b)
{
ctf_helem_t *hep_a = (ctf_helem_t *) a;
ctf_helem_t *hep_b = (ctf_helem_t *) b;
ctf_type_id_key_t *key_a = (ctf_type_id_key_t *) hep_a->key;
ctf_type_id_key_t *key_b = (ctf_type_id_key_t *) hep_b->key;
return (key_a->ctii_input_num == key_b->ctii_input_num)
&& (key_a->ctii_type == key_b->ctii_type);
}
libctf, hash: introduce the ctf_dynset There are many places in the deduplicator which use hashtables as tiny sets: keys with no value (and usually, but not always, no freeing function) often with only one or a few members. For each of these, even after the last change to not store the freeing functions, we are storing a little malloced block for each item just to track the key/value pair, and a little malloced block for the hash table itself just to track the freeing function because we can't use libiberty hashtab's freeing function because we are using that to free the little malloced per-item block. If we only have a key, we don't need any of that: we can ditch the per-malloced block because we don't have a value, and we can ditch the per-hashtab structure because we don't need to independently track the freeing functions since libiberty hashtab is doing it for us. That means we don't need an owner field in the (now nonexistent) item block either. Roughly speaking, this datatype saves about 25% in time and 20% in peak memory usage for normal links, even fairly big ones. So this might seem redundant, but it's really worth it. Instead of a _lookup function, a dynset has two distinct functions: ctf_dynset_exists, which returns true or false and an optional pointer to the set member, and ctf_dynhash_lookup_any, which is used if all members of the set are expected to be equivalent and we just want *any* member and we don't care which one. There is no iterator in this set of functions, not because we don't iterate over dynset members -- we do, a lot -- but because the iterator here is a member of an entirely new family of much more convenient iteration functions, introduced in the next commit. libctf/ * ctf-hash.c (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (DYNSET_EMPTY_ENTRY_REPLACEMENT): New. (DYNSET_DELETED_ENTRY_REPLACEMENT): New. (key_to_internal): New. (internal_to_key): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. (ctf_hash_insert_type): Coding style. (ctf_hash_define_type): Likewise. * ctf-impl.h (ctf_dynset_t): New. (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. * ctf-inlines.h (ctf_dynset_cinsert): New.
2020-06-03 05:26:38 +08:00
/* Hash and eq functions for the dynset. Most of these can just use the
underlying hashtab functions directly. */
int
ctf_dynset_eq_string (const void *a, const void *b)
{
return !strcmp((const char *) a, (const char *) b);
}
/* The dynhash, used for hashes whose size is not known at creation time. */
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
/* Free a single ctf_helem with arbitrary key/value functions. */
static void
ctf_dynhash_item_free (void *item)
{
ctf_helem_t *helem = item;
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
if (helem->owner->key_free && helem->key)
helem->owner->key_free (helem->key);
if (helem->owner->value_free && helem->value)
helem->owner->value_free (helem->value);
free (helem);
}
ctf_dynhash_t *
ctf_dynhash_create (ctf_hash_fun hash_fun, ctf_hash_eq_fun eq_fun,
ctf_hash_free_fun key_free, ctf_hash_free_fun value_free)
{
ctf_dynhash_t *dynhash;
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
htab_del del = ctf_dynhash_item_free;
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
if (key_free || value_free)
dynhash = malloc (sizeof (ctf_dynhash_t));
else
dynhash = malloc (offsetof (ctf_dynhash_t, key_free));
if (!dynhash)
return NULL;
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
if (key_free == NULL && value_free == NULL)
del = free;
/* 7 is arbitrary and untested for now. */
if ((dynhash->htab = htab_create_alloc (7, (htab_hash) hash_fun, eq_fun,
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
del, xcalloc, free)) == NULL)
{
free (dynhash);
return NULL;
}
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
if (key_free || value_free)
{
dynhash->key_free = key_free;
dynhash->value_free = value_free;
}
return dynhash;
}
static ctf_helem_t **
ctf_hashtab_lookup (struct htab *htab, const void *key, enum insert_option insert)
{
ctf_helem_t tmp = { .key = (void *) key };
return (ctf_helem_t **) htab_find_slot (htab, &tmp, insert);
}
static ctf_helem_t *
ctf_hashtab_insert (struct htab *htab, void *key, void *value,
ctf_hash_free_fun key_free,
ctf_hash_free_fun value_free)
{
ctf_helem_t **slot;
slot = ctf_hashtab_lookup (htab, key, INSERT);
if (!slot)
{
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
errno = ENOMEM;
return NULL;
}
if (!*slot)
{
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
/* Only spend space on the owner if we're going to use it: if there is a
key or value freeing function. */
if (key_free || value_free)
*slot = malloc (sizeof (ctf_helem_t));
else
*slot = malloc (offsetof (ctf_helem_t, owner));
if (!*slot)
return NULL;
(*slot)->key = key;
}
else
{
if (key_free)
key_free (key);
if (value_free)
value_free ((*slot)->value);
}
(*slot)->value = value;
return *slot;
}
int
ctf_dynhash_insert (ctf_dynhash_t *hp, void *key, void *value)
{
ctf_helem_t *slot;
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
ctf_hash_free_fun key_free = NULL, value_free = NULL;
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
if (hp->htab->del_f == ctf_dynhash_item_free)
{
key_free = hp->key_free;
value_free = hp->value_free;
}
slot = ctf_hashtab_insert (hp->htab, key, value,
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
key_free, value_free);
if (!slot)
return errno;
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
/* Keep track of the owner, so that the del function can get at the key_free
and value_free functions. Only do this if one of those functions is set:
if not, the owner is not even present in the helem. */
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
if (key_free || value_free)
slot->owner = hp;
return 0;
}
void
ctf_dynhash_remove (ctf_dynhash_t *hp, const void *key)
{
libctf, hash: save per-item space when no key/item freeing function The libctf dynhash hashtab abstraction supports per-hashtab arbitrary key/item freeing functions -- but it also has a constant slot type that holds both key and value requested by the user, so it needs to use its own freeing function to free that -- and it has nowhere to store the freeing functions the caller requested. So it copies them into every hash item, bloating every slot, even though all items in a given hash table must have the same key and value freeing functions. So point back to the owner using a back-pointer, but don't even spend space in the item or the hashtab allocating those freeing functions unless necessary: if none are needed, we can simply arrange to not pass in ctf_dynhash_item_free as a del_f to hashtab_create_alloc, and none of those fields will ever be accessed. The only downside is that this makes the code sensitive to the order of fields in the ctf_helem_t and ctf_hashtab_t: but the deduplicator allocates so many hash tables that doing this alone cuts memory usage during deduplication by about 10%. (libiberty hashtab itself has a lot of per-hashtab bloat: in the future we might trim that down, or make a trimmer version.) libctf/ * ctf-hash.c (ctf_helem_t) <key_free>: Remove. <value_free>: Likewise. <owner>: New. (ctf_dynhash_item_free): Indirect through the owner. (ctf_dynhash_create): Only pass in ctf_dynhash_item_free and allocate space for the key_free and value_free fields fields if necessary. (ctf_hashtab_insert): Likewise. Fix OOM errno value. (ctf_dynhash_insert): Only access ctf_hashtab's key_free and value_free if they will exist. Set the slot's owner, but only if it exists. (ctf_dynhash_remove): Adjust.
2020-06-03 05:00:14 +08:00
ctf_helem_t hep = { (void *) key, NULL, NULL };
htab_remove_elt (hp->htab, &hep);
}
libctf: map from old to corresponding newly-added types in ctf_add_type This lets you call ctf_type_mapping (dest_fp, src_fp, src_type_id) and get told what type ID the corresponding type has in the target ctf_file_t. This works even if it was added by a recursive call, and because it is stored in the target ctf_file_t it works even if we had to add one type to multiple ctf_file_t's as part of conflicting type handling. We empty out this mapping after every archive is linked: because it maps input to output fps, and we only visit each input fp once, its contents are rendered entirely useless every time the source fp changes. v3: add several missing mapping additions. Add ctf_dynhash_empty, and empty after every input archive. v5: fix tabdamage. libctf/ * ctf-impl.h (ctf_file_t): New field ctf_link_type_mapping. (struct ctf_link_type_mapping_key): New. (ctf_hash_type_mapping_key): Likewise. (ctf_hash_eq_type_mapping_key): Likewise. (ctf_add_type_mapping): Likewise. (ctf_type_mapping): Likewise. (ctf_dynhash_empty): Likewise. * ctf-open.c (ctf_file_close): Update accordingly. * ctf-create.c (ctf_update): Likewise. (ctf_add_type): Populate the mapping. * ctf-hash.c (ctf_hash_type_mapping_key): Hash a type mapping key. (ctf_hash_eq_type_mapping_key): Check the key for equality. (ctf_dynhash_insert): Fix comment typo. (ctf_dynhash_empty): New. * ctf-link.c (ctf_add_type_mapping): New. (ctf_type_mapping): Likewise. (empty_link_type_mapping): New. (ctf_link_one_input_archive): Call it.
2019-07-14 04:31:26 +08:00
void
ctf_dynhash_empty (ctf_dynhash_t *hp)
{
htab_empty (hp->htab);
}
size_t
ctf_dynhash_elements (ctf_dynhash_t *hp)
{
return htab_elements (hp->htab);
}
void *
ctf_dynhash_lookup (ctf_dynhash_t *hp, const void *key)
{
ctf_helem_t **slot;
slot = ctf_hashtab_lookup (hp->htab, key, NO_INSERT);
if (slot)
return (*slot)->value;
return NULL;
}
/* TRUE/FALSE return. */
int
ctf_dynhash_lookup_kv (ctf_dynhash_t *hp, const void *key,
const void **orig_key, void **value)
{
ctf_helem_t **slot;
slot = ctf_hashtab_lookup (hp->htab, key, NO_INSERT);
if (slot)
{
if (orig_key)
*orig_key = (*slot)->key;
if (value)
*value = (*slot)->value;
return 1;
}
return 0;
}
typedef struct ctf_traverse_cb_arg
{
ctf_hash_iter_f fun;
void *arg;
} ctf_traverse_cb_arg_t;
static int
ctf_hashtab_traverse (void **slot, void *arg_)
{
ctf_helem_t *helem = *((ctf_helem_t **) slot);
ctf_traverse_cb_arg_t *arg = (ctf_traverse_cb_arg_t *) arg_;
arg->fun (helem->key, helem->value, arg->arg);
return 1;
}
void
ctf_dynhash_iter (ctf_dynhash_t *hp, ctf_hash_iter_f fun, void *arg_)
{
ctf_traverse_cb_arg_t arg = { fun, arg_ };
htab_traverse (hp->htab, ctf_hashtab_traverse, &arg);
}
typedef struct ctf_traverse_find_cb_arg
{
ctf_hash_iter_find_f fun;
void *arg;
void *found_key;
} ctf_traverse_find_cb_arg_t;
static int
ctf_hashtab_traverse_find (void **slot, void *arg_)
{
ctf_helem_t *helem = *((ctf_helem_t **) slot);
ctf_traverse_find_cb_arg_t *arg = (ctf_traverse_find_cb_arg_t *) arg_;
if (arg->fun (helem->key, helem->value, arg->arg))
{
arg->found_key = helem->key;
return 0;
}
return 1;
}
void *
ctf_dynhash_iter_find (ctf_dynhash_t *hp, ctf_hash_iter_find_f fun, void *arg_)
{
ctf_traverse_find_cb_arg_t arg = { fun, arg_, NULL };
htab_traverse (hp->htab, ctf_hashtab_traverse_find, &arg);
return arg.found_key;
}
typedef struct ctf_traverse_remove_cb_arg
{
struct htab *htab;
ctf_hash_iter_remove_f fun;
void *arg;
} ctf_traverse_remove_cb_arg_t;
static int
ctf_hashtab_traverse_remove (void **slot, void *arg_)
{
ctf_helem_t *helem = *((ctf_helem_t **) slot);
ctf_traverse_remove_cb_arg_t *arg = (ctf_traverse_remove_cb_arg_t *) arg_;
if (arg->fun (helem->key, helem->value, arg->arg))
htab_clear_slot (arg->htab, slot);
return 1;
}
void
ctf_dynhash_iter_remove (ctf_dynhash_t *hp, ctf_hash_iter_remove_f fun,
void *arg_)
{
ctf_traverse_remove_cb_arg_t arg = { hp->htab, fun, arg_ };
htab_traverse (hp->htab, ctf_hashtab_traverse_remove, &arg);
}
/* Traverse a dynhash in arbitrary order, in _next iterator form.
Mutating the dynhash while iterating is not supported (just as it isn't for
htab_traverse).
Note: unusually, this returns zero on success and a *positive* value on
error, because it does not take an fp, taking an error pointer would be
incredibly clunky, and nearly all error-handling ends up stuffing the result
of this into some sort of errno or ctf_errno, which is invariably
positive. So doing this simplifies essentially all callers. */
int
ctf_dynhash_next (ctf_dynhash_t *h, ctf_next_t **it, void **key, void **value)
{
ctf_next_t *i = *it;
ctf_helem_t *slot;
if (!i)
{
size_t size = htab_size (h->htab);
/* If the table has too many entries to fit in an ssize_t, just give up.
This might be spurious, but if any type-related hashtable has ever been
nearly as large as that then something very odd is going on. */
if (((ssize_t) size) < 0)
return EDOM;
if ((i = ctf_next_create ()) == NULL)
return ENOMEM;
i->u.ctn_hash_slot = h->htab->entries;
i->cu.ctn_h = h;
i->ctn_n = 0;
i->ctn_size = (ssize_t) size;
i->ctn_iter_fun = (void (*) (void)) ctf_dynhash_next;
*it = i;
}
if ((void (*) (void)) ctf_dynhash_next != i->ctn_iter_fun)
return ECTF_NEXT_WRONGFUN;
if (h != i->cu.ctn_h)
return ECTF_NEXT_WRONGFP;
if ((ssize_t) i->ctn_n == i->ctn_size)
goto hash_end;
while ((ssize_t) i->ctn_n < i->ctn_size
&& (*i->u.ctn_hash_slot == HTAB_EMPTY_ENTRY
|| *i->u.ctn_hash_slot == HTAB_DELETED_ENTRY))
{
i->u.ctn_hash_slot++;
i->ctn_n++;
}
if ((ssize_t) i->ctn_n == i->ctn_size)
goto hash_end;
slot = *i->u.ctn_hash_slot;
if (key)
*key = slot->key;
if (value)
*value = slot->value;
i->u.ctn_hash_slot++;
i->ctn_n++;
return 0;
hash_end:
ctf_next_destroy (i);
*it = NULL;
return ECTF_NEXT_END;
}
libctf: symbol type linking support This adds facilities to write out the function info and data object sections, which efficiently map from entries in the symbol table to types. The write-side code is entirely new: the read-side code was merely significantly changed and support for indexed tables added (pointed to by the no-longer-unused cth_objtidxoff and cth_funcidxoff header fields). With this in place, you can use ctf_lookup_by_symbol to look up the types of symbols of function and object type (and, as before, you can use ctf_lookup_variable to look up types of file-scope variables not present in the symbol table, as long as you know their name: but variables that are also data objects are now found in the data object section instead.) (Compatible) file format change: The CTF spec has always said that the function info section looks much like the CTF_K_FUNCTIONs in the type section: an info word (including an argument count) followed by a return type and N argument types. This format is suboptimal: it means function symbols cannot be deduplicated and it causes a lot of ugly code duplication in libctf. But conveniently the compiler has never emitted this! Because it has always emitted a rather different format that libctf has never accepted, we can be sure that there are no instances of this function info section in the wild, and can freely change its format without compatibility concerns or a file format version bump. (And since it has never been emitted in any code that generated any older file format version, either, we need keep no code to read the format as specified at all!) So the function info section is now specified as an array of uint32_t, exactly like the object data section: each entry is a type ID in the type section which must be of kind CTF_K_FUNCTION, the prototype of this function. This allows function types to be deduplicated and also correctly encodes the fact that all functions declared in C really are types available to the program: so they should be stored in the type section like all other types. (In format v4, we will be able to represent the types of static functions as well, but that really does require a file format change.) We introduce a new header flag, CTF_F_NEWFUNCINFO, which is set if the new function info format is in use. A sufficiently new compiler will always set this flag. New libctf will always set this flag: old libctf will refuse to open any CTF dicts that have this flag set. If the flag is not set on a dict being read in, new libctf will disregard the function info section. Format v4 will remove this flag (or, rather, the flag has no meaning there and the bit position may be recycled for some other purpose). New API: Symbol addition: ctf_add_func_sym: Add a symbol with a given name and type. The type must be of kind CTF_K_FUNCTION (a function pointer). Internally this adds a name -> type mapping to the ctf_funchash in the ctf_dict. ctf_add_objt_sym: Add a symbol with a given name and type. The type kind can be anything, including function pointers. This adds to ctf_objthash. These both treat symbols as name -> type mappings: the linker associates symbol names with symbol indexes via the ctf_link_shuffle_syms callback, which sets up the ctf_dynsyms/ctf_dynsymidx/ctf_dynsymmax fields in the ctf_dict. Repeated relinks can add more symbols. Variables that are also exposed as symbols are removed from the variable section at serialization time. CTF symbol type sections which have enough pads, defined by CTF_INDEX_PAD_THRESHOLD (whether because they are in dicts with symbols where most types are unknown, or in archive where most types are defined in some child or parent dict, not in this specific dict) are sorted by name rather than symidx and accompanied by an index which associates each symbol type entry with a name: the existing ctf_lookup_by_symbol will map symbol indexes to symbol names and look the names up in the index automatically. (This is currently ELF-symbol-table-dependent, but there is almost nothing specific to ELF in here and we can add support for other symbol table formats easily). The compiler also uses index sections to communicate the contents of object file symbol tables without relying on any specific ordering of symbols: it doesn't need to sort them, and libctf will detect an unsorted index section via the absence of the new CTF_F_IDXSORTED header flag, and sort it if needed. Iteration: ctf_symbol_next: Iterator which returns the types and names of symbols one by one, either for function or data symbols. This does not require any sorting: the ctf_link machinery uses it to pull in all the compiler-provided symbols cheaply, but it is not restricted to that use. (Compatible) changes in API: ctf_lookup_by_symbol: can now be called for object and function symbols: never returns ECTF_NOTDATA (which is now not thrown by anything, but is kept for compatibility and because it is a plausible error that we might start throwing again at some later date). Internally we also have changes to the ctf-string functionality so that "external" strings (those where we track a string -> offset mapping, but only write out an offset) can be consulted via the usual means (ctf_strptr) before the strtab is written out. This is important because ctf_link_add_linker_symbol can now be handed symbols named via strtab offsets, and ctf_link_shuffle_syms must figure out their actual names by looking in the external symtab we have just been fed by the ctf_link_add_strtab callback, long before that strtab is written out. include/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctf-api.h (ctf_symbol_next): New. (ctf_add_objt_sym): Likewise. (ctf_add_func_sym): Likewise. * ctf.h: Document new function info section format. (CTF_F_NEWFUNCINFO): New. (CTF_F_IDXSORTED): New. (CTF_F_MAX): Adjust accordingly. libctf/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctf-impl.h (CTF_INDEX_PAD_THRESHOLD): New. (_libctf_nonnull_): Likewise. (ctf_in_flight_dynsym_t): New. (ctf_dict_t) <ctf_funcidx_names>: Likewise. <ctf_objtidx_names>: Likewise. <ctf_nfuncidx>: Likewise. <ctf_nobjtidx>: Likewise. <ctf_funcidx_sxlate>: Likewise. <ctf_objtidx_sxlate>: Likewise. <ctf_objthash>: Likewise. <ctf_funchash>: Likewise. <ctf_dynsyms>: Likewise. <ctf_dynsymidx>: Likewise. <ctf_dynsymmax>: Likewise. <ctf_in_flight_dynsym>: Likewise. (struct ctf_next) <u.ctn_next>: Likewise. (ctf_symtab_skippable): New prototype. (ctf_add_funcobjt_sym): Likewise. (ctf_dynhash_sort_by_name): Likewise. (ctf_sym_to_elf64): Rename to... (ctf_elf32_to_link_sym): ... this, and... (ctf_elf64_to_link_sym): ... this. * ctf-open.c (init_symtab): Check for lack of CTF_F_NEWFUNCINFO flag, and presence of index sections. Refactor out ctf_symtab_skippable and ctf_elf*_to_link_sym, and use them. Use ctf_link_sym_t, not Elf64_Sym. Skip initializing objt or func sxlate sections if corresponding index section is present. Adjust for new func info section format. (ctf_bufopen_internal): Add ctf_err_warn to corrupt-file error handling. Report incorrect-length index sections. Always do an init_symtab, even if there is no symtab section (there may be index sections still). (flip_objts): Adjust comment: func and objt sections are actually identical in structure now, no need to caveat. (ctf_dict_close): Free newly-added data structures. * ctf-create.c (ctf_create): Initialize them. (ctf_symtab_skippable): New, refactored out of init_symtab, with st_nameidx_set check added. (ctf_add_funcobjt_sym): New, add a function or object symbol to the ctf_objthash or ctf_funchash, by name. (ctf_add_objt_sym): Call it. (ctf_add_func_sym): Likewise. (symtypetab_delete_nonstatic_vars): New, delete vars also present as data objects. (CTF_SYMTYPETAB_EMIT_FUNCTION): New flag to symtypetab emitters: this is a function emission, not a data object emission. (CTF_SYMTYPETAB_EMIT_PAD): New flag to symtypetab emitters: emit pads for symbols with no type (only set for unindexed sections). (CTF_SYMTYPETAB_FORCE_INDEXED): New flag to symtypetab emitters: always emit indexed. (symtypetab_density): New, figure out section sizes. (emit_symtypetab): New, emit a symtypetab. (emit_symtypetab_index): New, emit a symtypetab index. (ctf_serialize): Call them, emitting suitably sorted symtypetab sections and indexes. Set suitable header flags. Copy over new fields. * ctf-hash.c (ctf_dynhash_sort_by_name): New, used to impose an order on symtypetab index sections. * ctf-link.c (ctf_add_type_mapping): Delete erroneous comment relating to code that was never committed. (ctf_link_one_variable): Improve variable name. (check_sym): New, symtypetab analogue of check_variable. (ctf_link_deduplicating_one_symtypetab): New. (ctf_link_deduplicating_syms): Likewise. (ctf_link_deduplicating): Call them. (ctf_link_deduplicating_per_cu): Note that we don't call them in this case (yet). (ctf_link_add_strtab): Set the error on the fp correctly. (ctf_link_add_linker_symbol): New (no longer a do-nothing stub), add a linker symbol to the in-flight list. (ctf_link_shuffle_syms): New (no longer a do-nothing stub), turn the in-flight list into a mapping we can use, now its names are resolvable in the external strtab. * ctf-string.c (ctf_str_rollback_atom): Don't roll back atoms with external strtab offsets. (ctf_str_rollback): Adjust comment. (ctf_str_write_strtab): Migrate ctf_syn_ext_strtab population from writeout time... (ctf_str_add_external): ... to string addition time. * ctf-lookup.c (ctf_lookup_var_key_t): Rename to... (ctf_lookup_idx_key_t): ... this, now we use it for syms too. <clik_names>: New member, a name table. (ctf_lookup_var): Adjust accordingly. (ctf_lookup_variable): Likewise. (ctf_lookup_by_id): Shuffle further up in the file. (ctf_symidx_sort_arg_cb): New, callback for... (sort_symidx_by_name): ... this new function to sort a symidx found to be unsorted (likely originating from the compiler). (ctf_symidx_sort): New, sort a symidx. (ctf_lookup_symbol_name): Support dynamic symbols with indexes provided by the linker. Use ctf_link_sym_t, not Elf64_Sym. Check the parent if a child lookup fails. (ctf_lookup_by_symbol): Likewise. Work for function symbols too. (ctf_symbol_next): New, iterate over symbols with types (without sorting). (ctf_lookup_idx_name): New, bsearch for symbol names in indexes. (ctf_try_lookup_indexed): New, attempt an indexed lookup. (ctf_func_info): Reimplement in terms of ctf_lookup_by_symbol. (ctf_func_args): Likewise. (ctf_get_dict): Move... * ctf-types.c (ctf_get_dict): ... here. * ctf-util.c (ctf_sym_to_elf64): Re-express as... (ctf_elf64_to_link_sym): ... this. Add new st_symidx field, and st_nameidx_set (always 0, so st_nameidx can be ignored). Look in the ELF strtab for names. (ctf_elf32_to_link_sym): Likewise, for Elf32_Sym. (ctf_next_destroy): Destroy ctf_next_t.u.ctn_next if need be. * libctf.ver: Add ctf_symbol_next, ctf_add_objt_sym and ctf_add_func_sym.
2020-11-20 21:34:04 +08:00
int
ctf_dynhash_sort_by_name (const ctf_next_hkv_t *one, const ctf_next_hkv_t *two,
void *unused _libctf_unused_)
{
return strcmp ((char *) one->hkv_key, (char *) two->hkv_key);
}
/* Traverse a sorted dynhash, in _next iterator form.
See ctf_dynhash_next for notes on error returns, etc.
Sort keys before iterating over them using the SORT_FUN and SORT_ARG.
If SORT_FUN is null, thunks to ctf_dynhash_next. */
int
ctf_dynhash_next_sorted (ctf_dynhash_t *h, ctf_next_t **it, void **key,
void **value, ctf_hash_sort_f sort_fun, void *sort_arg)
{
ctf_next_t *i = *it;
if (sort_fun == NULL)
return ctf_dynhash_next (h, it, key, value);
if (!i)
{
size_t els = ctf_dynhash_elements (h);
ctf_next_t *accum_i = NULL;
void *key, *value;
int err;
ctf_next_hkv_t *walk;
if (((ssize_t) els) < 0)
return EDOM;
if ((i = ctf_next_create ()) == NULL)
return ENOMEM;
if ((i->u.ctn_sorted_hkv = calloc (els, sizeof (ctf_next_hkv_t))) == NULL)
{
ctf_next_destroy (i);
return ENOMEM;
}
walk = i->u.ctn_sorted_hkv;
i->cu.ctn_h = h;
while ((err = ctf_dynhash_next (h, &accum_i, &key, &value)) == 0)
{
walk->hkv_key = key;
walk->hkv_value = value;
walk++;
}
if (err != ECTF_NEXT_END)
{
ctf_next_destroy (i);
return err;
}
if (sort_fun)
ctf_qsort_r (i->u.ctn_sorted_hkv, els, sizeof (ctf_next_hkv_t),
(int (*) (const void *, const void *, void *)) sort_fun,
sort_arg);
i->ctn_n = 0;
i->ctn_size = (ssize_t) els;
i->ctn_iter_fun = (void (*) (void)) ctf_dynhash_next_sorted;
*it = i;
}
if ((void (*) (void)) ctf_dynhash_next_sorted != i->ctn_iter_fun)
return ECTF_NEXT_WRONGFUN;
if (h != i->cu.ctn_h)
return ECTF_NEXT_WRONGFP;
if ((ssize_t) i->ctn_n == i->ctn_size)
{
ctf_next_destroy (i);
*it = NULL;
return ECTF_NEXT_END;
}
if (key)
*key = i->u.ctn_sorted_hkv[i->ctn_n].hkv_key;
if (value)
*value = i->u.ctn_sorted_hkv[i->ctn_n].hkv_value;
i->ctn_n++;
return 0;
}
void
ctf_dynhash_destroy (ctf_dynhash_t *hp)
{
if (hp != NULL)
htab_delete (hp->htab);
free (hp);
}
libctf, hash: introduce the ctf_dynset There are many places in the deduplicator which use hashtables as tiny sets: keys with no value (and usually, but not always, no freeing function) often with only one or a few members. For each of these, even after the last change to not store the freeing functions, we are storing a little malloced block for each item just to track the key/value pair, and a little malloced block for the hash table itself just to track the freeing function because we can't use libiberty hashtab's freeing function because we are using that to free the little malloced per-item block. If we only have a key, we don't need any of that: we can ditch the per-malloced block because we don't have a value, and we can ditch the per-hashtab structure because we don't need to independently track the freeing functions since libiberty hashtab is doing it for us. That means we don't need an owner field in the (now nonexistent) item block either. Roughly speaking, this datatype saves about 25% in time and 20% in peak memory usage for normal links, even fairly big ones. So this might seem redundant, but it's really worth it. Instead of a _lookup function, a dynset has two distinct functions: ctf_dynset_exists, which returns true or false and an optional pointer to the set member, and ctf_dynhash_lookup_any, which is used if all members of the set are expected to be equivalent and we just want *any* member and we don't care which one. There is no iterator in this set of functions, not because we don't iterate over dynset members -- we do, a lot -- but because the iterator here is a member of an entirely new family of much more convenient iteration functions, introduced in the next commit. libctf/ * ctf-hash.c (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (DYNSET_EMPTY_ENTRY_REPLACEMENT): New. (DYNSET_DELETED_ENTRY_REPLACEMENT): New. (key_to_internal): New. (internal_to_key): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. (ctf_hash_insert_type): Coding style. (ctf_hash_define_type): Likewise. * ctf-impl.h (ctf_dynset_t): New. (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. * ctf-inlines.h (ctf_dynset_cinsert): New.
2020-06-03 05:26:38 +08:00
/* The dynset, used for sets of keys with no value. The implementation of this
can be much simpler, because without a value the slot can simply be the
stored key, which means we don't need to store the freeing functions and the
dynset itself is just a htab. */
ctf_dynset_t *
ctf_dynset_create (htab_hash hash_fun, htab_eq eq_fun,
ctf_hash_free_fun key_free)
{
/* 7 is arbitrary and untested for now. */
return (ctf_dynset_t *) htab_create_alloc (7, (htab_hash) hash_fun, eq_fun,
key_free, xcalloc, free);
}
/* The dynset has one complexity: the underlying implementation reserves two
values for internal hash table implementation details (empty versus deleted
entries). These values are otherwise very useful for pointers cast to ints,
so transform the ctf_dynset_inserted value to allow for it. (This
introduces an ambiguity in that one can no longer store these two values in
the dynset, but if we pick high enough values this is very unlikely to be a
problem.)
We leak this implementation detail to the freeing functions on the grounds
that any use of these functions is overwhelmingly likely to be in sets using
real pointers, which will be unaffected. */
#define DYNSET_EMPTY_ENTRY_REPLACEMENT ((void *) (uintptr_t) -64)
#define DYNSET_DELETED_ENTRY_REPLACEMENT ((void *) (uintptr_t) -63)
static void *
key_to_internal (const void *key)
{
if (key == HTAB_EMPTY_ENTRY)
return DYNSET_EMPTY_ENTRY_REPLACEMENT;
else if (key == HTAB_DELETED_ENTRY)
return DYNSET_DELETED_ENTRY_REPLACEMENT;
return (void *) key;
}
static void *
internal_to_key (const void *internal)
{
if (internal == DYNSET_EMPTY_ENTRY_REPLACEMENT)
return HTAB_EMPTY_ENTRY;
else if (internal == DYNSET_DELETED_ENTRY_REPLACEMENT)
return HTAB_DELETED_ENTRY;
return (void *) internal;
}
int
ctf_dynset_insert (ctf_dynset_t *hp, void *key)
{
struct htab *htab = (struct htab *) hp;
void **slot;
slot = htab_find_slot (htab, key, INSERT);
if (!slot)
{
errno = ENOMEM;
return -errno;
}
if (*slot)
{
if (htab->del_f)
(*htab->del_f) (*slot);
}
*slot = key_to_internal (key);
return 0;
}
void
ctf_dynset_remove (ctf_dynset_t *hp, const void *key)
{
htab_remove_elt ((struct htab *) hp, key_to_internal (key));
}
void
ctf_dynset_destroy (ctf_dynset_t *hp)
{
if (hp != NULL)
htab_delete ((struct htab *) hp);
}
void *
ctf_dynset_lookup (ctf_dynset_t *hp, const void *key)
{
void **slot = htab_find_slot ((struct htab *) hp,
key_to_internal (key), NO_INSERT);
if (slot)
return internal_to_key (*slot);
return NULL;
}
/* TRUE/FALSE return. */
int
ctf_dynset_exists (ctf_dynset_t *hp, const void *key, const void **orig_key)
{
void **slot = htab_find_slot ((struct htab *) hp,
key_to_internal (key), NO_INSERT);
if (orig_key && slot)
*orig_key = internal_to_key (*slot);
return (slot != NULL);
}
/* Look up a completely random value from the set, if any exist.
Keys with value zero cannot be distinguished from a nonexistent key. */
void *
ctf_dynset_lookup_any (ctf_dynset_t *hp)
{
struct htab *htab = (struct htab *) hp;
void **slot = htab->entries;
void **limit = slot + htab_size (htab);
while (slot < limit
&& (*slot == HTAB_EMPTY_ENTRY || *slot == HTAB_DELETED_ENTRY))
slot++;
if (slot < limit)
return internal_to_key (*slot);
return NULL;
}
/* Traverse a dynset in arbitrary order, in _next iterator form.
Otherwise, just like ctf_dynhash_next. */
int
ctf_dynset_next (ctf_dynset_t *hp, ctf_next_t **it, void **key)
{
struct htab *htab = (struct htab *) hp;
ctf_next_t *i = *it;
void *slot;
if (!i)
{
size_t size = htab_size (htab);
/* If the table has too many entries to fit in an ssize_t, just give up.
This might be spurious, but if any type-related hashtable has ever been
nearly as large as that then somthing very odd is going on. */
if (((ssize_t) size) < 0)
return EDOM;
if ((i = ctf_next_create ()) == NULL)
return ENOMEM;
i->u.ctn_hash_slot = htab->entries;
i->cu.ctn_s = hp;
i->ctn_n = 0;
i->ctn_size = (ssize_t) size;
i->ctn_iter_fun = (void (*) (void)) ctf_dynset_next;
*it = i;
}
if ((void (*) (void)) ctf_dynset_next != i->ctn_iter_fun)
return ECTF_NEXT_WRONGFUN;
if (hp != i->cu.ctn_s)
return ECTF_NEXT_WRONGFP;
if ((ssize_t) i->ctn_n == i->ctn_size)
goto set_end;
while ((ssize_t) i->ctn_n < i->ctn_size
&& (*i->u.ctn_hash_slot == HTAB_EMPTY_ENTRY
|| *i->u.ctn_hash_slot == HTAB_DELETED_ENTRY))
{
i->u.ctn_hash_slot++;
i->ctn_n++;
}
if ((ssize_t) i->ctn_n == i->ctn_size)
goto set_end;
slot = *i->u.ctn_hash_slot;
if (key)
*key = internal_to_key (slot);
i->u.ctn_hash_slot++;
i->ctn_n++;
return 0;
set_end:
ctf_next_destroy (i);
*it = NULL;
return ECTF_NEXT_END;
}
/* ctf_hash, used for fixed-size maps from const char * -> ctf_id_t without
removal. This is a straight cast of a hashtab. */
ctf_hash_t *
ctf_hash_create (unsigned long nelems, ctf_hash_fun hash_fun,
ctf_hash_eq_fun eq_fun)
{
return (ctf_hash_t *) htab_create_alloc (nelems, (htab_hash) hash_fun,
eq_fun, free, xcalloc, free);
}
uint32_t
ctf_hash_size (const ctf_hash_t *hp)
{
return htab_elements ((struct htab *) hp);
}
int
libctf, include, binutils, gdb, ld: rename ctf_file_t to ctf_dict_t The naming of the ctf_file_t type in libctf is a historical curiosity. Back in the Solaris days, CTF dictionaries were originally generated as a separate file and then (sometimes) merged into objects: hence the datatype was named ctf_file_t, and known as a "CTF file". Nowadays, raw CTF is essentially never written to a file on its own, and the datatype changed name to a "CTF dictionary" years ago. So the term "CTF file" refers to something that is never a file! This is at best confusing. The type has also historically been known as a 'CTF container", which is even more confusing now that we have CTF archives which are *also* a sort of container (they contain CTF dictionaries), but which are never referred to as containers in the source code. So fix this by completing the renaming, renaming ctf_file_t to ctf_dict_t throughout, and renaming those few functions that refer to CTF files by name (keeping compatibility aliases) to refer to dicts instead. Old users who still refer to ctf_file_t will see (harmless) pointer-compatibility warnings at compile time, but the ABI is unchanged (since C doesn't mangle names, and ctf_file_t was always an opaque type) and things will still compile fine as long as -Werror is not specified. All references to CTF containers and CTF files in the source code are fixed to refer to CTF dicts instead. Further (smaller) renamings of annoyingly-named functions to come, as part of the process of souping up queries across whole archives at once (needed for the function info and data object sections). binutils/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * objdump.c (dump_ctf_errs): Rename ctf_file_t to ctf_dict_t. (dump_ctf_archive_member): Likewise. (dump_ctf): Likewise. Use ctf_dict_close, not ctf_file_close. * readelf.c (dump_ctf_errs): Rename ctf_file_t to ctf_dict_t. (dump_ctf_archive_member): Likewise. (dump_section_as_ctf): Likewise. Use ctf_dict_close, not ctf_file_close. gdb/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctfread.c: Change uses of ctf_file_t to ctf_dict_t. (ctf_fp_info::~ctf_fp_info): Call ctf_dict_close, not ctf_file_close. include/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctf-api.h (ctf_file_t): Rename to... (ctf_dict_t): ... this. Keep ctf_file_t around for compatibility. (struct ctf_file): Likewise rename to... (struct ctf_dict): ... this. (ctf_file_close): Rename to... (ctf_dict_close): ... this, keeping compatibility function. (ctf_parent_file): Rename to... (ctf_parent_dict): ... this, keeping compatibility function. All callers adjusted. * ctf.h: Rename references to ctf_file_t to ctf_dict_t. (struct ctf_archive) <ctfa_nfiles>: Rename to... <ctfa_ndicts>: ... this. ld/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ldlang.c (ctf_output): This is a ctf_dict_t now. (lang_ctf_errs_warnings): Rename ctf_file_t to ctf_dict_t. (ldlang_open_ctf): Adjust comment. (lang_merge_ctf): Use ctf_dict_close, not ctf_file_close. * ldelfgen.h (ldelf_examine_strtab_for_ctf): Rename ctf_file_t to ctf_dict_t. Change opaque declaration accordingly. * ldelfgen.c (ldelf_examine_strtab_for_ctf): Adjust. * ldemul.h (examine_strtab_for_ctf): Likewise. (ldemul_examine_strtab_for_ctf): Likewise. * ldeuml.c (ldemul_examine_strtab_for_ctf): Likewise. libctf/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctf-impl.h: Rename ctf_file_t to ctf_dict_t: all declarations adjusted. (ctf_fileops): Rename to... (ctf_dictops): ... this. (ctf_dedup_t) <cd_id_to_file_t>: Rename to... <cd_id_to_dict_t>: ... this. (ctf_file_t): Fix outdated comment. <ctf_fileops>: Rename to... <ctf_dictops>: ... this. (struct ctf_archive_internal) <ctfi_file>: Rename to... <ctfi_dict>: ... this. * ctf-archive.c: Rename ctf_file_t to ctf_dict_t. Rename ctf_archive.ctfa_nfiles to ctfa_ndicts. Rename ctf_file_close to ctf_dict_close. All users adjusted. * ctf-create.c: Likewise. Refer to CTF dicts, not CTF containers. (ctf_bundle_t) <ctb_file>: Rename to... <ctb_dict): ... this. * ctf-decl.c: Rename ctf_file_t to ctf_dict_t. * ctf-dedup.c: Likewise. Rename ctf_file_close to ctf_dict_close. Refer to CTF dicts, not CTF containers. * ctf-dump.c: Likewise. * ctf-error.c: Likewise. * ctf-hash.c: Likewise. * ctf-inlines.h: Likewise. * ctf-labels.c: Likewise. * ctf-link.c: Likewise. * ctf-lookup.c: Likewise. * ctf-open-bfd.c: Likewise. * ctf-string.c: Likewise. * ctf-subr.c: Likewise. * ctf-types.c: Likewise. * ctf-util.c: Likewise. * ctf-open.c: Likewise. (ctf_file_close): Rename to... (ctf_dict_close): ...this. (ctf_file_close): New trivial wrapper around ctf_dict_close, for compatibility. (ctf_parent_file): Rename to... (ctf_parent_dict): ... this. (ctf_parent_file): New trivial wrapper around ctf_parent_dict, for compatibility. * libctf.ver: Add ctf_dict_close and ctf_parent_dict.
2020-11-20 21:34:04 +08:00
ctf_hash_insert_type (ctf_hash_t *hp, ctf_dict_t *fp, uint32_t type,
uint32_t name)
{
libctf: support getting strings from the ELF strtab The CTF file format has always supported "external strtabs", which internally are strtab offsets with their MSB on: such refs get their strings from the strtab passed in at CTF file open time: this is usually intended to be the ELF strtab, and that's what this implementation is meant to support, though in theory the external strtab could come from anywhere. This commit adds support for these external strings in the ctf-string.c strtab tracking layer. It's quite easy: we just add a field csa_offset to the atoms table that tracks all strings: this field tracks the offset of the string in the ELF strtab (with its MSB already on, courtesy of a new macro CTF_SET_STID), and adds a new function that sets the csa_offset to the specified offset (plus MSB). Then we just need to avoid writing out strings to the internal strtab if they have csa_offset set, and note that the internal strtab is shorter than it might otherwise be. (We could in theory save a little more time here by eschewing sorting such strings, since we never actually write the strings out anywhere, but that would mean storing them separately and it's just not worth the complexity cost until profiling shows it's worth doing.) We also have to go through a bit of extra effort at variable-sorting time. This was previously using direct references to the internal strtab: it couldn't use ctf_strptr or ctf_strraw because the new strtab is not yet ready to put in its usual field (in a ctf_file_t that hasn't even been allocated yet at this stage): but now we're using the external strtab, this will no longer do because it'll be looking things up in the wrong strtab, with disastrous results. Instead, pass the new internal strtab in to a new ctf_strraw_explicit function which is just like ctf_strraw except you can specify a ne winternal strtab to use. But even now that it is using a new internal strtab, this is not quite enough: it can't look up strings in the external strtab because ld hasn't written it out yet, and when it does will write it straight to disk. Instead, when we write the internal strtab, note all the offset -> string mappings that we have noted belong in the *external* strtab to a new "synthetic external strtab" dynhash, ctf_syn_ext_strtab, and look in there at ctf_strraw time if it is set. This uses minimal extra memory (because only strings in the external strtab that we actually use are stored, and even those come straight out of the atoms table), but let both variable sorting and name interning when ctf_bufopen is next called work fine. (This also means that we don't need to filter out spurious ECTF_STRTAB warnings from ctf_bufopen but can pass them back to the caller, once we wrap ctf_bufopen so that we have a new internal variant of ctf_bufopen etc that we can pass the synthetic external strtab to. That error has been filtered out since the days of Solaris libctf, which didn't try to handle the problem of getting external strtabs right at construction time at all.) v3: add the synthetic strtab and all associated machinery. v5: fix tabdamage. include/ * ctf.h (CTF_SET_STID): New. libctf/ * ctf-impl.h (ctf_str_atom_t) <csa_offset>: New field. (ctf_file_t) <ctf_syn_ext_strtab>: Likewise. (ctf_str_add_ref): Name the last arg. (ctf_str_add_external) New. (ctf_str_add_strraw_explicit): Likewise. (ctf_simple_open_internal): Likewise. (ctf_bufopen_internal): Likewise. * ctf-string.c (ctf_strraw_explicit): Split from... (ctf_strraw): ... here, with new support for ctf_syn_ext_strtab. (ctf_str_add_ref_internal): Return the atom, not the string. (ctf_str_add): Adjust accordingly. (ctf_str_add_ref): Likewise. Move up in the file. (ctf_str_add_external): New: update the csa_offset. (ctf_str_count_strtab): Only account for strings with no csa_offset in the internal strtab length. (ctf_str_write_strtab): If the csa_offset is set, update the string's refs without writing the string out, and update the ctf_syn_ext_strtab. Make OOM handling less ugly. * ctf-create.c (struct ctf_sort_var_arg_cb): New. (ctf_update): Handle failure to populate the strtab. Pass in the new ctf_sort_var arg. Adjust for ctf_syn_ext_strtab addition. Call ctf_simple_open_internal, not ctf_simple_open. (ctf_sort_var): Call ctf_strraw_explicit rather than looking up strings by hand. * ctf-hash.c (ctf_hash_insert_type): Likewise (but using ctf_strraw). Adjust to diagnose ECTF_STRTAB nonetheless. * ctf-open.c (init_types): No longer filter out ECTF_STRTAB. (ctf_file_close): Destroy the ctf_syn_ext_strtab. (ctf_simple_open): Rename to, and reimplement as a wrapper around... (ctf_simple_open_internal): ... this new function, which calls ctf_bufopen_internal. (ctf_bufopen): Rename to, and reimplement as a wrapper around... (ctf_bufopen_internal): ... this new function, which sets ctf_syn_ext_strtab.
2019-07-14 03:33:01 +08:00
const char *str = ctf_strraw (fp, name);
if (type == 0)
return EINVAL;
libctf: support getting strings from the ELF strtab The CTF file format has always supported "external strtabs", which internally are strtab offsets with their MSB on: such refs get their strings from the strtab passed in at CTF file open time: this is usually intended to be the ELF strtab, and that's what this implementation is meant to support, though in theory the external strtab could come from anywhere. This commit adds support for these external strings in the ctf-string.c strtab tracking layer. It's quite easy: we just add a field csa_offset to the atoms table that tracks all strings: this field tracks the offset of the string in the ELF strtab (with its MSB already on, courtesy of a new macro CTF_SET_STID), and adds a new function that sets the csa_offset to the specified offset (plus MSB). Then we just need to avoid writing out strings to the internal strtab if they have csa_offset set, and note that the internal strtab is shorter than it might otherwise be. (We could in theory save a little more time here by eschewing sorting such strings, since we never actually write the strings out anywhere, but that would mean storing them separately and it's just not worth the complexity cost until profiling shows it's worth doing.) We also have to go through a bit of extra effort at variable-sorting time. This was previously using direct references to the internal strtab: it couldn't use ctf_strptr or ctf_strraw because the new strtab is not yet ready to put in its usual field (in a ctf_file_t that hasn't even been allocated yet at this stage): but now we're using the external strtab, this will no longer do because it'll be looking things up in the wrong strtab, with disastrous results. Instead, pass the new internal strtab in to a new ctf_strraw_explicit function which is just like ctf_strraw except you can specify a ne winternal strtab to use. But even now that it is using a new internal strtab, this is not quite enough: it can't look up strings in the external strtab because ld hasn't written it out yet, and when it does will write it straight to disk. Instead, when we write the internal strtab, note all the offset -> string mappings that we have noted belong in the *external* strtab to a new "synthetic external strtab" dynhash, ctf_syn_ext_strtab, and look in there at ctf_strraw time if it is set. This uses minimal extra memory (because only strings in the external strtab that we actually use are stored, and even those come straight out of the atoms table), but let both variable sorting and name interning when ctf_bufopen is next called work fine. (This also means that we don't need to filter out spurious ECTF_STRTAB warnings from ctf_bufopen but can pass them back to the caller, once we wrap ctf_bufopen so that we have a new internal variant of ctf_bufopen etc that we can pass the synthetic external strtab to. That error has been filtered out since the days of Solaris libctf, which didn't try to handle the problem of getting external strtabs right at construction time at all.) v3: add the synthetic strtab and all associated machinery. v5: fix tabdamage. include/ * ctf.h (CTF_SET_STID): New. libctf/ * ctf-impl.h (ctf_str_atom_t) <csa_offset>: New field. (ctf_file_t) <ctf_syn_ext_strtab>: Likewise. (ctf_str_add_ref): Name the last arg. (ctf_str_add_external) New. (ctf_str_add_strraw_explicit): Likewise. (ctf_simple_open_internal): Likewise. (ctf_bufopen_internal): Likewise. * ctf-string.c (ctf_strraw_explicit): Split from... (ctf_strraw): ... here, with new support for ctf_syn_ext_strtab. (ctf_str_add_ref_internal): Return the atom, not the string. (ctf_str_add): Adjust accordingly. (ctf_str_add_ref): Likewise. Move up in the file. (ctf_str_add_external): New: update the csa_offset. (ctf_str_count_strtab): Only account for strings with no csa_offset in the internal strtab length. (ctf_str_write_strtab): If the csa_offset is set, update the string's refs without writing the string out, and update the ctf_syn_ext_strtab. Make OOM handling less ugly. * ctf-create.c (struct ctf_sort_var_arg_cb): New. (ctf_update): Handle failure to populate the strtab. Pass in the new ctf_sort_var arg. Adjust for ctf_syn_ext_strtab addition. Call ctf_simple_open_internal, not ctf_simple_open. (ctf_sort_var): Call ctf_strraw_explicit rather than looking up strings by hand. * ctf-hash.c (ctf_hash_insert_type): Likewise (but using ctf_strraw). Adjust to diagnose ECTF_STRTAB nonetheless. * ctf-open.c (init_types): No longer filter out ECTF_STRTAB. (ctf_file_close): Destroy the ctf_syn_ext_strtab. (ctf_simple_open): Rename to, and reimplement as a wrapper around... (ctf_simple_open_internal): ... this new function, which calls ctf_bufopen_internal. (ctf_bufopen): Rename to, and reimplement as a wrapper around... (ctf_bufopen_internal): ... this new function, which sets ctf_syn_ext_strtab.
2019-07-14 03:33:01 +08:00
if (str == NULL
&& CTF_NAME_STID (name) == CTF_STRTAB_1
&& fp->ctf_syn_ext_strtab == NULL
&& fp->ctf_str[CTF_NAME_STID (name)].cts_strs == NULL)
return ECTF_STRTAB;
libctf: support getting strings from the ELF strtab The CTF file format has always supported "external strtabs", which internally are strtab offsets with their MSB on: such refs get their strings from the strtab passed in at CTF file open time: this is usually intended to be the ELF strtab, and that's what this implementation is meant to support, though in theory the external strtab could come from anywhere. This commit adds support for these external strings in the ctf-string.c strtab tracking layer. It's quite easy: we just add a field csa_offset to the atoms table that tracks all strings: this field tracks the offset of the string in the ELF strtab (with its MSB already on, courtesy of a new macro CTF_SET_STID), and adds a new function that sets the csa_offset to the specified offset (plus MSB). Then we just need to avoid writing out strings to the internal strtab if they have csa_offset set, and note that the internal strtab is shorter than it might otherwise be. (We could in theory save a little more time here by eschewing sorting such strings, since we never actually write the strings out anywhere, but that would mean storing them separately and it's just not worth the complexity cost until profiling shows it's worth doing.) We also have to go through a bit of extra effort at variable-sorting time. This was previously using direct references to the internal strtab: it couldn't use ctf_strptr or ctf_strraw because the new strtab is not yet ready to put in its usual field (in a ctf_file_t that hasn't even been allocated yet at this stage): but now we're using the external strtab, this will no longer do because it'll be looking things up in the wrong strtab, with disastrous results. Instead, pass the new internal strtab in to a new ctf_strraw_explicit function which is just like ctf_strraw except you can specify a ne winternal strtab to use. But even now that it is using a new internal strtab, this is not quite enough: it can't look up strings in the external strtab because ld hasn't written it out yet, and when it does will write it straight to disk. Instead, when we write the internal strtab, note all the offset -> string mappings that we have noted belong in the *external* strtab to a new "synthetic external strtab" dynhash, ctf_syn_ext_strtab, and look in there at ctf_strraw time if it is set. This uses minimal extra memory (because only strings in the external strtab that we actually use are stored, and even those come straight out of the atoms table), but let both variable sorting and name interning when ctf_bufopen is next called work fine. (This also means that we don't need to filter out spurious ECTF_STRTAB warnings from ctf_bufopen but can pass them back to the caller, once we wrap ctf_bufopen so that we have a new internal variant of ctf_bufopen etc that we can pass the synthetic external strtab to. That error has been filtered out since the days of Solaris libctf, which didn't try to handle the problem of getting external strtabs right at construction time at all.) v3: add the synthetic strtab and all associated machinery. v5: fix tabdamage. include/ * ctf.h (CTF_SET_STID): New. libctf/ * ctf-impl.h (ctf_str_atom_t) <csa_offset>: New field. (ctf_file_t) <ctf_syn_ext_strtab>: Likewise. (ctf_str_add_ref): Name the last arg. (ctf_str_add_external) New. (ctf_str_add_strraw_explicit): Likewise. (ctf_simple_open_internal): Likewise. (ctf_bufopen_internal): Likewise. * ctf-string.c (ctf_strraw_explicit): Split from... (ctf_strraw): ... here, with new support for ctf_syn_ext_strtab. (ctf_str_add_ref_internal): Return the atom, not the string. (ctf_str_add): Adjust accordingly. (ctf_str_add_ref): Likewise. Move up in the file. (ctf_str_add_external): New: update the csa_offset. (ctf_str_count_strtab): Only account for strings with no csa_offset in the internal strtab length. (ctf_str_write_strtab): If the csa_offset is set, update the string's refs without writing the string out, and update the ctf_syn_ext_strtab. Make OOM handling less ugly. * ctf-create.c (struct ctf_sort_var_arg_cb): New. (ctf_update): Handle failure to populate the strtab. Pass in the new ctf_sort_var arg. Adjust for ctf_syn_ext_strtab addition. Call ctf_simple_open_internal, not ctf_simple_open. (ctf_sort_var): Call ctf_strraw_explicit rather than looking up strings by hand. * ctf-hash.c (ctf_hash_insert_type): Likewise (but using ctf_strraw). Adjust to diagnose ECTF_STRTAB nonetheless. * ctf-open.c (init_types): No longer filter out ECTF_STRTAB. (ctf_file_close): Destroy the ctf_syn_ext_strtab. (ctf_simple_open): Rename to, and reimplement as a wrapper around... (ctf_simple_open_internal): ... this new function, which calls ctf_bufopen_internal. (ctf_bufopen): Rename to, and reimplement as a wrapper around... (ctf_bufopen_internal): ... this new function, which sets ctf_syn_ext_strtab.
2019-07-14 03:33:01 +08:00
if (str == NULL)
return ECTF_BADNAME;
if (str[0] == '\0')
return 0; /* Just ignore empty strings on behalf of caller. */
if (ctf_hashtab_insert ((struct htab *) hp, (char *) str,
(void *) (ptrdiff_t) type, NULL, NULL) != NULL)
return 0;
return errno;
}
/* if the key is already in the hash, override the previous definition with
this new official definition. If the key is not present, then call
libctf, hash: introduce the ctf_dynset There are many places in the deduplicator which use hashtables as tiny sets: keys with no value (and usually, but not always, no freeing function) often with only one or a few members. For each of these, even after the last change to not store the freeing functions, we are storing a little malloced block for each item just to track the key/value pair, and a little malloced block for the hash table itself just to track the freeing function because we can't use libiberty hashtab's freeing function because we are using that to free the little malloced per-item block. If we only have a key, we don't need any of that: we can ditch the per-malloced block because we don't have a value, and we can ditch the per-hashtab structure because we don't need to independently track the freeing functions since libiberty hashtab is doing it for us. That means we don't need an owner field in the (now nonexistent) item block either. Roughly speaking, this datatype saves about 25% in time and 20% in peak memory usage for normal links, even fairly big ones. So this might seem redundant, but it's really worth it. Instead of a _lookup function, a dynset has two distinct functions: ctf_dynset_exists, which returns true or false and an optional pointer to the set member, and ctf_dynhash_lookup_any, which is used if all members of the set are expected to be equivalent and we just want *any* member and we don't care which one. There is no iterator in this set of functions, not because we don't iterate over dynset members -- we do, a lot -- but because the iterator here is a member of an entirely new family of much more convenient iteration functions, introduced in the next commit. libctf/ * ctf-hash.c (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (DYNSET_EMPTY_ENTRY_REPLACEMENT): New. (DYNSET_DELETED_ENTRY_REPLACEMENT): New. (key_to_internal): New. (internal_to_key): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. (ctf_hash_insert_type): Coding style. (ctf_hash_define_type): Likewise. * ctf-impl.h (ctf_dynset_t): New. (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. * ctf-inlines.h (ctf_dynset_cinsert): New.
2020-06-03 05:26:38 +08:00
ctf_hash_insert_type and hash it in. */
int
libctf, include, binutils, gdb, ld: rename ctf_file_t to ctf_dict_t The naming of the ctf_file_t type in libctf is a historical curiosity. Back in the Solaris days, CTF dictionaries were originally generated as a separate file and then (sometimes) merged into objects: hence the datatype was named ctf_file_t, and known as a "CTF file". Nowadays, raw CTF is essentially never written to a file on its own, and the datatype changed name to a "CTF dictionary" years ago. So the term "CTF file" refers to something that is never a file! This is at best confusing. The type has also historically been known as a 'CTF container", which is even more confusing now that we have CTF archives which are *also* a sort of container (they contain CTF dictionaries), but which are never referred to as containers in the source code. So fix this by completing the renaming, renaming ctf_file_t to ctf_dict_t throughout, and renaming those few functions that refer to CTF files by name (keeping compatibility aliases) to refer to dicts instead. Old users who still refer to ctf_file_t will see (harmless) pointer-compatibility warnings at compile time, but the ABI is unchanged (since C doesn't mangle names, and ctf_file_t was always an opaque type) and things will still compile fine as long as -Werror is not specified. All references to CTF containers and CTF files in the source code are fixed to refer to CTF dicts instead. Further (smaller) renamings of annoyingly-named functions to come, as part of the process of souping up queries across whole archives at once (needed for the function info and data object sections). binutils/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * objdump.c (dump_ctf_errs): Rename ctf_file_t to ctf_dict_t. (dump_ctf_archive_member): Likewise. (dump_ctf): Likewise. Use ctf_dict_close, not ctf_file_close. * readelf.c (dump_ctf_errs): Rename ctf_file_t to ctf_dict_t. (dump_ctf_archive_member): Likewise. (dump_section_as_ctf): Likewise. Use ctf_dict_close, not ctf_file_close. gdb/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctfread.c: Change uses of ctf_file_t to ctf_dict_t. (ctf_fp_info::~ctf_fp_info): Call ctf_dict_close, not ctf_file_close. include/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctf-api.h (ctf_file_t): Rename to... (ctf_dict_t): ... this. Keep ctf_file_t around for compatibility. (struct ctf_file): Likewise rename to... (struct ctf_dict): ... this. (ctf_file_close): Rename to... (ctf_dict_close): ... this, keeping compatibility function. (ctf_parent_file): Rename to... (ctf_parent_dict): ... this, keeping compatibility function. All callers adjusted. * ctf.h: Rename references to ctf_file_t to ctf_dict_t. (struct ctf_archive) <ctfa_nfiles>: Rename to... <ctfa_ndicts>: ... this. ld/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ldlang.c (ctf_output): This is a ctf_dict_t now. (lang_ctf_errs_warnings): Rename ctf_file_t to ctf_dict_t. (ldlang_open_ctf): Adjust comment. (lang_merge_ctf): Use ctf_dict_close, not ctf_file_close. * ldelfgen.h (ldelf_examine_strtab_for_ctf): Rename ctf_file_t to ctf_dict_t. Change opaque declaration accordingly. * ldelfgen.c (ldelf_examine_strtab_for_ctf): Adjust. * ldemul.h (examine_strtab_for_ctf): Likewise. (ldemul_examine_strtab_for_ctf): Likewise. * ldeuml.c (ldemul_examine_strtab_for_ctf): Likewise. libctf/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctf-impl.h: Rename ctf_file_t to ctf_dict_t: all declarations adjusted. (ctf_fileops): Rename to... (ctf_dictops): ... this. (ctf_dedup_t) <cd_id_to_file_t>: Rename to... <cd_id_to_dict_t>: ... this. (ctf_file_t): Fix outdated comment. <ctf_fileops>: Rename to... <ctf_dictops>: ... this. (struct ctf_archive_internal) <ctfi_file>: Rename to... <ctfi_dict>: ... this. * ctf-archive.c: Rename ctf_file_t to ctf_dict_t. Rename ctf_archive.ctfa_nfiles to ctfa_ndicts. Rename ctf_file_close to ctf_dict_close. All users adjusted. * ctf-create.c: Likewise. Refer to CTF dicts, not CTF containers. (ctf_bundle_t) <ctb_file>: Rename to... <ctb_dict): ... this. * ctf-decl.c: Rename ctf_file_t to ctf_dict_t. * ctf-dedup.c: Likewise. Rename ctf_file_close to ctf_dict_close. Refer to CTF dicts, not CTF containers. * ctf-dump.c: Likewise. * ctf-error.c: Likewise. * ctf-hash.c: Likewise. * ctf-inlines.h: Likewise. * ctf-labels.c: Likewise. * ctf-link.c: Likewise. * ctf-lookup.c: Likewise. * ctf-open-bfd.c: Likewise. * ctf-string.c: Likewise. * ctf-subr.c: Likewise. * ctf-types.c: Likewise. * ctf-util.c: Likewise. * ctf-open.c: Likewise. (ctf_file_close): Rename to... (ctf_dict_close): ...this. (ctf_file_close): New trivial wrapper around ctf_dict_close, for compatibility. (ctf_parent_file): Rename to... (ctf_parent_dict): ... this. (ctf_parent_file): New trivial wrapper around ctf_parent_dict, for compatibility. * libctf.ver: Add ctf_dict_close and ctf_parent_dict.
2020-11-20 21:34:04 +08:00
ctf_hash_define_type (ctf_hash_t *hp, ctf_dict_t *fp, uint32_t type,
uint32_t name)
{
libctf, hash: introduce the ctf_dynset There are many places in the deduplicator which use hashtables as tiny sets: keys with no value (and usually, but not always, no freeing function) often with only one or a few members. For each of these, even after the last change to not store the freeing functions, we are storing a little malloced block for each item just to track the key/value pair, and a little malloced block for the hash table itself just to track the freeing function because we can't use libiberty hashtab's freeing function because we are using that to free the little malloced per-item block. If we only have a key, we don't need any of that: we can ditch the per-malloced block because we don't have a value, and we can ditch the per-hashtab structure because we don't need to independently track the freeing functions since libiberty hashtab is doing it for us. That means we don't need an owner field in the (now nonexistent) item block either. Roughly speaking, this datatype saves about 25% in time and 20% in peak memory usage for normal links, even fairly big ones. So this might seem redundant, but it's really worth it. Instead of a _lookup function, a dynset has two distinct functions: ctf_dynset_exists, which returns true or false and an optional pointer to the set member, and ctf_dynhash_lookup_any, which is used if all members of the set are expected to be equivalent and we just want *any* member and we don't care which one. There is no iterator in this set of functions, not because we don't iterate over dynset members -- we do, a lot -- but because the iterator here is a member of an entirely new family of much more convenient iteration functions, introduced in the next commit. libctf/ * ctf-hash.c (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (DYNSET_EMPTY_ENTRY_REPLACEMENT): New. (DYNSET_DELETED_ENTRY_REPLACEMENT): New. (key_to_internal): New. (internal_to_key): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. (ctf_hash_insert_type): Coding style. (ctf_hash_define_type): Likewise. * ctf-impl.h (ctf_dynset_t): New. (ctf_dynset_eq_string): New. (ctf_dynset_create): New. (ctf_dynset_insert): New. (ctf_dynset_remove): New. (ctf_dynset_destroy): New. (ctf_dynset_lookup): New. (ctf_dynset_exists): New. (ctf_dynset_lookup_any): New. * ctf-inlines.h (ctf_dynset_cinsert): New.
2020-06-03 05:26:38 +08:00
/* This matches the semantics of ctf_hash_insert_type in this
implementation anyway. */
return ctf_hash_insert_type (hp, fp, type, name);
}
ctf_id_t
libctf, include, binutils, gdb, ld: rename ctf_file_t to ctf_dict_t The naming of the ctf_file_t type in libctf is a historical curiosity. Back in the Solaris days, CTF dictionaries were originally generated as a separate file and then (sometimes) merged into objects: hence the datatype was named ctf_file_t, and known as a "CTF file". Nowadays, raw CTF is essentially never written to a file on its own, and the datatype changed name to a "CTF dictionary" years ago. So the term "CTF file" refers to something that is never a file! This is at best confusing. The type has also historically been known as a 'CTF container", which is even more confusing now that we have CTF archives which are *also* a sort of container (they contain CTF dictionaries), but which are never referred to as containers in the source code. So fix this by completing the renaming, renaming ctf_file_t to ctf_dict_t throughout, and renaming those few functions that refer to CTF files by name (keeping compatibility aliases) to refer to dicts instead. Old users who still refer to ctf_file_t will see (harmless) pointer-compatibility warnings at compile time, but the ABI is unchanged (since C doesn't mangle names, and ctf_file_t was always an opaque type) and things will still compile fine as long as -Werror is not specified. All references to CTF containers and CTF files in the source code are fixed to refer to CTF dicts instead. Further (smaller) renamings of annoyingly-named functions to come, as part of the process of souping up queries across whole archives at once (needed for the function info and data object sections). binutils/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * objdump.c (dump_ctf_errs): Rename ctf_file_t to ctf_dict_t. (dump_ctf_archive_member): Likewise. (dump_ctf): Likewise. Use ctf_dict_close, not ctf_file_close. * readelf.c (dump_ctf_errs): Rename ctf_file_t to ctf_dict_t. (dump_ctf_archive_member): Likewise. (dump_section_as_ctf): Likewise. Use ctf_dict_close, not ctf_file_close. gdb/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctfread.c: Change uses of ctf_file_t to ctf_dict_t. (ctf_fp_info::~ctf_fp_info): Call ctf_dict_close, not ctf_file_close. include/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctf-api.h (ctf_file_t): Rename to... (ctf_dict_t): ... this. Keep ctf_file_t around for compatibility. (struct ctf_file): Likewise rename to... (struct ctf_dict): ... this. (ctf_file_close): Rename to... (ctf_dict_close): ... this, keeping compatibility function. (ctf_parent_file): Rename to... (ctf_parent_dict): ... this, keeping compatibility function. All callers adjusted. * ctf.h: Rename references to ctf_file_t to ctf_dict_t. (struct ctf_archive) <ctfa_nfiles>: Rename to... <ctfa_ndicts>: ... this. ld/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ldlang.c (ctf_output): This is a ctf_dict_t now. (lang_ctf_errs_warnings): Rename ctf_file_t to ctf_dict_t. (ldlang_open_ctf): Adjust comment. (lang_merge_ctf): Use ctf_dict_close, not ctf_file_close. * ldelfgen.h (ldelf_examine_strtab_for_ctf): Rename ctf_file_t to ctf_dict_t. Change opaque declaration accordingly. * ldelfgen.c (ldelf_examine_strtab_for_ctf): Adjust. * ldemul.h (examine_strtab_for_ctf): Likewise. (ldemul_examine_strtab_for_ctf): Likewise. * ldeuml.c (ldemul_examine_strtab_for_ctf): Likewise. libctf/ChangeLog 2020-11-20 Nick Alcock <nick.alcock@oracle.com> * ctf-impl.h: Rename ctf_file_t to ctf_dict_t: all declarations adjusted. (ctf_fileops): Rename to... (ctf_dictops): ... this. (ctf_dedup_t) <cd_id_to_file_t>: Rename to... <cd_id_to_dict_t>: ... this. (ctf_file_t): Fix outdated comment. <ctf_fileops>: Rename to... <ctf_dictops>: ... this. (struct ctf_archive_internal) <ctfi_file>: Rename to... <ctfi_dict>: ... this. * ctf-archive.c: Rename ctf_file_t to ctf_dict_t. Rename ctf_archive.ctfa_nfiles to ctfa_ndicts. Rename ctf_file_close to ctf_dict_close. All users adjusted. * ctf-create.c: Likewise. Refer to CTF dicts, not CTF containers. (ctf_bundle_t) <ctb_file>: Rename to... <ctb_dict): ... this. * ctf-decl.c: Rename ctf_file_t to ctf_dict_t. * ctf-dedup.c: Likewise. Rename ctf_file_close to ctf_dict_close. Refer to CTF dicts, not CTF containers. * ctf-dump.c: Likewise. * ctf-error.c: Likewise. * ctf-hash.c: Likewise. * ctf-inlines.h: Likewise. * ctf-labels.c: Likewise. * ctf-link.c: Likewise. * ctf-lookup.c: Likewise. * ctf-open-bfd.c: Likewise. * ctf-string.c: Likewise. * ctf-subr.c: Likewise. * ctf-types.c: Likewise. * ctf-util.c: Likewise. * ctf-open.c: Likewise. (ctf_file_close): Rename to... (ctf_dict_close): ...this. (ctf_file_close): New trivial wrapper around ctf_dict_close, for compatibility. (ctf_parent_file): Rename to... (ctf_parent_dict): ... this. (ctf_parent_file): New trivial wrapper around ctf_parent_dict, for compatibility. * libctf.ver: Add ctf_dict_close and ctf_parent_dict.
2020-11-20 21:34:04 +08:00
ctf_hash_lookup_type (ctf_hash_t *hp, ctf_dict_t *fp __attribute__ ((__unused__)),
const char *key)
{
ctf_helem_t **slot;
slot = ctf_hashtab_lookup ((struct htab *) hp, key, NO_INSERT);
if (slot)
return (ctf_id_t) (uintptr_t) ((*slot)->value);
return 0;
}
void
ctf_hash_destroy (ctf_hash_t *hp)
{
if (hp != NULL)
htab_delete ((struct htab *) hp);
}