From: Paul N. Hilfinger Date: Thu, 7 Oct 2010 06:53:44 +0000 (+0000) Subject: Extend hashed symbol dictionaries to work with Ada X-Git-Url: https://git.libre-soc.org/?a=commitdiff_plain;h=c4d840bdd6f9e19843bd35835cc08e992fe5e1db;p=binutils-gdb.git Extend hashed symbol dictionaries to work with Ada This patch allows Ada to speed up symbol lookup by using the facilities in dictionary.[ch] for hashed lookups. First, we generalize dictionary search to allow clients to specify any matching function compatible with the hashing function. Next, we modify the hashing algorithm so that symbols that wild-match a name hash to the same value. Finally, we modify Ada symbol lookup to use these facilities. Because this patch touches on a hashing algorithm used by other languages, I took the precaution of doing a speed test on a list of about 12000 identifiers (repeatedly inserting all of them into a table and then doing a lookup on a million names at random, thus testing the speed of the hashing algorithm and how well it distributed names). There was actually a slight speedup, probably as a result of open- coding some of the tests in msymbol_hash_iw. By design, the revised hashing algorithm produces the same results as the original on most "normal" C identifiers. We considered augmenting the dictionary interface still further by allowing different hashing algorithms for different dictionaries, based on the (supposed) language of the symbols in that dictionary. While this produced better isolation of the changes to Ada programs, the additional flexibility also complicated the dictionary interface. I'd prefer to keep things simple for now. Tested w/o regressions on Linux i686. ChangeLog: gdb/ * ada-lang.c (ada_match_name): Use new API for wild_match. (wild_match): Change API to be consistent with that of strcmp_iw; return 0 for a match, and switch operand order. (full_match): New function. (ada_add_block_symbols): Use dict_iter_match_{first,next} for matching to allow use of hashing. * dictionary.c (struct dict_vector): Generalize iter_name_first, iter_name_next ot iter_match_first, iter_match_next. (iter_name_first_hashed): Replace with iter_match_first_hashed. (iter_name_next_hashed): Replace with iter_match_next_hashed. (iter_name_first_linear): Replace with iter_match_first_linear. (iter_name_next_linear): Replace with iter_match_next_linear. (dict_iter_name_first): Re-implement to use dict_iter_match_first. (dict_iter_name_next): Re-implement to use dict_iter_match_next. (dict_iter_match_first): New function. (dict_iter_match_next): New function. (dict_hash): New function. * dictionary.h (dict_iter_match_first, dict_iter_match_next): Declare. * psymtab.c (ada_lookup_partial_symbol): Use new wild_match API. --- diff --git a/gdb/ChangeLog b/gdb/ChangeLog index a7dad2647b2..9e5a615b76c 100644 --- a/gdb/ChangeLog +++ b/gdb/ChangeLog @@ -1,3 +1,25 @@ +2010-10-06 Paul Hilfinger + + * ada-lang.c (ada_match_name): Use new API for wild_match. + (wild_match): Change API to be consistent with that of strcmp_iw; + return 0 for a match, and switch operand order. + (full_match): New function. + (ada_add_block_symbols): Use dict_iter_match_{first,next} for + matching to allow use of hashing. + * dictionary.c (struct dict_vector): Generalize iter_name_first, + iter_name_next ot iter_match_first, iter_match_next. + (iter_name_first_hashed): Replace with iter_match_first_hashed. + (iter_name_next_hashed): Replace with iter_match_next_hashed. + (iter_name_first_linear): Replace with iter_match_first_linear. + (iter_name_next_linear): Replace with iter_match_next_linear. + (dict_iter_name_first): Re-implement to use dict_iter_match_first. + (dict_iter_name_next): Re-implement to use dict_iter_match_next. + (dict_iter_match_first): New function. + (dict_iter_match_next): New function. + (dict_hash): New function. + * dictionary.h (dict_iter_match_first, dict_iter_match_next): Declare. + * psymtab.c (ada_lookup_partial_symbol): Use new wild_match API. + 2010-10-06 Doug Evans * data-directory/Makefile.in: Remove @host_makefile_frag@, @frags@. diff --git a/gdb/ada-lang.c b/gdb/ada-lang.c index 5fdf8511f00..c111e40356e 100644 --- a/gdb/ada-lang.c +++ b/gdb/ada-lang.c @@ -5062,6 +5062,13 @@ wild_match (const char *name, const char *patn) } } +static int +full_match (const char *sym_name, const char *search_name) +{ + return !ada_match_name (sym_name, search_name, 0); +} + + /* Add symbols from BLOCK matching identifier NAME in DOMAIN to vector *defn_symbols, updating the list of symbols in OBSTACKP (if necessary). If WILD, treat as NAME with a wildcard prefix. @@ -5086,9 +5093,9 @@ ada_add_block_symbols (struct obstack *obstackp, found_sym = 0; if (wild) { - struct symbol *sym; - - ALL_BLOCK_SYMBOLS (block, iter, sym) + for (sym = dict_iter_match_first (BLOCK_DICT (block), name, + wild_match, &iter); + sym != NULL; sym = dict_iter_match_next (name, wild_match, &iter)) { if (symbol_matches_domain (SYMBOL_LANGUAGE (sym), SYMBOL_DOMAIN (sym), domain) @@ -5110,29 +5117,25 @@ ada_add_block_symbols (struct obstack *obstackp, } else { - ALL_BLOCK_SYMBOLS (block, iter, sym) + for (sym = dict_iter_match_first (BLOCK_DICT (block), name, + full_match, &iter); + sym != NULL; sym = dict_iter_match_next (name, full_match, &iter)) { if (symbol_matches_domain (SYMBOL_LANGUAGE (sym), SYMBOL_DOMAIN (sym), domain)) { - int cmp = strncmp (name, SYMBOL_LINKAGE_NAME (sym), name_len); - - if (cmp == 0 - && is_name_suffix (SYMBOL_LINKAGE_NAME (sym) + name_len)) - { - if (SYMBOL_CLASS (sym) != LOC_UNRESOLVED) + if (SYMBOL_CLASS (sym) != LOC_UNRESOLVED) + { + if (SYMBOL_IS_ARGUMENT (sym)) + arg_sym = sym; + else { - if (SYMBOL_IS_ARGUMENT (sym)) - arg_sym = sym; - else - { - found_sym = 1; - add_defn_to_vec (obstackp, - fixup_symbol_section (sym, objfile), - block); - } + found_sym = 1; + add_defn_to_vec (obstackp, + fixup_symbol_section (sym, objfile), + block); } - } + } } } } diff --git a/gdb/dictionary.c b/gdb/dictionary.c index e1c2010bc1c..f3ac3069adb 100644 --- a/gdb/dictionary.c +++ b/gdb/dictionary.c @@ -21,6 +21,7 @@ along with this program. If not, see . */ #include "defs.h" +#include #include "gdb_obstack.h" #include "symtab.h" #include "buildsym.h" @@ -116,11 +117,15 @@ struct dict_vector struct dict_iterator *iterator); struct symbol *(*iterator_next) (struct dict_iterator *iterator); /* Functions to iterate over symbols with a given name. */ - struct symbol *(*iter_name_first) (const struct dictionary *dict, + struct symbol *(*iter_match_first) (const struct dictionary *dict, const char *name, + int (*equiv) (const char *, + const char *), + struct dict_iterator *iterator); + struct symbol *(*iter_match_next) (const char *name, + int (*equiv) (const char *, + const char *), struct dict_iterator *iterator); - struct symbol *(*iter_name_next) (const char *name, - struct dict_iterator *iterator); /* A size function, for maint print symtabs. */ int (*size) (const struct dictionary *dict); }; @@ -236,12 +241,18 @@ static struct symbol *iterator_first_hashed (const struct dictionary *dict, static struct symbol *iterator_next_hashed (struct dict_iterator *iterator); -static struct symbol *iter_name_first_hashed (const struct dictionary *dict, - const char *name, +static struct symbol *iter_match_first_hashed (const struct dictionary *dict, + const char *name, + int (*compare) (const char *, + const char *), struct dict_iterator *iterator); -static struct symbol *iter_name_next_hashed (const char *name, - struct dict_iterator *iterator); +static struct symbol *iter_match_next_hashed (const char *name, + int (*compare) (const char *, + const char *), + struct dict_iterator *iterator); + +static unsigned int dict_hash (const char *string); /* Functions only for DICT_HASHED. */ @@ -264,12 +275,16 @@ static struct symbol *iterator_first_linear (const struct dictionary *dict, static struct symbol *iterator_next_linear (struct dict_iterator *iterator); -static struct symbol *iter_name_first_linear (const struct dictionary *dict, - const char *name, - struct dict_iterator *iterator); +static struct symbol *iter_match_first_linear (const struct dictionary *dict, + const char *name, + int (*compare) (const char *, + const char *), + struct dict_iterator *iterator); -static struct symbol *iter_name_next_linear (const char *name, - struct dict_iterator *iterator); +static struct symbol *iter_match_next_linear (const char *name, + int (*compare) (const char *, + const char *), + struct dict_iterator *iterator); static int size_linear (const struct dictionary *dict); @@ -289,8 +304,8 @@ static const struct dict_vector dict_hashed_vector = add_symbol_nonexpandable, /* add_symbol */ iterator_first_hashed, /* iterator_first */ iterator_next_hashed, /* iterator_next */ - iter_name_first_hashed, /* iter_name_first */ - iter_name_next_hashed, /* iter_name_next */ + iter_match_first_hashed, /* iter_name_first */ + iter_match_next_hashed, /* iter_name_next */ size_hashed, /* size */ }; @@ -301,8 +316,8 @@ static const struct dict_vector dict_hashed_expandable_vector = add_symbol_hashed_expandable, /* add_symbol */ iterator_first_hashed, /* iterator_first */ iterator_next_hashed, /* iterator_next */ - iter_name_first_hashed, /* iter_name_first */ - iter_name_next_hashed, /* iter_name_next */ + iter_match_first_hashed, /* iter_name_first */ + iter_match_next_hashed, /* iter_name_next */ size_hashed_expandable, /* size */ }; @@ -313,8 +328,8 @@ static const struct dict_vector dict_linear_vector = add_symbol_nonexpandable, /* add_symbol */ iterator_first_linear, /* iterator_first */ iterator_next_linear, /* iterator_next */ - iter_name_first_linear, /* iter_name_first */ - iter_name_next_linear, /* iter_name_next */ + iter_match_first_linear, /* iter_name_first */ + iter_match_next_linear, /* iter_name_next */ size_linear, /* size */ }; @@ -325,8 +340,8 @@ static const struct dict_vector dict_linear_expandable_vector = add_symbol_linear_expandable, /* add_symbol */ iterator_first_linear, /* iterator_first */ iterator_next_linear, /* iterator_next */ - iter_name_first_linear, /* iter_name_first */ - iter_name_next_linear, /* iter_name_next */ + iter_match_first_linear, /* iter_name_first */ + iter_match_next_linear, /* iter_name_next */ size_linear, /* size */ }; @@ -516,14 +531,31 @@ dict_iter_name_first (const struct dictionary *dict, const char *name, struct dict_iterator *iterator) { - return (DICT_VECTOR (dict))->iter_name_first (dict, name, iterator); + return dict_iter_match_first (dict, name, strcmp_iw, iterator); } struct symbol * dict_iter_name_next (const char *name, struct dict_iterator *iterator) +{ + return dict_iter_match_next (name, strcmp_iw, iterator); +} + +struct symbol * +dict_iter_match_first (const struct dictionary *dict, + const char *name, + int (*compare) (const char *, const char *), + struct dict_iterator *iterator) +{ + return (DICT_VECTOR (dict))->iter_match_first (dict, name, compare, iterator); +} + +struct symbol * +dict_iter_match_next (const char *name, + int (*compare) (const char *, const char *), + struct dict_iterator *iterator) { return (DICT_VECTOR (DICT_ITERATOR_DICT (iterator))) - ->iter_name_next (name, iterator); + ->iter_match_next (name, compare, iterator); } int @@ -614,12 +646,12 @@ iterator_hashed_advance (struct dict_iterator *iterator) } static struct symbol * -iter_name_first_hashed (const struct dictionary *dict, - const char *name, - struct dict_iterator *iterator) +iter_match_first_hashed (const struct dictionary *dict, + const char *name, + int (*compare) (const char *, const char *), + struct dict_iterator *iterator) { - unsigned int hash_index - = msymbol_hash_iw (name) % DICT_HASHED_NBUCKETS (dict); + unsigned int hash_index = dict_hash (name) % DICT_HASHED_NBUCKETS (dict); struct symbol *sym; DICT_ITERATOR_DICT (iterator) = dict; @@ -632,8 +664,8 @@ iter_name_first_hashed (const struct dictionary *dict, sym != NULL; sym = sym->hash_next) { - /* Warning: the order of arguments to strcmp_iw matters! */ - if (strcmp_iw (SYMBOL_SEARCH_NAME (sym), name) == 0) + /* Warning: the order of arguments to compare matters! */ + if (compare (SYMBOL_SEARCH_NAME (sym), name) == 0) { break; } @@ -645,7 +677,9 @@ iter_name_first_hashed (const struct dictionary *dict, } static struct symbol * -iter_name_next_hashed (const char *name, struct dict_iterator *iterator) +iter_match_next_hashed (const char *name, + int (*compare) (const char *, const char *), + struct dict_iterator *iterator) { struct symbol *next; @@ -653,7 +687,7 @@ iter_name_next_hashed (const char *name, struct dict_iterator *iterator) next != NULL; next = next->hash_next) { - if (strcmp_iw (SYMBOL_SEARCH_NAME (next), name) == 0) + if (compare (SYMBOL_SEARCH_NAME (next), name) == 0) break; } @@ -671,8 +705,8 @@ insert_symbol_hashed (struct dictionary *dict, unsigned int hash_index; struct symbol **buckets = DICT_HASHED_BUCKETS (dict); - hash_index = (msymbol_hash_iw (SYMBOL_SEARCH_NAME (sym)) - % DICT_HASHED_NBUCKETS (dict)); + hash_index = + dict_hash (SYMBOL_SEARCH_NAME (sym)) % DICT_HASHED_NBUCKETS (dict); sym->hash_next = buckets[hash_index]; buckets[hash_index] = sym; } @@ -746,6 +780,60 @@ expand_hashtable (struct dictionary *dict) xfree (old_buckets); } +/* Produce an unsigned hash value from STRING0 that is consistent + with strcmp_iw, strcmp, and, at least on Ada symbols, wild_match. + That is, two identifiers equivalent according to any of those three + comparison operators hash to the same value. */ + +static unsigned int +dict_hash (const char *string) +{ + /* The Ada-encoded version of a name P1.P2...Pn has either the form + P1__P2__...Pn or _ada_P1__P2__...Pn (where the Pi + are lower-cased identifiers). The (which can be empty) + encodes additional information about the denoted entity. This + routine hashes such names to msymbol_hash_iw(Pn). It actually + does this for a superset of both valid Pi and of , but + in other cases it simply returns msymbol_hash_iw(STRING0). */ + + unsigned int hash; + int c; + + if (*string == '_' && strncmp (string, "_ada_", 5) == 0) + string += 5; + + hash = 0; + while (*string) + { + switch (*string) + { + case '$': + case '.': + case 'X': + case '(': + return hash; + case ' ': + string += 1; + break; + case '_': + if (string[1] == '_') + { + if (((c = string[2]) < 'a' || c > 'z') && c != 'O') + return hash; + hash = 0; + string += 2; + break; + } + /* FALL THROUGH */ + default: + hash = hash * 67 + *string - 113; + string += 1; + break; + } + } + return hash; +} + /* Functions for DICT_LINEAR and DICT_LINEAR_EXPANDABLE. */ static struct symbol * @@ -769,18 +857,21 @@ iterator_next_linear (struct dict_iterator *iterator) } static struct symbol * -iter_name_first_linear (const struct dictionary *dict, - const char *name, - struct dict_iterator *iterator) +iter_match_first_linear (const struct dictionary *dict, + const char *name, + int (*compare) (const char *, const char *), + struct dict_iterator *iterator) { DICT_ITERATOR_DICT (iterator) = dict; DICT_ITERATOR_INDEX (iterator) = -1; - return iter_name_next_linear (name, iterator); + return iter_match_next_linear (name, compare, iterator); } static struct symbol * -iter_name_next_linear (const char *name, struct dict_iterator *iterator) +iter_match_next_linear (const char *name, + int (*compare) (const char *, const char *), + struct dict_iterator *iterator) { const struct dictionary *dict = DICT_ITERATOR_DICT (iterator); int i, nsyms = DICT_LINEAR_NSYMS (dict); @@ -789,7 +880,7 @@ iter_name_next_linear (const char *name, struct dict_iterator *iterator) for (i = DICT_ITERATOR_INDEX (iterator) + 1; i < nsyms; ++i) { sym = DICT_LINEAR_SYM (dict, i); - if (strcmp_iw (SYMBOL_SEARCH_NAME (sym), name) == 0) + if (compare (SYMBOL_SEARCH_NAME (sym), name) == 0) { retval = sym; break; diff --git a/gdb/dictionary.h b/gdb/dictionary.h index 2242a791a18..f7d30350ed7 100644 --- a/gdb/dictionary.h +++ b/gdb/dictionary.h @@ -134,6 +134,31 @@ extern struct symbol *dict_iter_name_first (const struct dictionary *dict, extern struct symbol *dict_iter_name_next (const char *name, struct dict_iterator *iterator); +/* Initialize ITERATOR to point at the first symbol in DICT whose + SYMBOL_SEARCH_NAME is NAME, as tested using COMPARE (which must use + the same conventions as strcmp_iw and be compatible with any + dictionary hashing function), and return that first symbol, or NULL + if there are no such symbols. */ + +extern struct symbol *dict_iter_match_first (const struct dictionary *dict, + const char *name, + int (*compare) (const char*, + const char *), + struct dict_iterator *iterator); + +/* Advance ITERATOR to point at the next symbol in DICT whose + SYMBOL_SEARCH_NAME is NAME, as tested using COMPARE (see + dict_iter_match_first), or NULL if there are no more such symbols. + Don't call this if you've previously received NULL from + dict_iterator_match_first or dict_iterator_match_next on this + iteration. And don't call it unless ITERATOR was created by a + previous call to dict_iter_match_first with the same NAME and COMPARE. */ + +extern struct symbol *dict_iter_match_next (const char *name, + int (*compare) (const char*, + const char *), + struct dict_iterator *iterator); + /* Return some notion of the size of the dictionary: the number of symbols if we have that, the number of hash buckets otherwise. */