+ That is lengths are checked before doing the memcmp.
+
+ For GDB debugging GDB, it turned out that all lengths were 24 bytes
+ (no C++ so only psymbols were cached) and hence, all compares
+ required a call to memcmp. As a hack, two bytes of padding
+ (mentioned above) are used to store the upper 16 bits of the
+ string's hash value and then that is used in the comparison vis:
+
+ a.half_hash == b.half_hash && a.length == b.length && memcmp
+ (a.data, b.data, a.length)
+
+ The numbers from GDB debugging GDB show this to be a remarkable
+ 100% effective (only necessary length and memcmp tests being
+ performed).
+
+ Mind you, looking at the wall clock, the same GDB debugging GDB
+ showed only marginal speed up (0.780 vs 0.773s). Seems GDB is too
+ busy doing something else :-(
+
+*/
+
+namespace gdb {
+
+struct bstring;
+
+struct bcache
+{
+ /* Allocate a bcache. HASH_FN and COMPARE_FN can be used to pass in
+ custom hash, and compare functions to be used by this bcache. If
+ HASH_FUNCTION is NULL fast_hash() is used and if COMPARE_FUNCTION is
+ NULL memcmp() is used. */
+
+ explicit bcache (unsigned long (*hash_fn)(const void *,
+ int length) = nullptr,
+ int (*compare_fn)(const void *, const void *,
+ int length) = nullptr)
+ : m_hash_function (hash_fn == nullptr ? default_hash : hash_fn),
+ m_compare_function (compare_fn == nullptr ? compare : compare_fn)
+ {
+ }
+
+ ~bcache ();
+
+ /* Find a copy of the LENGTH bytes at ADDR in BCACHE. If BCACHE has
+ never seen those bytes before, add a copy of them to BCACHE. In
+ either case, return a pointer to BCACHE's copy of that string.
+ Since the cached value is ment to be read-only, return a const
+ buffer. If ADDED is not NULL, set *ADDED to true if the bytes
+ were newly added to the cache, or to false if the bytes were
+ found in the cache. */
+
+ const void *insert (const void *addr, int length, int *added = nullptr);
+
+ /* Print statistics on this bcache's memory usage and efficacity at
+ eliminating duplication. TYPE should be a string describing the
+ kind of data this bcache holds. Statistics are printed using
+ `printf_filtered' and its ilk. */
+ void print_statistics (const char *type);
+ int memory_used ();
+
+private:
+
+ /* All the bstrings are allocated here. */
+ struct obstack m_cache {};
+
+ /* How many hash buckets we're using. */
+ unsigned int m_num_buckets = 0;
+
+ /* Hash buckets. This table is allocated using malloc, so when we
+ grow the table we can return the old table to the system. */
+ struct bstring **m_bucket = nullptr;
+
+ /* Statistics. */
+ unsigned long m_unique_count = 0; /* number of unique strings */
+ long m_total_count = 0; /* total number of strings cached, including dups */
+ long m_unique_size = 0; /* size of unique strings, in bytes */
+ long m_total_size = 0; /* total number of bytes cached, including dups */
+ long m_structure_size = 0; /* total size of bcache, including infrastructure */
+ /* Number of times that the hash table is expanded and hence
+ re-built, and the corresponding number of times that a string is
+ [re]hashed as part of entering it into the expanded table. The
+ total number of hashes can be computed by adding TOTAL_COUNT to
+ expand_hash_count. */
+ unsigned long m_expand_count = 0;
+ unsigned long m_expand_hash_count = 0;
+ /* Number of times that the half-hash compare hit (compare the upper
+ 16 bits of hash values) hit, but the corresponding combined
+ length/data compare missed. */
+ unsigned long m_half_hash_miss_count = 0;
+
+ /* Hash function to be used for this bcache object. */
+ unsigned long (*m_hash_function)(const void *addr, int length);
+
+ /* Compare function to be used for this bcache object. */
+ int (*m_compare_function)(const void *, const void *, int length);
+
+ /* Default compare function. */
+ static int compare (const void *addr1, const void *addr2, int length);
+
+ /* Default hash function. */
+ static unsigned long default_hash (const void *ptr, int length)
+ {
+ return fast_hash (ptr, length, 0);
+ }
+
+ /* Expand the hash table. */
+ void expand_hash_table ();
+};