1 /*
   2  * Copyright (c) 1997, 2013, Oracle and/or its affiliates. All rights reserved.
   3  * DO NOT ALTER OR REMOVE COPYRIGHT NOTICES OR THIS FILE HEADER.
   4  *
   5  * This code is free software; you can redistribute it and/or modify it
   6  * under the terms of the GNU General Public License version 2 only, as
   7  * published by the Free Software Foundation.
   8  *
   9  * This code is distributed in the hope that it will be useful, but WITHOUT
  10  * ANY WARRANTY; without even the implied warranty of MERCHANTABILITY or
  11  * FITNESS FOR A PARTICULAR PURPOSE.  See the GNU General Public License
  12  * version 2 for more details (a copy is included in the LICENSE file that
  13  * accompanied this code).
  14  *
  15  * You should have received a copy of the GNU General Public License version
  16  * 2 along with this work; if not, write to the Free Software Foundation,
  17  * Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA.
  18  *
  19  * Please contact Oracle, 500 Oracle Parkway, Redwood Shores, CA 94065 USA
  20  * or visit www.oracle.com if you need additional information or have any
  21  * questions.
  22  *
  23  */
  24 
  25 
  26 #include "precompiled.hpp"
  27 #include "classfile/altHashing.hpp"
  28 #include "classfile/classLoaderData.hpp"
  29 #include "oops/symbol.hpp"
  30 #include "runtime/atomic.inline.hpp"
  31 #include "runtime/os.hpp"
  32 #include "memory/allocation.inline.hpp"
  33 #include "memory/resourceArea.hpp"
  34 
  35 Symbol::Symbol(const u1* name, int length, int refcount) : _refcount(refcount), _length(length) {
  36   _identity_hash = os::random();
  37   for (int i = 0; i < _length; i++) {
  38     byte_at_put(i, name[i]);
  39   }
  40 }
  41 
  42 void* Symbol::operator new(size_t sz, int len, TRAPS) {
  43   int alloc_size = size(len)*HeapWordSize;
  44   address res = (address) AllocateHeap(alloc_size, mtSymbol);
  45   return res;
  46 }
  47 
  48 void* Symbol::operator new(size_t sz, int len, Arena* arena, TRAPS) {
  49   int alloc_size = size(len)*HeapWordSize;
  50   address res = (address)arena->Amalloc(alloc_size);
  51   return res;
  52 }
  53 
  54 void* Symbol::operator new(size_t sz, int len, ClassLoaderData* loader_data, TRAPS) {
  55   address res;
  56   int alloc_size = size(len)*HeapWordSize;
  57   res = (address) Metaspace::allocate(loader_data, size(len), true,
  58                                       Metaspace::NonClassType, CHECK_NULL);
  59   return res;
  60 }
  61 
  62 void Symbol::operator delete(void *p) {
  63   assert(((Symbol*)p)->refcount() == 0, "should not call this");
  64   FreeHeap(p);
  65 }
  66 
  67 // ------------------------------------------------------------------
  68 // Symbol::equals
  69 //
  70 // Compares the symbol with a string of the given length.
  71 bool Symbol::equals(const char* str, int len) const {
  72   int l = utf8_length();
  73   if (l != len) return false;
  74   while (l-- > 0) {
  75     if (str[l] != (char) byte_at(l))
  76       return false;
  77   }
  78   assert(l == -1, "we should be at the beginning");
  79   return true;
  80 }
  81 
  82 
  83 // ------------------------------------------------------------------
  84 // Symbol::starts_with
  85 //
  86 // Tests if the symbol starts with the specified prefix of the given
  87 // length.
  88 bool Symbol::starts_with(const char* prefix, int len) const {
  89   if (len > utf8_length()) return false;
  90   while (len-- > 0) {
  91     if (prefix[len] != (char) byte_at(len))
  92       return false;
  93   }
  94   assert(len == -1, "we should be at the beginning");
  95   return true;
  96 }
  97 
  98 
  99 // ------------------------------------------------------------------
 100 // Symbol::index_of
 101 //
 102 // Finds if the given string is a substring of this symbol's utf8 bytes.
 103 // Return -1 on failure.  Otherwise return the first index where str occurs.
 104 int Symbol::index_of_at(int i, const char* str, int len) const {
 105   assert(i >= 0 && i <= utf8_length(), "oob");
 106   if (len <= 0)  return 0;
 107   char first_char = str[0];
 108   address bytes = (address) ((Symbol*)this)->base();
 109   address limit = bytes + utf8_length() - len;  // inclusive limit
 110   address scan = bytes + i;
 111   if (scan > limit)
 112     return -1;
 113   for (; scan <= limit; scan++) {
 114     scan = (address) memchr(scan, first_char, (limit + 1 - scan));
 115     if (scan == NULL)
 116       return -1;  // not found
 117     assert(scan >= bytes+i && scan <= limit, "scan oob");
 118     if (memcmp(scan, str, len) == 0)
 119       return (int)(scan - bytes);
 120   }
 121   return -1;
 122 }
 123 
 124 
 125 char* Symbol::as_C_string(char* buf, int size) const {
 126   if (size > 0) {
 127     int len = MIN2(size - 1, utf8_length());
 128     for (int i = 0; i < len; i++) {
 129       buf[i] = byte_at(i);
 130     }
 131     buf[len] = '\0';
 132   }
 133   return buf;
 134 }
 135 
 136 char* Symbol::as_C_string() const {
 137   int len = utf8_length();
 138   char* str = NEW_RESOURCE_ARRAY(char, len + 1);
 139   return as_C_string(str, len + 1);
 140 }
 141 
 142 char* Symbol::as_C_string_flexible_buffer(Thread* t,
 143                                                  char* buf, int size) const {
 144   char* str;
 145   int len = utf8_length();
 146   int buf_len = len + 1;
 147   if (size < buf_len) {
 148     str = NEW_RESOURCE_ARRAY(char, buf_len);
 149   } else {
 150     str = buf;
 151   }
 152   return as_C_string(str, buf_len);
 153 }
 154 
 155 void Symbol::print_symbol_on(outputStream* st) const {
 156   ResourceMark rm;
 157   st = st ? st : tty;
 158   st->print("%s", as_quoted_ascii());
 159 }
 160 
 161 char* Symbol::as_quoted_ascii() const {
 162   const char *ptr = (const char *)&_body[0];
 163   int quoted_length = UTF8::quoted_ascii_length(ptr, utf8_length());
 164   char* result = NEW_RESOURCE_ARRAY(char, quoted_length + 1);
 165   UTF8::as_quoted_ascii(ptr, result, quoted_length + 1);
 166   return result;
 167 }
 168 
 169 jchar* Symbol::as_unicode(int& length) const {
 170   Symbol* this_ptr = (Symbol*)this;
 171   length = UTF8::unicode_length((char*)this_ptr->bytes(), utf8_length());
 172   jchar* result = NEW_RESOURCE_ARRAY(jchar, length);
 173   if (length > 0) {
 174     UTF8::convert_to_unicode((char*)this_ptr->bytes(), result, length);
 175   }
 176   return result;
 177 }
 178 
 179 const char* Symbol::as_klass_external_name(char* buf, int size) const {
 180   if (size > 0) {
 181     char* str    = as_C_string(buf, size);
 182     int   length = (int)strlen(str);
 183     // Turn all '/'s into '.'s (also for array klasses)
 184     for (int index = 0; index < length; index++) {
 185       if (str[index] == '/') {
 186         str[index] = '.';
 187       }
 188     }
 189     return str;
 190   } else {
 191     return buf;
 192   }
 193 }
 194 
 195 const char* Symbol::as_klass_external_name() const {
 196   char* str    = as_C_string();
 197   int   length = (int)strlen(str);
 198   // Turn all '/'s into '.'s (also for array klasses)
 199   for (int index = 0; index < length; index++) {
 200     if (str[index] == '/') {
 201       str[index] = '.';
 202     }
 203   }
 204   return str;
 205 }
 206 
 207 // Alternate hashing for unbalanced symbol tables.
 208 unsigned int Symbol::new_hash(jint seed) {
 209   ResourceMark rm;
 210   // Use alternate hashing algorithm on this symbol.
 211   return AltHashing::murmur3_32(seed, (const jbyte*)as_C_string(), utf8_length());
 212 }
 213 
 214 void Symbol::increment_refcount() {
 215   // Only increment the refcount if positive.  If negative either
 216   // overflow has occurred or it is a permanent symbol in a read only
 217   // shared archive.
 218   if (_refcount >= 0) {
 219     Atomic::inc(&_refcount);
 220     NOT_PRODUCT(Atomic::inc(&_total_count);)
 221   }
 222 }
 223 
 224 void Symbol::decrement_refcount() {
 225   if (_refcount >= 0) {
 226     Atomic::dec(&_refcount);
 227 #ifdef ASSERT
 228     if (_refcount < 0) {
 229       print();
 230       assert(false, "reference count underflow for symbol");
 231     }
 232 #endif
 233   }
 234 }
 235 
 236 void Symbol::print_on(outputStream* st) const {
 237   if (this == NULL) {
 238     st->print_cr("NULL");
 239   } else {
 240     st->print("Symbol: '");
 241     print_symbol_on(st);
 242     st->print("'");
 243     st->print(" count %d", refcount());
 244   }
 245 }
 246 
 247 // The print_value functions are present in all builds, to support the
 248 // disassembler and error reporting.
 249 void Symbol::print_value_on(outputStream* st) const {
 250   if (this == NULL) {
 251     st->print("NULL");
 252   } else {
 253     st->print("'");
 254     for (int i = 0; i < utf8_length(); i++) {
 255       st->print("%c", byte_at(i));
 256     }
 257     st->print("'");
 258   }
 259 }
 260 
 261 // SymbolTable prints this in its statistics
 262 NOT_PRODUCT(int Symbol::_total_count = 0;)