// stringpool.cc -- a string pool for gold

// Copyright (C) 2006-2024 Free Software Foundation, Inc.
// Written by Ian Lance Taylor <[email protected]>.

// This file is part of gold.

// This program 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 of the License, 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; if not, write to the Free Software
// Foundation, Inc., 51 Franklin Street - Fifth Floor, Boston,
// MA 02110-1301, USA.

#include "gold.h"

#include <cstring>
#include <algorithm>
#include <vector>
#include <uchar.h>

#include "output.h"
#include "parameters.h"
#include "stringpool.h"

namespace gold
{

template<typename Stringpool_char>
Stringpool_template<Stringpool_char>::Stringpool_template(uint64_t addralign)
 : string_set_(), key_to_offset_(), strings_(), strtab_size_(0),
   zero_null_(true), optimize_(false), offset_(sizeof(Stringpool_char)),
   addralign_(addralign)
{
 if (parameters->options_valid()
     && parameters->options().optimize() >= 2
     && addralign <= sizeof(Stringpool_char))
   this->optimize_ = true;
}

template<typename Stringpool_char>
void
Stringpool_template<Stringpool_char>::clear()
{
 for (typename std::list<Stringdata*>::iterator p = this->strings_.begin();
      p != this->strings_.end();
      ++p)
   delete[] reinterpret_cast<char*>(*p);
 this->strings_.clear();
 this->key_to_offset_.clear();
 this->string_set_.clear();
}

template<typename Stringpool_char>
Stringpool_template<Stringpool_char>::~Stringpool_template()
{
 this->clear();
}

// Resize the internal hashtable with the expectation we'll get n new
// elements.  Note that the hashtable constructor takes a "number of
// buckets you'd like," rather than "number of elements you'd like,"
// but that's the best we can do.

template<typename Stringpool_char>
void
Stringpool_template<Stringpool_char>::reserve(unsigned int n)
{
 this->key_to_offset_.reserve(n);

#if defined(HAVE_UNORDERED_MAP)
 this->string_set_.rehash(this->string_set_.size() + n);
 return;
#elif defined(HAVE_TR1_UNORDERED_MAP)
 // rehash() implementation is broken in gcc 4.0.3's stl
 //this->string_set_.rehash(this->string_set_.size() + n);
 //return;
#elif defined(HAVE_EXT_HASH_MAP)
 this->string_set_.resize(this->string_set_.size() + n);
 return;
#endif

 // This is the generic "reserve" code, if no #ifdef above triggers.
 String_set_type new_string_set(this->string_set_.size() + n);
 new_string_set.insert(this->string_set_.begin(), this->string_set_.end());
 this->string_set_.swap(new_string_set);
}

// Compare two strings of arbitrary character type for equality.

template<typename Stringpool_char>
bool
Stringpool_template<Stringpool_char>::string_equal(const Stringpool_char* s1,
                                                  const Stringpool_char* s2)
{
 while (*s1 != 0)
   if (*s1++ != *s2++)
     return false;
 return *s2 == 0;
}

// Specialize string_equal for char.

template<>
inline bool
Stringpool_template<char>::string_equal(const char* s1, const char* s2)
{
 return strcmp(s1, s2) == 0;
}

// Equality comparison function for the hash table.

template<typename Stringpool_char>
inline bool
Stringpool_template<Stringpool_char>::Stringpool_eq::operator()(
   const Hashkey& h1,
   const Hashkey& h2) const
{
 return (h1.hash_code == h2.hash_code
         && h1.length == h2.length
         && (h1.string == h2.string
             || memcmp(h1.string, h2.string,
                       h1.length * sizeof(Stringpool_char)) == 0));
}

// Hash function.  The length is in characters, not bytes.

template<typename Stringpool_char>
size_t
Stringpool_template<Stringpool_char>::string_hash(const Stringpool_char* s,
                                                 size_t length)
{
 return gold::string_hash<Stringpool_char>(s, length);
}

// Add the string S to the list of canonical strings.  Return a
// pointer to the canonical string.  If PKEY is not NULL, set *PKEY to
// the key.  LENGTH is the length of S in characters.  Note that S may
// not be NUL terminated.

template<typename Stringpool_char>
const Stringpool_char*
Stringpool_template<Stringpool_char>::add_string(const Stringpool_char* s,
                                                size_t len)
{
 // We are in trouble if we've already computed the string offsets.
 gold_assert(this->strtab_size_ == 0);

 // The size we allocate for a new Stringdata.
 const size_t buffer_size = 1000;
 // The amount we multiply the Stringdata index when calculating the
 // key.
 const size_t key_mult = 1024;
 gold_assert(key_mult >= buffer_size);

 // Convert len to the number of bytes we need to allocate, including
 // the null character.
 len = (len + 1) * sizeof(Stringpool_char);

 size_t alc;
 bool front = true;
 if (len > buffer_size)
   {
     alc = sizeof(Stringdata) + len;
     front = false;
   }
 else if (this->strings_.empty())
   alc = sizeof(Stringdata) + buffer_size;
 else
   {
     Stringdata* psd = this->strings_.front();
     if (len > psd->alc - psd->len)
       alc = sizeof(Stringdata) + buffer_size;
     else
       {
         char* ret = psd->data + psd->len;
         memcpy(ret, s, len - sizeof(Stringpool_char));
         memset(ret + len - sizeof(Stringpool_char), 0,
                sizeof(Stringpool_char));

         psd->len += len;

         return reinterpret_cast<const Stringpool_char*>(ret);
       }
   }

 Stringdata* psd = reinterpret_cast<Stringdata*>(new char[alc]);
 psd->alc = alc - sizeof(Stringdata);
 memcpy(psd->data, s, len - sizeof(Stringpool_char));
 memset(psd->data + len - sizeof(Stringpool_char), 0,
        sizeof(Stringpool_char));
 psd->len = len;

 if (front)
   this->strings_.push_front(psd);
 else
   this->strings_.push_back(psd);

 return reinterpret_cast<const Stringpool_char*>(psd->data);
}

// Add a string to a string pool.

template<typename Stringpool_char>
const Stringpool_char*
Stringpool_template<Stringpool_char>::add(const Stringpool_char* s, bool copy,
                                         Key* pkey)
{
 return this->add_with_length(s, string_length(s), copy, pkey);
}

// Add a new key offset entry.

template<typename Stringpool_char>
void
Stringpool_template<Stringpool_char>::new_key_offset(size_t length)
{
 section_offset_type offset;
 if (this->zero_null_ && length == 0)
   offset = 0;
 else
   {
     offset = this->offset_;
     // Align strings.
     offset = align_address(offset, this->addralign_);
     this->offset_ = offset + (length + 1) * sizeof(Stringpool_char);
   }
 this->key_to_offset_.push_back(offset);
}

template<typename Stringpool_char>
const Stringpool_char*
Stringpool_template<Stringpool_char>::add_with_length(const Stringpool_char* s,
                                                     size_t length,
                                                     bool copy,
                                                     Key* pkey)
{
 typedef std::pair<typename String_set_type::iterator, bool> Insert_type;

 // We add 1 so that 0 is always invalid.
 const Key k = this->key_to_offset_.size() + 1;

 if (!copy)
   {
     // When we don't need to copy the string, we can call insert
     // directly.

     std::pair<Hashkey, Hashval> element(Hashkey(s, length), k);

     Insert_type ins = this->string_set_.insert(element);

     typename String_set_type::const_iterator p = ins.first;

     if (ins.second)
       {
         // We just added the string.  The key value has now been
         // used.
         this->new_key_offset(length);
       }
     else
       {
         gold_assert(k != p->second);
       }

     if (pkey != NULL)
       *pkey = p->second;
     return p->first.string;
   }

 // When we have to copy the string, we look it up twice in the hash
 // table.  The problem is that we can't insert S before we
 // canonicalize it by copying it into the canonical list. The hash
 // code will only be computed once.

 Hashkey hk(s, length);
 typename String_set_type::const_iterator p = this->string_set_.find(hk);
 if (p != this->string_set_.end())
   {
     if (pkey != NULL)
       *pkey = p->second;
     return p->first.string;
   }

 this->new_key_offset(length);

 hk.string = this->add_string(s, length);
 // The contents of the string stay the same, so we don't need to
 // adjust hk.hash_code or hk.length.

 std::pair<Hashkey, Hashval> element(hk, k);

 Insert_type ins = this->string_set_.insert(element);
 gold_assert(ins.second);

 if (pkey != NULL)
   *pkey = k;
 return hk.string;
}

template<typename Stringpool_char>
const Stringpool_char*
Stringpool_template<Stringpool_char>::find(const Stringpool_char* s,
                                          Key* pkey) const
{
 Hashkey hk(s);
 typename String_set_type::const_iterator p = this->string_set_.find(hk);
 if (p == this->string_set_.end())
   return NULL;

 if (pkey != NULL)
   *pkey = p->second;

 return p->first.string;
}

// Comparison routine used when sorting into an ELF strtab.  We want
// to sort this so that when one string is a suffix of another, we
// always see the shorter string immediately after the longer string.
// For example, we want to see these strings in this order:
//   abcd
//   cd
//   d
// When strings are not suffixes, we don't care what order they are
// in, but we need to ensure that suffixes wind up next to each other.
// So we do a reversed lexicographic sort on the reversed string.

template<typename Stringpool_char>
bool
Stringpool_template<Stringpool_char>::Stringpool_sort_comparison::operator()(
 const Stringpool_sort_info& sort_info1,
 const Stringpool_sort_info& sort_info2) const
{
 const Hashkey& h1(sort_info1->first);
 const Hashkey& h2(sort_info2->first);
 const Stringpool_char* s1 = h1.string;
 const Stringpool_char* s2 = h2.string;
 const size_t len1 = h1.length;
 const size_t len2 = h2.length;
 const size_t minlen = len1 < len2 ? len1 : len2;
 const Stringpool_char* p1 = s1 + len1 - 1;
 const Stringpool_char* p2 = s2 + len2 - 1;
 for (size_t i = minlen; i > 0; --i, --p1, --p2)
   {
     if (*p1 != *p2)
       return *p1 > *p2;
   }
 return len1 > len2;
}

// Return whether s1 is a suffix of s2.

template<typename Stringpool_char>
bool
Stringpool_template<Stringpool_char>::is_suffix(const Stringpool_char* s1,
                                               size_t len1,
                                               const Stringpool_char* s2,
                                               size_t len2)
{
 if (len1 > len2)
   return false;
 return memcmp(s1, s2 + len2 - len1, len1 * sizeof(Stringpool_char)) == 0;
}

// Turn the stringpool into an ELF strtab: determine the offsets of
// each string in the table.

template<typename Stringpool_char>
void
Stringpool_template<Stringpool_char>::set_string_offsets()
{
 if (this->strtab_size_ != 0)
   {
     // We've already computed the offsets.
     return;
   }

 const size_t charsize = sizeof(Stringpool_char);

 // Offset 0 may be reserved for the empty string.
 section_offset_type offset = this->zero_null_ ? charsize : 0;

 // Sorting to find suffixes can take over 25% of the total CPU time
 // used by the linker.  Since it's merely an optimization to reduce
 // the strtab size, and gives a relatively small benefit (it's
 // typically rare for a symbol to be a suffix of another), we only
 // take the time to sort when the user asks for heavy optimization.
 if (!this->optimize_)
   {
     // If we are not optimizing, the offsets are already assigned.
     offset = this->offset_;
   }
 else
   {
     size_t count = this->string_set_.size();

     std::vector<Stringpool_sort_info> v;
     v.reserve(count);

     for (typename String_set_type::iterator p = this->string_set_.begin();
          p != this->string_set_.end();
          ++p)
       v.push_back(Stringpool_sort_info(p));

     std::sort(v.begin(), v.end(), Stringpool_sort_comparison());

     section_offset_type last_offset = -1;
     for (typename std::vector<Stringpool_sort_info>::iterator last = v.end(),
            curr = v.begin();
          curr != v.end();
          last = curr++)
       {
         section_offset_type this_offset;
         if (this->zero_null_ && (*curr)->first.string[0] == 0)
           this_offset = 0;
         else if (last != v.end()
                  && ((((*curr)->first.length - (*last)->first.length)
                       % this->addralign_) == 0)
                  && is_suffix((*curr)->first.string,
                               (*curr)->first.length,
                               (*last)->first.string,
                               (*last)->first.length))
           this_offset = (last_offset
                          + (((*last)->first.length - (*curr)->first.length)
                             * charsize));
         else
           {
             this_offset = align_address(offset, this->addralign_);
             offset = this_offset + ((*curr)->first.length + 1) * charsize;
           }
         this->key_to_offset_[(*curr)->second - 1] = this_offset;
         last_offset = this_offset;
       }
   }

 this->strtab_size_ = offset;
}

// Get the offset of a string in the ELF strtab.  The string must
// exist.

template<typename Stringpool_char>
section_offset_type
Stringpool_template<Stringpool_char>::get_offset(const Stringpool_char* s)
 const
{
 return this->get_offset_with_length(s, string_length(s));
}

template<typename Stringpool_char>
section_offset_type
Stringpool_template<Stringpool_char>::get_offset_with_length(
   const Stringpool_char* s,
   size_t length) const
{
 gold_assert(this->strtab_size_ != 0);
 Hashkey hk(s, length);
 typename String_set_type::const_iterator p = this->string_set_.find(hk);
 if (p != this->string_set_.end())
   return this->key_to_offset_[p->second - 1];
 gold_unreachable();
}

// Write the ELF strtab into the buffer.

template<typename Stringpool_char>
void
Stringpool_template<Stringpool_char>::write_to_buffer(
   unsigned char* buffer,
   section_size_type bufsize)
{
 gold_assert(this->strtab_size_ != 0);
 gold_assert(bufsize >= this->strtab_size_);
 if (this->zero_null_)
   buffer[0] = '\0';
 for (typename String_set_type::const_iterator p = this->string_set_.begin();
      p != this->string_set_.end();
      ++p)
   {
     const int len = (p->first.length + 1) * sizeof(Stringpool_char);
     const section_offset_type offset = this->key_to_offset_[p->second - 1];
     gold_assert(static_cast<section_size_type>(offset) + len
                 <= this->strtab_size_);
     memcpy(buffer + offset, p->first.string, len);
   }
}

// Write the ELF strtab into the output file at the specified offset.

template<typename Stringpool_char>
void
Stringpool_template<Stringpool_char>::write(Output_file* of, off_t offset)
{
 gold_assert(this->strtab_size_ != 0);
 unsigned char* view = of->get_output_view(offset, this->strtab_size_);
 this->write_to_buffer(view, this->strtab_size_);
 of->write_output_view(offset, this->strtab_size_, view);
}

// Print statistical information to stderr.  This is used for --stats.

template<typename Stringpool_char>
void
Stringpool_template<Stringpool_char>::print_stats(const char* name) const
{
#if defined(HAVE_UNORDERED_MAP) || defined(HAVE_TR1_UNORDERED_MAP) || defined(HAVE_EXT_HASH_MAP)
 fprintf(stderr, _("%s: %s entries: %zu; buckets: %zu\n"),
         program_name, name, this->string_set_.size(),
         this->string_set_.bucket_count());
#else
 fprintf(stderr, _("%s: %s entries: %zu\n"),
         program_name, name, this->table_.size());
#endif
 fprintf(stderr, _("%s: %s Stringdata structures: %zu\n"),
         program_name, name, this->strings_.size());
}

// Instantiate the templates we need.

template
class Stringpool_template<char>;

template
class Stringpool_template<char16_t>;

template
class Stringpool_template<char32_t>;

} // End namespace gold.

You are viewing proxied material from ftp.icm.edu.pl. The copyright of proxied material belongs to its original authors. Any comments or complaints in relation to proxied material should be directed to the original authors of the content concerned. Please see the disclaimer for more details.