/*

  *

  * Copyright (c) 1994

  * Hewlett-Packard Company

  *

  * Copyright (c) 1996,1997

  * Silicon Graphics Computer Systems, Inc.

  *

  * Copyright (c) 1997

  * Moscow Center for SPARC Technology

  *

  * Copyright (c) 1999 

  * Boris Fomitchev

  *

  * This material is provided "as is", with absolutely no warranty expressed

  * or implied. Any use is at your own risk.

  *

  * Permission to use or copy this software for any purpose is hereby granted 

  * without fee, provided the above notices are retained on all copies.

  * Permission to modify the code and to distribute modified code is granted,

  * provided the above notices are retained, and a notice that the code was

  * modified is included with the above copyright notice.

  *

  */

 #ifndef _STLP_PTHREAD_ALLOC_H

 #define _STLP_PTHREAD_ALLOC_H

 // Pthread-specific node allocator.

 // This is similar to the default allocator, except that free-list

 // information is kept separately for each thread, avoiding locking.

 // This should be reasonably fast even in the presence of threads.

 // The down side is that storage may not be well-utilized.

 // It is not an error to allocate memory in thread A and deallocate

 // it in thread B.  But this effectively transfers ownership of the memory,

 // so that it can only be reallocated by thread B.  Thus this can effectively

 // result in a storage leak if it's done on a regular basis.

 // It can also result in frequent sharing of

 // cache lines among processors, with potentially serious performance

 // consequences.

 #include <pthread.h>

 #ifndef _STLP_INTERNAL_ALLOC_H

 #include <stl/_alloc.h>

 #endif

 #ifndef __RESTRICT

 #  define __RESTRICT

 #endif

 _STLP_BEGIN_NAMESPACE

 #define _STLP_DATA_ALIGNMENT 8

 union _Pthread_alloc_obj {

     union _Pthread_alloc_obj * __free_list_link;

     char __client_data[_STLP_DATA_ALIGNMENT];    /* The client sees this.    */

 };

 // Pthread allocators don't appear to the client to have meaningful

 // instances.  We do in fact need to associate some state with each

 // thread.  That state is represented by

 // _Pthread_alloc_per_thread_state<_Max_size>.

 template<size_t _Max_size>

 struct _Pthread_alloc_per_thread_state {

   typedef _Pthread_alloc_obj __obj;

   enum { _S_NFREELISTS = _Max_size/_STLP_DATA_ALIGNMENT };

   // Free list link for list of available per thread structures.

   // When one of these becomes available for reuse due to thread

   // termination, any objects in its free list remain associated

   // with it.  The whole structure may then be used by a newly

   // created thread.

   _Pthread_alloc_per_thread_state() : __next(0)

   {

     memset((void *)__free_list, 0, (size_t)_S_NFREELISTS * sizeof(__obj *));

   }

   // Returns an object of size __n, and possibly adds to size n free list.

   void *_M_refill(size_t __n);

   _Pthread_alloc_obj* volatile __free_list[_S_NFREELISTS]; 

   _Pthread_alloc_per_thread_state<_Max_size> * __next; 

   // this data member is only to be used by per_thread_allocator, which returns memory to the originating thread.

   _STLP_mutex _M_lock;

  };

 // Pthread-specific allocator.

 // The argument specifies the largest object size allocated from per-thread

 // free lists.  Larger objects are allocated using malloc_alloc.

 // Max_size must be a power of 2.

 template < __DFL_NON_TYPE_PARAM(size_t, _Max_size, _MAX_BYTES) >

 class _Pthread_alloc {

 public: // but only for internal use:

   typedef _Pthread_alloc_obj __obj;

   typedef _Pthread_alloc_per_thread_state<_Max_size> __state_type;

   typedef char value_type;

   // Allocates a chunk for nobjs of size size.  nobjs may be reduced

   // if it is inconvenient to allocate the requested number.

   static char *_S_chunk_alloc(size_t __size, size_t &__nobjs);

   enum {_S_ALIGN = _STLP_DATA_ALIGNMENT};

   static size_t _S_round_up(size_t __bytes) {

         return (((__bytes) + (int)_S_ALIGN-1) & ~((int)_S_ALIGN - 1));

   }

   static size_t _S_freelist_index(size_t __bytes) {

         return (((__bytes) + (int)_S_ALIGN-1)/(int)_S_ALIGN - 1);

   }

 private:

   // Chunk allocation state. And other shared state.

   // Protected by _S_chunk_allocator_lock.

   static _STLP_mutex_base _S_chunk_allocator_lock;

   static char *_S_start_free;

   static char *_S_end_free;

   static size_t _S_heap_size;

   static _Pthread_alloc_per_thread_state<_Max_size>* _S_free_per_thread_states;

   static pthread_key_t _S_key;

   static bool _S_key_initialized;

         // Pthread key under which per thread state is stored. 

         // Allocator instances that are currently unclaimed by any thread.

   static void _S_destructor(void *instance);

         // Function to be called on thread exit to reclaim per thread

         // state.

   static _Pthread_alloc_per_thread_state<_Max_size> *_S_new_per_thread_state();

 public:

         // Return a recycled or new per thread state.

   static _Pthread_alloc_per_thread_state<_Max_size> *_S_get_per_thread_state();

 private:

         // ensure that the current thread has an associated

         // per thread state.

   class _M_lock;

   friend class _M_lock;

   class _M_lock {

       public:

         _M_lock () { _S_chunk_allocator_lock._M_acquire_lock(); }

         ~_M_lock () { _S_chunk_allocator_lock._M_release_lock(); }

   };

 public:

   /* n must be > 0      */

   static void * allocate(size_t __n)

   {

     __obj * volatile * __my_free_list;

     __obj * __RESTRICT __result;

     __state_type* __a;

     if (__n > _Max_size) {

         return(__malloc_alloc<0>::allocate(__n));

     }

     __a = _S_get_per_thread_state();

     __my_free_list = __a -> __free_list + _S_freelist_index(__n);

     __result = *__my_free_list;

     if (__result == 0) {

         void *__r = __a -> _M_refill(_S_round_up(__n));

         return __r;

     }

     *__my_free_list = __result -> __free_list_link;

     return (__result);

   };

   /* p may not be 0 */

   static void deallocate(void *__p, size_t __n)

   {

     __obj *__q = (__obj *)__p;

     __obj * volatile * __my_free_list;

     __state_type* __a;

     if (__n > _Max_size) {

         __malloc_alloc<0>::deallocate(__p, __n);

         return;

     }

     __a = _S_get_per_thread_state();

     __my_free_list = __a->__free_list + _S_freelist_index(__n);

     __q -> __free_list_link = *__my_free_list;

     *__my_free_list = __q;

   }

   // boris : versions for per_thread_allocator

   /* n must be > 0      */

   static void * allocate(size_t __n, __state_type* __a)

   {

     __obj * volatile * __my_free_list;

     __obj * __RESTRICT __result;

     if (__n > _Max_size) {

         return(__malloc_alloc<0>::allocate(__n));

     }

     // boris : here, we have to lock per thread state, as we may be getting memory from

     // different thread pool.

     _STLP_mutex_lock __lock(__a->_M_lock);

     __my_free_list = __a -> __free_list + _S_freelist_index(__n);

     __result = *__my_free_list;

     if (__result == 0) {

         void *__r = __a -> _M_refill(_S_round_up(__n));

         return __r;

     }

     *__my_free_list = __result -> __free_list_link;

     return (__result);

   };

   /* p may not be 0 */

   static void deallocate(void *__p, size_t __n, __state_type* __a)

   {

     __obj *__q = (__obj *)__p;

     __obj * volatile * __my_free_list;

     if (__n > _Max_size) {

         __malloc_alloc<0>::deallocate(__p, __n);

         return;

     }

     // boris : here, we have to lock per thread state, as we may be returning memory from

     // different thread.

     _STLP_mutex_lock __lock(__a->_M_lock);

     __my_free_list = __a->__free_list + _S_freelist_index(__n);

     __q -> __free_list_link = *__my_free_list;

     *__my_free_list = __q;

   }

   static void * reallocate(void *__p, size_t __old_sz, size_t __new_sz);

 } ;

 # if defined (_STLP_USE_TEMPLATE_EXPORT)

 _STLP_EXPORT_TEMPLATE_CLASS _Pthread_alloc<_MAX_BYTES>;

 # endif

 typedef _Pthread_alloc<_MAX_BYTES> __pthread_alloc;

 typedef __pthread_alloc pthread_alloc;

 template <class _Tp>

 class pthread_allocator {

   typedef pthread_alloc _S_Alloc;          // The underlying allocator.

 public:

   typedef size_t     size_type;

   typedef ptrdiff_t  difference_type;

   typedef _Tp*       pointer;

   typedef const _Tp* const_pointer;

   typedef _Tp&       reference;

   typedef const _Tp& const_reference;

   typedef _Tp        value_type;

 #ifdef _STLP_MEMBER_TEMPLATE_CLASSES

   template <class _NewType> struct rebind {

     typedef pthread_allocator<_NewType> other;

   };

 #endif

   pthread_allocator() _STLP_NOTHROW {}

   pthread_allocator(const pthread_allocator<_Tp>& a) _STLP_NOTHROW {}

 #if defined (_STLP_MEMBER_TEMPLATES) /* && defined (_STLP_FUNCTION_PARTIAL_ORDER) */

   template <class _OtherType> pthread_allocator(const pthread_allocator<_OtherType>&)

 		_STLP_NOTHROW {}

 #endif

   ~pthread_allocator() _STLP_NOTHROW {}

   pointer address(reference __x) const { return &__x; }

   const_pointer address(const_reference __x) const { return &__x; }

   // __n is permitted to be 0.  The C++ standard says nothing about what

   // the return value is when __n == 0.

   _Tp* allocate(size_type __n, const void* = 0) {

     return __n != 0 ? __STATIC_CAST(_Tp*,_S_Alloc::allocate(__n * sizeof(_Tp)))

                     : 0;

   }

   // p is not permitted to be a null pointer.

   void deallocate(pointer __p, size_type __n)

     { _S_Alloc::deallocate(__p, __n * sizeof(_Tp)); }

   size_type max_size() const _STLP_NOTHROW 

     { return size_t(-1) / sizeof(_Tp); }

   void construct(pointer __p, const _Tp& __val) { _STLP_PLACEMENT_NEW (__p) _Tp(__val); }

   void destroy(pointer _p) { _p->~_Tp(); }

 };

 _STLP_TEMPLATE_NULL

 class _STLP_CLASS_DECLSPEC pthread_allocator<void> {

 public:

   typedef size_t      size_type;

   typedef ptrdiff_t   difference_type;

   typedef void*       pointer;

   typedef const void* const_pointer;

   typedef void        value_type;

 #ifdef _STLP_MEMBER_TEMPLATE_CLASSES

   template <class _NewType> struct rebind {

     typedef pthread_allocator<_NewType> other;

   };

 #endif

 };

 template <class _T1, class _T2>

 inline bool operator==(const pthread_allocator<_T1>&,

                        const pthread_allocator<_T2>& a2) 

 {

   return true;

 }

 #ifdef _STLP_FUNCTION_TMPL_PARTIAL_ORDER

 template <class _T1, class _T2>

 inline bool operator!=(const pthread_allocator<_T1>&,

                        const pthread_allocator<_T2>&)

 {

   return false;

 }

 #endif

 #ifdef _STLP_CLASS_PARTIAL_SPECIALIZATION

 # ifdef _STLP_USE_RAW_SGI_ALLOCATORS

 template <class _Tp, size_t _Max_size>

 struct _Alloc_traits<_Tp, _Pthread_alloc<_Max_size> >

 {

   typedef __allocator<_Tp, _Pthread_alloc<_Max_size> > 

           allocator_type;

 };

 # endif

 template <class _Tp, class _Atype>

 struct _Alloc_traits<_Tp, pthread_allocator<_Atype> >

 {

   typedef pthread_allocator<_Tp> allocator_type;

 };

 #endif

 #if !defined (_STLP_USE_NESTED_TCLASS_THROUGHT_TPARAM)

 template <class _Tp1, class _Tp2>

 inline pthread_allocator<_Tp2>&

 __stl_alloc_rebind(pthread_allocator<_Tp1>& __x, const _Tp2*) {

   return (pthread_allocator<_Tp2>&)__x;

 }

 template <class _Tp1, class _Tp2>

 inline pthread_allocator<_Tp2>

 __stl_alloc_create(pthread_allocator<_Tp1>&, const _Tp2*) {

   return pthread_allocator<_Tp2>();

 }

 #endif /* _STLP_USE_NESTED_TCLASS_THROUGHT_TPARAM */

 //

 // per_thread_allocator<> : this allocator always return memory to the same thread 

 // it was allocated from.

 //

 template <class _Tp>

 class per_thread_allocator {

   typedef pthread_alloc _S_Alloc;          // The underlying allocator.

   typedef pthread_alloc::__state_type __state_type;

 public:

   typedef size_t     size_type;

   typedef ptrdiff_t  difference_type;

   typedef _Tp*       pointer;

   typedef const _Tp* const_pointer;

   typedef _Tp&       reference;

   typedef const _Tp& const_reference;

   typedef _Tp        value_type;

 #ifdef _STLP_MEMBER_TEMPLATE_CLASSES

   template <class _NewType> struct rebind {

     typedef per_thread_allocator<_NewType> other;

   };

 #endif

   per_thread_allocator() _STLP_NOTHROW { 

     _M_state = _S_Alloc::_S_get_per_thread_state();

   }

   per_thread_allocator(const per_thread_allocator<_Tp>& __a) _STLP_NOTHROW : _M_state(__a._M_state){}

 #if defined (_STLP_MEMBER_TEMPLATES) /* && defined (_STLP_FUNCTION_PARTIAL_ORDER) */

   template <class _OtherType> per_thread_allocator(const per_thread_allocator<_OtherType>& __a)

 		_STLP_NOTHROW : _M_state(__a._M_state) {}

 #endif

   ~per_thread_allocator() _STLP_NOTHROW {}

   pointer address(reference __x) const { return &__x; }

   const_pointer address(const_reference __x) const { return &__x; }

   // __n is permitted to be 0.  The C++ standard says nothing about what

   // the return value is when __n == 0.

   _Tp* allocate(size_type __n, const void* = 0) {

     return __n != 0 ? __STATIC_CAST(_Tp*,_S_Alloc::allocate(__n * sizeof(_Tp), _M_state)): 0;

   }

   // p is not permitted to be a null pointer.

   void deallocate(pointer __p, size_type __n)

     { _S_Alloc::deallocate(__p, __n * sizeof(_Tp), _M_state); }

   size_type max_size() const _STLP_NOTHROW 

     { return size_t(-1) / sizeof(_Tp); }

   void construct(pointer __p, const _Tp& __val) { _STLP_PLACEMENT_NEW (__p) _Tp(__val); }

   void destroy(pointer _p) { _p->~_Tp(); }

   // state is being kept here

   __state_type* _M_state;

 };

 _STLP_TEMPLATE_NULL

 class _STLP_CLASS_DECLSPEC per_thread_allocator<void> {

 public:

   typedef size_t      size_type;

   typedef ptrdiff_t   difference_type;

   typedef void*       pointer;

   typedef const void* const_pointer;

   typedef void        value_type;

 #ifdef _STLP_MEMBER_TEMPLATE_CLASSES

   template <class _NewType> struct rebind {

     typedef per_thread_allocator<_NewType> other;

   };

 #endif

 };

 template <class _T1, class _T2>

 inline bool operator==(const per_thread_allocator<_T1>& __a1,

                        const per_thread_allocator<_T2>& __a2) 

 {

   return __a1._M_state == __a2._M_state;

 }

 #ifdef _STLP_FUNCTION_TMPL_PARTIAL_ORDER

 template <class _T1, class _T2>

 inline bool operator!=(const per_thread_allocator<_T1>& __a1,

                        const per_thread_allocator<_T2>& __a2)

 {

   return __a1._M_state != __a2._M_state;

 }

 #endif

 #ifdef _STLP_CLASS_PARTIAL_SPECIALIZATION

 template <class _Tp, class _Atype>

 struct _Alloc_traits<_Tp, per_thread_allocator<_Atype> >

 {

   typedef per_thread_allocator<_Tp> allocator_type;

 };

 #endif

 #if !defined (_STLP_USE_NESTED_TCLASS_THROUGHT_TPARAM)

 template <class _Tp1, class _Tp2>

 inline per_thread_allocator<_Tp2>&

 __stl_alloc_rebind(per_thread_allocator<_Tp1>& __x, const _Tp2*) {

   return (per_thread_allocator<_Tp2>&)__x;

 }

 template <class _Tp1, class _Tp2>

 inline per_thread_allocator<_Tp2>

 __stl_alloc_create(per_thread_allocator<_Tp1>&, const _Tp2*) {

   return per_thread_allocator<_Tp2>();

 }

 #endif /* _STLP_USE_NESTED_TCLASS_THROUGHT_TPARAM */

 _STLP_END_NAMESPACE

 # if defined (_STLP_EXPOSE_GLOBALS_IMPLEMENTATION) && !defined (_STLP_LINK_TIME_INSTANTIATION)

 #  include <stl/_pthread_alloc.c>

 # endif

 #endif /* _STLP_PTHREAD_ALLOC */

 // Local Variables:

 // mode:C++

 // End: