/************************************************************************

MODULE: BasicThreadPool

SUMMARY:

A simple thread pool class BasicThreadPool, as well as some higher-level macros
which facilitite simple parallel for loops.


***************************************************************************/


// ********************** Simple parallel for loops **************************
//
// We begin with a description of the higher-level macros for writing simple
// parallel for loops.  These facilitaties are activated only when NTL is
// configured with NTL_THREAD_BOOST=on (which implies NTL_THREADS=on).
// However, code that uses these facilties should still compile and run
// correctly even when NTL_THREAD_BOOST=off, or even when NTL_THREADS=off, so
// this is the simplest way to write parallel for loops across a range of
// compile-time and run-time environments.  Note that if NTL_THREADS=on, C++11
// features are reqired, but when NTL_THREADS=off, these features are not
// required, so the code should compile on older C++ compilers.
//
// Here is a simple recipe for writing parallel for loop.
//
// At the start of program execution, your program should execute

   SetNumThreads(nt);

// You can choose nt to be any positive integer, but for best results, it
// should correspond to the number of available cores on your machine.
// [NOTE: if NTL_THREAD_BOOST=off, this function is still defined, but does
// nothing.]
//
// Now consider the following routine:

   void mul(ZZ *x, const ZZ *a, const ZZ *b, long n)
   {
      for (long i = 0; i < n; i++)
         mul(x[i], a[i], b[i]);
   }

// We can parallelize it as follows:

   void mul(ZZ *x, const ZZ *a, const ZZ *b, long n)
   {
      NTL_EXEC_RANGE(n, first, last)

         for (long i = first; i < last; i++)
            mul(x[i], a[i], b[i]);

      NTL_EXEC_RANGE_END
   }

// NTL_EXEC_RANGE and NTL_EXEC_RANGE_END are macros that just "do the right
// thing".  If there are nt threads available, the interval [0..n) will be
// partitioned into (up to)  nt subintervals, and a different thread will be
// used to process each subinterval. You still have to write the for loop
// yourself: the macro just declares and initializes variables "first" and
// "last" (or whatever you want to call them) of type long that represent the
// subinterval [first..last) to be processed by one thread.
//
// Note that the current thread participates as one of the nt available
// threads, and that the current thread will wait for all other participating threads
// to finish their task before proceeding. The current thread can be identified
// as the one with first == 0.
//
// Withing the "body" of this construct, you can freely reference any variables
// that are visible at this point.  This is implemented using the C++ lambda
// feature (capturing all variables by reference).
//
// This construct will still work even if threads are disabled, in which case
// it runs single-threaded with first=0 and last=n.
//
// Note that the code within the EXEC_RANGE body could call other routines that
// themselves attempt to execute an EXEC_RANGE: if this happens, the latter
// EXEC_RANGE will detect this and run single-threaded.
//
// You may wish to do other things within the EXEC_RANGE body than just execute
// a loop.  One thing you may want to do is to declare variables.  Another
// thing you may want to do is setup a local context for a ZZ_p modulus (or
// other type of modulus).  Here is an example of doing this:


   void mul(ZZ_p *x, const ZZ_p *a, const ZZ_p *b, long n)
   {
      ZZ_pContext context;
      context.save();

      NTL_EXEC_RANGE(n, first, last)
      
         context.restore();

         for (long i = first; i < last; i++)
            mul(x[i], a[i], b[i]);

      NTL_EXEC_RANGE_END
   }


// Another useful function is AvailableThreads(), which will return the number
// of available threads.  If threads or thread boosting is not enabled, this
// will return 1.  Even if thread boosting is enabled, this may return 1 if for
// whatever reason, the thread pool is not available for use (for example,
// SetNumThreads was never called, or the thread pool is already active).
//
// A lower-level set of tools is available, which allow you to simply run a
// specified number of threads.  Assuming nt <= AvailableThreads(), the code

   NTL_EXEC_INDEX(nt, index)

      ... code ...

   NTL_EXEC_INDEX_END

// will execute the body on nt different threads, each with a unique index in
// the range [0..nt).  A variable named "index" (or whatever name you specify)
// of type long will hold the given index.  Just as with EXEC_RANGE, the current
// thread will participate as one of the nt threads, and will always be
// assigned an index of 0.
//
// This tool is useful if you need to manage memory a bit more carefully.  For
// example, the following code will compute an inner product using all
// available threads:

   ZZ InnerProd(const ZZ *a, const ZZ *b, long n)
   {
      PartitionInfo pinfo(n);

      long cnt = pinfo.NumIntervals();

      Vec<ZZ> acc;
      acc.SetLength(cnt);

      NTL_EXEC_INDEX(cnt, index)

         long first, last;
         pinfo.interval(first, last, index);

         ZZ& sum = acc[index];
         sum = 0;
         for (long i = first; i < last; i++)
            MulAddTo(sum, a[i], b[i]);

      NTL_EXEC_INDEX_END

      ZZ sum;
      sum = 0;
      for (long i = 0; i < cnt; i++)
         sum += acc[i];

      return sum;
   }

// This example also illustrates the class PartitionInfo, which is useful for
// partitioning a large interval into smaller intervals (it is used internally
// by EXEC_RANGE).  The constructor takes a single argument (in this example n)
// and computes a partition of [0..n) into nearly equally sized subintervals.
// The method NumIntervals() returns the number of subintervals, and the method
// interval(first, last, index) sets first and last according to the endpoints
// of the subinterval [first..last) with the given index.
//
// So in this example, cnt threads will run, each accumulating a sum into a
// corresponding element of the vector acc, and afterwords, these elements are
// summed.
//
// Note that if threads are not enabled or otherwise unavailable, the above
// code will compile and run correctly (just using one thread).
//
// Finally, there is a "guarded" version of NTL_EXEC_RANGE called
// NTL_GEXEC_RANGE.  This allows one to dynamically "guard" against parallel
// execution. For example, on very small problems the runtime overhead of a
// parallel for loop may not be worthwhile, or in other situations parallel
// execution could cause incorrect behavior.  See below for details.


// ************************** Thread Pools ******************************
//
// The above facilities are built on top of a more general thread pool class,
// which you may use for your own purposes.
//    
// You create a thread pool by constructing a BasicThreadPool object.  For
// example:

   long nthreads = 4;
   BasicThreadPool pool(nthreads);

// creates a thread pool of 4 threads.  These threads will exist until the
// destructor for pool is called.  
//
// The simplest way to use a thread pools is as follows.  Suppose you have a
// task that consists of sz subtasks, indexed 0..sz-1.  Then you can write:

   pool.exec_range(sz,
      [&](long first, long last) {
         for (long i = first; i < last; i++) {
            ... code to process subtask i ...
         }
      }
   );

// The second argument to exec_range is a C++11 "lambda".  The "[&]" indicates
// that all local variables in the calling context are captured by reference,
// so the lambda body can reference all visible local variables directly.
// C++11 provides other methods for capturing local variables.  The interval
// [0..sz) is partitioned into subintervals of the form [first..last), which
// are processed by the code in the supplied lambda.
//
// A lower-level interface is also provided.  One can write:

   pool.exec_index(cnt,
      [&](long index) {
         ... code to process index i ...
      }
   );

// This will activate exactly cnt threads with indices 0..cnt-1, and execute
// the given code on each index.  The parameter cnt must not exceed nthreads,
// otherwise an error is raised.


// ====================================================================
//
// NOTES:
//
// When one activates a thread pool with nthreads threads, the *current* thread
// (the one activating the pool) will also participate in the computation.
// This means that the thread pool only contains nthreads-1 other threads.
//
// If, during an activation, any thread throws an exception, it will be caught
// and rethrown in the activating thread when all the threads complete.  If
// more than one thread throws an exception, the first one that is caught is
// the one that is rethrown.
//
// Methods are also provided for adding, deleting, and moving threads in and
// among thread pools.
//
// If NTL_THREADS=off, the corresponding header file may be included, but the
// BasicThreadPool class is not defined.
//
// Unlike most classes in NTL, the BasicThreadPool is not relocatable and hence
// cannot be used in a Vec.  One should first wrap it in a pointer class, such
// as UniquePtr.



// class BasicThreadPool: provided basic functionality for thread pools

class BasicThreadPool {
private:

  BasicThreadPool(const BasicThreadPool&); // disabled
  void operator=(const BasicThreadPool&); // disabled

public:

  explicit
  BasicThreadPool(long nthreads);
  // creates a pool with nthreads threads, including the current thread
  // (so nthreads-1 other threads get created)

  template<class Fct>
  void exec_range(long sz, const Fct& fct);
  // activate by range (see example usage above)

  template<class Fct>
  void exec_index(long cnt, const Fct& fct);
  // activate by index (see example usage above)

  void add(long n = 1);
  // add n threads to the pool

  long NumThreads() const;
  // return number of threads (including current thread)

  void remove(long n = 1);
  // remove n threads from the pool
  
  void move(BasicThreadPool& other, long n = 1)
  // move n threads from other pool to this pool

  bool active() const;
  // indicates an activation is in process: invoking any of the methods
  // exec_index, exec_range, add, remove, move, or the destructor
  // whie active will raise an error

  template<class Fct>
  static void relaxed_exec_range(BasicThreadPool *pool, long sz, const Fct& fct);
  // similar to pool->exec_range(sz, fct), but will still work even
  // if !pool or pool->active(), using just the current thread

  template<class Fct>
  static void relaxed_exec_index(BasicThreadPool *pool, long cnt, const Fct& fct);
  // similar to pool->exec_index(cnt, fct), but will still work even
  // if !pool or pool->active(), provided cnt <= 1, using just the current thread

};




// THREAD BOOSTING FEATURES:

void SetNumThreads(long nt);
// convenience routine to set NTL's thread pool.
// If called more than once, the old thread pool is destroyed and
// replaced by a new one.
// If NTL_THREAD_BOOST=off, then this is still defined, but does nothing.

long AvailableThreads();
// Number of threads currently availble to use in NTL's thread pool.  This is
// always at least 1 (for the current thread).  
// If NTL_THREAD_BOOST=off, then this is still defined, and always returns 1.

BasicThreadPool *GetThreadPool();
void ResetThreadPool(BasicThreadPool *pool = 0);
BasicThreadPool *ReleaseThreadPool();
// Routines to get and set NTL's thread pool.  The interfaces parallel NTL's
// UniquePtr class, and indeed, behind the scenes, NTL's thread pool is stored
// as a UniquePtr<BasicThreadPool>.
// These are only declared when NTL_THREAD_BOOST=on.  


#define NTL_EXEC_RANGE(sz, first, last) ...
#define NTL_EXEC_RANGE_END ...
#define NTL_EXEC_INDEX(cnt, index) ...
#define NTL_EXEC_INDEX_END ...
// convenience macros to implement "parallel for loops" using NTL's thread
// pool.  See examples above for usage.  If NTL_THREAD_BOOST=off, then these
// are still defined, and code will run on a single thread


#define NTL_GEXEC_RANGE(seq, sz, first, last) ...
#define NTL_GEXEC_RANGE_END ...
// "guarded" version of NTL_EXEC_RANGE: if seq evaluates to true, the code runs
// on a single thread.  This is useful in avoiding situations where the
// overhead of a parallel loop is too high.  If seq evaluates to the constant
// true, a good compiler will optimize code to run on a single thread, with no
// overhead.

#define NTL_IMPORT(x)
// To be used in conjunction with NTL_EXEC_RANGE and friends.  When
// NTL_THREAD_BOOST=on, this will copy the variable named x from the enclosing
// scope to a local copy.  This should only be used for types with cheap
// copies, such as scalars and pointers.  In some situations, this allows the
// compiler to optimize a bit more aggressively.  One or more of these may be
// placed right after an NTL_EXEC_RANGE.
// When NTL_THREAD_BOOST=off, this is still defined, and does nothing.


// class PartitionInfo: A helper class to facilitate partitioning an interval
// into subintervals.  NOTE: this class is available, even when
// NTL_THREAD_BOOST=off.

class PartitionInfo {
public:

   explicit
   PartitionInfo(long sz, long nt = AvailableThreads());
   // partitions [0..sz) into at most nt subintervals.  sz may be 0 or
   // negative, in which case the number of subintervals is 0.

   long NumIntervals() const;
   // return the number of subintervals

   void interval(long& first, long& last, long i) const;
   // [first..last) is the ith interval, where i in [0..NumInvervals()).  No
   // range checking is performed.

};