Merge pull request #47 from IntelPython/adjust-workers-behavior-in-ff…

…t-backend

Adjusted workes to threads logic to agree with what is in scipy.fft

mkl_fft/_scipy_fft_backend.py

@@ -39,6 +39,35 @@
 )
 
 from numpy.core import (array, asarray, shape, conjugate, take, sqrt, prod)
+from os import cpu_count as os_cpu_count
+import warnings
+
+class _cpu_max_threads_count:
+    def __init__(self):
+        self.cpu_count = None
+        self.max_threads_count = None
+
+    def get_cpu_count(self):
+        max_threads = self.get_max_threads_count()
+        if self.cpu_count is None:
+            self.cpu_count = os_cpu_count()
+            if self.cpu_count > max_threads:
+                warnings.warn(
+                    ("os.cpu_count() returned value of {} greater than mkl.get_max_threads()'s value of {}. "
+                               "Using negative values of worker option may amount to requesting more threads than "
+                               "Intel(R) MKL can acommodate."
+                    ).format(self.cpu_count, max_threads))
+        return self.cpu_count
+
+    def get_max_threads_count(self):
+        if self.max_threads_count is None:
+            self.max_threads_count = mkl.get_max_threads()
+
+        return self.max_threads_count
+
+
+_hardware_counts = _cpu_max_threads_count()
+
 
 __all__ = ['fft', 'ifft', 'fft2', 'ifft2', 'fftn', 'ifftn',
            'rfft', 'irfft', 'rfft2', 'irfft2', 'rfftn', 'irfftn',
@@ -101,9 +130,20 @@ def _tot_size(x, axes):
 
 
 def _workers_to_num_threads(w):
+    """Handle conversion of workers to a positive number of threads in the
+    same way as scipy.fft.helpers._workers.
+    """
     if w is None:
-        return mkl.domain_get_max_threads(domain='fft')
-    return int(w)
+        return get_workers()
+    _w = int(w)
+    if (_w == 0):
+        raise ValueError("Number of workers must be nonzero")
+    if (_w < 0):
+        _w += _hardware_counts.get_cpu_count() + 1
+        if _w <= 0:
+            raise ValueError("workers value out of range; got {}, must not be"
+                             " less than {}".format(w, -_hardware_counts.get_cpu_count()))
+    return _w
 
 
 class Workers:
@@ -119,8 +159,8 @@ def __enter__(self):
 
     def __exit__(self, *args):
         # restore default
-        max_num_threads = mkl.domain_get_max_threads(domain='fft')
-        mkl.domain_set_num_threads(max_num_threads, domain='fft')
+        n_threads = _hardware_counts.get_max_threads_count()
+        mkl.domain_set_num_threads(n_threads, domain='fft')
 
 
 @_implements(_fft.fft)