Added an explicit cast in c_code for AdvancedSubtensor1 Op to match logic in…

Added an explicit cast in c_code for AdvancedSubtensor1 Op to match logic in perform() and avoid test failures on 32 bit platforms. The real issue seem to be that x.shape is a 64-bit tensor regardless of the size of platform's size_t type. This leads to int64 type appearance in any computation involving shape and resulting index cannot be passed back to numpy.take() necessitating the cast.

Added an explicit cast in c_code for AdvancedSubtensor1 Op to match logic in…
967ba312 · abalkin · 856c24a6 · 967ba312
--- a/theano/tensor/basic.py
+++ b/theano/tensor/basic.py
@@ -6559,23 +6559,30 @@ class AdvancedSubtensor1(Op):
        output_name = output_names[0]
        fail = sub['fail']
        return """
+            PyObject *indices;
+            // This cast makes c_code mimic the logic in perform(), but
+            // also makes theano code less safe.
+            indices = PyArray_Cast(%(i_name)s, NPY_INTP);
+            if (indices == NULL) {
+                %(fail)s;
+            }
            if (%(output_name)s != NULL) {
                npy_intp nd, i, *shape;
-                nd = PyArray_NDIM(%(a_name)s) + PyArray_NDIM(%(i_name)s) - 1;
+                nd = PyArray_NDIM(%(a_name)s) + PyArray_NDIM(indices) - 1;
                if (PyArray_NDIM(%(output_name)s) != nd) {
                    Py_CLEAR(%(output_name)s);
                }
                else {
                    shape = PyArray_DIMS(%(output_name)s);
-                    for (i = 0; i < PyArray_NDIM(%(i_name)s); i++) {
-                        if (shape[i] != PyArray_DIMS(%(i_name)s)[i]) {
+                    for (i = 0; i < PyArray_NDIM(indices); i++) {
+                        if (shape[i] != PyArray_DIMS(indices)[i]) {
                            Py_CLEAR(%(output_name)s);
                            break;
                        }
                    }
                    if (%(output_name)s != NULL) {
                        for (; i < nd; i++) {
-                            if (shape[i] != PyArray_DIMS(%(a_name)s)[i-PyArray_NDIM(%(i_name)s)+1]) {
+                            if (shape[i] != PyArray_DIMS(%(a_name)s)[i-PyArray_NDIM(indices)+1]) {
                                Py_CLEAR(%(output_name)s);
                                break;
                            }
@@ -6583,13 +6590,14 @@ class AdvancedSubtensor1(Op):
                    }
                }
            }
-            %(output_name)s = (PyArrayObject*)PyArray_TakeFrom(%(a_name)s, (PyObject*)%(i_name)s, 0,
+            %(output_name)s = (PyArrayObject*)PyArray_TakeFrom(%(a_name)s, indices, 0,
                                                               %(output_name)s, NPY_RAISE);
+            Py_DECREF(indices);
            if (%(output_name)s == NULL) %(fail)s;
        """ % locals()

    def c_code_cache_version(self):
-        return (0, 0, 2)
+        return (0, 0, 3)

 advanced_subtensor1 = AdvancedSubtensor1()