Fix crash when we more code to the gpu.

theano/sandbox/gpuarray/opt.py

@@ -347,10 +347,21 @@ def local_gpua_careduce(node):
                 else:
                     new_mask.append(reduce_mask[i])
                     new_in_shp.append(x_shape[i])
+            new_axis = []
+            for idx, m in enumerate(new_mask):
+                if m == 1:
+                    new_axis.append(idx)
+            new_greduce = GpuCAReduceCuda(
+                node.op.scalar_op,
+                axis=new_axis, reduce_mask=new_mask,
+                dtype=getattr(node.op, 'dtype', None),
+                acc_dtype=getattr(node.op, 'acc_dtype', None))
 
-            new_greduce = GpuCAReduceCuda(new_mask, scalar_op)
             reshaped_x = x.reshape(tensor.stack(*new_in_shp))
             gpu_reshaped_x = gpu_from_host(reshaped_x)
+            gvar = greduce(gpu_reshaped_x)
+            #We need to have the make node called, otherwise the mask can
+            #be None
             reshaped_gpu_inputs = [gpu_reshaped_x]
             if new_greduce.supports_c_code(reshaped_gpu_inputs):
                 reduce_reshaped_x = host_from_gpu(

theano/sandbox/gpuarray/tests/test_elemwise.py

@@ -133,12 +133,12 @@ class test_GpuCAReduceCuda(test_GpuCAReduceCPY):
              ((4100,2,3,4),[0,1,2,3]),((2,4100,3,4),[0,1,2,3]),((2,3,4100,4),[0,1,2,3]),((2,3,4,4100),[0,1,2,3]),((128,1,3,3), [0,1,2,3]),#1111
 
              #test pattern implemented by reshape
-#             ((4100,4,3,2),[0]),((4,4100,3,2),[0]),((4,3,4100,2),[0]),((4,3,2,4100),[0]),#1000
-#             ((4100,4,3,2),[1]),((4,4100,3,2),[1]),((4,3,4100,2),[1]),((4,3,2,4100),[1]),#0100
-#             ((4100,4,3,2),[2]),((4,4100,3,2),[2]),((4,3,4100,2),[2]),((4,3,2,4100),[2]),#0010
-#             ((4100,4,3,2),[3]),((4,4100,3,2),[3]),((4,3,4100,2),[3]),((4,3,2,4100),[3]),#0001
-#             ((1100,2,3,4,5),[0,1,2,3,4]),((2,1100,3,4,5),[0,1,2,3,4]),((2,3,1100,4,5),[0,1,2,3,4]),((2,3,4,1100,5),[0,1,2,3,4]),((2,3,4,5,1100),[0,1,2,3,4]),#11111
-#             ((5,4,3,10,11),[1,2]),
+             ((4100,4,3,2),[0]),((4,4100,3,2),[0]),((4,3,4100,2),[0]),((4,3,2,4100),[0]),#1000
+             ((4100,4,3,2),[1]),((4,4100,3,2),[1]),((4,3,4100,2),[1]),((4,3,2,4100),[1]),#0100
+             ((4100,4,3,2),[2]),((4,4100,3,2),[2]),((4,3,4100,2),[2]),((4,3,2,4100),[2]),#0010
+             ((4100,4,3,2),[3]),((4,4100,3,2),[3]),((4,3,4100,2),[3]),((4,3,2,4100),[3]),#0001
+             ((1100,2,3,4,5),[0,1,2,3,4]),((2,1100,3,4,5),[0,1,2,3,4]),((2,3,1100,4,5),[0,1,2,3,4]),((2,3,4,1100,5),[0,1,2,3,4]),((2,3,4,5,1100),[0,1,2,3,4]),#11111
+             ((5,4,3,10,11),[1,2]),
         ]
     op = GpuCAReduceCuda
     reds = [scalar.add, scalar.mul,