Adding the documentation.

doc/library/config.txt

@@ -397,6 +397,17 @@ import theano and print the config variable, as in:
     the user and skip this optimization ('warn'), raise the exception
     ('raise'), or fall into the pdb debugger ('pdb').
 
+
+.. attribute:: assert_no_cpu_op
+
+    String value: 'ignore' or 'warn' or 'raise' or 'pdb'
+
+    Default: 'ignore'
+
+    If there is a CPU op in the computational graph, depending on its value;
+    this flag can either raise a warning, an exception or stop the 
+    compilation with pdb.
+
 .. attribute:: on_shape_error
 
     String value: 'warn' or 'raise'
@@ -407,6 +418,7 @@ import theano and print the config variable, as in:
     node, either warn the user and use a default value ('warn'), or
     raise the exception ('raise').
 
+
 .. attribute:: config.warn.ignore_bug_before
 
     String value: 'None', 'all', '0.3', '0.4', '0.4.1', '0.5', '0.6', '0.7'

doc/tutorial/debug_faq.txt

@@ -295,6 +295,23 @@ Tips:
   of type *float64*.
 
 
+"Why is my GPU function seems to be slow?"
+------------------------------------
+
+When you compile a theano function, if you do not get the speedup that you expect over the 
+CPU performance of the same code. It is oftentimes due to the fact that some Ops might be running
+on CPU instead GPU. If that is the case, you can use assert_no_cpu_op to check if there 
+is a CPU Op on your computational graph. assert_no_cpu_op can take the following one of the three 
+options:
+
+* ``warn``: Raise a warning
+* ``pdb``: Stop with a pdb in the computational graph during the compilation
+* ``raise``: Raise an error,
+if there is a CPU Op in the computational graph.
+
+It is possible to use this mode by providing the flag in THEANO_FLAGS, such as:
+``THEANO_FLAGS="float32,assert_no_cpu_op='raise'" python test.py``
+
 .. _faq_monitormode:
 
 "How do I Step through a Compiled Function?"

doc/tutorial/using_gpu.txt

@@ -215,6 +215,10 @@ Tips for Improving Performance on GPU
   enable all of them with the `nvcc.flags=--use_fast_math` Theano
   flag or you can enable them individually as in this example:
   `nvcc.flags=-ftz=true --prec-div=false`.
+* To investigate whether if all the Ops in the computational graph are running on GPU. 
+  It is possible to debug or check your code by providing a value to `assert_no_cpu_op`
+  flag, i.e. `warn`, for warning `raise` for raising an error or `pdb` for putting a breakpoint
+  in the computational graph if there is a CPU Op.
 
 .. _gpu_async: