Resolve #530 (#531)

* Add a test for fbp method

* Add support for fbp on gpu

* Add comment to test as per PR review

scico/linop/xray/astra.py

@@ -191,7 +191,6 @@ def fbp(self, sino: jax.Array, filter_type: str = "Ram-Lak") -> jax.Array:
                <https://www.astra-toolbox.com/docs/algs/FBP_CUDA.html>`__.
         """
 
-        # Just use the CPU FBP alg for now; hitting memory issues with GPU one.
         def f(sino):
             sino = _ensure_writeable(sino)
             sino_id = astra.data2d.create("-sino", self.proj_geom, sino)
@@ -200,7 +199,7 @@ def f(sino):
             rec_id = astra.data2d.create("-vol", self.vol_geom)
 
             # start to populate config
-            cfg = astra.astra_dict("FBP")
+            cfg = astra.astra_dict("FBP_CUDA" if self.device == "gpu" else "FBP")
             cfg["ReconstructionDataId"] = rec_id
             cfg["ProjectorId"] = self.proj_id
             cfg["ProjectionDataId"] = sino_id

scico/test/linop/xray/test_astra.py

@@ -122,6 +122,13 @@ def test_adjoint_typical_input(testobj):
     adjoint_test(A, x=x, rtol=get_tol())
 
 
+def test_fbp(testobj):
+    x = testobj.A.fbp(testobj.y)
+    # Test for a bug (related to calling the Astra CPU FBP implementation
+    # when using a FPU device) that resulted in a constant zero output.
+    assert np.sum(np.abs(x)) > 0.0
+
+
 def test_jit_in_DiagonalStack():
     """See https://github.com/lanl/scico/issues/331"""
     N = 10