diff --git a/python/tvm/relay/frontend/paddlepaddle.py b/python/tvm/relay/frontend/paddlepaddle.py
old mode 100644
new mode 100755
index 4927a362522e..0842cd55dae2
--- a/python/tvm/relay/frontend/paddlepaddle.py
+++ b/python/tvm/relay/frontend/paddlepaddle.py
@@ -1141,6 +1141,32 @@ def convert_mv(g, op, block):
     g.add_node(op.output("Out")[0], out)
 
 
+def convert_norm(g, op, block):
+    """Operator converter for norm."""
+
+    x = g.get_node(op.input("X")[0])
+    axis = op.attr("axis")
+    axis_l = [axis]
+    epsilon = op.attr("epsilon")
+    out = _op.nn.l2_normalize(x, epsilon, axis_l)
+    g.add_node(op.output("Out")[0], out)
+
+
+def convert_one_hot_v2(g, op, block):
+    """Operator converter for one_hot_v2."""
+
+    x = g.get_node(op.input("X")[0])
+    depth = op.attr("depth")
+    dtype = op.attr("dtype")
+    dtype = _convert_dtype_value(dtype)
+    ndim = len(infer_shape(x))
+    on_value = _op.const(1)
+    off_value = _op.const(0)
+    axis = ndim
+    out = _op.one_hot(x, on_value, off_value, depth, axis, dtype)
+    g.add_node(op.output("Out")[0], out)
+
+
 def convert_padding(g, op, block):
     """Operator converter for padding."""
 
@@ -2135,7 +2161,9 @@ def convert_unsqueeze(g, op, block):
     "mul": convert_mul,
     "mv": convert_mv,
     "nearest_interp_v2": convert_interpolate,
+    "norm": convert_norm,
     "not_equal": convert_elementwise_op,
+    "one_hot_v2": convert_one_hot_v2,
     "pad1d": convert_padding,
     "pad2d": convert_padding,
     "pad3d": convert_padding,
diff --git a/tests/python/frontend/paddlepaddle/test_forward.py b/tests/python/frontend/paddlepaddle/test_forward.py
old mode 100644
new mode 100755
index de6ea1dcf1da..cd2c0be7ef36
--- a/tests/python/frontend/paddlepaddle/test_forward.py
+++ b/tests/python/frontend/paddlepaddle/test_forward.py
@@ -1723,5 +1723,31 @@ def topk6(inputs):
     verify_model(topk6, input_data=input_data)
 
 
+@tvm.testing.uses_gpu
+def test_forward_one_hot_v2():
+    @paddle.jit.to_static
+    def one_hot_v2_1(inputs):
+        return nn.functional.one_hot(inputs, num_classes=4)
+
+    input_data = paddle.to_tensor([1, 1, 3, 0], dtype=paddle.int32)
+    verify_model(one_hot_v2_1, input_data=input_data)
+
+
+@tvm.testing.uses_gpu
+def test_forward_norm():
+    @paddle.jit.to_static
+    def norm_1(inputs):
+        return paddle.fluid.layers.l2_normalize(inputs, -1, 1e-12)
+
+    def norm_2(inputs):
+        return paddle.fluid.layers.l2_normalize(inputs, 1, 1e-12)
+
+    input_data = paddle.to_tensor(
+        [[[1, 2], [3, 1], [4, 5]], [[3, 1], [3, 5], [2, 4]]], dtype=paddle.float32
+    )
+    verify_model(norm_1, input_data=input_data)
+    verify_model(norm_2, input_data=input_data)
+
+
 if __name__ == "__main__":
     tvm.testing.main()