CUTLASS 3.0.0 (#786)

* CUTLASS 3.0.0

tools/library/scripts/pycutlass/test/frontend/run_test.sh

@@ -1 +1,33 @@
+#################################################################################################
+#
+# Copyright (c) 2017 - 2023 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
+# SPDX-License-Identifier: BSD-3-Clause
+#
+# Redistribution and use in source and binary forms, with or without
+# modification, are permitted provided that the following conditions are met:
+#
+# 1. Redistributions of source code must retain the above copyright notice, this
+# list of conditions and the following disclaimer.
+#
+# 2. Redistributions in binary form must reproduce the above copyright notice,
+# this list of conditions and the following disclaimer in the documentation
+# and/or other materials provided with the distribution.
+#
+# 3. Neither the name of the copyright holder nor the names of its
+# contributors may be used to endorse or promote products derived from
+# this software without specific prior written permission.
+#
+# THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
+# AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
+# IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
+# DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
+# FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
+# DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
+# SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
+# CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
+# OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
+# OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
+#
+#################################################################################################
+
 CUPY_CACHE_DIR=./ python test_frontend.py

tools/library/scripts/pycutlass/test/frontend/test_frontend.py

@@ -29,13 +29,15 @@
 # OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
 #
 #################################################################################################
-## Test case for Pytorch
+
+"""
+Test cases for frontends
+"""
+
 import pycutlass
 import unittest
 from pycutlass import *
 from pycutlass.utils.device import device_cc
-import torch
-import cupy as cp
 
 
 class Test_Frontend(unittest.TestCase):
@@ -49,9 +51,7 @@ class Test_Frontend(unittest.TestCase):
             cutlass.OpClass.Simt, MathOperation.multiply_add
         )
 
-        # Stages > 2 is supported only for compute capability 80 and beyond
-        stages = 4 if cc >= 80 else 2
-
+        stages = 2
         tile_description = TileDescription(
             [128, 128, 8], stages, [2, 4, 1],
             math_inst
@@ -84,6 +84,11 @@ class Test_Frontend(unittest.TestCase):
 
 
     def test_torch_frontend(self):
+        try:
+            import torch
+        except:
+            self.assertTrue(False, "Unable to import torch")
+
         problem_size = cutlass.gemm.GemmCoord(512, 256, 128)
 
         tensor_A = torch.ceil(torch.empty(size=(problem_size.m(), problem_size.k()), dtype=torch.float32, device="cuda").uniform_(-8.5, 7.5))
@@ -111,6 +116,11 @@ class Test_Frontend(unittest.TestCase):
         self.assertTrue(torch.equal(tensor_D, tensor_D_ref))
     
     def test_cupy_frontend(self):
+        try:
+            import cupy as cp
+        except:
+            self.assertTrue(False, "Unable to import cupy")
+
         cp.cuda.set_allocator(rmm.rmm_cupy_allocator)
 
         problem_size = cutlass.gemm.GemmCoord(512, 256, 128)
@@ -139,7 +149,6 @@ class Test_Frontend(unittest.TestCase):
         self.assertTrue(cp.array_equal(tensor_D, tensor_D_ref))
 
 
-
 if __name__ == '__main__':
     pycutlass.get_memory_pool(2**32, 2**32)
     unittest.main()