CUTLASS 3.5.1 (#1623)

* CUTLASS 3.5.1 * updates, optimizations, fixes
2024-07-29 08:46:24 -04:00
parent 56b46e2d13
commit be60a0b272
312 changed files with 19793 additions and 6775 deletions
@@ -29,6 +29,7 @@
 add_custom_target(
  cutlass_test_unit_cute_hopper
  DEPENDS
+  cutlass_test_unit_cute_hopper_cooperative_gemm
  cutlass_test_unit_cute_hopper_stsm
  cutlass_test_unit_cute_hopper_tma_load
  cutlass_test_unit_cute_hopper_tma_store
@@ -46,6 +47,11 @@ add_custom_target(
  test_unit_cute_hopper_bulk_store
 )

+cutlass_test_unit_add_executable(
+  cutlass_test_unit_cute_hopper_cooperative_gemm
+  cooperative_gemm.cu
+)
+
 cutlass_test_unit_add_executable(
  cutlass_test_unit_cute_hopper_stsm
  stsm.cu
@@ -0,0 +1,132 @@
+/***************************************************************************************************
+ * Copyright (c) 2023 - 2024 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.
+ *
+ **************************************************************************************************/
+
+#include "cutlass_unit_test.h"
+
+#include <cute/tensor.hpp>
+
+#include "../cooperative_gemm_common.hpp"
+
+using namespace cute;
+
+#define USE_FP8 1
+
+#if USE_FP8
+TEST(SM90_CuTe_Hopper, CooperativeGemmTilingF8) {
+
+  using TA = uint8_t;
+  using TB = uint8_t;
+  using TC = uint32_t;
+
+  constexpr uint32_t thread_block_size = 128;
+  constexpr int MaxVecBits = 16;
+
+  using tiled_mma_t =
+      TiledMMA<
+        MMA_Atom<SM80_16x8x32_S32S8S8S32_TN>,
+        Layout<Shape<_2, _2, _1>, Stride<_1, _2, _0>>,
+        Tile<_32, _32, _32>
+      >;
+
+  using swizzle = Swizzle<2, 4, 3>;
+
+  // This is for A row major, B col major according to CUTLASS default configs
+  using ALayout = decltype(composition(swizzle{}, Layout<Shape<_64, _64>, Stride<_64, _1>>{}));
+  using BLayout = decltype(composition(swizzle{}, Layout<Shape<_64, _64>, Stride<_1, _64>>{}));
+
+  using CLayout = decltype(make_layout(Shape<_64, _64>{}, LayoutLeft{}));
+
+  test_cooperative_gemm<ALayout,
+                        BLayout,
+                        CLayout,
+                        ALayout,
+                        BLayout,
+                        CLayout,
+                        AutoVectorizingCopyWithAssumedAlignment<MaxVecBits>, // A
+                        AutoVectorizingCopyWithAssumedAlignment<MaxVecBits>, // B
+                        AutoVectorizingCopyWithAssumedAlignment<MaxVecBits>, // C
+                        thread_block_size,
+                        tiled_mma_t,
+                        MaxVecBits,
+                        TA,
+                        TB,
+                        TC>();
+
+}
+
+#else
+
+TEST(SM90_CuTe_Hopper, CooperativeGemmTilingF16) {
+
+  using TA = half_t;
+  using TB = half_t;
+  using TC = half_t;
+
+  constexpr uint32_t thread_block_size = 64;
+  constexpr int MaxVecBits = 16;
+
+  using tiled_mma_t =
+      TiledMMA<
+        MMA_Atom<SM80_16x8x16_F16F16F16F16_TN>,
+        Layout<Shape<_2, _1, _1>, Stride<_1, _0, _0>>,
+        Tile<_32, _32, _32>
+      >;
+
+  using swizzle = Swizzle<3, 3, 3>;
+
+  // This is for A row major, B col major according to CUTLASS default configs
+  using ALayout = decltype(composition(swizzle{},
+      Layout<Shape<_64, _64>, Stride<_64, _1>>{}));
+
+  using BLayout = decltype(composition(swizzle{},
+      Layout<Shape<_64, _64>, Stride<_1, _64>>{}));
+
+  using CLayout = decltype(make_layout(Shape<_64, _64>{}, LayoutLeft{}));
+
+  test_cooperative_gemm<ALayout,
+                        BLayout,
+                        CLayout,
+                        ALayout,
+                        BLayout,
+                        CLayout,
+                        AutoVectorizingCopyWithAssumedAlignment<MaxVecBits>, // A
+                        AutoVectorizingCopyWithAssumedAlignment<MaxVecBits>, // B
+                        AutoVectorizingCopyWithAssumedAlignment<MaxVecBits>, // C
+                        thread_block_size,
+                        tiled_mma_t,
+                        MaxVecBits,
+                        TA,
+                        TB,
+                        TC>();
+
+}
+
+#endif
@@ -122,6 +122,11 @@ tma_test_device_cute(T const* g_in, T* g_out,
  }
 #endif

+  // Test L2 prefetch
+  if (threadIdx.x == 0) {
+    prefetch(tma, tAgA);
+  }
+
  // Loop over the TMA stages, using smem as our buffer
  for (int stage = 0; stage < size<1>(tAgA); ++stage)
  {
@@ -117,6 +117,9 @@ tma_test_device_cute(T const* g_in, T* g_out,
  }
 #endif

+  // Test L2 prefetch
+  cooperative_prefetch<128>(threadIdx.x, gA);
+
  // Loop over the TMA stages, using smem as our buffer
  for (int stage = 0; stage < size<1>(tBgB); ++stage)
  {