leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 1 Packages Projects Releases Wiki Activity

CUTLASS 3.2.1 (#1113)

Browse Source 
* Updates for 3.2.1 release.

* Minor fix in gemm op profiler for raster order.

* Add scheduler mapping for raster order in the kernels.
This commit is contained in:
ANIKET SHIVAM
2023-09-26 14:24:26 -07:00
committed by GitHub
parent e0aaa3c3b3
commit 90d3b0fb18
428 changed files with 22253 additions and 21762 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
test/unit/gemm/threadblock/mma_pipelined_sm75.cu
View File

@@ -569,7 +569,6 @@ TEST(SM75_gemm_threadblock_crosswise,
}
////////////////////////////////////////////////////////////////////////////////
TEST(SM75_gemm_threadblock_interleaved, tensor_op_32x32x64_16x16x64_8x8x16) {
using ElementA = uint8_t;
using LayoutA = cutlass::layout::ColumnMajorInterleaved<32>;
@@ -1793,6 +1792,7 @@ TEST(SM75_gemm_threadblock_interleaved,
}
////////////////////////////////////////////////////////////////////////////////
TEST(SM75_gemm_threadblock_crosswise, tensor_op_64x64x512_64x64x512_8x8x128) {
using ElementA = cutlass::uint1b_t;
using LayoutA = cutlass::layout::RowMajor;

2
test/unit/gemm/threadblock/mma_pipelined_wmma_sm75.cu
View File

@@ -193,7 +193,6 @@ TEST(SM75_gemm_threadblock_wmma_tensor_op_col_row_row_s8, 64x64x64_64x64x64_16x1
///////////////////////////////////////////////////////////////////////
#if defined(CUTLASS_SUBBYTE_INTEGER_MATRIX_MULTIPLY_ENABLED)
TEST(SM75_gemm_threadblock_wmma_tensor_op_row_col_row_s4, 64x64x128_64x64x128_8x8x32) {
using ElementA = cutlass::int4b_t;
using LayoutA = cutlass::layout::RowMajor;
@@ -262,6 +261,7 @@ TEST(SM75_gemm_threadblock_wmma_tensor_op_row_col_col_s4, 64x64x64_64x64x64_8x8x
problem_size.k(), alpha, beta)
.run(grid, block);
}
TEST(SM75_gemm_threadblock_wmma_tensor_op_row_col_row_b1, 64x64x512_64x64x512_8x8x128) {
using ElementA = cutlass::uint1b_t;
using LayoutA = cutlass::layout::RowMajor;

2
test/unit/gemm/threadblock/mma_singlestage_wmma_sm75.cu
View File

@@ -193,7 +193,6 @@ TEST(SM75_gemm_threadblock_singlestage_wmma_tensor_op_col_row_row_s8, 64x64x64_6
///////////////////////////////////////////////////////////////////////
#if defined(CUTLASS_SUBBYTE_INTEGER_MATRIX_MULTIPLY_ENABLED)
TEST(SM75_gemm_threadblock_singlestage_wmma_tensor_op_row_col_row_s4, 64x64x128_64x64x128_8x8x32) {
using ElementA = cutlass::int4b_t;
using LayoutA = cutlass::layout::RowMajor;
@@ -262,6 +261,7 @@ TEST(SM75_gemm_threadblock_singlestage_wmma_tensor_op_row_col_col_s4, 64x64x64_6
problem_size.k(), alpha, beta)
.run(grid, block);
}
TEST(SM75_gemm_threadblock_singlestage_wmma_tensor_op_row_col_row_b1, 64x64x512_64x64x512_8x8x128) {
using ElementA = cutlass::uint1b_t;
using LayoutA = cutlass::layout::RowMajor;