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

Add definitions for tag structs. (#752)

Browse Source 
This commit changes the declarations of MMA operator class (SIMT, Tensor Core, WMMA Tensor Core) and operator type (multiply-add and so on) to definitions. This is done so that these tag structs are no longer incomplete types, which allows the `typeid` operator to be used on these tag structs. This is necessary for these tag structs to be used as type parameters in [GoogleTest typed tests](https://google.github.io/googletest/advanced.html#typed-tests).
This commit is contained in:
Gregory Meyer (gregjm) 2023-01-06 06:46:52 -08:00 committed by GitHub
parent c54ede3a9e
commit 7bdba07310
No known key found for this signature in database
GPG Key ID: 4AEE18F83AFDEB23
1 changed files with 12 additions and 12 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

24
include/cutlass/arch/mma.h
View File

@ -49,61 +49,61 @@ namespace arch {
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag indicating the operation implied by MMA.
struct OpMultiplyAdd;
struct OpMultiplyAdd {};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag indicating the result is saturated to MAX_FLOAT|MIN_FLOAT or MAX_INT|MIN_INT
struct OpMultiplyAddSaturate;
struct OpMultiplyAddSaturate {};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag indicating the input is converted to a narrower type (BF16)
struct OpMultiplyAddFastBF16;
struct OpMultiplyAddFastBF16 {};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag indicating the input is converted to a narrower type (F16)
struct OpMultiplyAddFastF16;
struct OpMultiplyAddFastF16 {};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag indicating the input is converted to 2 (big and small) TF32 components
// Perform 3xTF32 or 4xTF32 for every F32 output element
struct OpMultiplyAddFastF32;
struct OpMultiplyAddFastF32 {}
/// Tag indicating the input is converted to 2 (big and small) TF32 components
// Perform 3xTF32 or 4xTF32 for every complex<F32> output element
struct OpMultiplyAddComplexFastF32;
struct OpMultiplyAddComplexFastF32 {};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag indicating the complex multiply-add operation
struct OpMultiplyAddComplex;
struct OpMultiplyAddComplex {};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag indicating the gaussian complex multiply-add operation
struct OpMultiplyAddGaussianComplex;
struct OpMultiplyAddGaussianComplex {};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag indicating the inner product is defined by (XOR, POPC)
struct OpXorPopc;
struct OpXorPopc {};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag classifying math operators as thread-level operations.
struct OpClassSimt;
struct OpClassSimt {};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag classifing operators as Tensor Core operations.
struct OpClassTensorOp;
struct OpClassTensorOp {};
/////////////////////////////////////////////////////////////////////////////////////////////////
/// Tag classifing operators as WMMA Tensor Core operations
struct OpClassWmmaTensorOp;
struct OpClassWmmaTensorOp {};
/////////////////////////////////////////////////////////////////////////////////////////////////