CUTLASS 2.1 (#83)

CUTLASS 2.1 contributes:
- BLAS-style host-side API added to CUTLASS Library
- Planar Complex GEMM kernels targeting Volta and Turing Tensor Cores
- Minor enhancements and bug fixes
This commit is contained in:
Andrew Kerr
2020-04-07 13:51:25 -07:00
committed by GitHub
parent 7c0cd26d13
commit 96dab34ad9
196 changed files with 20653 additions and 1995 deletions

View File

@@ -35,6 +35,7 @@
// CUTLASS Library includes
#include "cutlass/library/library.h"
#include "cutlass/library/util.h"
#include "cutlass/library/manifest.h"
// Profiler includes
@@ -43,6 +44,7 @@
#include "performance_result.h"
#include "performance_report.h"
#include "problem_space.h"
#include "debug.h"
/////////////////////////////////////////////////////////////////////////////////////////////////
@@ -192,8 +194,8 @@ public:
DeviceContext &device_context,
Options const &options,
library::OperationDescription const &desc,
Provider provider,
Provider verification_provider = Provider::kInvalid);
library::Provider provider,
library::Provider verification_provider = library::Provider::kInvalid);
protected: