leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 26 Packages Projects Releases Wiki Activity

CUTLASS 2.6.1 - functional and performance enhancements to strided DGRAD, fixes, and tuning

Browse Source 
* cutlass 2.6 update

* remove debug prints

* cutlass 2.6.1 (minor update)

* Updated CHANGELOG.

* Minor edit to readme to indicate patch version.

* Minor edit to readme.

Co-authored-by:  Haicheng Wu <haichengw@nvidia.com>, Andrew Kerr <akerr@nvidia.com>
This commit is contained in:
Manish Gupta
2021-09-03 10:26:15 -07:00
committed by GitHub
co-authored by Haicheng Wu <haichengw@nvidia.com>, Andrew Kerr <akerr@nvidia.com>
parent a01feb93d9
commit 6c2f8f2fb8
55 changed files with 317 additions and 315 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/library/src/reference/gemm_reference_operation.h
+7 -13
View File
@@ -186,12 +186,6 @@ public:
GemmUniversalConfiguration const &config = *static_cast<GemmUniversalConfiguration const *>(host_workspace);
GemmUniversalArguments const &args = *static_cast<GemmUniversalArguments const *>(arguments);
ElementCompute alpha;
ElementCompute beta;
alpha = *static_cast<ElementCompute const *>(args.alpha);
beta = *static_cast<ElementCompute const *>(args.beta);
TensorRefA ref_A{static_cast<ElementA *>(const_cast<void *>(args.A)), LayoutA(int(config.lda))};
TensorRefB ref_B{static_cast<ElementB *>(const_cast<void *>(args.B)), LayoutB(int(config.ldb))};
TensorRefC ref_C{static_cast<ElementC *>(const_cast<void *>(args.C)), LayoutC(int(config.ldc))};
@@ -212,16 +206,16 @@ public:
InnerProductOp
>(
config.problem_size,
alpha,
*static_cast<ElementCompute const *>(args.alpha),
ref_A,
kTransformA,
ref_B,
kTransformB,
beta,
*static_cast<ElementCompute const *>(args.beta),
ref_C,
ref_D,
ElementAccumulator(),
config.batch_count,
((config.mode == library::GemmUniversalMode::kBatched) ? config.batch_count : 1),
args.batch_stride_A,
args.batch_stride_B,
args.batch_stride_C,
@@ -245,16 +239,16 @@ public:
InnerProductOp
>(
config.problem_size,
alpha,
*static_cast<ElementCompute const *>(args.alpha),
ref_A,
kTransformA,
ref_B,
kTransformB,
beta,
*static_cast<ElementCompute const *>(args.beta),
ref_C,
ref_D,
ElementAccumulator(),
config.batch_count,
((config.mode == library::GemmUniversalMode::kBatched) ? config.batch_count : 1),
args.batch_stride_A,
args.batch_stride_B,
args.batch_stride_C,
@@ -263,7 +257,7 @@ public:
return Status::kSuccess;
}
return Status::kErrorNotSupported;
}
};