CUTLASS 2.6 (#298)

CUTLASS 2.6
This commit is contained in:
Manish Gupta
2021-07-23 00:40:53 -04:00
committed by GitHub
parent 6c29fe20ba
commit e5d51840e8
308 changed files with 32408 additions and 4722 deletions
+3 -3
View File
@@ -52,7 +52,7 @@ DeviceAllocation *DeviceContext::allocate_tensor(
library::NumericTypeID type,
library::LayoutTypeID layout_id,
std::vector<int> const &extent,
std::vector<int> const &stride,
std::vector<int64_t> const &stride,
int batch_count) {
device_memory_.emplace_back(type, layout_id, extent, stride, batch_count);
@@ -69,7 +69,7 @@ DeviceAllocation *DeviceContext::allocate_tensor(
library::NumericTypeID type,
library::LayoutTypeID layout_id,
std::vector<int> const &extent,
std::vector<int> const &stride,
std::vector<int64_t> const &stride,
int batch_count) {
DeviceAllocation *allocation =
@@ -133,7 +133,7 @@ DeviceAllocation *DeviceContext::allocate_sparsemeta_tensor(
library::LayoutTypeID layout_id,
library::NumericTypeID type_a,
std::vector<int> const &extent,
std::vector<int> const &stride,
std::vector<int64_t> const &stride,
int batch_count) {
DeviceAllocation *allocation =