Repository navigation
MetalPerformancePrimitives macOS xcode27.3 b1
Alex Soto edited this page Nov 4, 2026
·
1 revision
#MetalPerformancePrimitives.framework
diff -ruN /Applications/Xcode_27.2.0.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/MPPTensorOpsMatMul2d.h /Applications/Xcode_27.3.0-beta1.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/MPPTensorOpsMatMul2d.h
--- /Applications/Xcode_27.2.0.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/MPPTensorOpsMatMul2d.h 2026-10-09 07:35:08
+++ /Applications/Xcode_27.3.0-beta1.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/MPPTensorOpsMatMul2d.h 2026-11-04 09:32:08
@@ -8,6 +8,8 @@
// C = A*B + C;
// A and B can be tensor_handle, tensor_offset, and tensor_inline.
// C can be tensor_handle, tensor_offset, tensor_inline or cooperative_tensor.
+// With a deployment target of 27.3 or later, A and B can also be tensors in the constant address space, and
+// cooperative tensors can be loaded from them. C cannot be in the constant address space.
// Data type combinations supported by this operation are as follows:
//
// Left Right Destination
@@ -278,15 +280,15 @@
// // It is imperative for performance to include "unroll pragma" so compiler
// // fully unrolls the loop.
//
-// #pragma unroll full
-// for (uint16_t i = 0, i < cT.get_capacity(); ++i) {
-// if(cT.get_mask(i))
+// #pragma clang loop unroll(full)
+// for (uint16_t i = 0; i < cT.get_capacity(); ++i) {
+// if(cT.is_valid_element(i))
// cT[i] = 0;
// }
//
// // execute the operation. All threads compute the matmul cooperatively
// // and results are written to cooperative_tensor.
-// op.run(mA, mB, cT);
+// matmulOp.run(mA, mB, cT);
//
// // create cooperative bias tensor with same layout as destination
// // cooperative_tensor of matmul
@@ -296,17 +298,17 @@
// // layout and distribution of element among threads of scope on which matmul was created.
// biasT.load(bias);
//
-// #pragma unroll full
-// for (uint16_t i = 0, i < cT.get_capacity(); ++i) {
+// #pragma clang loop unroll(full)
+// for (uint16_t i = 0; i < cT.get_capacity(); ++i) {
//
-// if(cT.get_mask(i)) {
+// if(cT.is_valid_element(i)) {
// //add bias
// cT[i] += biasT[i];
//
// // get the 2-dimensional local coordinate of this thread's i-th
// // element in destination local coordinate system (in this example
// // 32 x 64 tile).
-// auto ids = cT.get_multidimensional_index(i);
+// auto idx = cT.get_multidimensional_index(i);
// cT[i] = foo(cT[i], idx); // do some operation based on coordinate values
// }
// }
@@ -315,6 +317,117 @@
// cT.store(mC);
// }
//
+// Linear cooperative tensors
+// ==========================
+// By default (matmul2d_cooperative_tensor_layout::automatic) the distribution of a
+// cooperative tensor's elements among threads is implementation defined. With
+// matmul2d_cooperative_tensor_layout::linear, passed as the Layout template argument of
+// get_left_input_cooperative_tensor, get_destination_cooperative_tensor and the row /
+// column reduction destination getters, each thread owns whole rows:
+//
+// - The left input (M x K) or destination (M x N) is distributed in blocks of 32 rows.
+// Each thread of the SIMD group owns one row of every block, so a thread owns M / 32
+// rows and get_capacity() == (M / 32) * C, where C = K for a left input and C = N for
+// a destination. Every element is valid.
+// - The C columns of an owned row are contiguous in the thread's storage: &t[r * C] points
+// to this thread's r-th row, r = 0 .. get_capacity() / C - 1, and (&t[r * C])[c] is
+// column c for c < C. Different rows are not contiguous with each other: form each row
+// pointer separately. (Note that t[0, i] is not a row access: cooperative_tensor only
+// has a single-index subscript and t[0, i] is the comma expression t[i].)
+// - Do not assume that the thread with lane id L owns row L. The implementation may map
+// each lane to an arbitrary row of the cooperative tensor for performance. Get the row of
+// element i from t.get_multidimensional_index(i)[1] (index [0] is the column); for this
+// thread's r-th row that is t.get_multidimensional_index(r * C)[1].
+// - Reductions of a linear M x N destination t use linear reduction destinations:
+// * Row reduction rr (get_row_reduction_destination_cooperative_tensor with
+// Layout = linear): each thread holds M / 32 elements, one per row it owns in t.
+// After reduce_rows(t, rr), rr[r] is the reduction of this thread's r-th row of t,
+// i.e. of the N values at &t[r * N]. rr.get_multidimensional_index(r)[0] is that row
+// (equal to t.get_multidimensional_index(r * N)[1]). No data moves between threads.
+// * Column reduction cr (get_column_reduction_destination_cooperative_tensor with
+// Layout = linear): each thread holds all N elements. After reduce_columns(t, cr),
+// cr[c] is the reduction of column c over all M rows, and every thread holds the
+// same values.
+// - Linear cooperative tensors of the same M in one SIMD group (left input, destination,
+// row reduction destination) own the same rows in the same order. Element-wise
+// operations between them by index, map_iterator, and handing a linear destination to
+// the next matmul as its linear left input
+// (get_left_input_cooperative_tensor<...>(destination)) are always valid. The handoff's
+// requirements (matching M, destination N == left input K, element type) are checked at
+// compile time, so a linear -> linear handoff needs no is_compatible_as_left_input check
+// (it returns true for a linear source).
+//
+// Requirements: deployment target 27.3 or later, execution_simdgroup, M a multiple of 32,
+// for a left input a static K that is a multiple of 16 and transpose_left == false, for a
+// destination N a multiple of 16, and no int2/int4/fp4/fp8 format element types. The right
+// input cannot be linear. run() accepts any combination of linear, automatic and
+// non-cooperative operands. A linear tensor cannot be converted to an automatic one or
+// vice versa; use the same layout on both sides of a destination -> left input handoff.
+//
+// run() converts between the linear layout and the hardware layout internally. That is
+// cheap when a linear tensor is consumed or produced once (for example the first input and
+// the last output of a chain of matmuls), but a linear destination that is accumulated by
+// many run() calls (multiply_accumulate in a loop) pays the conversion on every call. Prefer
+// automatic for intermediate results and use linear where whole rows are needed.
+//
+// Example: a 2-layer MLP as used in neural rendering. Each SIMD group evaluates 32 pixels: it
+// fills a linear left input row by row from a feature buffer (a texture read or any other
+// per-row gather works the same way; no load() is needed), runs the hidden layer with an
+// automatic destination, and gets a linear output so each thread can post-process its own
+// pixels before the store.
+//
+// kernel void mlp(device const half *features, // pixels x 64
+// tensor<device half, extents<int32_t, 128, 64>> W1, // 64 x 128
+// tensor<device half, extents<int32_t, 64, 128>> W2, // 128 x 64
+// tensor<device half, extents<int32_t, 64, dynamic_extent>> Y, // pixels x 64
+// uint tgid [[threadgroup_position_in_grid]])
+// {
+// using layout = matmul2d_cooperative_tensor_layout;
+// constexpr int K = 64, H = 128, N = 64;
+// constexpr auto desc1 = matmul2d_descriptor(32, H, K);
+// constexpr auto desc2 = matmul2d_descriptor(32, N, H);
+// matmul2d<desc1, execution_simdgroup> op1;
+// matmul2d<desc2, execution_simdgroup> op2;
+//
+// // Linear left input: write this thread's rows directly
+// auto x = op1.get_left_input_cooperative_tensor<half, half, half, int, layout::linear>();
+// for (uint16_t r = 0; r < x.get_capacity() / K; ++r) {
+// int pixel = tgid * 32 + x.get_multidimensional_index(r * K)[1]; // row id of this row
+// thread half *row = &x[r * K]; // K contiguous columns
+// for (int c = 0; c < K; ++c)
+// row[c] = features[pixel * K + c];
+// }
+//
+// // Hidden layer: automatic destination, activation applied element-wise
+// auto h = op1.get_destination_cooperative_tensor<decltype(W1), decltype(W1), half>();
+// op1.run(x, W1, h);
+// #pragma clang loop unroll(full)
+// for (uint16_t i = 0; i < h.get_capacity(); ++i)
+// if (h.is_valid_element(i))
+// h[i] = max(h[i], half(0));
+//
+// // Automatic destination -> automatic left input. Unlike a linear -> linear handoff, this
+// // depends on the hardware layout and is checked at run time.
+// if (!op2.is_compatible_as_left_input<half, half, half>(h))
+// return; // not expected for these shapes; fall back to memory otherwise
+// auto h2 = op2.get_left_input_cooperative_tensor<half, half, half>(h);
+//
+// // Linear output: RMS-normalize each of this thread's rows, then store the 64 x 32 tile
+// auto y = op2.get_destination_cooperative_tensor<decltype(W2), decltype(W2), half, int, layout::linear>();
+// op2.run(h2, W2, y);
+// for (uint16_t r = 0; r < y.get_capacity() / N; ++r) {
+// thread half *row = &y[r * N];
+// float sumSq = 0;
+// for (int c = 0; c < N; ++c)
+// sumSq += float(row[c]) * float(row[c]);
+// half scale = half(rsqrt(sumSq / N + 1e-6f));
+// for (int c = 0; c < N; ++c)
+// row[c] *= scale;
+// }
+// auto mY = Y.slice<N, 32>(0, tgid * 32); // static 64 x 32 slice
+// y.store(mY);
+// }
+//
// Note on scope of operation
// ==========================
// A tensor operation may be executed on a single thread entirely or
@@ -398,13 +511,26 @@
}
};
+// Distribution of a cooperative tensor's elements among the threads of the
+// execution scope.
+// automatic: implementation defined. Default.
+// linear: each thread owns whole rows; the columns of an owned row are
+// contiguous in the thread's storage. Requires execution_simdgroup.
+// For both layouts, the element -> coordinate mapping (including which rows a
+// thread owns under linear) is obtained with get_multidimensional_index().
+enum class matmul2d_cooperative_tensor_layout
+{
+ automatic = 0,
+ linear = 1,
+};
+
template <typename ElementType>
struct reduction_operation_identity
{
- static const constant ElementType sum_identity = (ElementType)0;
- static const constant ElementType max_identity =
+ static constexpr constant ElementType sum_identity = (ElementType)0;
+ static constexpr constant ElementType max_identity =
metal::numeric_limits<ElementType>::lowest();
- static const constant ElementType min_identity =
+ static constexpr constant ElementType min_identity =
metal::numeric_limits<ElementType>::max();
};
@@ -439,42 +565,45 @@
RunArgs...>(left, right, destination);
}
- template <typename LeftElementType, typename RightElementType, typename ElementType, typename CoordType = int, typename... CoopArgs>
+ template <typename LeftElementType, typename RightElementType, typename ElementType, typename CoordType = int,
+ matmul2d_cooperative_tensor_layout Layout = matmul2d_cooperative_tensor_layout::automatic, typename... CoopArgs>
using cooperative_tensor_left_input_t =
__mutmul2d_detail::__cooperative_tensor_left_input_t<
- Descriptor, Scope, LeftElementType, RightElementType, ElementType, CoordType, CoopArgs...>;
+ Descriptor, Scope, LeftElementType, RightElementType, ElementType, CoordType, Layout, CoopArgs...>;
template <typename LeftElementType, typename RightElementType,
typename ElementType, typename CoordType = int,
+ matmul2d_cooperative_tensor_layout Layout = matmul2d_cooperative_tensor_layout::automatic,
+ typename... CoopArgs,
typename U = __tensor_ops_detail::__enable_if_t<
__tensor_ops_detail::__is_unqualified_v<LeftElementType> &&
__tensor_ops_detail::__is_unqualified_v<RightElementType> &&
__tensor_ops_detail::__is_unqualified_v<ElementType> &&
- __tensor_ops_detail::__is_integral_v<CoordType>>,
- typename... CoopArgs>
- INLINE cooperative_tensor_left_input_t<LeftElementType, RightElementType, ElementType, CoordType, CoopArgs...>
+ __tensor_ops_detail::__is_integral_v<CoordType>>>
+ INLINE cooperative_tensor_left_input_t<LeftElementType, RightElementType, ElementType, CoordType, Layout, CoopArgs...>
get_left_input_cooperative_tensor() thread const
{
return __mutmul2d_detail::__get_left_input_cooperative_tensor<
- Descriptor, Scope, LeftElementType, RightElementType, ElementType, CoordType, CoopArgs...>();
+ Descriptor, Scope, LeftElementType, RightElementType, ElementType, CoordType, Layout, CoopArgs...>();
}
template <typename LeftElementType, typename RightElementType,
typename ElementType, typename CoordType = int,
typename SrcElemType, typename SrcExtents, typename SrcLayout,
+ matmul2d_cooperative_tensor_layout Layout = __mutmul2d_detail::__matmul2d_layout_of<SrcLayout>::value,
+ typename... CoopArgs,
typename U = __tensor_ops_detail::__enable_if_t<
__tensor_ops_detail::__is_unqualified_v<LeftElementType> &&
__tensor_ops_detail::__is_unqualified_v<RightElementType> &&
__tensor_ops_detail::__is_unqualified_v<ElementType> &&
__tensor_ops_detail::__is_unqualified_v<SrcElemType> &&
- __tensor_ops_detail::__is_integral_v<CoordType>>,
- typename... CoopArgs>
- INLINE cooperative_tensor_left_input_t<LeftElementType, RightElementType, ElementType, CoordType, CoopArgs...>
+ __tensor_ops_detail::__is_integral_v<CoordType>>>
+ INLINE cooperative_tensor_left_input_t<LeftElementType, RightElementType, ElementType, CoordType, Layout, CoopArgs...>
get_left_input_cooperative_tensor(const thread metal::cooperative_tensor<SrcElemType, SrcExtents, SrcLayout> & src) thread const
{
return __mutmul2d_detail::__get_left_input_cooperative_tensor<
SrcElemType, SrcExtents, SrcLayout, Descriptor, Scope, LeftElementType, RightElementType,
- ElementType, CoordType, CoopArgs...>(src);
+ ElementType, CoordType, Layout, CoopArgs...>(src);
}
template <typename LeftElementType, typename RightElementType, typename ElementType,
@@ -492,42 +621,45 @@
SrcElemType, SrcExtents, SrcLayout>(src);
}
- template <typename LeftElementType, typename RightElementType, typename ElementType, typename CoordType = int, typename... CoopArgs>
+ template <typename LeftElementType, typename RightElementType, typename ElementType, typename CoordType = int,
+ matmul2d_cooperative_tensor_layout Layout = matmul2d_cooperative_tensor_layout::automatic, typename... CoopArgs>
using cooperative_tensor_right_input_t =
__mutmul2d_detail::__cooperative_tensor_right_input_t<
- Descriptor, Scope, LeftElementType, RightElementType, ElementType, CoordType, CoopArgs...>;
+ Descriptor, Scope, LeftElementType, RightElementType, ElementType, CoordType, Layout, CoopArgs...>;
template <typename LeftElementType, typename RightElementType,
typename ElementType, typename CoordType = int,
+ matmul2d_cooperative_tensor_layout Layout = matmul2d_cooperative_tensor_layout::automatic,
+ typename... CoopArgs,
typename U = __tensor_ops_detail::__enable_if_t<
__tensor_ops_detail::__is_unqualified_v<LeftElementType> &&
__tensor_ops_detail::__is_unqualified_v<RightElementType> &&
__tensor_ops_detail::__is_unqualified_v<ElementType> &&
- __tensor_ops_detail::__is_integral_v<CoordType>>,
- typename... CoopArgs>
- INLINE cooperative_tensor_right_input_t<LeftElementType, RightElementType, ElementType, CoordType, CoopArgs...>
+ __tensor_ops_detail::__is_integral_v<CoordType>>>
+ INLINE cooperative_tensor_right_input_t<LeftElementType, RightElementType, ElementType, CoordType, Layout, CoopArgs...>
get_right_input_cooperative_tensor() thread const
{
return __mutmul2d_detail::__get_right_input_cooperative_tensor<
- Descriptor, Scope, LeftElementType, RightElementType, ElementType, CoordType, CoopArgs...>();
+ Descriptor, Scope, LeftElementType, RightElementType, ElementType, CoordType, Layout, CoopArgs...>();
}
template <typename LeftElementType, typename RightElementType,
typename ElementType, typename CoordType = int,
typename SrcElemType, typename SrcExtents, typename SrcLayout,
+ matmul2d_cooperative_tensor_layout Layout = __mutmul2d_detail::__matmul2d_layout_of<SrcLayout>::value,
+ typename... CoopArgs,
typename U = __tensor_ops_detail::__enable_if_t<
__tensor_ops_detail::__is_unqualified_v<LeftElementType> &&
__tensor_ops_detail::__is_unqualified_v<RightElementType> &&
__tensor_ops_detail::__is_unqualified_v<ElementType> &&
__tensor_ops_detail::__is_unqualified_v<SrcElemType> &&
- __tensor_ops_detail::__is_integral_v<CoordType>>,
- typename... CoopArgs>
- INLINE cooperative_tensor_right_input_t<LeftElementType, RightElementType, ElementType, CoordType, CoopArgs...>
+ __tensor_ops_detail::__is_integral_v<CoordType>>>
+ INLINE cooperative_tensor_right_input_t<LeftElementType, RightElementType, ElementType, CoordType, Layout, CoopArgs...>
get_right_input_cooperative_tensor(const thread metal::cooperative_tensor<SrcElemType, SrcExtents, SrcLayout> & src) thread const
{
return __mutmul2d_detail::__get_right_input_cooperative_tensor<
SrcElemType, SrcExtents, SrcLayout, Descriptor, Scope, LeftElementType, RightElementType,
- ElementType, CoordType, CoopArgs...>(src);
+ ElementType, CoordType, Layout, CoopArgs...>(src);
}
template <typename LeftElementType, typename RightElementType, typename ElementType,
@@ -545,64 +677,69 @@
SrcElemType, SrcExtents, SrcLayout>(src);
}
- template <typename LeftOperand, typename RightOperand, typename ElementType, typename CoordType = int, typename... CoopArgs>
+ template <typename LeftOperand, typename RightOperand, typename ElementType, typename CoordType = int,
+ matmul2d_cooperative_tensor_layout Layout = matmul2d_cooperative_tensor_layout::automatic, typename... CoopArgs>
using cooperative_tensor_destination_t =
__mutmul2d_detail::__cooperative_tensor_destination_t<
- Descriptor, Scope, LeftOperand, RightOperand, ElementType, CoordType, CoopArgs...>;
+ Descriptor, Scope, LeftOperand, RightOperand, ElementType, CoordType, Layout, CoopArgs...>;
template <typename LeftOperandType, typename RightOperandType,
typename ElementType, typename CoordType = int,
+ matmul2d_cooperative_tensor_layout Layout = matmul2d_cooperative_tensor_layout::automatic,
+ typename... CoopArgs,
typename U = __tensor_ops_detail::__enable_if_t<
(__tensor_ops_detail::__is_tensor_type_v<LeftOperandType> || __tensor_ops_detail::__is_cooperative_tensor_type_v<LeftOperandType>) &&
(__tensor_ops_detail::__is_tensor_type_v<RightOperandType> || __tensor_ops_detail::__is_cooperative_tensor_type_v<RightOperandType>) &&
__tensor_ops_detail::__is_unqualified_v<ElementType> &&
- __tensor_ops_detail::__is_integral_v<CoordType>>,
- typename... CoopArgs>
- INLINE cooperative_tensor_destination_t<LeftOperandType, RightOperandType, ElementType, CoordType, CoopArgs...>
+ __tensor_ops_detail::__is_integral_v<CoordType>>>
+ INLINE cooperative_tensor_destination_t<LeftOperandType, RightOperandType, ElementType, CoordType, Layout, CoopArgs...>
get_destination_cooperative_tensor() thread const
{
return __mutmul2d_detail::__get_destination_cooperative_tensor<
Descriptor, Scope, LeftOperandType, RightOperandType, ElementType,
- CoordType, CoopArgs...>();
+ CoordType, Layout, CoopArgs...>();
}
template <typename LeftOperandType, typename RightOperandType, typename ElementType,
- typename CoordType = int, typename... CoopArgs>
+ typename CoordType = int, matmul2d_cooperative_tensor_layout Layout = matmul2d_cooperative_tensor_layout::automatic,
+ typename... CoopArgs>
using cooperative_tensor_row_reduction_destination_t =
__mutmul2d_detail::__cooperative_tensor_row_reduction_destination_t<
- Descriptor, Scope, LeftOperandType, RightOperandType, ElementType, CoordType, CoopArgs...>;
+ Descriptor, Scope, LeftOperandType, RightOperandType, ElementType, CoordType, Layout, CoopArgs...>;
template <typename LeftOperandType, typename RightOperandType, typename ElementType,
- typename CoordType = int,
+ typename CoordType = int, matmul2d_cooperative_tensor_layout Layout = matmul2d_cooperative_tensor_layout::automatic,
+ typename... CoopArgs,
typename U = __tensor_ops_detail::__enable_if_t<
- __tensor_ops_detail::__is_integral_v<CoordType>>,
- typename... CoopArgs>
+ __tensor_ops_detail::__is_integral_v<CoordType>>>
INLINE cooperative_tensor_row_reduction_destination_t<LeftOperandType, RightOperandType,
- ElementType, CoordType, CoopArgs...>
+ ElementType, CoordType, Layout, CoopArgs...>
get_row_reduction_destination_cooperative_tensor() thread const
{
return __mutmul2d_detail::__get_row_reduction_destination_cooperative_tensor<
- Descriptor, Scope, LeftOperandType, RightOperandType, ElementType, CoordType, CoopArgs...>();
+ Descriptor, Scope, LeftOperandType, RightOperandType, ElementType, CoordType, Layout, CoopArgs...>();
}
template <typename LeftOperandType, typename RightOperandType, typename ElementType,
- typename CoordType = int, typename... CoopArgs>
+ typename CoordType = int, matmul2d_cooperative_tensor_layout Layout = matmul2d_cooperative_tensor_layout::automatic,
+ typename... CoopArgs>
using cooperative_tensor_column_reduction_destination_t =
__mutmul2d_detail::__cooperative_tensor_column_reduction_destination_t<
- Descriptor, Scope, LeftOperandType, RightOperandType, ElementType, CoordType, CoopArgs...>;
+ Descriptor, Scope, LeftOperandType, RightOperandType, ElementType, CoordType, Layout, CoopArgs...>;
template <typename LeftOperandType, typename RightOperandType, typename ElementType,
- typename CoordType = int, typename U = __tensor_ops_detail::__enable_if_t<
- __tensor_ops_detail::__is_integral_v<CoordType>>,
- typename... CoopArgs>
+ typename CoordType = int, matmul2d_cooperative_tensor_layout Layout = matmul2d_cooperative_tensor_layout::automatic,
+ typename... CoopArgs,
+ typename U = __tensor_ops_detail::__enable_if_t<
+ __tensor_ops_detail::__is_integral_v<CoordType>>>
INLINE cooperative_tensor_column_reduction_destination_t<LeftOperandType, RightOperandType,
- ElementType, CoordType, CoopArgs...>
+ ElementType, CoordType, Layout, CoopArgs...>
get_column_reduction_destination_cooperative_tensor() thread const
{
return __mutmul2d_detail::__get_column_reduction_destination_cooperative_tensor<
- Descriptor, Scope, LeftOperandType, RightOperandType, ElementType, CoordType, CoopArgs...>();
+ Descriptor, Scope, LeftOperandType, RightOperandType, ElementType, CoordType, Layout, CoopArgs...>();
}
};
@@ -636,7 +773,7 @@
//
// if (is_iterator_compatible(sourceCT, destCT)) {
// for (auto it = sourceCT.begin(); it != sourceCT.end(); it++) {
-// auto dst_it = destCT.map_iterator(sourceCT)
+// auto dst_it = destCT.map_iterator(it);
//
// *it += *dst_it;
// }
diff -ruN /Applications/Xcode_27.2.0.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/__impl/MPPTensorOpsAvailability.h /Applications/Xcode_27.3.0-beta1.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/__impl/MPPTensorOpsAvailability.h
--- /Applications/Xcode_27.2.0.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/__impl/MPPTensorOpsAvailability.h 2026-10-09 07:35:08
+++ /Applications/Xcode_27.3.0-beta1.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/__impl/MPPTensorOpsAvailability.h 2026-11-04 09:32:08
@@ -9,5 +9,6 @@
#define __TENSOR_OPS_SUPPORT_DEPLOYMENT_TARGET_26_2 ((__ENVIRONMENT_OS_VERSION_MIN_REQUIRED__) >= 260200)
#define __TENSOR_OPS_SUPPORT_DEPLOYMENT_TARGET_27_0 ((__ENVIRONMENT_OS_VERSION_MIN_REQUIRED__) >= 270000)
+#define __TENSOR_OPS_SUPPORT_DEPLOYMENT_TARGET_27_3 ((__ENVIRONMENT_OS_VERSION_MIN_REQUIRED__) >= 270300)
#endif
diff -ruN /Applications/Xcode_27.2.0.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/__impl/MPPTensorOpsMatMul2dImpl.h /Applications/Xcode_27.3.0-beta1.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/__impl/MPPTensorOpsMatMul2dImpl.h
--- /Applications/Xcode_27.2.0.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/__impl/MPPTensorOpsMatMul2dImpl.h 2026-10-09 06:42:30
+++ /Applications/Xcode_27.3.0-beta1.app/Contents/Developer/Platforms/MacOSX.platform/Developer/SDKs/MacOSX.sdk/System/Library/Frameworks/MetalPerformancePrimitives.framework/Headers/__impl/MPPTensorOpsMatMul2dImpl.h 2026-11-04 09:32:08
@@ -27,6 +27,113 @@
using __reduction_operation = reduction_operation;
+using __matmul2d_cooperative_tensor_layout = matmul2d_cooperative_tensor_layout;
+
+template <class T>
+struct __matmul2d_void
+{
+ using type = void;
+};
+
+// Layout of a matmul2d cooperative tensor's layout type; automatic for any other type, so functions that
+// require a matmul2d source keep reporting it through their own static_asserts.
+template <class L, class = void>
+struct __matmul2d_layout_of
+{
+ static constexpr constant __matmul2d_cooperative_tensor_layout value = __matmul2d_cooperative_tensor_layout::automatic;
+};
+
+template <class L>
+struct __matmul2d_layout_of<L, typename __matmul2d_void<decltype(L::__layout)>::type>
+{
+ static constexpr constant __matmul2d_cooperative_tensor_layout value = L::__layout;
+};
+
+// Trailing argument of the _v3 entrypoints. Its size, alignment and field offsets never change.
+struct alignas(128) __matmul2d_ext
+{
+ uint8_t version;
+ uint8_t left_layout;
+ uint8_t right_layout;
+ uint8_t destination_layout;
+ uint8_t flags; // __matmul2d_ext_flag_* bits
+};
+
+// Bits of __matmul2d_ext::flags (bits 3..7 are reserved)
+constexpr constant uint8_t __matmul2d_ext_flag_left_constant = 1 << 0; // left operand is a constant address space tensor
+constexpr constant uint8_t __matmul2d_ext_flag_right_constant = 1 << 1; // right operand is a constant address space tensor
+constexpr constant uint8_t __matmul2d_ext_flag_destination_constant = 1 << 2; // destination is a constant address space tensor
+
+constexpr constant uint8_t __matmul2d_ext_v3_version = 3;
+
+static_assert(sizeof(__matmul2d_ext) == 128 && alignof(__matmul2d_ext) == 128, "__matmul2d_ext is ABI");
+static_assert(__builtin_offsetof(__matmul2d_ext, version) == 0 && __builtin_offsetof(__matmul2d_ext, left_layout) == 1 &&
+ __builtin_offsetof(__matmul2d_ext, right_layout) == 2 && __builtin_offsetof(__matmul2d_ext, destination_layout) == 3,
+ "__matmul2d_ext is ABI");
+static_assert(__builtin_offsetof(__matmul2d_ext, flags) == 4, "__matmul2d_ext is ABI");
+
+// The extension struct the header builds (the newest version the deployment target supports)
+using __matmul2d_ext_t = __matmul2d_ext;
+
+// The part of an extension struct that is passed to the _v3 entrypoints
+inline const thread __matmul2d_ext &__matmul2d_ext_base(const thread __matmul2d_ext &ext)
+{
+ return ext;
+}
+
+inline __matmul2d_ext_t __make_matmul2d_ext(__matmul2d_cooperative_tensor_layout left_layout,
+ __matmul2d_cooperative_tensor_layout right_layout,
+ __matmul2d_cooperative_tensor_layout destination_layout,
+ bool left_is_constant = false,
+ bool right_is_constant = false,
+ bool destination_is_constant = false)
+{
+ __matmul2d_ext_t ext;
+ ext.version = __matmul2d_ext_v3_version;
+ ext.left_layout = static_cast<uint8_t>(left_layout);
+ ext.right_layout = static_cast<uint8_t>(right_layout);
+ ext.destination_layout = static_cast<uint8_t>(destination_layout);
+ ext.flags = (left_is_constant ? __matmul2d_ext_flag_left_constant : 0) |
+ (right_is_constant ? __matmul2d_ext_flag_right_constant : 0) |
+ (destination_is_constant ? __matmul2d_ext_flag_destination_constant : 0);
+ return ext;
+}
+
+// Version 3, all layouts automatic, no constant operands
+inline __matmul2d_ext __make_matmul2d_ext_automatic()
+{
+ __matmul2d_ext ext;
+ ext.version = __matmul2d_ext_v3_version;
+ ext.left_layout = static_cast<uint8_t>(__matmul2d_cooperative_tensor_layout::automatic);
+ ext.right_layout = static_cast<uint8_t>(__matmul2d_cooperative_tensor_layout::automatic);
+ ext.destination_layout = static_cast<uint8_t>(__matmul2d_cooperative_tensor_layout::automatic);
+ ext.flags = 0;
+ return ext;
+}
+
+// Address spaces served by the device ("dv") entrypoints: device, and from 27.3 constant (the operand's
+// __matmul2d_ext_flag_*_constant bit in __matmul2d_ext::flags is set)
+#if __TENSOR_OPS_SUPPORT_DEPLOYMENT_TARGET_27_3
+template <class PtrType>
+constexpr constant bool __matmul2d_is_dv_addrspace_v =
+ __tensor_ops_detail::__is_device_addrspace_v<PtrType> || __tensor_ops_detail::__is_constant_addrspace_v<PtrType>;
+#else
+template <class PtrType>
+constexpr constant bool __matmul2d_is_dv_addrspace_v = __tensor_ops_detail::__is_device_addrspace_v<PtrType>;
+#endif
+
+#if __TENSOR_OPS_SUPPORT_DEPLOYMENT_TARGET_27_3
+#define __MATMUL2D_UTIL_CALL(fn, ext, ...) fn##_v3(__VA_ARGS__, __matmul2d_ext_base(ext))
+#else
+#define __MATMUL2D_UTIL_CALL(fn, ext, ...) fn(__VA_ARGS__)
+#endif
+
+#if __TENSOR_OPS_SUPPORT_DEPLOYMENT_TARGET_27_3
+#define __MATMUL2D_RUN_CALL(fn, ext, ...) fn##_v3(__VA_ARGS__, __matmul2d_ext_base(ext))
+#else
+#define __MATMUL2D_RUN_CALL(fn, ext, ...) fn##_v2(__VA_ARGS__)
+#endif
+
enum class __matmul2d_cooperative_operand_index
{
left,
@@ -4372,6 +4479,2638 @@
extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_single_thread_dv_ui8_dv_b16_dv_f32(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType);
#endif
+#if __TENSOR_OPS_SUPPORT_DEPLOYMENT_TARGET_27_3
+// 27.3 run entrypoints with a cooperative operand: the _v2 signature plus a trailing __matmul2d_ext.
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_i8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_i8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_i8_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_i8_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_i8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_i8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_i8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_i8_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_i8_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_i8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i8_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i8_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i8_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i8_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_i8_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_i8_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_i8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_ui8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_ui8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_ui8_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_ui8_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_ui8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_ui8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_ui8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_ui8_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_ui8_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_ui8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui8_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui8_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui8_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui8_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_ui8_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_ui8_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_ui8_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_dv_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_tg_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_dv_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_tg_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_dv_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_tg_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_dv_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_tg_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_dv_i8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_tg_i8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_i8_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_i8_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_i8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_dv_i8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_tg_i8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_i8_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_i8_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_i8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_i8_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_i8_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_i8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_i8_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_i8_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_i8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_i8_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_i8_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_i8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_dv_ui8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_tg_ui8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_ui8_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_ui8_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_ui8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_dv_ui8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_tg_ui8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_ui8_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_ui8_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_ui8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_ui8_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_ui8_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_ui8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_ui8_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_ui8_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_ui8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_ui8_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_ui8_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_ui8_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f32_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f32_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f32_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_i8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_i8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_i8_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_i8_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_i8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_i8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_i8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_i8_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_i8_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_i8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i8_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i8_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i8_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i8_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_i8_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_i8_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_i8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_i8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_i8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_i8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f32_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f32_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f32_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_f16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_f16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_f16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_f16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_f16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_b16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_b16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_b16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_b16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_b16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_b16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_b16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_b16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_b16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_b16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_b16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_b16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_b16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_b16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_b16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_b16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_b16_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_b16_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_b16_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_ui8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_ui8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_ui8_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_ui8_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_ui8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_ui8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_ui8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_ui8_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_ui8_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_ui8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui8_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui8_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui8_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui8_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_ui8_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_ui8_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_ui8_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_ui8_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_ui8_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_ui8_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_b16_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_b16_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_b16_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_b16_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_b16_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_b16_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_i4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_i4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_i4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_i4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i4_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i4_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i4_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i4_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i4_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i4_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i4_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i4_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_ui4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_ui4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_ui4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_ui4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui4_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui4_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui4_dv_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui4_tg_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui4_f16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_dv_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_f16_tg_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui4_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui4_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_dv_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui4_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui4_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_f16_tg_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_dv_i4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_tg_i4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_dv_i4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_tg_i4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_i4_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_i4_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_i4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_i4_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_i4_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_i4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_dv_ui4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_tg_ui4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_dv_ui4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_tg_ui4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_ui4_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_ui4_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_ui4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_ui4_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_ui4_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_ui4_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_i4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_i4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_i4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_i4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i4_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i4_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i4_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i4_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_ui4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_ui4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_ui4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_ui4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui4_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui4_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui4_dv_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui4_tg_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui4_b16_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i4_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i4_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i4_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i4_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_i4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_dv_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_b16_tg_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_dv_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_b16_tg_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui4_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui4_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_dv_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui4_dv_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui4_tg_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_b16_tg_ui4_f32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_dv_i2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_i8_tg_i2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_dv_i2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_i8_tg_i2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_i2_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_i2_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_dv_i2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_i2_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_i2_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_i8_tg_i2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_dv_ui2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_ui8_tg_ui2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_dv_ui2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_tg_ui8_tg_ui2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_ui2_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_ui2_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_dv_ui2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_ui2_dv_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_ui2_tg_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type destinationDescType, int destinationStaticExtent0, int destinationStaticExtent1, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_ui8_tg_ui2_i32_v3(thread matmul2d_descriptor & desc, thread void *left, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_dv_i2_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type leftDescType, __tensor_ops_detail::__tensor_ops_datatype leftScaleDataType, int leftScaleBlockSize0, int leftScaleBlockSize1, int leftStaticExtent0, int leftStaticExtent1, thread void *right, __tensor_ops_detail::__tensor_ops_tensor_descriptor_type rightDescType, __tensor_ops_detail::__tensor_ops_datatype rightScaleDataType, int rightScaleBlockSize0, int rightScaleBlockSize1, int rightStaticExtent0, int rightStaticExtent1, thread void *destination, int threads, const thread __matmul2d_ext & ext);
+extern "C" EXTERNALLY_DEFINED_ATTR void __tensorops_impl_matmul2d_op_run_cooperative_dv_f16_tg_i2_f16_v3(thread matmul2d_descriptor & desc, thread void *left, __tensor_ops_detail: