rename gridwise ops

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_kloop_v1.hpp

@@ -641,7 +641,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Kloop_Xdl_CShuffle_V1
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V1<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Kloop_Xdl_CShuffle_V1<
         InputDataType, // TODO: distinguish A/B datatype
         OutputDataType,
         ZDataType,

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_kloop_v2.hpp

@@ -640,7 +640,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Kloop_Xdl_CShuffle_V2
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V2<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Kloop_Xdl_CShuffle_V2<
         InputDataType, // TODO: distinguish A/B datatype
         OutputDataType,
         ZDataType,

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_qloop_phased_v1.hpp

@@ -637,7 +637,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Phased_Xdl_CShuffle_V1
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V1<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V1<
         InputDataType, // TODO: distinguish A/B datatype
         OutputDataType,
         ZDataType,

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_qloop_v1.hpp

@@ -627,7 +627,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V1
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V1<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V1<
         InputDataType, // TODO: distinguish A/B datatype
         OutputDataType,
         ZDataType,

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_bwd_xdl_cshuffle_qloop_v2.hpp

@@ -633,7 +633,7 @@ struct DeviceBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V2
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V2<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V2<
         InputDataType, // TODO: distinguish A/B datatype
         OutputDataType,
         ZDataType,

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_fwd_xdl_cshuffle_v1.hpp

@@ -449,7 +449,7 @@ struct DeviceBatchedMultiheadAttentionForward_Xdl_CShuffle_V1
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionForward_Xdl_CShuffle<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionForward_Xdl_CShuffle_V1<
         ADataType, // TODO: distinguish A/B datatype
         ZDataType,
         GemmDataType,

include/ck/tensor_operation/gpu/device/impl/device_batched_mha_fwd_xdl_cshuffle_v2.hpp

@@ -457,7 +457,7 @@ struct DeviceBatchedMultiheadAttentionForward_Xdl_CShuffle_V2
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionForward_Xdl_CShuffle<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionForward_Xdl_CShuffle_V2<
         ADataType, // TODO: distinguish A/B datatype
         ZDataType,
         GemmDataType,

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_kloop_v1.hpp

@@ -576,7 +576,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Kloop_Xdl_CShuffle_V1
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V1<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Kloop_Xdl_CShuffle_V1<
         InputDataType, // TODO: distinguish A/B datatype
         OutputDataType,
         ZDataType,

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_kloop_v2.hpp

@@ -569,7 +569,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Kloop_Xdl_CShuffle_V2
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V2<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Kloop_Xdl_CShuffle_V2<
         InputDataType, // TODO: distinguish A/B datatype
         OutputDataType,
         ZDataType,

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_qloop_v1.hpp

@@ -573,7 +573,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V1
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V1<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V1<
         InputDataType, // TODO: distinguish A/B datatype
         OutputDataType,
         ZDataType,

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_bwd_xdl_cshuffle_qloop_v2.hpp

@@ -573,7 +573,7 @@ struct DeviceGroupedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V2
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V2<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V2<
         InputDataType, // TODO: distinguish A/B datatype
         OutputDataType,
         ZDataType,

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_fwd_xdl_cshuffle_v1.hpp

@@ -465,7 +465,7 @@ struct DeviceGroupedMultiheadAttentionForward_Xdl_CShuffle_V1
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionForward_Xdl_CShuffle<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionForward_Xdl_CShuffle_V1<
         ADataType, // TODO: distinguish A/B datatype
         ZDataType,
         GemmDataType,

include/ck/tensor_operation/gpu/device/impl/device_grouped_mha_fwd_xdl_cshuffle_v2.hpp

@@ -471,7 +471,7 @@ struct DeviceGroupedMultiheadAttentionForward_Xdl_CShuffle_V2
     };
 
     // GridwiseGemm
-    using GridwiseGemm = GridwiseBatchedMultiheadAttentionForward_Xdl_CShuffle<
+    using GridwiseGemm = GridwiseBatchedMultiheadAttentionForward_Xdl_CShuffle_V2<
         ADataType, // TODO: distinguish A/B datatype
         ZDataType,
         GemmDataType,

include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_bwd_xdl_cshuffle_kloop_v1.hpp

@@ -88,7 +88,7 @@ template <typename InputDataType,
           bool MaskOutUpperTriangle,
           bool Deterministic,
           PipelineVersion PipelineVer = PipelineVersion::v1>
-struct GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V1
+struct GridwiseBatchedMultiheadAttentionBackward_Kloop_Xdl_CShuffle_V1
 {
     static_assert(LoopSched == LoopScheduler::Default,
                   "Non-default loop scheduler is currently not supported");

include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_bwd_xdl_cshuffle_kloop_v2.hpp

@@ -88,7 +88,7 @@ template <typename InputDataType,
           bool MaskOutUpperTriangle,
           bool Deterministic,
           PipelineVersion PipelineVer = PipelineVersion::v1>
-struct GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V2
+struct GridwiseBatchedMultiheadAttentionBackward_Kloop_Xdl_CShuffle_V2
 {
     static_assert(LoopSched == LoopScheduler::Default,
                   "Non-default loop scheduler is currently not supported");

include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_bwd_xdl_cshuffle_qloop_b2t_v1.hpp

@@ -81,7 +81,7 @@ template <typename InputDataType,
           bool MaskOutUpperTriangle,
           bool Deterministic,
           PipelineVersion PipelineVer = PipelineVersion::v1>
-struct GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V1
+struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V1
 {
     static_assert(LoopSched == LoopScheduler::Default,
                   "Non-default loop scheduler is currently not supported");

include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_bwd_xdl_cshuffle_qloop_b2t_v2.hpp

@@ -89,7 +89,7 @@ template <typename InputDataType,
           bool MaskOutUpperTriangle,
           bool Deterministic,
           PipelineVersion PipelineVer = PipelineVersion::v1>
-struct GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V2
+struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V2
 {
     static_assert(LoopSched == LoopScheduler::Default,
                   "Non-default loop scheduler is currently not supported");

include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_bwd_xdl_cshuffle_qloop_t2b_phased_v1.hpp

@@ -89,7 +89,7 @@ template <typename InputDataType,
           bool MaskOutUpperTriangle,
           bool Deterministic,
           PipelineVersion PipelineVer = PipelineVersion::v1>
-struct GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V1
+struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V1
 {
     static_assert(LoopSched == LoopScheduler::Default,
                   "Non-default loop scheduler is currently not supported");

include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_bwd_xdl_cshuffle_qloop_t2b_v1.hpp

@@ -81,7 +81,7 @@ template <typename InputDataType,
           bool MaskOutUpperTriangle,
           bool Deterministic,
           PipelineVersion PipelineVer = PipelineVersion::v1>
-struct GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V1
+struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V1
 {
     static_assert(LoopSched == LoopScheduler::Default,
                   "Non-default loop scheduler is currently not supported");

include/ck/tensor_operation/gpu/grid/gridwise_batched_mha_bwd_xdl_cshuffle_qloop_t2b_v2.hpp

@@ -89,7 +89,7 @@ template <typename InputDataType,
           bool MaskOutUpperTriangle,
           bool Deterministic,
           PipelineVersion PipelineVer = PipelineVersion::v1>
-struct GridwiseBatchedMultiheadAttentionBackward_Xdl_CShuffle_V2
+struct GridwiseBatchedMultiheadAttentionBackward_Qloop_Xdl_CShuffle_V2
 {
     static_assert(LoopSched == LoopScheduler::Default,
                   "Non-default loop scheduler is currently not supported");