Fix format

d9ce3a57 · Po-Yen, Chen · b5f11e02 · d9ce3a57 · d9ce3a57 · d9ce3a57
Commit d9ce3a57 authored Jul 05, 2023 by Po-Yen, Chen
20 changed files
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_add_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_add_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_default_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_default_pipeline_v1_instance.cpp
@@ -29,8 +29,7 @@ using Instances =
        >;
 void add_device_gemm_xdl_f16_f16_f16_km_kn_mn_default_pipeline_v1_instances(
-    OwnerList<InstanceNT>&
+    OwnerList<InstanceNT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_default_pipeline_v2_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_default_pipeline_v2_instance.cpp
@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for a[k, m] * b[k, n] = c[m, n]
 using Instances =
    std::tuple<
-        // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES        
        // pipeline v2, 1 wave
        //##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|          LoopScheduler| Pipeline|
@@ -31,8 +31,7 @@ using Instances =
        >;
 void add_device_gemm_xdl_f16_f16_f16_km_kn_mn_default_pipeline_v2_instances(
-    OwnerList<InstanceNT>&
+    OwnerList<InstanceNT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_interwave_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_interwave_pipeline_v1_instance.cpp
@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for a[k, m] * b[k, n] = c[m, n]
 using Instances =
    std::tuple<
-        // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES        
        // pipeline v1, 2 waves
        //##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|          LoopScheduler| Pipeline|
@@ -31,8 +31,7 @@ using Instances =
        >;
 void add_device_gemm_xdl_f16_f16_f16_km_kn_mn_interwave_pipeline_v1_instances(
-    OwnerList<InstanceNT>&
+    OwnerList<InstanceNT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_irregular_default_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_irregular_default_pipeline_v1_instance.cpp
@@ -21,8 +21,7 @@ using Instances = std::tuple<
    >;
 void add_device_gemm_xdl_f16_f16_f16_km_kn_mn_irregular_default_pipeline_v1_instances(
-    OwnerList<InstanceNT>&
+    OwnerList<InstanceNT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_irregular_default_pipeline_v2_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_irregular_default_pipeline_v2_instance.cpp
@@ -10,7 +10,7 @@ namespace instance {
 // irregular tile size
 using Instances = std::tuple<
-    // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES        
        // pipeline v2, 1 wave
        //###########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|          LoopScheduler|                     Pipeline|
@@ -23,8 +23,7 @@ using Instances = std::tuple<
    >;
 void add_device_gemm_xdl_f16_f16_f16_km_kn_mn_irregular_default_pipeline_v2_instances(
-       OwnerList<InstanceNT>&
+    OwnerList<InstanceNT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_irregular_interwave_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_kn_mn_irregular_interwave_pipeline_v1_instance.cpp
@@ -10,7 +10,7 @@ namespace instance {
 // irregular tile size
 using Instances = std::tuple<
-    // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES        
        // pipeline v1, 2 waves
        //###########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|            LoopScheduler|                   Pipeline|
@@ -23,8 +23,7 @@ using Instances = std::tuple<
    >;
 void add_device_gemm_xdl_f16_f16_f16_km_kn_mn_irregular_interwave_pipeline_v1_instances(
-    OwnerList<InstanceNT>&
+    OwnerList<InstanceNT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_add_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_add_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_default_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_default_pipeline_v1_instance.cpp
@@ -29,8 +29,7 @@ using Instances =
        >;
 void add_device_gemm_xdl_f16_f16_f16_km_nk_mn_default_pipeline_v1_instances(
-    OwnerList<InstanceNN>&
+    OwnerList<InstanceNN>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_default_pipeline_v2_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_default_pipeline_v2_instance.cpp
@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for a[k, m] * b[n, k] = c[m, n]
 using Instances =
    std::tuple<
-        // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES        
        // pipeline v2, 1 wave
        //##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|          LoopScheduler|                     Pipeline|
@@ -31,8 +31,7 @@ using Instances =
        >;
 void add_device_gemm_xdl_f16_f16_f16_km_nk_mn_default_pipeline_v2_instances(
-    OwnerList<InstanceNN>&
+    OwnerList<InstanceNN>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_interwave_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_interwave_pipeline_v1_instance.cpp
@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for a[k, m] * b[n, k] = c[m, n]
 using Instances =
    std::tuple<
-        // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES        
        // pipeline v1, 2 waves
        //##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|          LoopScheduler|                     Pipeline|
@@ -31,8 +31,7 @@ using Instances =
        >;
 void add_device_gemm_xdl_f16_f16_f16_km_nk_mn_interwave_pipeline_v1_instances(
-    OwnerList<InstanceNN>&
+    OwnerList<InstanceNN>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_irregular_default_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_irregular_default_pipeline_v1_instance.cpp
@@ -21,8 +21,7 @@ using Instances = std::tuple<
    >;
 void add_device_gemm_xdl_f16_f16_f16_km_nk_mn_irregular_default_pipeline_v1_instances(
-    OwnerList<InstanceNN>&
+    OwnerList<InstanceNN>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_irregular_default_pipeline_v2_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_irregular_default_pipeline_v2_instance.cpp
@@ -10,7 +10,7 @@ namespace instance {
 // irregular tile size
 using Instances = std::tuple<
-    // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES        
        // pipeline v2, 1 wave
        //###########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|          LoopScheduler|                     Pipeline|
@@ -23,8 +23,7 @@ using Instances = std::tuple<
    >;
 void add_device_gemm_xdl_f16_f16_f16_km_nk_mn_irregular_default_pipeline_v2_instances(
-    OwnerList<InstanceNN>&
+    OwnerList<InstanceNN>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_irregular_interwave_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/km_nk_mn_irregular_interwave_pipeline_v1_instance.cpp
@@ -10,7 +10,7 @@ namespace instance {
 // irregular tile size
 using Instances = std::tuple<
-    // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES        
        // pipeline v1, 2 waves
        //###########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|          LoopScheduler|                     Pipeline|
@@ -23,8 +23,7 @@ using Instances = std::tuple<
    >;
 void add_device_gemm_xdl_f16_f16_f16_km_nk_mn_irregular_interwave_pipeline_v1_instances(
-    OwnerList<InstanceNN>&
+    OwnerList<InstanceNN>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_add_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_add_instance.cpp
--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_default_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_default_pipeline_v1_instance.cpp
@@ -38,8 +38,7 @@ using Instances =
        >;
 void add_device_gemm_xdl_f16_f16_f16_mk_kn_mn_default_pipeline_v1_instances(
-        OwnerList<InstanceTT>&
+    OwnerList<InstanceTT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_default_pipeline_v2_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_default_pipeline_v2_instance.cpp
@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for a[m, k] * b[k, n] = c[m, n]
 using Instances =
    std::tuple<
-        // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES        
        // pipeline v2, 1 wave
        //##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|          LoopScheduler|                     Pipeline|
@@ -40,8 +40,7 @@ using Instances =
        >;
 void add_device_gemm_xdl_f16_f16_f16_mk_kn_mn_default_pipeline_v2_instances(
-        OwnerList<InstanceTT>&
+    OwnerList<InstanceTT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_interwave_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_interwave_pipeline_v1_instance.cpp
@@ -11,7 +11,7 @@ namespace instance {
 // Compilation parameters for a[m, k] * b[k, n] = c[m, n]
 using Instances =
    std::tuple<
-        // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_INTER_WAVE_INSTANCES        
        // pipeline v1, 2 waves
        //##########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|          LoopScheduler|                     Pipeline|
@@ -40,8 +40,7 @@ using Instances =
        >;
 void add_device_gemm_xdl_f16_f16_f16_mk_kn_mn_interwave_pipeline_v1_instances(
-        OwnerList<InstanceTT>&
+    OwnerList<InstanceTT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_irregular_default_pipeline_v1_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_irregular_default_pipeline_v1_instance.cpp
@@ -21,8 +21,7 @@ using Instances = std::tuple<
    >;
 void add_device_gemm_xdl_f16_f16_f16_mk_kn_mn_irregular_default_pipeline_v1_instances(
-        OwnerList<InstanceTT>&
+    OwnerList<InstanceTT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }

--- a/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_irregular_default_pipeline_v2_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm/device_gemm_xdl_f16_f16_f16/mk_kn_mn_irregular_default_pipeline_v2_instance.cpp
@@ -10,7 +10,7 @@ namespace instance {
 // irregular tile size
 using Instances = std::tuple<
-    // clang-format off
+// clang-format off
 #if CK_EXPERIMENTAL_PIPELINE_V2_INSTANCES        
        // pipeline v2, 1 wave
        //###########| AData| BData| CData| AccData| ALayout| BLayout| CLayout|           A|           B|           C|          GEMM| Block|  MPer|  NPer| K0Per| K1| MPer| NPer| MXdl| NXdl|  ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockTransfer| ABlockLds|  BBlockTransfer| BBlockTransfer| BBlockTransfer| BlockTransfer| BBlockTransfer| BBlockTransfer| BBlockLds| CThreadTransfer| CThreadTransfer| NumPrefetch|          LoopScheduler|                     Pipeline|
@@ -23,8 +23,7 @@ using Instances = std::tuple<
    >;
 void add_device_gemm_xdl_f16_f16_f16_mk_kn_mn_irregular_default_pipeline_v2_instances(
-        OwnerList<InstanceTT>&
+    OwnerList<InstanceTT>& instances)
-        instances)
 {
    add_device_operation_instances(instances, Instances{});
 }