From 5627fdd90e38640c82879ff8ba2ffebf0916c213 Mon Sep 17 00:00:00 2001
From: letaoqin <letaoqin@amd.com>
Date: Mon, 24 Mar 2025 07:27:40 +0000
Subject: [PATCH] bf16 add NPadding for library

---
 ..._mk_mfma16x16_mn_compute_default_instance_p1_padding.cpp | 5 +++++
 ..._mk_mfma16x16_mn_compute_default_instance_p2_padding.cpp | 5 +++++
 ..._mk_mfma16x16_mn_compute_default_instance_p3_padding.cpp | 5 +++++
 ..._mk_mfma16x16_mn_compute_default_instance_p4_padding.cpp | 5 +++++
 ..._mk_mfma16x16_mn_compute_default_instance_p5_padding.cpp | 5 +++++
 ..._mk_mfma16x16_mn_compute_default_instance_p6_padding.cpp | 5 +++++
 ..._gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn.hpp | 1 +
 ..._bf16_mk_mfma_mn_compute_default_instance_p1_padding.cpp | 5 +++++
 ..._bf16_mk_mfma_mn_compute_default_instance_p2_padding.cpp | 5 +++++
 ...dl_f8_f8_bf16_mk_mfma_mn_p1_default_instance_padding.cpp | 6 ++++++
 ...f8_f8_bf16_mk_mfma_mn_p1_default_instance_v2_padding.cpp | 6 ++++++
 ...dl_f8_f8_bf16_mk_mfma_mn_p2_default_instance_padding.cpp | 6 ++++++
 ...f8_f8_bf16_mk_mfma_mn_p2_default_instance_v2_padding.cpp | 6 ++++++
 ...dl_f8_f8_bf16_mk_mfma_mn_p3_default_instance_padding.cpp | 6 ++++++
 ...f8_f8_bf16_mk_mfma_mn_p3_default_instance_v2_padding.cpp | 6 ++++++
 ...dl_f8_f8_bf16_mk_mfma_mn_p4_default_instance_padding.cpp | 6 ++++++
 ...f8_f8_bf16_mk_mfma_mn_p4_default_instance_v2_padding.cpp | 6 ++++++
 ...dl_f8_f8_bf16_mk_mfma_mn_p5_default_instance_padding.cpp | 6 ++++++
 ...f8_f8_bf16_mk_mfma_mn_p5_default_instance_v2_padding.cpp | 6 ++++++
 19 files changed, 101 insertions(+)

diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p1_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p1_padding.cpp
index 912117213a..84913b011f 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p1_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p1_padding.cpp
@@ -25,6 +25,11 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma1
         instances,
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p1<
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p1<
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p2_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p2_padding.cpp
index 49fe8e53d9..31e525587d 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p2_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p2_padding.cpp
@@ -25,6 +25,11 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma1
         instances,
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p2<
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p2<
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p3_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p3_padding.cpp
index c7bf7fe895..fc0c735b29 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p3_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p3_padding.cpp
@@ -25,6 +25,11 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma1
         instances,
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p3<
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p3<
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p4_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p4_padding.cpp
index b77918d6b7..429d65f3b5 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p4_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p4_padding.cpp
@@ -25,6 +25,11 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma1
         instances,
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p4<
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p4<
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p5_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p5_padding.cpp
index 8338ead7e3..611af32591 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p5_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p5_padding.cpp
@@ -25,6 +25,11 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma1
         instances,
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p5<
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p5<
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p6_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p6_padding.cpp
index 431749a67e..5c08311dfd 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p6_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_default_instance_p6_padding.cpp
@@ -25,6 +25,11 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma1
         instances,
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p6<
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma16x16_mn_compute_instances_p6<
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn.hpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn.hpp
index 4c29aa7bd9..aa7305c694 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn.hpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn.hpp
@@ -28,6 +28,7 @@ using MultiplyMultiply = element_wise::MultiplyMultiply;
 
 static constexpr auto GemmDefault    = GemmSpecialization::Default;
 static constexpr auto GemmKPadding   = GemmSpecialization::KPadding;
+static constexpr auto GemmNPadding   = GemmSpecialization::NPadding;
 static constexpr auto GemmNKPadding  = GemmSpecialization::NKPadding;
 static constexpr auto GemmMNPadding  = GemmSpecialization::MNPadding;
 static constexpr auto GemmMNKPadding = GemmSpecialization::MNKPadding;
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_compute_default_instance_p1_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_compute_default_instance_p1_padding.cpp
index 3d76063dde..98f2b011bc 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_compute_default_instance_p1_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_compute_default_instance_p1_padding.cpp
@@ -25,6 +25,11 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         instances,
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_compute_instances_p1<
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_compute_instances_p1<
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_compute_default_instance_p2_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_compute_default_instance_p2_padding.cpp
index d2e3decc04..16c3f500c6 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_compute_default_instance_p2_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_compute_default_instance_p2_padding.cpp
@@ -25,6 +25,11 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         instances,
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_compute_instances_p2<
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_compute_instances_p2<
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p1_default_instance_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p1_default_instance_padding.cpp
index dc31720da9..9dddcb05b9 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p1_default_instance_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p1_default_instance_padding.cpp
@@ -26,6 +26,12 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p1_instances<
             v1,
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p1_instances<
+            v1,
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p1_default_instance_v2_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p1_default_instance_v2_padding.cpp
index 9e91b37e5c..ba00e97a71 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p1_default_instance_v2_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p1_default_instance_v2_padding.cpp
@@ -26,6 +26,12 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p1_instances<
             v2,
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p1_instances<
+            v2,
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p2_default_instance_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p2_default_instance_padding.cpp
index f8b514304a..2c361d49e0 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p2_default_instance_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p2_default_instance_padding.cpp
@@ -26,6 +26,12 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p2_instances<
             v1,
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p2_instances<
+            v1,
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p2_default_instance_v2_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p2_default_instance_v2_padding.cpp
index 66f0e1460a..fd57bc4899 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p2_default_instance_v2_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p2_default_instance_v2_padding.cpp
@@ -26,6 +26,12 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p2_instances<
             v2,
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p2_instances<
+            v2,
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p3_default_instance_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p3_default_instance_padding.cpp
index d515456bc2..b96571032a 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p3_default_instance_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p3_default_instance_padding.cpp
@@ -26,6 +26,12 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p3_instances<
             v1,
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p3_instances<
+            v1,
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p3_default_instance_v2_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p3_default_instance_v2_padding.cpp
index 6586cae5f9..6fdd277dda 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p3_default_instance_v2_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p3_default_instance_v2_padding.cpp
@@ -26,6 +26,12 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p3_instances<
             v2,
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p3_instances<
+            v2,
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p4_default_instance_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p4_default_instance_padding.cpp
index d6bb82e965..1f18bf868b 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p4_default_instance_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p4_default_instance_padding.cpp
@@ -26,6 +26,12 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p4_instances<
             v1,
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p4_instances<
+            v1,
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p4_default_instance_v2_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p4_default_instance_v2_padding.cpp
index 4e71a77604..e1a9a12a8f 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p4_default_instance_v2_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p4_default_instance_v2_padding.cpp
@@ -26,6 +26,12 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p4_instances<
             v2,
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p4_instances<
+            v2,
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p5_default_instance_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p5_default_instance_padding.cpp
index 18d0167e1b..f58b1ffaef 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p5_default_instance_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p5_default_instance_padding.cpp
@@ -26,6 +26,12 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p5_instances<
             v1,
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p5_instances<
+            v1,
+            GemmNPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p5_default_instance_v2_padding.cpp b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p5_default_instance_v2_padding.cpp
index 9a4b995865..b43c7f9f50 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p5_default_instance_v2_padding.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_multiply_multiply_wp/f8_f8_bf16/device_gemm_multiply_multiply_wp_xdl_f8_f8_bf16_mk_mfma_mn_p5_default_instance_v2_padding.cpp
@@ -26,6 +26,12 @@ void add_device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_
         device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p5_instances<
             v2,
             GemmKPadding>{});
+
+    add_device_operation_instances(
+        instances,
+        device_gemm_multiply_multiply_weight_preshuffle_xdl_f8_f8_bf16_mk_mfma_mn_p5_instances<
+            v2,
+            GemmNPadding>{});
 }
 
 } // namespace instance