From e2855633eed935fb97d9bcbc956bb7bc704c7a5c Mon Sep 17 00:00:00 2001
From: Anca Hamuraru <anca@streamhpc.com>
Date: Wed, 14 May 2025 10:50:34 +0000
Subject: [PATCH] Fixes for padding instances for f16xf16xf16

---
 ...universal_f16_f16_f16_mk_nk_mn_comp_kpadding_instance.cpp | 4 ++--
 ...iversal_f16_f16_f16_mk_nk_mn_comp_mnkpadding_instance.cpp | 5 +++--
 ...niversal_f16_f16_f16_mk_nk_mn_comp_mnpadding_instance.cpp | 4 ++--
 3 files changed, 7 insertions(+), 6 deletions(-)

diff --git a/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_kpadding_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_kpadding_instance.cpp
index 331ca8b2ff..6a3c9159ed 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_kpadding_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_kpadding_instance.cpp
@@ -9,13 +9,13 @@ namespace tensor_operation {
 namespace device {
 namespace instance {
 
-void add_device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_default_instances(
+void add_device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_kpadding_instances(
     std::vector<std::unique_ptr<
         DeviceGemmV2<Row, Col, Row, F16, F16, F16, PassThrough, PassThrough, PassThrough>>>&
         instances)
 {
     add_device_operation_instances(
-        instances, device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_instances<GemmDefault>{});
+        instances, device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_instances<GemmKPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_mnkpadding_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_mnkpadding_instance.cpp
index 331ca8b2ff..bad4851eac 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_mnkpadding_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_mnkpadding_instance.cpp
@@ -9,13 +9,14 @@ namespace tensor_operation {
 namespace device {
 namespace instance {
 
-void add_device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_default_instances(
+void add_device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_mnkpadding_instances(
     std::vector<std::unique_ptr<
         DeviceGemmV2<Row, Col, Row, F16, F16, F16, PassThrough, PassThrough, PassThrough>>>&
         instances)
 {
     add_device_operation_instances(
-        instances, device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_instances<GemmDefault>{});
+        instances,
+        device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_instances<GemmMNKPadding>{});
 }
 
 } // namespace instance
diff --git a/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_mnpadding_instance.cpp b/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_mnpadding_instance.cpp
index 331ca8b2ff..3f9c34c83e 100644
--- a/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_mnpadding_instance.cpp
+++ b/library/src/tensor_operation_instance/gpu/gemm_universal/device_gemm_wmma_universal_f16_f16_f16/device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_mnpadding_instance.cpp
@@ -9,13 +9,13 @@ namespace tensor_operation {
 namespace device {
 namespace instance {
 
-void add_device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_default_instances(
+void add_device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_mnpadding_instances(
     std::vector<std::unique_ptr<
         DeviceGemmV2<Row, Col, Row, F16, F16, F16, PassThrough, PassThrough, PassThrough>>>&
         instances)
 {
     add_device_operation_instances(
-        instances, device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_instances<GemmDefault>{});
+        instances, device_gemm_wmma_universal_f16_f16_f16_mk_nk_mn_comp_instances<GemmMNPadding>{});
 }
 
 } // namespace instance