diff --git a/CMakeLists.txt b/CMakeLists.txt
index 9f1bdf86899..356491d9c12 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -259,6 +259,11 @@ if ((SUPPORTED_GPU_TARGETS MATCHES "gfx94" OR SUPPORTED_GPU_TARGETS MATCHES "gfx
     add_definitions(-DCK_USE_GFX94)
     set(CK_USE_GFX94 "ON")
 endif()
+if (SUPPORTED_GPU_TARGETS MATCHES "gfx950" AND NOT FORCE_DISABLE_XDL)
+    message(STATUS "Enabling XDL FP8 gemms on gfx950")
+    add_definitions(-DCK_USE_GFX950)
+    set(CK_USE_GFX950 "ON")
+endif()
 
 # new macro CK_TILE_USE_WMMA in order to separately compile examples for MFMA/WMMA
 set(CK_TILE_USE_WMMA 0)
diff --git a/include/ck/config.h.in b/include/ck/config.h.in
index f5421e7d5ee..306a6c2ff1a 100644
--- a/include/ck/config.h.in
+++ b/include/ck/config.h.in
@@ -55,9 +55,6 @@
 #ifndef CK_ENABLE_FP32
 #define CK_ENABLE_FP32 "ON"
 #endif
-#ifndef CK_ENABLE_TF32
-#define CK_ENABLE_TF32 "ON"
-#endif
 #ifndef CK_ENABLE_FP64
 #define CK_ENABLE_FP64 "ON"
 #endif
@@ -88,10 +85,6 @@
 #cmakedefine CK_ENABLE_FP32 @CK_ENABLE_FP32@
 #endif
 
-#ifndef CK_ENABLE_TF32
-#cmakedefine CK_ENABLE_TF32 @CK_ENABLE_TF32@
-#endif
-
 #ifndef CK_ENABLE_FP64
 #cmakedefine CK_ENABLE_FP64 @CK_ENABLE_FP64@
 #endif
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_data/device_grouped_conv_bwd_data_xdl_instance.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_data/device_grouped_conv_bwd_data_xdl_instance.hpp
index 745f8cbd321..970bcb04393 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_data/device_grouped_conv_bwd_data_xdl_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_data/device_grouped_conv_bwd_data_xdl_instance.hpp
@@ -376,7 +376,7 @@ using device_grouped_conv_bwd_data_xdl_f32_optimized_loads_instances =
         // clang-format on
         >;
 
-#if defined(__gfx950__)
+#if defined(CK_USE_GFX950)
 constexpr auto _k_per_block = 32;
 #else
 constexpr auto _k_per_block = 16;
diff --git a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_merged_groups_instance.hpp b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_merged_groups_instance.hpp
index 944e68f1927..51aab2d4bf2 100644
--- a/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_merged_groups_instance.hpp
+++ b/library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_merged_groups_instance.hpp
@@ -143,7 +143,7 @@ using device_grouped_conv_fwd_xdl_merged_groups_f32_instances = std::tuple<
     // clang-format on
     >;
 
-#if defined(__gfx950__)
+#if defined(CK_USE_GFX950)
 constexpr auto _k_per_block = 32;
 #else
 constexpr auto _k_per_block = 16;