ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-07-01 04:07:56 +00:00
Code Issues Packages Projects Releases Wiki Activity

fix some compiling errors

Browse Source
This commit is contained in:
joye
2025-05-15 14:36:02 +08:00
parent 726ae62113
commit 2fdcb55cf1
3 changed files with 8 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
include/ck_tile/core/arch/amd_buffer_addressing.hpp
View File

@@ -77,6 +77,9 @@ struct buffer_store;
template <index_t bytes>
struct buffer_store_if;
template <index_t bytes, bool pre_nop = false>
struct async_buffer_load;
#pragma clang diagnostic push
#pragma clang diagnostic ignored "-Wundefined-reinterpret-cast"
// TODO: strict aliasing rule seems fail when reinterpret_cast between vector type

12
include/ck_tile/core/tensor/load_tile.hpp
View File

@@ -92,18 +92,14 @@ CK_TILE_DEVICE auto load_tile_raw(T& tile,
template <typename LdsTileWindow_,
typename TileWindow_,
index_t i_access = -1,
bool oob_conditional_check = true,
bool pre_nop = false>
bool oob_conditional_check = true>
CK_TILE_DEVICE auto async_load_tile(LdsTileWindow_&& lds_tile,
const TileWindow_& tile_window,
number<i_access> = {},
bool_constant<oob_conditional_check> = {},
bool_constant<pre_nop> = {})
bool_constant<oob_conditional_check> = {})
{
return tile_window.async_load(lds_tile,
number<i_access>{},
bool_constant<oob_conditional_check>{},
bool_constant<pre_nop>{});
return tile_window.async_load(
lds_tile, number<i_access>{}, bool_constant<oob_conditional_check>{});
}
template <typename LdsTileWindow_,

2
include/ck_tile/core/tensor/tile_window_linear.hpp
View File

@@ -748,7 +748,7 @@ struct tile_window_linear
auto bottom_tensor_flag = cached_flags_[IAccess];
auto lds_bottom_tensor_thread_idx =
lds_tile.get_window_origin() + window_adaptor_thread_coord.get_bottom_index();
lds_tile.get_window_origin() + window_adaptor_coord.get_bottom_index();
const auto lds_coord =
make_tensor_coordinate(lds_tile.get_bottom_tensor_view().get_tensor_descriptor(),