ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-19 20:40:07 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge commit '9fcc1ee9fd9730efd865f530afde505f2556954d' into develop

Browse Source
This commit is contained in:
assistant-librarian[bot]
2025-08-18 17:12:50 +00:00
parent d436787ed0
commit 68b20e1d4f
113 changed files with 610 additions and 531 deletions
Download Patch File Download Diff File Expand all files Collapse all files

24
example/ck_tile/05_reduce/reduce.cpp
View File

@@ -94,18 +94,18 @@ bool run(const ck_tile::ArgParser& arg_parser)
throw std::runtime_error("Wrong! Arguments not supported!\n");
}
float ave_time = launch_kernel(ck_tile::stream_config{nullptr, true, 0, warmup, repeat},
ck_tile::make_kernel<kBlockSize, kBlockPerCu>(
Kernel{},
kGridSize,
kBlockSize,
0,
static_cast<XDataType*>(x_buf.GetDeviceBuffer()),
static_cast<YDataType*>(y_buf.GetDeviceBuffer()),
input_shape,
input_strides,
kept_dim,
reduce_dims));
float ave_time = launch_kernel(
ck_tile::stream_config{nullptr, true, 0, warmup, repeat},
ck_tile::make_kernel<kBlockPerCu>(Kernel{},
kGridSize,
kBlockSize,
0,
static_cast<XDataType*>(x_buf.GetDeviceBuffer()),
static_cast<YDataType*>(y_buf.GetDeviceBuffer()),
input_shape,
input_strides,
kept_dim,
reduce_dims));
std::size_t num_btype = sizeof(XDataType) * N * C * H * W + sizeof(YDataType) * N * C;