ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-19 22:39:03 +00:00
Code Issues Packages Projects Releases Wiki Activity

Support Wave32 in CK_TILE - Part 1 (#2594)

Browse Source 
* Support wave32/wave64 in CK_TILE - Part 1

* remove blocksize in kernel launch

* fix build error

* fix clang format

* fix clang format 2

* fix clang format 3

* fix fmha build error

* fix fmha build 2

* fix fmha build 3

* fix build error 4

* address review comment

* update change log

* replace KernelBlockSize with kBlockSize

* fix CI fail

* fix clang format

* address review comment and rebase code.

* fix universal test fail

---------

Co-authored-by: Lin, Qun <Quentin.Lin+amdeng@amd.com>
Co-authored-by: Thomas Ning <Thomas.Ning@amd.com>
This commit is contained in:
linqunAMD
2025-08-19 01:08:31 +08:00
committed by GitHub
parent 26d3300930
commit 9fcc1ee9fd
113 changed files with 610 additions and 531 deletions
Download Patch File Download Diff File Expand all files Collapse all files

5
example/ck_tile/12_smoothquant/example_smoothquant.cpp
View File

@@ -126,12 +126,11 @@ bool run(const ck_tile::ArgParser& arg_parser)
auto kargs = Kernel::MakeKargs(args);
const dim3 grids = Kernel::GridSize(args);
constexpr dim3 blocks = Kernel::BlockSize();
const dim3 blocks = Kernel::BlockSize();
constexpr ck_tile::index_t kBlockPerCu = 1;
auto s = ck_tile::stream_config{nullptr, true, 1, warmup, repeat};
ck_tile::launch_kernel(
s, ck_tile::make_kernel<blocks.x, kBlockPerCu>(Kernel{}, grids, blocks, 0, kargs));
ck_tile::launch_kernel(s, ck_tile::make_kernel<kBlockPerCu>(Kernel{}, grids, blocks, 0, kargs));
bool pass = true;

4
example/ck_tile/12_smoothquant/instances/smoothquant_instance_common.hpp
View File

@@ -50,7 +50,7 @@ float smoothquant_(const S& s, A a)
using Kernel = ck_tile::Smoothquant<Pipeline>;
const dim3 grids = Kernel::GridSize(a);
constexpr dim3 blocks = Kernel::BlockSize();
const dim3 blocks = Kernel::BlockSize();
constexpr ck_tile::index_t kBlockPerCu = 1;
auto kargs = Kernel::MakeKargs(a);
@@ -58,5 +58,5 @@ float smoothquant_(const S& s, A a)
std::cout << ", " << Kernel::GetName() << std::flush;
return ck_tile::launch_kernel(
s, ck_tile::make_kernel<blocks.x, kBlockPerCu>(Kernel{}, grids, blocks, 0, kargs));
s, ck_tile::make_kernel<kBlockPerCu>(Kernel{}, grids, blocks, 0, kargs));
}