ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-20 06:49:15 +00:00
Code Issues Packages Projects Releases Wiki Activity

Shuffle fix for gfx950 (#3491)

Browse Source 
* solve compiler issue

* solve the gfx950 mfma shuffle regression

* refactor jenkinsfile to handle arch name better

* [CK TILE] set divisor to count of thread along k dimension

* fix the compiler error

* solve degradation

* Finish the multiplies fix

* fix the scales

* solve compilation error

* solve the composes

* solve the error of tile sweeper

* fix the test and example

* fix for gfx950

---------

Co-authored-by: Max Podkorytov <4273004+tenpercent@users.noreply.github.com>
Co-authored-by: illsilin_amdeng <Illia.Silin@amd.com>
Co-authored-by: Cong Ma <congma13@amd.com>
This commit is contained in:
Thomas Ning
2026-01-14 01:21:29 +08:00
committed by GitHub
parent 9908a87c31
commit 00c46785a8
33 changed files with 161 additions and 152 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
include/ck_tile/host/device_prop.hpp
View File

@@ -65,7 +65,7 @@ inline bool is_gfx12_supported()
return get_device_name() == "gfx1200" || get_device_name() == "gfx1201";
}
inline bool is_load_tr_supported()
inline bool is_gfx95_supported()
{
// Check if load transpose is supported.
return get_device_name() == "gfx950";

5
include/ck_tile/host/tensor_shuffle_utils.hpp
View File

@@ -2,6 +2,7 @@
// SPDX-License-Identifier: MIT
#pragma once
#include "device_prop.hpp"
#include <stdexcept>
namespace ck_tile {
@@ -98,7 +99,7 @@ auto shuffle_b(const ck_tile::HostTensor<T>& t, const GemmConfig& gemmConfig)
else
{
assert(is_wave32() == false);
divisor = gemmConfig.N_Warp_Tile == 32 ? 2 : 4;
divisor = get_warp_size() / gemmConfig.N_Warp_Tile;
}
ck_tile::HostTensor<T> t_view({n_ / gemmConfig.N_Warp_Tile,
gemmConfig.N_Warp_Tile,
@@ -167,7 +168,7 @@ auto shuffle_b_permuteN(const ck_tile::HostTensor<T>& t, const GemmConfig& gemmC
else
{
assert(is_wave32() == false);
divisor = gemmConfig.N_Warp_Tile == 32 ? 2 : 4;
divisor = get_warp_size() / gemmConfig.N_Warp_Tile;
}
ck_tile::HostTensor<T> t_view({n_ / gemmConfig.N_Tile,
gemmConfig.N_Warp,