ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-03 21:21:22 +00:00
Code Issues Packages Projects Releases Wiki Activity

Add read_first_lane function for int64 (#1347)

Browse Source
This commit is contained in:
Bartłomiej Kocot
2024-06-18 22:05:30 +02:00
committed by GitHub
parent e2d139201b
commit 8faec23cb4
10 changed files with 105 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
include/ck/tensor_operation/gpu/grid/gridwise_gemm_multiple_d_wmma_cshuffle.hpp
View File

@@ -60,9 +60,12 @@ __global__ void
__builtin_amdgcn_readfirstlane(get_grid_size() / batch_count);
const index_t g_idx = __builtin_amdgcn_readfirstlane(get_block_1d_id() / num_blocks_per_batch);
const long_index_t a_batch_offset = compute_ptr_offset_of_batch.GetAPtrOffset(g_idx);
const long_index_t b_batch_offset = compute_ptr_offset_of_batch.GetBPtrOffset(g_idx);
const long_index_t e_batch_offset = compute_ptr_offset_of_batch.GetEPtrOffset(g_idx);
const long_index_t a_batch_offset =
amd_wave_read_first_lane(compute_ptr_offset_of_batch.GetAPtrOffset(g_idx));
const long_index_t b_batch_offset =
amd_wave_read_first_lane(compute_ptr_offset_of_batch.GetBPtrOffset(g_idx));
const long_index_t e_batch_offset =
amd_wave_read_first_lane(compute_ptr_offset_of_batch.GetEPtrOffset(g_idx));
const auto ds_batch_offset = compute_ptr_offset_of_batch.GetDsPtrOffset(g_idx);
@@ -152,9 +155,12 @@ __global__ void
__builtin_amdgcn_readfirstlane(get_grid_size() / batch_count);
const index_t g_idx = __builtin_amdgcn_readfirstlane(get_block_1d_id() / num_blocks_per_batch);
const long_index_t a_batch_offset = compute_ptr_offset_of_batch.GetAPtrOffset(g_idx);
const long_index_t b_batch_offset = compute_ptr_offset_of_batch.GetBPtrOffset(g_idx);
const long_index_t e_batch_offset = compute_ptr_offset_of_batch.GetEPtrOffset(g_idx);
const long_index_t a_batch_offset =
amd_wave_read_first_lane(compute_ptr_offset_of_batch.GetAPtrOffset(g_idx));
const long_index_t b_batch_offset =
amd_wave_read_first_lane(compute_ptr_offset_of_batch.GetBPtrOffset(g_idx));
const long_index_t e_batch_offset =
amd_wave_read_first_lane(compute_ptr_offset_of_batch.GetEPtrOffset(g_idx));
const auto ds_batch_offset = compute_ptr_offset_of_batch.GetDsPtrOffset(g_idx);