CS348Project/llama.cpp
5
0
Fork 0
mirror of https://github.com/ggml-org/llama.cpp.git synced 2025-11-17 11:37:10 +00:00
Code Issues Packages Projects Releases Wiki Activity

Revert "sycl: add usage of enqueue_functions extension (#14244)" (#15910)

Browse Source 
* Revert "sycl: add usage of enqueue_functions extension (#14244)"

This reverts commit 8308f98c7f.

* fix missed revert code, format the code
This commit is contained in:
Neo Zhang Jianyu
2025-09-12 09:15:12 +08:00
committed by GitHub
parent 360d6533db
commit 704d90c987
20 changed files with 845 additions and 674 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
ggml/src/ggml-sycl/softmax.cpp
View File

@@ -127,11 +127,11 @@ static void soft_max_f32_submitter(const float * x, const T * mask, float * dst,
const int nrows_y, const float scale, const float max_bias, const float m0,
const float m1, uint32_t n_head_log2, sycl::range<3> block_nums, sycl::range<3> block_dims,
const size_t n_local_scratch, queue_ptr stream) {
sycl_launch(stream, [&](sycl::handler & cgh) {
stream->submit([&](sycl::handler &cgh) {
sycl::local_accessor<float, 1> local_buf_acc(n_local_scratch, cgh);
sycl_parallel_for(
cgh, sycl::nd_range<3>(block_nums * block_dims, block_dims),
cgh.parallel_for(
sycl::nd_range<3>(block_nums * block_dims, block_dims),
[=](sycl::nd_item<3> item_ct1) [[sycl::reqd_sub_group_size(WARP_SIZE)]] {
soft_max_f32<vals_smem, ncols_template, block_size_template>(x, mask, dst, ncols_par,
nrows_y, scale, max_bias, m0,