Lumpiasty/llama.cpp
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

[SYCL] Support gpt-oss by OPs add-id, mul_mat for mxfp4, swiglu_oai (#17826)

Browse Source 
* support gpt-oss GPU by OP add-id, mul_mat for mxfp4, swiglu_oai, fix warning

* fix fault ut case, update ops.md

* rebase, fix format issue
This commit is contained in:
Neo Zhang Jianyu
2025-12-15 10:35:15 +08:00
committed by GitHub
parent 745fa0e78b
commit 4aced7a631
15 changed files with 1201 additions and 385 deletions
Download Patch File Download Diff File Expand all files Collapse all files
ggml/src/ggml-sycl/mmvq.cpp
+22
View File
@@ -595,6 +595,25 @@ static void mul_mat_vec_q4_1_q8_1_sycl(const void *vx, const void *vy,
}
}
static void mul_mat_vec_mxfp4_q8_1_sycl(const void * vx, const void * vy, float * dst, const int ncols, const int nrows,
dpct::queue_ptr stream) {
GGML_ASSERT(ncols % QK_MXFP4 == 0);
const int block_num_y = (nrows + GGML_SYCL_MMV_Y - 1) / GGML_SYCL_MMV_Y;
const sycl::range<3> block_nums(1, 1, block_num_y);
const sycl::range<3> block_dims(1, GGML_SYCL_MMV_Y, WARP_SIZE);
{
stream->submit([&](sycl::handler & cgh) {
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)]] {
mul_mat_vec_q<QK_MXFP4, QI_MXFP4, block_mxfp4, VDR_MXFP4_Q8_1_MMVQ, vec_dot_mxfp4_q8_1>(
vx, vy, dst, ncols, nrows, item_ct1);
});
});
}
}
static void mul_mat_vec_q5_0_q8_1_sycl(const void *vx, const void *vy,
float *dst, const int ncols,
const int nrows,
@@ -1123,6 +1142,9 @@ void ggml_sycl_op_mul_mat_vec_q(ggml_backend_sycl_context & ctx, const ggml_tens
case GGML_TYPE_IQ4_XS:
mul_mat_vec_iq4_xs_q8_1_sycl(src0_dd_i, src1_ddq_i_bs, dst_dd_i_bs, ne00, row_diff, stream);
break;
case GGML_TYPE_MXFP4:
mul_mat_vec_mxfp4_q8_1_sycl(src0_dd_i, src1_ddq_i_bs, dst_dd_i_bs, ne00, row_diff, stream);
break;
default:
GGML_ABORT("fatal error");
}