Skip to content
Open
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 2 additions & 0 deletions src/ggml-cuda/ggml-cuda.cu
Original file line number Diff line number Diff line change
Expand Up @@ -1195,6 +1195,8 @@ static void ggml_cuda_op_mul_mat_cublas(
const bool use_fp16 = (src0->type == GGML_TYPE_F16 || ggml_is_quantized(src0->type)) && ggml_is_contiguous(src0) && row_diff == src0->ne[1] && dst->op_params[0] == GGML_PREC_DEFAULT;

if (src0->type == GGML_TYPE_BF16 && ggml_is_contiguous(src0) && row_diff == src0->ne[1]) {
// MIT licensed. Copyright (C) 2024 Iwan Kawrakow
// https://github.com/ikawrakow/ik_llama.cpp/pull/40
ggml_cuda_pool_alloc<nv_bfloat16> src1_as_bf16(ctx.pool(id));
if (src1->type != GGML_TYPE_BF16) {
const to_bf16_cuda_t to_bf16_cuda = ggml_get_to_bf16_cuda(src1->type);
Expand Down