From 17366df842e358768c0df7024484fffecfc7865b Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Johannes=20G=C3=A4=C3=9Fler?= Date: Tue, 6 Jun 2023 21:33:23 +0200 Subject: Multi GPU support, CUDA refactor, CUDA scratch buffer (#1703) * CUDA multi GPU + scratch ggml_cuda_compute_forward Tensor parallelism ggml_cuda_add ggml_cuda_rms_norm ggml_cuda_silu CUDA scratch buffer --main-gpu CLI option --- examples/common.h | 16 +++++++++------- 1 file changed, 9 insertions(+), 7 deletions(-) (limited to 'examples/common.h') diff --git a/examples/common.h b/examples/common.h index 66bdeb5e..12b49734 100644 --- a/examples/common.h +++ b/examples/common.h @@ -21,13 +21,15 @@ int32_t get_num_physical_cores(); struct gpt_params { - int32_t seed = -1; // RNG seed - int32_t n_threads = get_num_physical_cores(); - int32_t n_predict = -1; // new tokens to predict - int32_t n_ctx = 512; // context size - int32_t n_batch = 512; // batch size for prompt processing (must be >=32 to use BLAS) - int32_t n_keep = 0; // number of tokens to keep from initial prompt - int32_t n_gpu_layers = 0; // number of layers to store in VRAM + int32_t seed = -1; // RNG seed + int32_t n_threads = get_num_physical_cores(); + int32_t n_predict = -1; // new tokens to predict + int32_t n_ctx = 512; // context size + int32_t n_batch = 512; // batch size for prompt processing (must be >=32 to use BLAS) + int32_t n_keep = 0; // number of tokens to keep from initial prompt + int32_t n_gpu_layers = 0; // number of layers to store in VRAM + int32_t main_gpu = 0; // the GPU that is used for scratch and small tensors + float tensor_split[LLAMA_MAX_DEVICES] = {0}; // how split tensors should be distributed across GPUs // sampling parameters std::unordered_map logit_bias; // logit bias for specific tokens -- cgit v1.2.3