vrr/koboldcpp
2
0
Fork 0
mirror of https://github.com/LostRuins/koboldcpp.git synced 2025-09-12 09:59:41 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 3

reduce some spamminess

Browse source
This commit is contained in:
Concedo 2025-02-08 22:49:48 +08:00
parent 816d9b7989
commit a83f2d5fce
2 changed files with 5 additions and 4 deletions
Download patch file Download diff file Expand all files Collapse all files

2
src/llama-kv-cache.cpp
View file

@ -75,7 +75,7 @@ bool llama_kv_cache_init(
const uint32_t n_embd_k_gqa = hparams.n_embd_k_gqa(i) + hparams.n_embd_k_s(); const uint32_t n_embd_k_gqa = hparams.n_embd_k_gqa(i) + hparams.n_embd_k_s();
const uint32_t n_embd_v_gqa = hparams.n_embd_v_gqa(i) + hparams.n_embd_v_s(); const uint32_t n_embd_v_gqa = hparams.n_embd_v_gqa(i) + hparams.n_embd_v_s();
LLAMA_LOG_DEBUG("%s: layer %d: n_embd_k_gqa = %d, n_embd_v_gqa = %d\n", __func__, i, n_embd_k_gqa, n_embd_v_gqa); // LLAMA_LOG_DEBUG("%s: layer %d: n_embd_k_gqa = %d, n_embd_v_gqa = %d\n", __func__, i, n_embd_k_gqa, n_embd_v_gqa);
ggml_backend_buffer_type_t buft; ggml_backend_buffer_type_t buft;
if (offload) { if (offload) {

7
src/llama-model.cpp
View file

@ -1280,7 +1280,8 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
bool use_mmap_buffer = true; bool use_mmap_buffer = true;
LLAMA_LOG_INFO("%s: loading model tensors, this can take a while... (mmap = %s)\n", __func__, use_mmap_buffer ? "true" : "false"); LLAMA_LOG_INFO("%s: loading model tensors, this can take a while...", __func__);
// LLAMA_LOG_INFO("%s: loading model tensors, this can take a while... (mmap = %s)\n", __func__, use_mmap_buffer ? "true" : "false");
// build a list of buffer types for the CPU and GPU devices // build a list of buffer types for the CPU and GPU devices
pimpl->cpu_buft_list = make_cpu_buft_list(devices); pimpl->cpu_buft_list = make_cpu_buft_list(devices);
@ -1328,12 +1329,12 @@ bool llama_model::load_tensors(llama_model_loader & ml) {
const int act_gpu_layers = devices.empty() ? 0 : std::min(n_gpu_layers, (int)n_layer + 1); const int act_gpu_layers = devices.empty() ? 0 : std::min(n_gpu_layers, (int)n_layer + 1);
auto get_layer_buft_list = [&](int il) -> llama_model::impl::layer_dev { auto get_layer_buft_list = [&](int il) -> llama_model::impl::layer_dev {
if (il < i_gpu_start || (il - i_gpu_start) >= act_gpu_layers) { if (il < i_gpu_start || (il - i_gpu_start) >= act_gpu_layers) {
LLAMA_LOG_DEBUG("load_tensors: layer %3d assigned to device %s\n", il, ggml_backend_dev_name(cpu_dev)); // LLAMA_LOG_DEBUG("load_tensors: layer %3d assigned to device %s\n", il, ggml_backend_dev_name(cpu_dev));
return {cpu_dev, &pimpl->cpu_buft_list}; return {cpu_dev, &pimpl->cpu_buft_list};
} }
const int layer_gpu = std::upper_bound(splits.begin(), splits.begin() + n_devices(), float(il - i_gpu_start)/act_gpu_layers) - splits.begin(); const int layer_gpu = std::upper_bound(splits.begin(), splits.begin() + n_devices(), float(il - i_gpu_start)/act_gpu_layers) - splits.begin();
auto * dev = devices.at(layer_gpu); auto * dev = devices.at(layer_gpu);
LLAMA_LOG_DEBUG("load_tensors: layer %3d assigned to device %s\n", il, ggml_backend_dev_name(dev)); // LLAMA_LOG_DEBUG("load_tensors: layer %3d assigned to device %s\n", il, ggml_backend_dev_name(dev));
return {dev, &pimpl->gpu_buft_list.at(dev)}; return {dev, &pimpl->gpu_buft_list.at(dev)};
}; };