vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

llama : minor fixes for up llama load model speed (#11448)

Browse source 
* impl::load change map bpe_ranks to onordered map for reduce time of impl::load on 30%

* llama_model_loader::init_mapping - replace new llama_mmap to std::make_unique<llama_mmap> for clean code & reduce (/2) time of running init_mappings

* Update src/llama-vocab.cpp

---------

Co-authored-by: lexasub <empty@empty.ru>
Co-authored-by: Diego Devesa <slarengh@gmail.com>
This commit is contained in:
lexasub 2025-01-27 17:42:09 +04:00 • committed by GitHub
parent df984e0147
commit a5203b4465
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
2 changed files with 8 additions and 3 deletions
Download patch file Download diff file Expand all files Collapse all files

2
src/llama-model-loader.cpp
View file

@ -819,7 +819,7 @@ void llama_model_loader::init_mappings(bool prefetch, llama_mlocks * mlock_mmaps
for (const auto & file : files) {
auto * reg = ggml_backend_dev_backend_reg(ggml_backend_dev_by_type(GGML_BACKEND_DEVICE_TYPE_CPU));
auto * is_numa_fn = (decltype(ggml_is_numa) *) ggml_backend_reg_get_proc_address(reg, "ggml_backend_cpu_is_numa");
std::unique_ptr<llama_mmap> mapping(new llama_mmap(file.get(), prefetch ? -1 : 0, is_numa_fn()));
std::unique_ptr<llama_mmap> mapping = std::make_unique<llama_mmap>(file.get(), prefetch ? -1 : 0, is_numa_fn());
mmaps_used.emplace_back(mapping->size(), 0);
if (mlock_mmaps) {
std::unique_ptr<llama_mlock> mlock_mmap(new llama_mlock());