Skip to content

Minor fixes for up llama load model speed #11448

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
merged 3 commits into from
Jan 27, 2025
Merged

Minor fixes for up llama load model speed #11448

Show file tree
Hide file tree
Changes from 2 commits
Commits
Show all changes
3 commits
Select commit Hold shift + click to select a range
5144a18
impl::load change map bpe_ranks to onordered map for reduce time of i…
Jan 27, 2025
7c2b924
llama_model_loader::init_mapping - replace new llama_mmap to std::mak…
Jan 27, 2025
723fc66
Update src/llama-vocab.cpp
slaren Jan 27, 2025
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: 1 addition & 1 deletion src/llama-model-loader.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -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());
Expand Down
9 changes: 7 additions & 2 deletions src/llama-vocab.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1245,8 +1245,13 @@ struct llama_vocab::impl {

std::vector<llama_token> cache_special_tokens;
std::vector<std::string> cache_token_to_piece; // llama_token_to_piece(special = true);

std::map<std::pair<std::string, std::string>, int> bpe_ranks;
struct PairHash {
size_t operator()(const std::pair<std::string, std::string>& p) const {
return std::hash<std::string>{}(p.first) ^ //create some hash for pair
(std::hash<std::string>{}(p.second) << 1);
}
};
std::unordered_map<std::pair<std::string, std::string>, int, PairHash> bpe_ranks;
Copy link
Member

@ggerganov ggerganov Jan 27, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What if there is a hash collision of 2 string pairs?

Copy link
Contributor Author

@lexasub lexasub Jan 27, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

What if there is a hash collision of 2 string pairs?

https://en.cppreference.com/w/cpp/container/unordered_map#:~:text=Internally%2C%20the%20elements,for%20both%20keys.

Copy link
Contributor Author

@lexasub lexasub Jan 27, 2025 •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

We can use a custom comparator for reliability, but std::unordered_map uses the same comparator (std::pair is comparable) as std::map.

Copy link
Contributor Author

@lexasub lexasub Jan 27, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

comparator in unordered_map use only equals, but in map used equals, <, > (tree)

Copy link
Member

@ggerganov ggerganov Jan 27, 2025

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yes, I see. I was a bit confused how this works. It should be OK.


// set of all tokens that cause "end of generation"
std::set<llama_token> special_eog_ids;
Expand Down
Loading