From e8a309050803a1f1e1db41a0d34c4f08b4b01ba1 Mon Sep 17 00:00:00 2001 From: Cebtenzzre Date: Fri, 15 Sep 2023 15:07:40 -0400 Subject: [PATCH] fix a few missing 'static' specifiers --- .../convert-llama2c-to-ggml.cpp | 6 +-- examples/perplexity/perplexity.cpp | 5 +- examples/quantize-stats/quantize-stats.cpp | 2 +- examples/server/server.cpp | 52 ++++++++++--------- tests/test-opt.cpp | 9 ++-- 5 files changed, 36 insertions(+), 38 deletions(-) diff --git a/examples/convert-llama2c-to-ggml/convert-llama2c-to-ggml.cpp b/examples/convert-llama2c-to-ggml/convert-llama2c-to-ggml.cpp index ea0f41ee4..c291f0adf 100644 --- a/examples/convert-llama2c-to-ggml/convert-llama2c-to-ggml.cpp +++ b/examples/convert-llama2c-to-ggml/convert-llama2c-to-ggml.cpp @@ -444,7 +444,7 @@ __attribute__((format(gnu_printf, 1, 2))) __attribute__((format(printf, 1, 2))) #endif #endif -std::string format(const char * fmt, ...) { +static std::string format(const char * fmt, ...) { va_list ap, ap2; va_start(ap, fmt); va_copy(ap2, ap); @@ -531,7 +531,7 @@ struct llama_file { } }; -bool is_ggml_file(const char *filename) { +static bool is_ggml_file(const char * filename) { llama_file file(filename, "rb"); if (file.size < 4) { return false; @@ -540,7 +540,7 @@ bool is_ggml_file(const char *filename) { return magic == GGUF_MAGIC; } -std::string llama_escape_whitespaces(const std::string& text) { +static std::string llama_escape_whitespaces(const std::string & text) { std::ostringstream out; for (char c : text) { if (c == ' ') out << "\xe2\x96\x81"; diff --git a/examples/perplexity/perplexity.cpp b/examples/perplexity/perplexity.cpp index 85505e459..4620c43ac 100644 --- a/examples/perplexity/perplexity.cpp +++ b/examples/perplexity/perplexity.cpp @@ -132,7 +132,7 @@ static void process_logits( } -results_perplexity perplexity_v2(llama_context * ctx, const gpt_params & params) { +static results_perplexity perplexity_v2(llama_context * ctx, const gpt_params & params) { // Download: https://s3.amazonaws.com/research.metamind.io/wikitext/wikitext-2-raw-v1.zip?ref=salesforce-research // Run `./perplexity -m models/7B/ggml-model-q4_0.bin -f wiki.test.raw` // Output: `perplexity: 13.5106 [114/114]` @@ -262,8 +262,7 @@ results_perplexity perplexity_v2(llama_context * ctx, const gpt_params & params) return {tokens, std::exp(nll / count), logit_history, prob_history}; } -results_perplexity perplexity(llama_context * ctx, const gpt_params & params) { - +static results_perplexity perplexity(llama_context * ctx, const gpt_params & params) { if (params.ppl_stride > 0) { return perplexity_v2(ctx, params); } diff --git a/examples/quantize-stats/quantize-stats.cpp b/examples/quantize-stats/quantize-stats.cpp index b2887ff81..bfe70889a 100644 --- a/examples/quantize-stats/quantize-stats.cpp +++ b/examples/quantize-stats/quantize-stats.cpp @@ -102,7 +102,7 @@ static void combine_error_stats(error_stats & into, const error_stats & from) { for (size_t i=0; i & a, const std::vector & b) +static size_t common_part(const std::vector &a, const std::vector &b) { size_t i; for (i = 0; i < a.size() && i < b.size() && a[i] == b[i]; i++) @@ -63,13 +63,13 @@ enum stop_type STOP_PARTIAL, }; -static bool ends_with(const std::string & str, const std::string & suffix) +static bool ends_with(const std::string &str, const std::string &suffix) { return str.size() >= suffix.size() && 0 == str.compare(str.size() - suffix.size(), suffix.size(), suffix); } -size_t find_partial_stop_string(const std::string & stop, const std::string & text) +static size_t find_partial_stop_string(const std::string &stop, const std::string &text) { if (!text.empty() && !stop.empty()) { @@ -90,7 +90,7 @@ size_t find_partial_stop_string(const std::string & stop, const std::string & te } template -std::string tokens_to_str(llama_context *ctx, Iter begin, Iter end) +static std::string tokens_to_str(llama_context *ctx, Iter begin, Iter end) { std::string ret; for (; begin != end; ++begin) @@ -101,7 +101,7 @@ std::string tokens_to_str(llama_context *ctx, Iter begin, Iter end) } static void server_log( - const char * level, const char * function, int line, const char * message, const nlohmann::ordered_json & extra + const char *level, const char *function, int line, const char *message, const nlohmann::ordered_json &extra ) { nlohmann::ordered_json log{ {"timestamp", time(nullptr)}, @@ -122,7 +122,7 @@ static void server_log( } // format incomplete utf-8 multibyte character for output -std::string tokens_to_output_formatted_string(const llama_context * ctx, llama_token token) +static std::string tokens_to_output_formatted_string(const llama_context *ctx, llama_token token) { std::string out = token == -1 ? "" : llama_token_to_piece(ctx, token); // if the size is 1 and first bit is 1, meaning it's a partial character @@ -138,7 +138,7 @@ std::string tokens_to_output_formatted_string(const llama_context * ctx, llama_t } // convert a vector of completion_token_output to json -json probs_vector_to_json(const llama_context * ctx, const std::vector & probs) +static json probs_vector_to_json(const llama_context *ctx, const std::vector &probs) { json out = json::array(); for (const auto &prob : probs) @@ -690,7 +690,7 @@ struct llama_server_context } }; -static void server_print_usage(const char * argv0, const gpt_params & params, const server_params & sparams) +static void server_print_usage(const char *argv0, const gpt_params ¶ms, const server_params &sparams) { printf("usage: %s [options]\n", argv0); printf("\n"); @@ -738,7 +738,7 @@ static void server_print_usage(const char * argv0, const gpt_params & params, co printf("\n"); } -static void server_params_parse(int argc, char ** argv, server_params & sparams, gpt_params & params) +static void server_params_parse(int argc, char **argv, server_params &sparams, gpt_params ¶ms) { gpt_params default_params; server_params default_sparams; @@ -992,7 +992,7 @@ static void server_params_parse(int argc, char ** argv, server_params & sparams, } } -json format_generation_settings(llama_server_context & llama) +static json format_generation_settings(llama_server_context &llama) { const auto eos_bias = llama.params.logit_bias.find(llama_token_eos(llama.ctx)); const bool ignore_eos = eos_bias != llama.params.logit_bias.end() && @@ -1026,14 +1026,14 @@ json format_generation_settings(llama_server_context & llama) }; } -json format_embedding_response(llama_server_context & llama) +static json format_embedding_response(llama_server_context &llama) { return json{ {"embedding", llama.getEmbedding()}, }; } -json format_timings(llama_server_context & llama) +static json format_timings(llama_server_context &llama) { const auto timings = llama_get_timings(llama.ctx); @@ -1052,8 +1052,8 @@ json format_timings(llama_server_context & llama) }; } -json format_final_response( - llama_server_context & llama, const std::string & content, const std::vector & probs +static json format_final_response( + llama_server_context &llama, const std::string &content, const std::vector &probs ) { json res = json{ @@ -1081,8 +1081,8 @@ json format_final_response( return res; } -json format_partial_response( - llama_server_context & llama, const std::string & content, const std::vector & probs +static json format_partial_response( + llama_server_context &llama, const std::string &content, const std::vector &probs ) { json res = json{ {"content", content}, @@ -1097,20 +1097,20 @@ json format_partial_response( return res; } -json format_tokenizer_response(const std::vector & tokens) +static json format_tokenizer_response(const std::vector &tokens) { return json{ {"tokens", tokens}}; } -json format_detokenized_response(std::string content) +static json format_detokenized_response(std::string content) { return json{ {"content", content}}; } template -T json_value(const json & body, const std::string & key, const T & default_value) +static T json_value(const json &body, const std::string &key, const T &default_value) { // Fallback null to default value return body.contains(key) && !body.at(key).is_null() @@ -1118,7 +1118,7 @@ T json_value(const json & body, const std::string & key, const T & default_value : default_value; } -static void parse_options_completion(const json & body, llama_server_context & llama) +static void parse_options_completion(const json &body, llama_server_context &llama) { gpt_params default_params; @@ -1197,7 +1197,7 @@ static void parse_options_completion(const json & body, llama_server_context & l LOG_VERBOSE("completion parameters parsed", format_generation_settings(llama)); } -static void log_server_request(const Request & req, const Response & res) +static void log_server_request(const Request &req, const Response &res) { LOG_INFO("request", { {"remote_addr", req.remote_addr}, @@ -1214,7 +1214,7 @@ static void log_server_request(const Request & req, const Response & res) }); } -static bool is_at_eob(llama_server_context & server_context, const llama_token * tokens, const size_t n_tokens) { +static bool is_at_eob(llama_server_context &server_context, const llama_token *tokens, const size_t n_tokens) { return n_tokens && tokens[n_tokens-1] == llama_token_eos(server_context.ctx); } @@ -1224,7 +1224,7 @@ static bool is_at_eob(llama_server_context & server_context, const llama_token * // * When all beams converge to a common prefix, they are made available in beams_state.beams[0]. // This is also called when the stop condition is met. // Collect tokens into std::vector response which is pointed to by callback_data. -static void beam_search_callback(void * callback_data, llama_beams_state beams_state) { +static void beam_search_callback(void *callback_data, llama_beams_state beams_state) { auto & llama = *static_cast(callback_data); // Mark beams as EOS as needed. for (size_t i = 0 ; i < beams_state.n_beams ; ++i) { @@ -1257,7 +1257,8 @@ struct token_translator { std::string operator()(const completion_token_output & cto) const { return (*this)(cto.tok); } }; -static void append_to_generated_text_from_generated_token_probs(llama_server_context & llama) { +static void append_to_generated_text_from_generated_token_probs(llama_server_context &llama) +{ auto & gtps = llama.generated_token_probs; auto translator = token_translator{llama.ctx}; auto add_strlen = [=](size_t sum, const completion_token_output & cto) { return sum + translator(cto).size(); }; @@ -1270,7 +1271,8 @@ static void append_to_generated_text_from_generated_token_probs(llama_server_con } } -int main(int argc, char **argv) { +int main(int argc, char **argv) +{ // own arguments required by this example gpt_params params; server_params sparams; diff --git a/tests/test-opt.cpp b/tests/test-opt.cpp index 25f7a532e..ce4976858 100644 --- a/tests/test-opt.cpp +++ b/tests/test-opt.cpp @@ -61,12 +61,9 @@ static void get_random_dims_minmax(int64_t * dims, int ndims, int min, int max) } -struct ggml_tensor * get_random_tensor( - struct ggml_context * ctx0, - int ndims, - int64_t ne[], - float fmin, - float fmax) { +static struct ggml_tensor * get_random_tensor( + struct ggml_context * ctx0, int ndims, int64_t ne[], float fmin, float fmax +) { struct ggml_tensor * result = ggml_new_tensor(ctx0, GGML_TYPE_F32, ndims, ne); switch (ndims) {