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

build : fix most gcc and clang warnings (#2861)

Browse source 
* fix most gcc and clang warnings

* baby-llama : remove commented opt_params_adam

* fix some MinGW warnings

* fix more MinGW warnings
This commit is contained in:
Cebtenzzre 2023-09-01 09:34:50 -04:00 • committed by GitHub
parent d8d6977f48
commit ef15649972
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
9 changed files with 30 additions and 22 deletions
Download patch file Download diff file Expand all files Collapse all files

5
examples/baby-llama/baby-llama.cpp
View file

@ -1617,15 +1617,10 @@ int main(int argc, char ** argv) {
float error_before_opt = ggml_get_f32_1d(e, 0);
struct ggml_opt_params opt_params_adam = ggml_opt_default_params(GGML_OPT_ADAM);
struct ggml_opt_params opt_params_lbfgs = ggml_opt_default_params(GGML_OPT_LBFGS);
opt_params_adam.print_forward_graph = false;
opt_params_adam.print_backward_graph = false;
opt_params_lbfgs.print_forward_graph = false;
opt_params_lbfgs.print_backward_graph = false;
opt_params_adam.adam.n_iter = 16;
opt_params_lbfgs.lbfgs.n_iter = 16;
// ggml_opt(ctx0, opt_params_adam, e);
ggml_opt(ctx0, opt_params_lbfgs, e);
//
ggml_build_forward_expand(&gf, e);

8
examples/beam-search/beam-search.cpp
View file

@ -22,7 +22,9 @@
#include <unistd.h>
#elif defined (_WIN32)
#define WIN32_LEAN_AND_MEAN
#define NOMINMAX
#ifndef NOMINMAX
# define NOMINMAX
#endif
#include <windows.h>
#include <signal.h>
#endif
@ -73,7 +75,7 @@ void beam_search_callback(void * callback_data_ptr, llama_beams_state beams_stat
assert(0u < beams_state.n_beams);
const llama_token * tokens = beams_state.beam_views[0].tokens;
std::copy(tokens, tokens + n, callback_data.response.end() - n);
printf("%lu", n);
printf("%zu", n);
}
fflush(stdout);
#if 1 // DEBUG: print current beams for this iteration
@ -145,7 +147,7 @@ int main(int argc, char ** argv)
if (tokens_list.size() > max_tokens_list_size)
{
fprintf( stderr , "%s: error: prompt too long (%lu tokens, max %lu)\n" ,
fprintf( stderr , "%s: error: prompt too long (%zu tokens, max %zu)\n" ,
__func__ , tokens_list.size() , max_tokens_list_size );
return 1;
}

8
examples/server/server.cpp
View file

@ -17,6 +17,8 @@
#include "completion.js.hpp"
#include "json-schema-to-grammar.mjs.hpp"
#include <cstddef>
#ifndef SERVER_VERBOSE
#define SERVER_VERBOSE 1
#endif
@ -1038,7 +1040,7 @@ static json format_timings(llama_server_context &llama)
{
const auto timings = llama_get_timings(llama.ctx);
assert(timings.n_eval == llama.num_tokens_predicted);
assert(timings.n_eval == ptrdiff_t(llama.num_tokens_predicted));
return json{
{"prompt_n", timings.n_p_eval},
@ -1239,7 +1241,7 @@ void beam_search_callback(void * callback_data, llama_beams_state beams_state) {
const llama_token * tokens = beams_state.beam_views[0].tokens;
const auto map = [](llama_token tok) { return completion_token_output{{},tok}; };
std::transform(tokens, tokens + n, llama.generated_token_probs.end() - n, map);
printf("%lu", n);
printf("%zu", n);
}
fflush(stdout);
#if 0 // DEBUG: print current beams for this iteration
@ -1548,7 +1550,7 @@ int main(int argc, char **argv)
svr.set_exception_handler([](const Request &, Response &res, std::exception_ptr ep)
{
const auto * fmt = "500 Internal Server Error\n%s";
const char fmt[] = "500 Internal Server Error\n%s";
char buf[BUFSIZ];
try {
std::rethrow_exception(std::move(ep));