app : introduce the llama unified executable (#23296)

* app : introduce the llama unified executable

Signed-off-by: Adrien Gallouët <angt@huggingface.co>

* Use serve for server

Signed-off-by: Adrien Gallouët <angt@huggingface.co>

* Hide completion and bench, add help command

Signed-off-by: Adrien Gallouët <angt@huggingface.co>

* Remove STATIC

Signed-off-by: Adrien Gallouët <angt@huggingface.co>

* Use -impl targets instead of -lib

Signed-off-by: Adrien Gallouët <angt@huggingface.co>

* Revert "Remove STATIC"

This reverts commit cc44caccb9902b34a3531633edac911e5b3d65cd.

---------

Signed-off-by: Adrien Gallouët <angt@huggingface.co>

tools/llama-bench/CMakeLists.txt

@@ -1,6 +1,18 @@
+# llama-bench-impl: benchmark logic, reusable by app
+
+set(TARGET llama-bench-impl)
+
+add_library(${TARGET} STATIC llama-bench.cpp)
+
+target_include_directories(${TARGET} PUBLIC ${CMAKE_CURRENT_SOURCE_DIR})
+target_link_libraries(${TARGET} PUBLIC llama-common llama ${CMAKE_THREAD_LIBS_INIT})
+
+# llama-bench executable
+
 set(TARGET llama-bench)
-add_executable(${TARGET} llama-bench.cpp)
-target_link_libraries(${TARGET} PRIVATE llama-common llama ${CMAKE_THREAD_LIBS_INIT})
+
+add_executable(${TARGET} main.cpp)
+target_link_libraries(${TARGET} PRIVATE llama-bench-impl)
 target_compile_features(${TARGET} PRIVATE cxx_std_17)
 
 if(LLAMA_TOOLS_INSTALL)

tools/llama-bench/llama-bench.cpp

@@ -2136,7 +2136,10 @@ static std::unique_ptr<printer> create_printer(output_formats format) {
     GGML_ABORT("fatal error");
 }
 
-int main(int argc, char ** argv) {
+// satisfies -Wmissing-declarations
+int llama_bench(int argc, char ** argv);
+
+int llama_bench(int argc, char ** argv) {
     std::setlocale(LC_NUMERIC, "C");
     // try to set locale for unicode characters in markdown
     std::setlocale(LC_CTYPE, ".UTF-8");

tools/llama-bench/main.cpp

@@ -0,0 +1,5 @@
+int llama_bench(int argc, char ** argv);
+
+int main(int argc, char ** argv) {
+    return llama_bench(argc, argv);
+}