build : fix build info generation and cleanup Makefile

.gitignore

@@ -86,15 +86,16 @@ poetry.lock
 poetry.toml
 
 # Test binaries
-tests/test-grammar-parser
-tests/test-llama-grammar
-tests/test-double-float
-tests/test-grad0
-tests/test-opt
-tests/test-quantize-fns
-tests/test-quantize-perf
-tests/test-sampling
-tests/test-tokenizer-0-llama
-tests/test-tokenizer-0-falcon
-tests/test-tokenizer-1-llama
-tests/test-tokenizer-1-bpe
+/tests/test-grammar-parser
+/tests/test-llama-grammar
+/tests/test-double-float
+/tests/test-grad0
+/tests/test-opt
+/tests/test-quantize-fns
+/tests/test-quantize-perf
+/tests/test-sampling
+/tests/test-tokenizer-0-llama
+/tests/test-tokenizer-0-falcon
+/tests/test-tokenizer-1-llama
+/tests/test-tokenizer-1-bpe
+/tests/test-rope

Makefile

@@ -8,7 +8,7 @@ BUILD_TARGETS = \
 TEST_TARGETS = \
 	tests/test-llama-grammar tests/test-grammar-parser tests/test-double-float tests/test-grad0 tests/test-opt \
 	tests/test-quantize-fns tests/test-quantize-perf tests/test-sampling tests/test-tokenizer-0-llama          \
-	tests/test-tokenizer-0-falcon tests/test-tokenizer-1-llama tests/test-tokenizer-1-bpe
+	tests/test-tokenizer-0-falcon tests/test-tokenizer-1-llama tests/test-tokenizer-1-bpe tests/test-rope
 
 # Code coverage output files
 COV_TARGETS = *.gcno tests/*.gcno *.gcda tests/*.gcda *.gcov tests/*.gcov lcov-report gcovr-report
@@ -629,7 +629,7 @@ beam-search: examples/beam-search/beam-search.cpp ggml.o llama.o $(COMMON_DEPS)
 finetune: examples/finetune/finetune.cpp ggml.o llama.o $(COMMON_DEPS) train.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-export-lora: examples/export-lora/export-lora.cpp ggml.o llama.o $(COMMON_DEPS) $(OBJS)
+export-lora: examples/export-lora/export-lora.cpp ggml.o common/common.h $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
 speculative: examples/speculative/speculative.cpp ggml.o llama.o $(COMMON_DEPS) grammar-parser.o $(OBJS)
@@ -679,28 +679,28 @@ vdot: pocs/vdot/vdot.cpp ggml.o $(OBJS)
 q8dot: pocs/vdot/q8dot.cpp ggml.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $^ -o $@ $(LDFLAGS)
 
-tests/test-llama-grammar: tests/test-llama-grammar.cpp ggml.o $(COMMON_DEPS) grammar-parser.o $(OBJS)
+tests/test-llama-grammar: tests/test-llama-grammar.cpp ggml.o grammar-parser.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-tests/test-grammar-parser: tests/test-grammar-parser.cpp ggml.o llama.o $(COMMON_DEPS) grammar-parser.o $(OBJS)
+tests/test-grammar-parser: tests/test-grammar-parser.cpp ggml.o llama.o grammar-parser.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-tests/test-double-float: tests/test-double-float.cpp ggml.o llama.o $(COMMON_DEPS) $(OBJS)
+tests/test-double-float: tests/test-double-float.cpp ggml.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-tests/test-grad0: tests/test-grad0.cpp ggml.o llama.o $(COMMON_DEPS) $(OBJS)
+tests/test-grad0: tests/test-grad0.cpp ggml.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-tests/test-opt: tests/test-opt.cpp ggml.o llama.o $(COMMON_DEPS) $(OBJS)
+tests/test-opt: tests/test-opt.cpp ggml.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-tests/test-quantize-fns: tests/test-quantize-fns.cpp ggml.o llama.o $(COMMON_DEPS) $(OBJS)
+tests/test-quantize-fns: tests/test-quantize-fns.cpp ggml.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-tests/test-quantize-perf: tests/test-quantize-perf.cpp ggml.o llama.o $(COMMON_DEPS) $(OBJS)
+tests/test-quantize-perf: tests/test-quantize-perf.cpp ggml.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
-tests/test-sampling: tests/test-sampling.cpp ggml.o llama.o $(COMMON_DEPS) $(OBJS)
+tests/test-sampling: tests/test-sampling.cpp ggml.o llama.o $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
 tests/test-tokenizer-0-falcon: tests/test-tokenizer-0-falcon.cpp ggml.o llama.o $(COMMON_DEPS) $(OBJS)
@@ -715,5 +715,8 @@ tests/test-tokenizer-1-bpe: tests/test-tokenizer-1-bpe.cpp ggml.o llama.o $(COMM
 tests/test-tokenizer-1-llama: tests/test-tokenizer-1-llama.cpp ggml.o llama.o $(COMMON_DEPS) $(OBJS)
 	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
 
+tests/test-rope: tests/test-rope.cpp ggml.o $(OBJS)
+	$(CXX) $(CXXFLAGS) $(filter-out %.h,$^) -o $@ $(LDFLAGS)
+
 tests/test-c.o: tests/test-c.c llama.h
 	$(CC) $(CFLAGS) -c $(filter-out %.h,$^) -o $@

common/CMakeLists.txt

@@ -11,7 +11,12 @@ if(EXISTS "${CMAKE_CURRENT_SOURCE_DIR}/../.git")
     if(NOT IS_DIRECTORY "${GIT_DIR}")
         file(READ ${GIT_DIR} REAL_GIT_DIR_LINK)
         string(REGEX REPLACE "gitdir: (.*)\n$" "\\1" REAL_GIT_DIR ${REAL_GIT_DIR_LINK})
-        set(GIT_DIR "${CMAKE_CURRENT_SOURCE_DIR}/../${REAL_GIT_DIR}")
+        string(FIND "${REAL_GIT_DIR}" "/" SLASH_POS)
+        if (SLASH_POS EQUAL 0)
+            set(GIT_DIR "${REAL_GIT_DIR}")
+        else()
+            set(GIT_DIR "${CMAKE_CURRENT_SOURCE_DIR}/../${REAL_GIT_DIR}")
+        endif()
     endif()
 
     set(GIT_INDEX "${GIT_DIR}/index")

common/common.cpp

@@ -1,20 +1,28 @@
 #include "common.h"
+#include "ggml.h"
 #include "llama.h"
+#include "log.h"
+#include "sampling.h"
 
 #include <algorithm>
-#include <cassert>
+#include <cctype>
+#include <chrono>
+#include <cinttypes>
 #include <cmath>
-#include <cstring>
+#include <cstdlib>
 #include <ctime>
+#include <exception>
 #include <fstream>
 #include <iterator>
-#include <iostream>
 #include <regex>
 #include <sstream>
+#include <stdexcept>
 #include <string>
+#include <thread>
+#include <unordered_map>
 #include <unordered_set>
+#include <utility>
 #include <vector>
-#include <cinttypes>
 
 #if defined(__APPLE__) && defined(__MACH__)
 #include <sys/types.h>
@@ -32,9 +40,7 @@
 #include <fcntl.h>
 #include <io.h>
 #else
-#include <sys/ioctl.h>
 #include <sys/stat.h>
-#include <unistd.h>
 #endif
 
 #if defined(_MSC_VER)

common/common.h

@@ -3,19 +3,18 @@
 #pragma once
 
 #include "llama.h"
-
 #include "sampling.h"
 
 #define LOG_NO_FILE_LINE_FUNCTION
 #include "log.h"
 
 #include <cmath>
-#include <string>
-#include <vector>
+#include <cstdint>
+#include <cstdio>
 #include <random>
-#include <thread>
-#include <unordered_map>
+#include <string>
 #include <tuple>
+#include <vector>
 
 #ifdef _WIN32
 #define DIRECTORY_SEPARATOR '\\'
@@ -68,7 +67,7 @@ struct gpt_params {
     int32_t yarn_orig_ctx                   = 0;    // YaRN original context length
     int8_t  rope_scaling_type               = LLAMA_ROPE_SCALING_UNSPECIFIED;
 
-    // // sampling parameters
+    // sampling parameters
     struct llama_sampling_params sparams;
 
     std::string model             = "models/7B/ggml-model-f16.gguf"; // model path

common/console.cpp

@@ -14,14 +14,13 @@
 #define ENABLE_VIRTUAL_TERMINAL_PROCESSING 0x0004
 #endif
 #else
-#include <climits>
+#include <clocale>
+#include <cstdio>
+#include <cwchar>
+
 #include <sys/ioctl.h>
-#include <unistd.h>
-#include <wchar.h>
-#include <stdio.h>
-#include <stdlib.h>
-#include <signal.h>
 #include <termios.h>
+#include <unistd.h>
 #endif
 
 #define ANSI_COLOR_RED     "\x1b[31m"

common/grammar-parser.cpp

@@ -1,6 +1,5 @@
 #include "grammar-parser.h"
 #include <cstdint>
-#include <cwchar>
 #include <string>
 #include <utility>
 #include <stdexcept>

common/grammar-parser.h

@@ -10,11 +10,14 @@
 // space ::= [ \t\n]*
 
 #pragma once
+
 #include "llama.h"
-#include <vector>
-#include <map>
+
 #include <cstdint>
+#include <cstdio>
+#include <map>
 #include <string>
+#include <vector>
 
 namespace grammar_parser {
     struct parse_state {

common/sampling.cpp

@@ -1,5 +1,11 @@
+#include "common.h"
 #include "sampling.h"
 
+#include <algorithm>
+#include <cstdio>
+#include <map>
+#include <utility>
+
 struct llama_sampling_context * llama_sampling_init(const struct llama_sampling_params & params) {
     struct llama_sampling_context * result = new llama_sampling_context();
 

common/sampling.h

@@ -1,12 +1,12 @@
 #pragma once
 
-#include "llama.h"
-
 #include "grammar-parser.h"
+#include "llama.h"
 
+#include <cstdint>
 #include <string>
-#include <vector>
 #include <unordered_map>
+#include <vector>
 
 // sampling parameters
 typedef struct llama_sampling_params {
@@ -56,8 +56,6 @@ struct llama_sampling_context {
     std::vector<llama_token_data> cur;
 };
 
-#include "common.h"
-
 // Create a new sampling context instance.
 struct llama_sampling_context * llama_sampling_init(const struct llama_sampling_params & params);
 

common/train.cpp

@@ -1,9 +1,19 @@
-#include "train.h"
 #include "common.h"
+#include "ggml.h"
+#include "llama.h"
+#include "train.h"
 
+#include <algorithm>
+#include <cerrno>
+#include <cmath>
+#include <cstdint>
+#include <cstdio>
+#include <cstdlib>
+#include <cstring>
+#include <functional>
+#include <locale>
 #include <random>
 #include <sstream>
-#include <functional>
 
 struct random_normal_distribution {
     std::mt19937 gen;

common/train.h

@@ -2,13 +2,14 @@
 
 #pragma once
 
-#include <string>
+#include "llama.h"
+
+#include <cstddef>
+#include <cstdint>
 #include <random>
+#include <string>
 #include <vector>
 
-#include "ggml.h"
-#include "llama.h"
-
 typedef std::string mt19937_state;
 
 struct train_state {

examples/baby-llama/baby-llama.cpp

@@ -1,11 +1,13 @@
 #include "ggml.h"
 #include "train.h"
 
-#include <vector>
+#include <algorithm>
 #include <cassert>
+#include <cmath>
+#include <cstdint>
+#include <cstdio>
 #include <cstdlib>
 #include <cstring>
-#include <random>
 #include <vector>
 
 #if defined(_MSC_VER)

examples/batched-bench/batched-bench.cpp

@@ -1,10 +1,11 @@
 #include "common.h"
+#include "ggml.h"
 #include "llama.h"
 
 #include <algorithm>
-#include <cmath>
+#include <cstdint>
 #include <cstdio>
-#include <string>
+#include <cstdlib>
 #include <vector>
 
 // mutates the input string

examples/batched/batched.cpp

@@ -1,9 +1,11 @@
 #include "common.h"
+#include "ggml.h"
 #include "llama.h"
 
 #include <algorithm>
-#include <cmath>
+#include <cstdint>
 #include <cstdio>
+#include <cstdlib>
 #include <string>
 #include <vector>
 

examples/beam-search/beam-search.cpp

@@ -1,29 +1,14 @@
 #include "common.h"
 #include "llama.h"
 
+#include <algorithm>
 #include <cassert>
-#include <cinttypes>
-#include <cmath>
 #include <cstdio>
-#include <cstring>
-#include <ctime>
-#include <fstream>
 #include <iostream>
 #include <string>
+#include <tuple>
 #include <vector>
 
-#if defined (__unix__) || (defined (__APPLE__) && defined (__MACH__))
-#include <signal.h>
-#include <unistd.h>
-#elif defined (_WIN32)
-#define WIN32_LEAN_AND_MEAN
-#ifndef NOMINMAX
-#   define NOMINMAX
-#endif
-#include <windows.h>
-#include <signal.h>
-#endif
-
 // Used for debugging to print out beam tokens.
 struct ostream_beam_view {
     llama_context * ctx;