Using make to build llama.cpp:common/common.cpp:1108:55: error: expected primary-expression before ')' token 1108 | fprintf(stream, "build_number: %d\n", BUILD_NUMBER);

[OuXiao514]

D:/LLM/llama.cpp-master $ make
I llama.cpp build info:
I UNAME_S: Windows_NT
I UNAME_P: unknown
I UNAME_M: x86_64
I CFLAGS: -I. -Icommon -D_XOPEN_SOURCE=600 -DNDEBUG -DGGML_USE_K_QUANTS -std=c11 -fPIC -O3 -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wshadow -Wstrict-prototypes -Wpointer-arith -Wmissing-prototypes -Werror=implicit-int -Werror=implicit-function-declaration -Wdouble-promotion -march=native -mtune=native -Xassembler -muse-unaligned-vector-move
I CXXFLAGS: -I. -Icommon -D_XOPEN_SOURCE=600 -DNDEBUG -DGGML_USE_K_QUANTS -std=c++11 -fPIC -O3 -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wmissing-declarations -Wmissing-noreturn -Xassembler -muse-unaligned-vector-move -Wno-array-bounds -Wno-format-truncation -Wextra-semi -march=native -mtune=native
I NVCCFLAGS: -I. -Icommon -D_XOPEN_SOURCE=600 -DNDEBUG -DGGML_USE_K_QUANTS -std=c++11 -fPIC -O3 -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wmissing-declarations -Wmissing-noreturn -Xassembler -muse-unaligned-vector-move -Wno-pedantic -Xcompiler "-Wno-array-bounds -Wno-format-truncation -Wextra-semi -march=native -mtune=native "
I LDFLAGS:
I CC: cc (GCC) 13.2.0
I CXX: g++ (GCC) 13.2.0

g++ -I. -Icommon -D_XOPEN_SOURCE=600 -DNDEBUG -DGGML_USE_K_QUANTS -std=c++11 -fPIC -O3 -Wall -Wextra -Wpedantic -Wcast-qual -Wno-unused-function -Wmissing-declarations -Wmissing-noreturn -Xassembler -muse-unaligned-vector-move -Wno-array-bounds -Wno-format-truncation -Wextra-semi -march=native -mtune=native -c common/common.cpp -o common.o
common/common.cpp: In function 'void dump_non_result_info_yaml(FILE*, const gpt_params&, const llama_context*, const std::string&, const std::vector&, const char*)':
common/common.cpp:1108:55: error: expected primary-expression before ')' token
1108 | fprintf(stream, "build_number: %d\n", BUILD_NUMBER);
| ^
make: *** [Makefile:551: common.o] Error 1

[OuXiao514]

When I use cmake:
D:\LLM\llama.cpp-master\build>cmake --build . --config Release
适用于 .NET Framework MSBuild 版本 17.7.2+d6990bcfa

Checking Build System
Building Custom Rule D:/LLM/llama.cpp-master/CMakeLists.txt
ggml.c
ggml-alloc.c
ggml-backend.c
k_quants.c
正在生成代码...
ggml.vcxproj -> D:\LLM\llama.cpp-master\build\ggml.dir\Release\ggml.lib
Building Custom Rule D:/LLM/llama.cpp-master/CMakeLists.txt
llama.cpp
D:\LLM\llama.cpp-master\llama.cpp(1055,31): warning C4305: “初始化”: 从“double”到“float”截断 [D:\LLM\llama.cpp-master\build\llama.vcxproj]
D:\LLM\llama.cpp-master\llama.cpp(1205,40): warning C4566: 由通用字符名称“\u0120”表示的字符不能在当前代码页(936)中表示出来 [D:\LLM\llama.cpp-master\build\llama.vcxproj]
D:\LLM\llama.cpp-master\llama.cpp(1206,40): warning C4566: 由通用字符名称“\u010A”表示的字符不能在当前代码页(936)中表示出来 [D:\LLM\llama.cpp-master\build\llama.vcxproj]
D:\LLM\llama.cpp-master\llama.cpp(1207,40): warning C4566: 由通用字符名称“\u0120”表示的字符不能在当前代码页(936)中表示出来 [D:\LLM\llama.cpp-master\build\llama.vcxproj]
D:\LLM\llama.cpp-master\llama.cpp(1208,40): warning C4566: 由通用字符名称“\u010A”表示的字符不能在当前代码页(936)中表示出来 [D:\LLM\llama.cpp-master\build\llama.vcxproj]
D:\LLM\llama.cpp-master\llama.cpp(2238,60): warning C4566: 由通用字符名称“\u010A”表示的字符不能在当前代码页(936)中表示出来 [D:\LLM\llama.cpp-master\build\llama.vcxproj]
D:\LLM\llama.cpp-master\llama.cpp(9724,28): warning C4146: 一元负运算符应用于无符号类型，结果仍为无符号类型 [D:\LLM\llama.cpp-master\build\llama.vcxproj]
D:\LLM\llama.cpp-master\llama.cpp(9754,28): warning C4146: 一元负运算符应用于无符号类型，结果仍为无符号类型 [D:\LLM\llama.cpp-master\build\llama.vcxproj]
llama.vcxproj -> D:\LLM\llama.cpp-master\build\Release\llama.lib
Building Custom Rule D:/LLM/llama.cpp-master/common/CMakeLists.txt
common.cpp
D:\LLM\llama.cpp-master\common\common.cpp(1108,55): error C2059: 语法错误:“)” [D:\LLM\llama.cpp-master\build\common\common.vcxproj]
sampling.cpp
console.cpp
D:\LLM\llama.cpp-master\common\console.cpp(253,30): warning C4267: “初始化”: 从“size_t”转换到“DWORD”，可能丢失数据 [D:\LLM\llama.cpp-master\build\common\common.vcxproj]
D:\LLM\llama.cpp-master\common\console.cpp(407,28): warning C4267: “初始化”: 从“size_t”转换到“int”，可能丢失数据 [D:\LLM\llama.cpp-master\build\common\common.vcxproj]
grammar-parser.cpp
train.cpp
正在生成代码...

[shibe2]

There is probably a bug in generation of build-info.h when not building from Git tree. I have to manually edit it after running CMake.

[OuXiao514]

How to edit build-info.h?

[shibe2]

It's generated in the root of source tree. Open it with a text editor (like WordPad on Windows). Put some number after BUILD_NUMBER. It can be 0 or the number of commits in the branch you are building from. Current master has 1399 commits.

[OuXiao514]

I find!
#ifndef BUILD_INFO_H
#define BUILD_INFO_H

#define BUILD_NUMBER 1399
#define BUILD_COMMIT "004797f"
#define BUILD_COMPILER "MSVC 19.37.32822.0"
#define BUILD_TARGET "x64"

#endif // BUILD_INFO_H

[shibe2]

It looks correct. If you still have an error, it must be something else.

[staviq]

Possible duplicate of #3670

[shibe2]

@OuXiao514 Have you managed to build it? What was the problem?

[OuXiao514]

When I editor build-info.h, It is successful to build the project. But it has some warning. such as:
D:\LLM\llama.cpp-master.\llama.cpp(1055,31): warning C4305: “初始化”: 从“double”到“float”截断 [D:\LLM\llama.cpp-master\build\tests\test-llama-grammar.vcxproj]
D:\LLM\llama.cpp-master.\llama.cpp(1205,40): warning C4566: 由通用字符名称“\u0120”表示的字符不能在当前代码页(936)中表示出来 [D:\LLM\llama.cpp-master\build\tests\test-llama-grammar.vcxproj]
D:\LLM\llama.cpp-master.\llama.cpp(1206,40): warning C4566: 由通用字符名称“\u010A”表示的字符不能在当前代码页(936)中表示出来 [D:\LLM\llama.cpp-master\build\tests\test-llama-grammar.vcxproj]
D:\LLM\llama.cpp-master.\llama.cpp(1207,40): warning C4566: 由通用字符名称“\u0120”表示的字符不能在当前代码页(936)中表示出来 [D:\LLM\llama.cpp-master\build\tests\test-llama-grammar.vcxproj]

[shibe2]

Then we can indeed consider it a duplicate of #3670.