llama : Metal inference (#1642)

* mtl : export the LLaMA computation graph

* ci : disable temporary

* mtl : adapt the MNIST example as starter

* mtl : no need for mtl-export tool, add cli arg for main instead

* mtl : export just a small part of the graph for now to make it easier

* mtl : move MSL code into separate file for easy editing

* mtl : initial get_rows_q4_0 kernel

* mtl : confirmed get_rows_q4_0 is working correctly

* mtl : add rms_norm kernel + confirm working

* mtl : add mul kernel + confirm working

* mtl : initial mul_mat Q4 kernel (wrong results)

* mtl : mul_mat fixes (still wrong)

* mtl : another mul_mat Q4 (still does not work)

* mtl : working mul_mat q4

* ggml : fix handling of "view" ops in ggml_graph_import()

* mtl : add rope kernel

* mtl : add reshape and transpose handling

* ggml : store offset as opt arg for ggml_view_xd() operators

* mtl : add cpy kernel + handle view ops

* mtl : confirm f16 x f32 attention mul mat

* mtl : add scale kernel

* mtl : add diag_mask_inf kernel

* mtl : fix soft_max kernel

* ggml : update ggml_nbytes() to handle non-contiguous tensors

* mtl : verify V tensor contents

* mtl : add f32 -> f32 cpy kernel

* mtl : add silu kernel

* mtl : add non-broadcast mul kernel

* mtl : full GPU inference of the computation graph

* mtl : optimize rms_norm and soft_max kernels

* mtl : add f16 mat x f32 vec multiplication kernel

* mtl : fix bug in f16 x f32 mul mat + speed-up computation

* mtl : faster mul_mat_q4_0_f32 kernel

* mtl : fix kernel signature + roll inner loop

* mtl : more threads for rms_norm + better timing

* mtl : remove printfs from inner loop

* mtl : simplify implementation

* mtl : add save/load vocab to ggml file

* mtl : plug Metal inference into llama.cpp (very quick-n-dirty)

* mtl : make it work with main example

Lots of hacks but at least now it generates text

* mtl : preparing for merge

* mtl : clean-up ggml mtl interface + suport scratch / inplace

* mtl : remove temp / debug code

* metal : final refactoring and simplification

* Revert "ci : disable temporary"

This reverts commit 98c267f.

* metal : add comments

* metal : clean-up stuff, fix typos

* readme : add Metal instructions

* readme : add example for main

Author: ggerganov

.gitignore

@@ -17,6 +17,7 @@ build-release/
 build-static/
 build-cublas/
 build-opencl/
+build-metal/
 build-no-accel/
 build-sanitize-addr/
 build-sanitize-thread/

CMakeLists.txt

@@ -64,13 +64,14 @@ if (NOT MSVC)
 endif()
 
 # 3rd party libs
-option(LLAMA_ACCELERATE                 "llama: enable Accelerate framework"                    ON)
-option(LLAMA_BLAS                       "llama: use BLAS"                                       OFF)
+option(LLAMA_ACCELERATE                      "llama: enable Accelerate framework"               ON)
+option(LLAMA_BLAS                            "llama: use BLAS"                                  OFF)
 set(LLAMA_BLAS_VENDOR "Generic" CACHE STRING "llama: BLAS library vendor")
-option(LLAMA_CUBLAS                     "llama: use cuBLAS"                                     OFF)
-set(LLAMA_CUDA_DMMV_X "32" CACHE STRING "llama: x stride for dmmv CUDA kernels")
-set(LLAMA_CUDA_DMMV_Y "1" CACHE STRING  "llama: y block size for dmmv CUDA kernels")
-option(LLAMA_CLBLAST                    "llama: use CLBlast"                                    OFF)
+option(LLAMA_CUBLAS                          "llama: use cuBLAS"                                OFF)
+set(LLAMA_CUDA_DMMV_X      "32" CACHE STRING "llama: x stride for dmmv CUDA kernels")
+set(LLAMA_CUDA_DMMV_Y       "1" CACHE STRING "llama: y block size for dmmv CUDA kernels")
+option(LLAMA_CLBLAST                         "llama: use CLBlast"                               OFF)
+option(LLAMA_METAL                           "llama: use Metal"                                 OFF)
 
 option(LLAMA_BUILD_TESTS                "llama: build tests"    ${LLAMA_STANDALONE})
 option(LLAMA_BUILD_EXAMPLES             "llama: build examples" ${LLAMA_STANDALONE})
@@ -183,7 +184,7 @@ if (LLAMA_CUBLAS)
 
         enable_language(CUDA)
 
-        set(GGML_CUDA_SOURCES ggml-cuda.cu ggml-cuda.h)
+        set(GGML_SOURCES_CUDA ggml-cuda.cu ggml-cuda.h)
 
         add_compile_definitions(GGML_USE_CUBLAS)
         add_compile_definitions(GGML_CUDA_DMMV_X=${LLAMA_CUDA_DMMV_X})
@@ -200,12 +201,37 @@ if (LLAMA_CUBLAS)
     endif()
 endif()
 
+if (LLAMA_METAL)
+    find_library(FOUNDATION_LIBRARY         Foundation              REQUIRED)
+    find_library(METAL_FRAMEWORK            Metal                   REQUIRED)
+    find_library(METALKIT_FRAMEWORK         MetalKit                REQUIRED)
+    find_library(METALPERFORMANCE_FRAMEWORK MetalPerformanceShaders REQUIRED)
+
+    set(GGML_SOURCES_METAL ggml-metal.m ggml-metal.h)
+
+    add_compile_definitions(GGML_USE_METAL)
+    add_compile_definitions(GGML_METAL_NDEBUG)
+
+    # get full path to the file
+    #add_compile_definitions(GGML_METAL_DIR_KERNELS="${CMAKE_CURRENT_SOURCE_DIR}/")
+
+    # copy ggml-metal.metal to bin directory
+    configure_file(ggml-metal.metal bin/ggml-metal.metal COPYONLY)
+
+    set(LLAMA_EXTRA_LIBS ${LLAMA_EXTRA_LIBS}
+        ${FOUNDATION_LIBRARY}
+        ${METAL_FRAMEWORK}
+        ${METALKIT_FRAMEWORK}
+        ${METALPERFORMANCE_FRAMEWORK}
+        )
+endif()
+
 if (LLAMA_CLBLAST)
     find_package(CLBlast)
     if (CLBlast_FOUND)
         message(STATUS "CLBlast found")
 
-        set(GGML_OPENCL_SOURCES ggml-opencl.cpp ggml-opencl.h)
+        set(GGML_SOURCES_OPENCL ggml-opencl.cpp ggml-opencl.h)
 
         add_compile_definitions(GGML_USE_CLBLAST)
 
@@ -370,8 +396,10 @@ endif()
 add_library(ggml OBJECT
             ggml.c
             ggml.h
-            ${GGML_CUDA_SOURCES}
-            ${GGML_OPENCL_SOURCES})
+            ${GGML_SOURCES_CUDA}
+            ${GGML_SOURCES_OPENCL}
+            ${GGML_SOURCES_METAL}
+            )
 
 target_include_directories(ggml PUBLIC .)
 target_compile_features(ggml PUBLIC c_std_11) # don't bump
@@ -384,21 +412,25 @@ endif()
 add_library(llama
             llama.cpp
             llama.h
-            llama-util.h)
+            llama-util.h
+            )
 
 target_include_directories(llama PUBLIC .)
 target_compile_features(llama PUBLIC cxx_std_11) # don't bump
-target_link_libraries(llama PRIVATE ggml ${LLAMA_EXTRA_LIBS})
+target_link_libraries(llama PRIVATE
+    ggml
+    ${LLAMA_EXTRA_LIBS}
+    )
 
 if (BUILD_SHARED_LIBS)
     set_target_properties(llama PROPERTIES POSITION_INDEPENDENT_CODE ON)
     target_compile_definitions(llama PRIVATE LLAMA_SHARED LLAMA_BUILD)
 endif()
 
-if (GGML_CUDA_SOURCES)
+if (GGML_SOURCES_CUDA)
     message(STATUS "GGML CUDA sources found, configuring CUDA architecture")
-    set_property(TARGET ggml PROPERTY CUDA_ARCHITECTURES OFF)
-    set_property(TARGET ggml PROPERTY CUDA_SELECT_NVCC_ARCH_FLAGS "Auto")
+    set_property(TARGET ggml  PROPERTY CUDA_ARCHITECTURES OFF)
+    set_property(TARGET ggml  PROPERTY CUDA_SELECT_NVCC_ARCH_FLAGS "Auto")
     set_property(TARGET llama PROPERTY CUDA_ARCHITECTURES OFF)
 endif()
 

Makefile

@@ -105,6 +105,7 @@ ifeq ($(UNAME_M),$(filter $(UNAME_M),x86_64 i686))
 	#CFLAGS   += -mfma -mf16c -mavx
 	#CXXFLAGS += -mfma -mf16c -mavx
 endif
+
 ifneq ($(filter ppc64%,$(UNAME_M)),)
 	POWER9_M := $(shell grep "POWER9" /proc/cpuinfo)
 	ifneq (,$(findstring POWER9,$(POWER9_M)))
@@ -116,26 +117,30 @@ ifneq ($(filter ppc64%,$(UNAME_M)),)
 		CXXFLAGS += -std=c++23 -DGGML_BIG_ENDIAN
 	endif
 endif
+
 ifndef LLAMA_NO_ACCELERATE
 	# Mac M1 - include Accelerate framework.
 	# `-framework Accelerate` works on Mac Intel as well, with negliable performance boost (as of the predict time).
 	ifeq ($(UNAME_S),Darwin)
 		CFLAGS  += -DGGML_USE_ACCELERATE
 		LDFLAGS += -framework Accelerate
 	endif
-endif
+endif # LLAMA_NO_ACCELERATE
+
 ifdef LLAMA_OPENBLAS
 	CFLAGS  += -DGGML_USE_OPENBLAS -I/usr/local/include/openblas -I/usr/include/openblas
 	ifneq ($(shell grep -e "Arch Linux" -e "ID_LIKE=arch" /etc/os-release 2>/dev/null),)
 		LDFLAGS += -lopenblas -lcblas
 	else
 		LDFLAGS += -lopenblas
 	endif
-endif
+endif # LLAMA_OPENBLAS
+
 ifdef LLAMA_BLIS
 	CFLAGS += -DGGML_USE_OPENBLAS -I/usr/local/include/blis -I/usr/include/blis
 	LDFLAGS += -lblis -L/usr/local/lib
-endif
+endif # LLAMA_BLIS
+
 ifdef LLAMA_CUBLAS
 	CFLAGS    += -DGGML_USE_CUBLAS -I/usr/local/cuda/include -I/opt/cuda/include -I$(CUDA_PATH)/targets/x86_64-linux/include
 	CXXFLAGS  += -DGGML_USE_CUBLAS -I/usr/local/cuda/include -I/opt/cuda/include -I$(CUDA_PATH)/targets/x86_64-linux/include
@@ -156,33 +161,49 @@ endif # LLAMA_CUDA_DMMV_Y
 ggml-cuda.o: ggml-cuda.cu ggml-cuda.h
 	$(NVCC) $(NVCCFLAGS) $(CXXFLAGS) -Wno-pedantic -c $< -o $@
 endif # LLAMA_CUBLAS
+
 ifdef LLAMA_CLBLAST
-	CFLAGS  += -DGGML_USE_CLBLAST
-	CXXFLAGS  += -DGGML_USE_CLBLAST
+	CFLAGS   += -DGGML_USE_CLBLAST
+	CXXFLAGS += -DGGML_USE_CLBLAST
 	# Mac provides OpenCL as a framework
 	ifeq ($(UNAME_S),Darwin)
 		LDFLAGS += -lclblast -framework OpenCL
 	else
 		LDFLAGS += -lclblast -lOpenCL
 	endif
 	OBJS    += ggml-opencl.o
+
 ggml-opencl.o: ggml-opencl.cpp ggml-opencl.h
 	$(CXX) $(CXXFLAGS) -c $< -o $@
-endif
+endif # LLAMA_CLBLAST
+
+ifdef LLAMA_METAL
+	CFLAGS   += -DGGML_USE_METAL -DGGML_METAL_NDEBUG
+	CXXFLAGS += -DGGML_USE_METAL
+	LDFLAGS  += -framework Foundation -framework Metal -framework MetalKit -framework MetalPerformanceShaders
+	OBJS     += ggml-metal.o
+
+ggml-metal.o: ggml-metal.m ggml-metal.h
+	$(CC) $(CFLAGS) -c $< -o $@
+endif # LLAMA_METAL
+
 ifneq ($(filter aarch64%,$(UNAME_M)),)
 	# Apple M1, M2, etc.
 	# Raspberry Pi 3, 4, Zero 2 (64-bit)
 	CFLAGS   += -mcpu=native
 	CXXFLAGS += -mcpu=native
 endif
+
 ifneq ($(filter armv6%,$(UNAME_M)),)
 	# Raspberry Pi 1, Zero
 	CFLAGS += -mfpu=neon-fp-armv8 -mfp16-format=ieee -mno-unaligned-access
 endif
+
 ifneq ($(filter armv7%,$(UNAME_M)),)
 	# Raspberry Pi 2
 	CFLAGS += -mfpu=neon-fp-armv8 -mfp16-format=ieee -mno-unaligned-access -funsafe-math-optimizations
 endif
+
 ifneq ($(filter armv8%,$(UNAME_M)),)
 	# Raspberry Pi 3, 4, Zero 2 (32-bit)
 	CFLAGS += -mfp16-format=ieee -mno-unaligned-access

README.md

@@ -51,11 +51,10 @@ Inference of [LLaMA](https://arxiv.org/abs/2302.13971) model in pure C/C++
 The main goal of `llama.cpp` is to run the LLaMA model using 4-bit integer quantization on a MacBook
 
 - Plain C/C++ implementation without dependencies
-- Apple silicon first-class citizen - optimized via ARM NEON and Accelerate framework
+- Apple silicon first-class citizen - optimized via ARM NEON, Accelerate and Metal frameworks
 - AVX, AVX2 and AVX512 support for x86 architectures
 - Mixed F16 / F32 precision
 - 4-bit, 5-bit and 8-bit integer quantization support
-- Runs on the CPU
 - Supports OpenBLAS/Apple BLAS/ARM Performance Lib/ATLAS/BLIS/Intel MKL/NVHPC/ACML/SCSL/SGIMATH and [more](https://cmake.org/cmake/help/latest/module/FindBLAS.html#blas-lapack-vendors) in BLAS
 - cuBLAS and CLBlast support
 
@@ -236,6 +235,32 @@ In order to build llama.cpp you have three different options.
     zig build -Drelease-fast
     ```
 
+### Metal Build
+
+Using Metal allows the computation to be executed on the GPU for Apple devices:
+
+- Using `make`:
+
+  ```bash
+  LLAMA_METAL=1 make
+  ```
+
+- Using `CMake`:
+
+    ```bash
+    mkdir build-metal
+    cd build-metal
+    cmake -DLLAMA_METAL=ON ..
+    cmake --build . --config Release
+    ```
+
+When built with Metal support, you can enable GPU inference with the `--gpu-layers|-ngl` command-line argument.
+Any value larger than 0 will offload the computation to the GPU. For example:
+
+```bash
+./main -m ./models/7B/ggml-model-q4_0.bin -n 128 -ngl 1
+```
+
 ### BLAS Build
 
 Building the program with BLAS support may lead to some performance improvements in prompt processing using batch sizes higher than 32 (the default is 512). BLAS doesn't affect the normal generation performance. There are currently three different implementations of it:
@@ -369,7 +394,7 @@ Building the program with BLAS support may lead to some performance improvements
 
   Running:
 
-  The CLBlast build supports `--gpu-layers|-ngl` like  the CUDA version does.
+  The CLBlast build supports `--gpu-layers|-ngl` like the CUDA version does.
 
   To select the correct platform (driver) and device (GPU), you can use the environment variables `GGML_OPENCL_PLATFORM` and `GGML_OPENCL_DEVICE`.
   The selection can be a number (starting from 0) or a text string to search:

examples/CMakeLists.txt

@@ -37,7 +37,10 @@ else()
     add_subdirectory(save-load-state)
     add_subdirectory(benchmark)
     add_subdirectory(baby-llama)
-    if(LLAMA_BUILD_SERVER)
+    if (LLAMA_METAL)
+        add_subdirectory(metal)
+    endif()
+    if (LLAMA_BUILD_SERVER)
         add_subdirectory(server)
     endif()
 endif()

examples/common.cpp

@@ -299,6 +299,8 @@ bool gpt_params_parse(int argc, char ** argv, gpt_params & params) {
             params.use_mmap = false;
         } else if (arg == "--mtest") {
             params.mem_test = true;
+        } else if (arg == "--export") {
+            params.export_cgraph = true;
         } else if (arg == "--verbose-prompt") {
             params.verbose_prompt = true;
         } else if (arg == "-r" || arg == "--reverse-prompt") {
@@ -438,6 +440,7 @@ void gpt_print_usage(int /*argc*/, char ** argv, const gpt_params & params) {
     fprintf(stderr, "                        number of layers to store in VRAM\n");
 #endif
     fprintf(stderr, "  --mtest               compute maximum memory usage\n");
+    fprintf(stderr, "  --export              export the computation graph to 'llama.ggml'\n");
     fprintf(stderr, "  --verbose-prompt      print prompt before generation\n");
     fprintf(stderr, "  --lora FNAME          apply LoRA adapter (implies --no-mmap)\n");
     fprintf(stderr, "  --lora-base FNAME     optional model to use as a base for the layers modified by the LoRA adapter\n");

examples/common.h

@@ -71,6 +71,7 @@ struct gpt_params {
     bool use_mmap          = true;  // use mmap for faster loads
     bool use_mlock         = false; // use mlock to keep model in memory
     bool mem_test          = false; // compute maximum memory usage
+    bool export_cgraph     = false; // export the computation graph
     bool verbose_prompt    = false; // print prompt tokens before generation
 };
 

examples/main/main.cpp

@@ -134,6 +134,13 @@ int main(int argc, char ** argv) {
         return 0;
     }
 
+    // export the cgraph and exit
+    if (params.export_cgraph) {
+        llama_eval_export(ctx, "llama.ggml");
+        llama_free(ctx);
+
+        return 0;
+    }
 
     std::string path_session = params.path_prompt_cache;
     std::vector<llama_token> session_tokens;

examples/metal/CMakeLists.txt

@@ -0,0 +1,3 @@
+set(TEST_TARGET metal)
+add_executable(${TEST_TARGET} metal.cpp)
+target_link_libraries(${TEST_TARGET} PRIVATE ggml)