Rust port

.devops/nix/package.nix

@@ -207,7 +207,7 @@ effectiveStdenv.mkDerivation (finalAttrs: {
   # if they haven't been added yet.
   postInstall = ''
     mkdir -p $out/include
-    cp $src/include/llama.h $out/include/
+    cp $src/include/llama.h.inc $out/include/
   '';
 
   meta = {

.editorconfig

@@ -41,7 +41,7 @@ charset = unset
 trim_trailing_whitespace = unset
 insert_final_newline = unset
 
-[vendor/miniaudio/miniaudio.h]
+[vendor/miniaudio/miniaudio.h.inc]
 trim_trailing_whitespace = unset
 insert_final_newline = unset
 

.github/labeler.yml

@@ -2,35 +2,35 @@
 Apple Metal:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-metal.h
+            - ggml/include/ggml-metal.h.inc
             - ggml/src/ggml-metal/**
             - README-metal.md
 SYCL:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-sycl.h
+            - ggml/include/ggml-sycl.h.inc
             - ggml/src/ggml-sycl/**
             - docs/backend/SYCL.md
             - examples/sycl/**
 Nvidia GPU:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-cuda.h
+            - ggml/include/ggml-cuda.h.inc
             - ggml/src/ggml-cuda/**
 Vulkan:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-vulkan.h
+            - ggml/include/ggml-vulkan.h.inc
             - ggml/src/ggml-vulkan/**
 IBM zDNN:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-zdnn.h
+            - ggml/include/ggml-zdnn.h.inc
             - ggml/src/ggml-zdnn/**
 AMD ZenDNN:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-zendnn.h
+            - ggml/include/ggml-zendnn.h.inc
             - ggml/src/ggml-zendnn/**
 documentation:
     - changed-files:
@@ -109,28 +109,28 @@ jinja parser:
 Ascend NPU:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-cann.h
+            - ggml/include/ggml-cann.h.inc
             - ggml/src/ggml-cann/**
             - docs/backend/CANN.md
 OpenCL:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-opencl.h
+            - ggml/include/ggml-opencl.h.inc
             - ggml/src/ggml-opencl/**
             - docs/backend/OPENCL.md
 Hexagon:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-hexagon.h
+            - ggml/include/ggml-hexagon.h.inc
             - ggml/src/ggml-hexagon/**
 WebGPU:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-webgpu.h
+            - ggml/include/ggml-webgpu.h.inc
             - ggml/src/ggml-webgpu/**
 OpenVINO:
     - changed-files:
         - any-glob-to-any-file:
-            - ggml/include/ggml-openvino.h
+            - ggml/include/ggml-openvino.h.inc
             - ggml/src/ggml-openvino/**
             - docs/backend/OPENVINO.md

.github/workflows/build-and-test-snapdragon.yml

@@ -7,7 +7,7 @@ on:
       - master
     paths:
       - '.github/workflows/build-and-test-snapdragon.yml'
-      - 'ggml/include/ggml-hexagon.h'
+      - 'ggml/include/ggml-hexagon.h.inc'
       - 'ggml/src/ggml-hexagon/**'
       - 'docs/backend/snapdragon/**'
       - 'scripts/snapdragon/**'
@@ -17,7 +17,7 @@ on:
     types: [opened, synchronize, reopened]
     paths:
       - '.github/workflows/build-and-test-snapdragon.yml'
-      - 'ggml/include/ggml-hexagon.h'
+      - 'ggml/include/ggml-hexagon.h.inc'
       - 'ggml/src/ggml-hexagon/**'
       - 'docs/backend/snapdragon/**'
       - 'scripts/snapdragon/**'

.gitignore

@@ -157,3 +157,4 @@ a.out.*
 
 AGENTS.local.md
 .pi/SYSTEM.md
+target/

CMakeLists.txt

@@ -240,8 +240,7 @@ set(LLAMA_LIB_INSTALL_DIR     ${CMAKE_INSTALL_LIBDIR}     CACHE PATH "Location o
 set(LLAMA_BIN_INSTALL_DIR     ${CMAKE_INSTALL_BINDIR}     CACHE PATH "Location of binary  files")
 
 set(LLAMA_PUBLIC_HEADERS
-    ${CMAKE_CURRENT_SOURCE_DIR}/include/llama.h
-    ${CMAKE_CURRENT_SOURCE_DIR}/include/llama-cpp.h)
+    ${CMAKE_CURRENT_SOURCE_DIR}/include/llama.h.inc)
 
 set_target_properties(llama
     PROPERTIES

CODEOWNERS

@@ -55,11 +55,11 @@
 /ggml/include/                          @ggerganov
 /ggml/src/ggml-backend-meta.cpp         @JohannesGaessler
 /ggml/src/ggml-cann/                    @ggml-org/ggml-cann
-/ggml/src/ggml-common.h                 @ggerganov
+/ggml/src/ggml-common.h.inc                 @ggerganov
 /ggml/src/ggml-cpu/                     @ggerganov
 /ggml/src/ggml-cpu/spacemit/            @alex-spacemit
 /ggml/src/ggml-cuda/                    @ggml-org/ggml-cuda
-/ggml/src/ggml-cuda/vendors/hip.h       @IMbackK
+/ggml/src/ggml-cuda/vendors/hip.h.inc       @IMbackK
 /ggml/src/ggml-cuda/fattn-wmma*         @IMbackK
 /ggml/src/ggml-hexagon/                 @ggml-org/ggml-hexagon
 /ggml/src/ggml-hip/                     @IMbackK

CONTRIBUTING.md

@@ -93,7 +93,7 @@ Maintainers reserve the right to decline review or close pull requests for any r
 - Try to follow the existing patterns in the code (indentation, spaces, etc.). In case of doubt use `clang-format` (from clang-tools v15+) to format the added code
 - For anything not covered in the current guidelines, refer to the [C++ Core Guidelines](https://isocpp.github.io/CppCoreGuidelines/CppCoreGuidelines)
 - Tensors store data in row-major order. We refer to dimension 0 as columns, 1 as rows, 2 as matrices
-- Matrix multiplication is unconventional: [`C = ggml_mul_mat(ctx, A, B)`](https://github.com/ggml-org/llama.cpp/blob/880e352277fc017df4d5794f0c21c44e1eae2b84/ggml.h#L1058-L1064) means $C^T = A B^T \Leftrightarrow C = B A^T.$
+- Matrix multiplication is unconventional: [`C = ggml_mul_mat(ctx, A, B)`](https://github.com/ggml-org/llama.cpp/blob/880e352277fc017df4d5794f0c21c44e1eae2b84/ggml.h.inc#L1058-L1064) means $C^T = A B^T \Leftrightarrow C = B A^T.$
 
 ![matmul](media/matmul.png)