doc: code gen task 1

Author: RivinHD

.github/workflows/benchmarks.yml

@@ -7,13 +7,12 @@ on:
   pull_request:
     branches: [ "main" ]
 
-env:
-  # Customize the CMake build type here (Release, Debug, RelWithDebInfo, etc.)
-  BUILD_TYPE: Release
-
 jobs:
   build:
     runs-on: ubuntu-24.04-arm
+    strategy:
+      matrix:
+        build_type: ["Release", "Debug"]
 
     steps:
     - uses: actions/checkout@v4
@@ -22,14 +21,23 @@ jobs:
       # Choose CMakeLists.text from a specific source directory with -S.
       # Configure CMake in a 'build' subdirectory. `CMAKE_BUILD_TYPE` is only required if you are using a single-configuration generator such as make.
       # See https://cmake.org/cmake/help/latest/variable/CMAKE_BUILD_TYPE.html?highlight=cmake_build_type
-      run: cmake -S ${{github.workspace}}/submissions/submission_25_05_08 -B ${{github.workspace}}/build -DCMAKE_BUILD_TYPE=${{env.BUILD_TYPE}}
+      run: |
+        cmake -S ${{github.workspace}}/submissions/submission_25_05_01 -B ${{github.workspace}}/build/submission_25_05_01 -DCMAKE_BUILD_TYPE=${{matrix.build_type}}
+        cmake -S ${{github.workspace}}/submissions/submission_25_05_08 -B ${{github.workspace}}/build/submission_25_05_08 -DCMAKE_BUILD_TYPE=${{matrix.build_type}}
+        cmake -S ${{github.workspace}} -B ${{github.workspace}}/build -DCMAKE_BUILD_TYPE=${{matrix.build_type}}
 
     - name: Build
       # Build your program with the given configuration
-      run: cmake --build ${{github.workspace}}/build --config ${{env.BUILD_TYPE}}
+      run: |
+        cmake --build ${{github.workspace}}/build/submission_25_05_01 --config ${{matrix.build_type}}
+        cmake --build ${{github.workspace}}/build/submission_25_05_08 --config ${{matrix.build_type}}
+        cmake --build ${{github.workspace}}/build --config ${{matrix.build_type}}
 
     - name: Test
       working-directory: ${{github.workspace}}/build
       # Execute tests defined by the CMake configuration.
-      run: ctest -C ${{env.BUILD_TYPE}}
+      run: |
+        ctest -C ${{matrix.build_type}} --test-dir submission_25_05_01
+        ctest -C ${{matrix.build_type}} --test-dir submission_25_05_08
+        ctest -C ${{matrix.build_type}}
 

.github/workflows/ci.yml

@@ -0,0 +1,28 @@
+.. _api_mini_jit:
+
+mini_jit
+========
+.. doxygennamespace:: mini_jit
+   :project: Machine Learning Compilers
+   :content-only:
+   :members:
+   :private-members:
+   :undoc-members:
+
+arm_instructions
+----------------
+.. doxygennamespace:: mini_jit::arm_instructions
+   :project: Machine Learning Compilers
+   :content-only:
+   :members:
+   :private-members:
+   :undoc-members:
+
+internal
+^^^^^^^^
+.. doxygennamespace:: mini_jit::arm_instructions::internal
+   :project: Machine Learning Compilers
+   :content-only:
+   :members:
+   :private-members:
+   :undoc-members:

docs_sphinx/api/mini_jit.rst

@@ -22,6 +22,7 @@ Machine Learning Compilers
    submissions/report_25_04_17.rst
    submissions/report_25_04_24.rst
    submissions/report_25_05_01.rst
+   submissions/report_25_05_08.rst
 
 .. toctree::
    :maxdepth: 4

docs_sphinx/api/mlc.rst

@@ -218,7 +218,7 @@ Loops
 - File ``submissions/submission_25_05_01/neon_3_1.s``
 
 .. code-block:: asm
-    :linenos:
+  :linenos:
 
     ...
     // Offset the used leading dimension by the size of floats
@@ -347,7 +347,7 @@ Loops
 - File ``submissions/submission_25_05_01/neon_3_2.s``
 
 .. code-block:: asm
-    :linenos:
+  :linenos:
 
       // Offset the used leading dimension by the size of floats
       lsl x3, x3, #2 // x3 * 4 = x3 * sizeof(float)
@@ -364,7 +364,7 @@ Loops
   matmul_loop_over_M:
       sub x16, x16, #1
 
-      ... <logic of loop over K - neon_3_1>
+      // ... <logic of loop over K - neon_3_1>
 
       // next M iteration on the matrix c and matrix a, both need offset about 16 values
       // also matrix b needs to start at the initial location again
@@ -383,12 +383,13 @@ Loops
       // Loop back to M
       cbnz x16, matmul_loop_over_M
 
+
 3. Loop over N: Implement a kernel that computes C+=AB for M=64, N=48 and K=64.
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 - File ``submissions/submission_25_05_01/neon_3_3.s``
 
 .. code-block:: asm
-    :linenos:
+  :linenos:
   
       // Offset the used leading dimension by the size of floats
       lsl x3, x3, #2 // x3 * 4 = x3 * sizeof(float)
@@ -409,7 +410,7 @@ Loops
   matmul_loop_over_N:
       sub x17, x17, #1
 
-    ... <logic of loop over M - neon_3_2>
+    // ... <logic of loop over M - neon_3_2>
 
       // next M iteration on the matrix b and matrix c, both need offset about 6*ldb/ldc values
       // also matrix a needs to start at the initial location again
@@ -430,7 +431,7 @@ Loops
       // Loop back to N
       cbnz x17, matmul_loop_over_N
 
-4. Test and optimize the kernels. Report your performance in GFLOPS.
+1. Test and optimize the kernels. Report your performance in GFLOPS.
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 - File ``submissions/submission_25_05_01/neon_3.h``
 - Tests ``submissions/submission_25_05_01/neon_3.test.cpp``