data: cat 3 SGEMM repos — siboehm, wangzyon, edtallison (full clone, no --depth)

Sources:
  siboehm/SGEMM_CUDA        → upstream_ref/sgemm_siboehm/     (25 files)
  wangzyon/NVIDIA_SGEMM_PRACTICE → upstream_ref/nvidia_sgemm_practice/ (23 files, filled gaps)
  edtallison/sgemm-cuda      → upstream_ref/sgemm_edtallison/  (41 files)

All files cat'd one by one from git clone (no --depth).
These are the 3 public SGEMM repos that can compile on CUDA 10.2 + CoreX ivcore10.

Key files for BI-V100 porting:
  kernel 10 (warp tiling) — already proven on device with WARPSIZE=64
  kernel 11/12 (double buffering) — next optimization target
  sgemm.cu + runner.cu — complete build+benchmark harness
  CMakeLists.txt — build system reference

upstream_ref/sgemm_edtallison/run_on_colab.ipynb

@@ -0,0 +1,71 @@
+{
+  "cells": [
+    {
+      "cell_type": "markdown",
+      "id": "b9326784",
+      "metadata": {},
+      "source": [
+        "This notebook installs dependencies, builds the project, and runs a selected kernel on Colab's GPU.\n",
+        "\n",
+        "**Note:** Ensure Colab runtime type is set to GPU (Runtime → Change runtime type → GPU)"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "id": "35ef02c6",
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "# install system dependencies\n",
+        "!apt-get update && apt-get install -y cmake ninja-build"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "id": "37a03120",
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "# clone\n",
+        "!git clone https://github.com/edtallison/sgemm-cuda.git\n",
+        "%cd sgemm-cuda"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "id": "cd1855de",
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "# build\n",
+        "!mkdir -p build && cd build && cmake -G Ninja .. && ninja"
+      ]
+    },
+    {
+      "cell_type": "code",
+      "execution_count": null,
+      "id": "093a0b6d",
+      "metadata": {},
+      "outputs": [],
+      "source": [
+        "# run for kernel 1\n",
+        "!cd build && ./sgemm 1"
+      ]
+    }
+  ],
+  "metadata": {
+    "kernelspec": {
+      "display_name": "Python 3",
+      "language": "python",
+      "name": "python3"
+    },
+    "language_info": {
+      "name": "python"
+    }
+  },
+  "nbformat": 4,
+  "nbformat_minor": 5
+}