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/src/runner.cuh

@@ -0,0 +1,26 @@
+#pragma once
+#include <cublas_v2.h>
+#include <cuda_runtime.h>
+#include <fstream>
+#include <stdio.h>
+#include <stdlib.h>
+#include <sys/time.h>
+#include <time.h>
+#include <unistd.h>
+
+void cudaCheck(cudaError_t error, const char *file,
+               int line); // CUDA error check
+void CudaDeviceInfo();    // print CUDA information
+
+void range_init_matrix(float *mat, int N);
+void randomize_matrix(float *mat, int N);
+void zero_init_matrix(float *mat, int N);
+void copy_matrix(const float *src, float *dest, int N);
+void print_matrix(const float *A, int M, int N, std::ofstream &fs);
+bool verify_matrix(float *mat1, float *mat2, int N);
+
+float get_current_sec();                        // Get the current moment
+float cpu_elapsed_time(float &beg, float &end); // Calculate time difference
+
+void run_kernel(int kernel_num, int m, int n, int k, float alpha, float *A,
+                float *B, float beta, float *C, cublasHandle_t handle);