[SYCL][Matrix] Add initial get_coord API.

This patch adds initial API for retrieval of coordinates from a work item element.

Author: arnamoy.bhattacharyya

sycl/include/CL/__spirv/spirv_ops.hpp

@@ -123,6 +123,14 @@ template <typename T, std::size_t R, std::size_t C,
 extern SYCL_EXTERNAL size_t __spirv_JointMatrixWorkItemLengthINTEL(
     JOINT_MATRIX_INTEL(T, R, C, L, S, U) *);
 
+template <typename T, std::size_t R, std::size_t C,
+          __spv::MatrixUse U = __spv::MatrixUse::Unnecessary,
+          __spv::MatrixLayout L = __spv::MatrixLayout::RowMajor,
+          __spv::Scope::Flag S = __spv::Scope::Flag::Subgroup>
+extern SYCL_EXTERNAL std::tuple<T, T>
+__spirv_JointMatrixWorkItemElemCoord(JOINT_MATRIX_INTEL(T, R, C, L, S, U) *,
+                                     size_t i);
+
 template <typename T, std::size_t R, std::size_t C,
           __spv::MatrixUse U = __spv::MatrixUse::Unnecessary,
           __spv::MatrixLayout L = __spv::MatrixLayout::RowMajor,

sycl/include/sycl/ext/oneapi/matrix/matrix-jit-use.hpp

@@ -12,6 +12,7 @@
 #include <CL/__spirv/spirv_ops.hpp>
 #include <sycl/detail/defines_elementary.hpp>
 #include <sycl/feature_test.hpp>
+#include <tuple>
 
 namespace sycl {
 __SYCL_INLINE_VER_NAMESPACE(_V1) {
@@ -256,6 +257,18 @@ class wi_element {
   wi_element(joint_matrix<T, NumRows, NumCols, Use, Layout, Group> &Mat,
              std::size_t i)
       : M(Mat), idx(i) {}
+
+  // Functions
+  std::tuple<size_t, size_t> get_coord() {
+#ifdef __SYCL_DEVICE_ONLY__
+    return __spirv_JointMatrixWorkItemElemCoord(M.spvm, idx);
+#else
+    throw runtime_error("joint matrix is not supported on host device.",
+                        PI_ERROR_INVALID_DEVICE);
+#endif // __SYCL_DEVICE_ONLY__
+  }
+
+  // Various Operations
   operator T() {
 #ifdef __SYCL_DEVICE_ONLY__
     return __spirv_VectorExtractDynamic(M.spvm, idx);

sycl/test/matrix/matrix-bfloat16-test-coord.cpp

@@ -0,0 +1,205 @@
+// RUN: %clangxx -fsycl -fsycl-targets=spir64_gen -DSYCL_EXT_ONEAPI_MATRIX_VERSION=2 -S -emit-llvm %s -o %t.out
+#include <iostream>
+#include <sycl/sycl.hpp>
+
+using namespace sycl::ext::oneapi::experimental::matrix;
+using bfloat16 = sycl::ext::oneapi::experimental::bfloat16;
+
+static constexpr auto TILE_SZ = 16;
+static constexpr auto TM = TILE_SZ - 1;
+static constexpr auto TN = TILE_SZ - 1;
+static constexpr auto TK = 2 * TILE_SZ - 2;
+
+static constexpr auto SG_SZ = 16;
+
+template <typename T, size_t NUM_ROWS, size_t NUM_COLS> struct big_matrix {
+public:
+  T *mat;
+
+public:
+  T *get_data() { return mat; }
+  void set_data(T *data) { mat = data; }
+  big_matrix(T *data) : mat(data) {}
+};
+
+static constexpr size_t MATRIX_M = TM * 2;
+static constexpr size_t MATRIX_N = TN * 2;
+static constexpr size_t MATRIX_K = TK * 2;
+bfloat16 A[MATRIX_M][MATRIX_K];
+bfloat16 B[MATRIX_K / 2][MATRIX_N * 2];
+unsigned short Aref[MATRIX_M][MATRIX_K];
+unsigned short Bref[MATRIX_K / 2][MATRIX_N * 2];
+float C[MATRIX_M][MATRIX_N];
+float D[MATRIX_M][MATRIX_N];
+int32_t *res_local_row;
+
+template <typename T1, typename T2, size_t NUM_ROWS_A, size_t NUM_COLS_A,
+          size_t NUM_ROWS_B, size_t NUM_COLS_B, size_t NUM_ROWS_C,
+          size_t NUM_COLS_C>
+void matrix_multiply(big_matrix<T1, NUM_ROWS_C, NUM_COLS_C> &C,
+                     big_matrix<T2, NUM_ROWS_A, NUM_COLS_A> &A,
+                     big_matrix<T2, NUM_ROWS_B, NUM_COLS_B> &B) {
+  size_t M = NUM_ROWS_C;
+  size_t N = NUM_COLS_C;
+  size_t K = NUM_COLS_A;
+  // B => K/4 x N*4, A => M x K, C => M, N
+  // stride should be X's cols, e.g., B's stirde = N*4
+  assert(NUM_ROWS_C == NUM_ROWS_A && NUM_COLS_A == NUM_ROWS_B * 2);
+  size_t NDRangeM = M / TM;
+  size_t NDRangeN = N / TN;
+  sycl::buffer<bfloat16, 2> bufA(A.get_data(), sycl::range<2>(M, K));
+  sycl::buffer<bfloat16, 2> bufB(B.get_data(), sycl::range<2>(K, N));
+  sycl::buffer<float, 2> bufC((float *)C.get_data(), sycl::range<2>(M, N));
+
+  sycl::buffer<int32_t, 1> res_local_row_buf(res_local_row,
+                                             sycl::range<1>(MATRIX_M));
+
+  sycl::queue q;
+  q.submit([&](sycl::handler &cgh) {
+     auto accC = bufC.get_access<sycl::access::mode::read_write>(cgh);
+     auto accA = bufA.get_access<sycl::access::mode::read_write>(cgh);
+     auto accB = bufB.get_access<sycl::access::mode::read_write>(cgh);
+     auto res_local_row_acc =
+         res_local_row_buf.get_access<sycl::access::mode::read_write>(cgh);
+
+     cgh.parallel_for<class imatrix>(
+         sycl::nd_range<2>({NDRangeM, NDRangeN * SG_SZ}, {1, 1 * SG_SZ}),
+         [accA, accB, accC, M, N, K,
+          res_local_row_acc](sycl::nd_item<2> spmd_item)
+
+         {
+           // The submatrix API has to be accessed by all the workitems in a
+           // subgroup these functions will be called once by the subgroup no
+           // code divergence between the workitems
+           const auto global_idx = spmd_item.get_global_id(0);
+           const auto global_idy = spmd_item.get_global_id(1);
+           const auto sg_startx = global_idx - spmd_item.get_local_id(0);
+           const auto sg_starty = global_idy - spmd_item.get_local_id(1);
+
+           sycl::ext::oneapi::sub_group sg = spmd_item.get_sub_group();
+           joint_matrix<bfloat16, TM, TK, use::a> sub_a(sg);
+           // For B, since current implementation does not support non-packed
+           // layout, users need to specify the updated VNNI sizes along with
+           // the packed_b layout. By default, the layout is row_major and size
+           // is (TK, TN).
+           joint_matrix<bfloat16, TK, TN, use::b> sub_b(sg);
+           joint_matrix<float, TM, TN, use::accumulator> sub_c(sg);
+
+           joint_matrix_load(sg, sub_c,
+                             accC.get_pointer() + (sg_startx * TM) * N +
+                                 sg_starty / SG_SZ * TN,
+                             N, layout::row_major);
+           for (int k = 0; k < K / TK; k += 1) { //
+             joint_matrix_load(
+                 sg, sub_a, accA.get_pointer() + (sg_startx * TM) * K + k * TK,
+                 K, layout::row_major);
+             // Assuming B data is already in VNNI format.
+             joint_matrix_load(sg, sub_b,
+                               accB.get_pointer() + (k * TK / 2) * (N * 2) +
+                                   sg_starty / SG_SZ * TN * 2,
+                               N * 2, layout::packed_b);
+             sub_c = joint_matrix_mad(sg, sub_a, sub_b, sub_c);
+           }
+           joint_matrix_store(sg, sub_c,
+                              accC.get_pointer() + (sg_startx * TM) * N +
+                                  sg_starty / SG_SZ * TN,
+                              N, layout::row_major);
+           // Element wise operation
+           auto tCData = sub_c.get_wi_data();
+
+           for (int i = 0; i < tCData.length(); ++i) {
+             size_t row, col;
+             std::tie(row, col) = tCData[i].get_coord();
+             res_local_row_acc[row] += tCData[i];
+           }
+         }); // parallel for
+   }).wait();
+}
+
+float make_fp32(short x) {
+  unsigned int y = x;
+  y = y << 16;
+  float *res = reinterpret_cast<float *>(&y);
+  return *res;
+}
+
+unsigned short make_bf16(float x) {
+  int *res = reinterpret_cast<int *>(&x);
+  *res = *res >> 16;
+  return (unsigned short)*res;
+}
+
+void matrix_multiply_ref(int *A_mem, int *B_mem, int *C_mem, int M, int N,
+                         int K) {
+  // tiling
+  for (int m = 0; m < M; m++)
+    for (int n = 0; n < N; n++) {
+      for (int k = 0; k < K; k++) {
+        short *va = (short *)(A_mem + m * K + k);
+        short *vb = (short *)(B_mem + k * N + n);
+        float acc = *((float *)(C_mem + m * N + n));
+        // FIXME: Should we do reduce-add in another version?
+        for (int i = 0; i < 2; i++) {
+          acc += (make_fp32(va[i]) * make_fp32(vb[i]));
+        }
+        *((float *)(C_mem + m * N + n)) = acc;
+      }
+    }
+}
+
+int main() {
+  for (int i = 0; i < MATRIX_M; i++) {
+    for (int j = 0; j < MATRIX_K; j++) {
+      // Ee create bfloat16 from unsigned short since float-to-bfloat's
+      // conversion is not allowed.
+      A[i][j] = bfloat16::from_bits(make_bf16(1.0f * (i + j)));
+      Aref[i][j] = make_bf16(1.0f * (i + j));
+    }
+  }
+  for (int i = 0; i < MATRIX_K / 2; i++) {
+    for (int j = 0; j < MATRIX_N * 2; j++) {
+      B[i][j] = bfloat16::from_bits((make_bf16(2.0f * i + 3.0f * j)));
+      Bref[i][j] = make_bf16(2.0f * i + 3.0f * j);
+    }
+  }
+  for (int i = 0; i < MATRIX_M; i++) {
+    for (int j = 0; j < MATRIX_N; j++) {
+      C[i][j] = 1.0;
+      D[i][j] = 1.0;
+    }
+  }
+
+  big_matrix<float, MATRIX_M, MATRIX_N> MC((float *)&C);
+  big_matrix<float, MATRIX_M, MATRIX_N> MD((float *)&D);
+  big_matrix<bfloat16, MATRIX_M, MATRIX_K> MA((bfloat16 *)&A);
+  big_matrix<bfloat16, MATRIX_K / 2, MATRIX_N * 2> MB((bfloat16 *)&B);
+
+  res_local_row = (int32_t *)calloc(MATRIX_M, sizeof(int32_t));
+
+  matrix_multiply(MC, MA, MB);
+  matrix_multiply_ref((int32_t *)Aref, (int32_t *)Bref, (int32_t *)D, MATRIX_M,
+                      MATRIX_N, MATRIX_K / 2);
+
+  bool res = true;
+  for (int i = 0; i < MATRIX_M; i++) {
+    for (int j = 0; j < MATRIX_N; j++) {
+      if (C[i][j] != D[i][j])
+        res = false;
+    }
+  }
+  if (res)
+    std::cout << "passed\n";
+  else
+    std::cout << "failed\n";
+  for (int i = 0; i < MATRIX_M; i++) {
+    for (int j = 0; j < MATRIX_N; j++)
+      std::cout << C[i][j] << ", ";
+    std::cout << "\n";
+  }
+  std::cout << std::endl;
+  for (int i = 0; i < MATRIX_M; i++) {
+    for (int j = 0; j < MATRIX_N; j++)
+      std::cout << D[i][j] << ", ";
+    std::cout << "\n";
+  }
+}