From 3bfacb4130c46d663f8c8abcb278b8f57a518fa6 Mon Sep 17 00:00:00 2001
From: Dounia <dounia.khaldi@intel.com>
Date: Fri, 23 Oct 2026 09:01:36 -0700
Subject: [PATCH 1/2] [SYCL][E2E][Matrix] Use integer inputs for bfloat16
 accumulator tests when C or D is bfloat16

---
 .../Matrix/Inputs/joint_matrix_16bit_impl.hpp | 22 ++++++++++++++++---
 .../joint_matrix_bfloat16_accumulator.cpp     |  3 ---
 2 files changed, 19 insertions(+), 6 deletions(-)

diff --git a/sycl/test-e2e/Matrix/Inputs/joint_matrix_16bit_impl.hpp b/sycl/test-e2e/Matrix/Inputs/joint_matrix_16bit_impl.hpp
index 943e8b33ab207..4d8046b0d7226 100644
--- a/sycl/test-e2e/Matrix/Inputs/joint_matrix_16bit_impl.hpp
+++ b/sycl/test-e2e/Matrix/Inputs/joint_matrix_16bit_impl.hpp
@@ -100,8 +100,23 @@ void test() {
   TResult D[MATRIX_M][MATRIX_N];
   TResult DRef[MATRIX_M][MATRIX_N];
 
-  matrix_rand<Tab>(MATRIX_M, MATRIX_K, (Tab *)A, Tab(1));
-  matrix_rand<Tab>(MATRIX_K / VF, MATRIX_N * VF, (Tab *)B, Tab(1));
+  // A bfloat16 C or D is rounded after every DPAS while the reference rounds
+  // once, and with only 8 significant bits the difference can exceed the
+  // tolerance. Integer inputs keep every partial sum exactly representable, so
+  // the result does not depend on where the rounding happens.
+  constexpr bool bf16_acc =
+      std::is_same_v<TAcc, bfloat16> || std::is_same_v<TResult, bfloat16>;
+  if constexpr (!bf16_acc) {
+    matrix_rand<Tab>(MATRIX_M, MATRIX_K, (Tab *)A, Tab(1));
+    matrix_rand<Tab>(MATRIX_K / VF, MATRIX_N * VF, (Tab *)B, Tab(1));
+  } else {
+    std::mt19937 gen(0);
+    std::uniform_int_distribution<int> dist(-1, 1);
+    matrix_fill(MATRIX_M, MATRIX_K, (Tab *)A,
+                [&](int, int) { return Tab(dist(gen)); });
+    matrix_fill(MATRIX_K / VF, MATRIX_N * VF, (Tab *)B,
+                [&](int, int) { return Tab(dist(gen)); });
+  }
 
   matrix_fill(MATRIX_M, MATRIX_N, (TAcc *)C, TAcc(1));
   matrix_fill(MATRIX_M, MATRIX_N, (TResult *)D, TResult(1));
@@ -116,7 +131,8 @@ void test() {
                   B_layout, VF>(MD, MC, MA, MB);
   matrix_multiply_ref<Tab, Tab, TResult, VF>(
       (Tab *)A, (Tab *)B, (TResult *)DRef, MATRIX_M, MATRIX_N, MATRIX_K / VF);
-  assert(matrix_compare(MATRIX_M, MATRIX_N, (TResult *)D, (TResult *)DRef));
+  assert((matrix_compare<TResult, TResult, bf16_acc>(
+      MATRIX_M, MATRIX_N, (TResult *)D, (TResult *)DRef)));
 }
 
 template <typename TLow, typename THigh, size_t TM, size_t TN, size_t TK,
diff --git a/sycl/test-e2e/Matrix/joint_matrix_bfloat16_accumulator.cpp b/sycl/test-e2e/Matrix/joint_matrix_bfloat16_accumulator.cpp
index e1a9d1fbbc61f..18c14380b1682 100644
--- a/sycl/test-e2e/Matrix/joint_matrix_bfloat16_accumulator.cpp
+++ b/sycl/test-e2e/Matrix/joint_matrix_bfloat16_accumulator.cpp
@@ -20,9 +20,6 @@
 // XFAIL: windows && arch-intel_gpu_lnl_m && O0
 // XFAIL-TRACKER: CMPLRLLVM-72111
 
-// UNSUPPORTED: linux && (arch-intel_gpu_pvc || arch-intel_gpu_bmg_g21 || arch-intel_gpu_ptl_u || arch-intel_gpu_ptl_h)
-// UNSUPPORTED-TRACKER: https://github.com/intel/llvm/issues/21096
-
 // RUN: %{build} -o %t.out
 // RUN: %{run} %t.out
 // RUN: %if gpu %{ env IGC_JointMatrixLoadStoreOpt=2 %{run} %t.out %}

From a71ef057199ad685ff3ce1eed6dc13541a527242 Mon Sep 17 00:00:00 2001
From: Dounia <dounia.khaldi@intel.com>
Date: Thu, 29 Oct 2026 10:59:12 -0700
Subject: [PATCH 2/2] Retrigger CI

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>