From d12bd512653ed0d83a3ed07277b20584ba14b310 Mon Sep 17 00:00:00 2001
From: "assistant-librarian[bot]"
 <assistant-librarian[bot]@users.noreply.github.com>
Date: Tue, 7 Oct 2025 19:11:30 +0000
Subject: [PATCH] Merge commit 'ae9f29b7d514b0829256a0a3ca9ab4511e7a1e04' into
 develop

---
 include/ck_tile/ops/gemm/kernel/universal_gemm_kernel.hpp | 1 +
 1 file changed, 1 insertion(+)

diff --git a/include/ck_tile/ops/gemm/kernel/universal_gemm_kernel.hpp b/include/ck_tile/ops/gemm/kernel/universal_gemm_kernel.hpp
index 51ad4e3dd1..e77355ed3d 100644
--- a/include/ck_tile/ops/gemm/kernel/universal_gemm_kernel.hpp
+++ b/include/ck_tile/ops/gemm/kernel/universal_gemm_kernel.hpp
@@ -1134,6 +1134,7 @@ struct UniversalGemmKernel
 
         while(block_id < num_work)
         {
+            s_waitcnt_barrier();
             // Get the tile index for this block
             const auto tile_idx = amd_wave_read_first_lane(block_id % num_tiles);
             const auto [iM, iN] = TilePartitioner{kargs.M, kargs.N}.GetOutputTileIndex(tile_idx);