From 44234214fcd2e806cdc4531292a1fd52423f1406 Mon Sep 17 00:00:00 2001
From: Ash Vardanian <1983160+ashvardanian@users.noreply.github.com>
Date: Mon, 10 Feb 2025 23:17:47 +0000
Subject: [PATCH] Fix: Invoke `f16f32` in WGMMA

---
 less_slow.cu | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/less_slow.cu b/less_slow.cu
index bfa845c..6073fb7 100644
--- a/less_slow.cu
+++ b/less_slow.cu
@@ -603,7 +603,7 @@ __global__ void tops_f16f32_sm90wgmma_64x256x16_loop128_cuda_kernel() {
     std::uint64_t b_descriptor = wgmma_descriptor((std::uint64_t)b_shared, 128 * 256 / 8, 128, 0, 0);
     wgmma_fence();
     for (int i = 0; i != 128; ++i) {
-        wgmma_bf16f32_64x256x16(c_registers, a_descriptor, b_descriptor);
+        wgmma_f16f32_64x256x16(c_registers, a_descriptor, b_descriptor);
         wgmma_commit_group();
     }
     wgmma_sync_group();