From 2302dbed5763a7b98a204fbe5e3c9d2202b0feae Mon Sep 17 00:00:00 2001
From: Alexey <AlexeyAB@users.noreply.github.com>
Date: Mon, 26 Mar 2018 22:55:33 +0000
Subject: [PATCH] Merge pull request #505 from judwhite/patch-2
---
src/gemm.c | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/src/gemm.c b/src/gemm.c
index a4db8a4..c3154ec 100644
--- a/src/gemm.c
+++ b/src/gemm.c
@@ -177,6 +177,7 @@
float *C_gpu, int ldc)
{
cublasHandle_t handle = blas_handle();
+ cudaError_t stream_status = cublasSetStream(handle, get_cuda_stream());
cudaError_t status = cublasSgemm(handle, (TB ? CUBLAS_OP_T : CUBLAS_OP_N),
(TA ? CUBLAS_OP_T : CUBLAS_OP_N), N, M, K, &ALPHA, B_gpu, ldb, A_gpu, lda, &BETA, C_gpu, ldc);
check_error(status);
--
Gitblit v1.10.0