summaryrefslogtreecommitdiff
path: root/src/routines/level3/xgemm.cpp
diff options
context:
space:
mode:
Diffstat (limited to 'src/routines/level3/xgemm.cpp')
-rw-r--r--src/routines/level3/xgemm.cpp10
1 files changed, 2 insertions, 8 deletions
diff --git a/src/routines/level3/xgemm.cpp b/src/routines/level3/xgemm.cpp
index 9ea5559c..97e8db7e 100644
--- a/src/routines/level3/xgemm.cpp
+++ b/src/routines/level3/xgemm.cpp
@@ -118,12 +118,6 @@ StatusCode Xgemm<T>::DoGemm(const Layout layout,
const auto b_temp = (b_no_temp) ? b_buffer : Buffer<T>(context_, k_ceiled*n_ceiled);
const auto c_temp = (c_no_temp) ? c_buffer : Buffer<T>(context_, m_ceiled*n_ceiled);
- // Upload the scalar arguments as constant buffers to the device (needed for half-precision)
- auto alpha_buffer = Buffer<T>(context_, 1);
- auto beta_buffer = Buffer<T>(context_, 1);
- alpha_buffer.Write(queue_, 1, &alpha);
- beta_buffer.Write(queue_, 1, &beta);
-
// Events of all kernels (including pre/post processing kernels)
auto eventWaitList = std::vector<Event>();
auto emptyEventList = std::vector<Event>();
@@ -174,8 +168,8 @@ StatusCode Xgemm<T>::DoGemm(const Layout layout,
kernel.SetArgument(0, static_cast<int>(m_ceiled));
kernel.SetArgument(1, static_cast<int>(n_ceiled));
kernel.SetArgument(2, static_cast<int>(k_ceiled));
- kernel.SetArgument(3, alpha_buffer());
- kernel.SetArgument(4, beta_buffer());
+ kernel.SetArgument(3, GetRealArg(alpha));
+ kernel.SetArgument(4, GetRealArg(beta));
kernel.SetArgument(5, a_temp());
kernel.SetArgument(6, b_temp());
kernel.SetArgument(7, c_temp());