From 14303717dcddae43cdc55beb0685dae86f566fd8 Mon Sep 17 00:00:00 2001
From: Joseph Redmon <pjreddie@gmail.com>
Date: Sat, 25 Oct 2014 18:57:26 +0000
Subject: [PATCH] Fast, needs to be faster

---
 src/mini_blas.h |    8 ++++++++
 1 files changed, 8 insertions(+), 0 deletions(-)

diff --git a/src/mini_blas.h b/src/mini_blas.h
index 34905a1..923afc7 100644
--- a/src/mini_blas.h
+++ b/src/mini_blas.h
@@ -10,10 +10,16 @@
 void time_random_matrix(int TA, int TB, int m, int k, int n);
 
 #ifdef GPU
+void axpy_ongpu(int N, float ALPHA, cl_mem X, int INCX, cl_mem Y, int INCY);
+void copy_ongpu(int N, cl_mem X, int INCX, cl_mem Y, int INCY);
+void scal_ongpu(int N, float ALPHA, cl_mem X, int INCX);
 void im2col_ongpu(cl_mem data_im, int batch,
          int channels, int height, int width,
          int ksize, int stride, int pad, cl_mem data_col);
 
+void col2im_gpu(float *data_col,  int batch,
+         int channels,  int height,  int width,
+         int ksize,  int stride,  int pad, float *data_im);
 void col2im_ongpu(cl_mem data_col, int batch,
         int channels, int height, int width,
         int ksize, int stride, int pad, cl_mem data_im);
@@ -50,5 +56,7 @@
                     float BETA,
                     float *C, int ldc);
 void axpy_cpu(int N, float ALPHA, float *X, int INCX, float *Y, int INCY);
+void copy_cpu(int N, float *X, int INCX, float *Y, int INCY);
 void scal_cpu(int N, float ALPHA, float *X, int INCX);
+float dot_cpu(int N, float *X, int INCX, float *Y, int INCY);
 void test_gpu_blas();

--
Gitblit v1.10.0