From 08b757a0bf76efe8c76b453063a1bb19315bcaa6 Mon Sep 17 00:00:00 2001
From: Joseph Redmon <pjreddie@gmail.com>
Date: Wed, 14 Jan 2015 20:18:57 +0000
Subject: [PATCH] Stable, needs to be way faster
---
src/mini_blas.h | 49 +++++++++++++++++++++++++++++++++++--------------
1 files changed, 35 insertions(+), 14 deletions(-)
diff --git a/src/mini_blas.h b/src/mini_blas.h
index c80e6ad..ea73ed6 100644
--- a/src/mini_blas.h
+++ b/src/mini_blas.h
@@ -10,13 +10,31 @@
void time_random_matrix(int TA, int TB, int m, int k, int n);
#ifdef GPU
-void im2col_ongpu(cl_mem data_im, const int batch,
- const int channels, const int height, const int width,
- const int ksize, const int stride, cl_mem data_col);
+void axpy_ongpu(int N, float ALPHA, cl_mem X, int INCX, cl_mem Y, int INCY);
+void axpy_ongpu_offset(int N, float ALPHA, cl_mem X, int OFFX, int INCX, cl_mem Y, int OFFY, int INCY);
+void copy_ongpu(int N, cl_mem X, int INCX, cl_mem Y, int INCY);
+void copy_ongpu_offset(int N, cl_mem X, int OFFX, int INCX, cl_mem Y, int OFFY, int INCY);
+void scal_ongpu(int N, float ALPHA, cl_mem X, int INCX);
+void im2col_ongpu(cl_mem data_im, int offset,
+ int channels, int height, int width,
+ int ksize, int stride, int pad, cl_mem data_col);
+
+void col2im_gpu(float *data_col, int offset,
+ 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);
void im2col_gpu(float *data_im,
- const int batch, const int channels, const int height, const int width,
- const int ksize, const int stride, float *data_col);
+ int channels, int height, int width,
+ int ksize, int stride, int pad, float *data_col);
+
+void gemm_ongpu_offset(int TA, int TB, int M, int N, int K, float ALPHA,
+ cl_mem A_gpu, int a_off, int lda,
+ cl_mem B_gpu, int b_off, int ldb,
+ float BETA,
+ cl_mem C_gpu, int c_off, int ldc);
void gemm_ongpu(int TA, int TB, int M, int N, int K, float ALPHA,
cl_mem A_gpu, int lda,
@@ -25,13 +43,14 @@
cl_mem C_gpu, int ldc);
#endif
-void im2col_cpu(float* data_im, const int batch,
- const int channels, const int height, const int width,
- const int ksize, const int stride, int pad, float* data_col);
+void im2col_cpu(float* data_im,
+ int channels, int height, int width,
+ int ksize, int stride, int pad, float* data_col);
void col2im_cpu(float* data_col,
- const int channels, const int height, const int width,
- const int ksize, const int stride, int pad, float* data_im);
+ int channels, int height, int width,
+ int ksize, int stride, int pad, float* data_im);
+
void test_blas();
void gemm_gpu(int TA, int TB, int M, int N, int K, float ALPHA,
@@ -40,10 +59,12 @@
float BETA,
float *C, int ldc);
void gemm_cpu(int TA, int TB, int M, int N, int K, float ALPHA,
- float *A, int lda,
- float *B, int ldb,
- float BETA,
- float *C, int ldc);
+ float *A, int lda,
+ float *B, int ldb,
+ 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