From e36182cd8c5dd5c6d0aa1f77cf5cdca87e8bb1f0 Mon Sep 17 00:00:00 2001
From: Joseph Redmon <pjreddie@gmail.com>
Date: Fri, 21 Nov 2014 23:35:19 +0000
Subject: [PATCH] cleaned up data parsing a lot. probably nothing broken?

---
 src/mini_blas.h |   16 ++++++++++++----
 1 files changed, 12 insertions(+), 4 deletions(-)

diff --git a/src/mini_blas.h b/src/mini_blas.h
index a155c35..07b7cc6 100644
--- a/src/mini_blas.h
+++ b/src/mini_blas.h
@@ -11,7 +11,9 @@
 
 #ifdef GPU
 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 batch,
          int channels, int height, int width,
@@ -28,6 +30,12 @@
          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, 
         cl_mem B_gpu, int ldb,
@@ -55,8 +63,8 @@
                     float *B, int ldb,
                     float BETA,
                     float *C, int ldc);
-inline void axpy_cpu(int N, float ALPHA, float *X, int INCX, float *Y, int INCY);
-inline void copy_cpu(int N, float *X, int INCX, float *Y, int INCY);
-inline void scal_cpu(int N, float ALPHA, float *X, int INCX);
-inline float dot_cpu(int N, float *X, int INCX, float *Y, int INCY);
+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