ICODE-ADC/lammps
4
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

git-svn-id: svn://svn.icms.temple.edu/lammps-ro/trunk@6053 f3b2605a-c512-4ea7-a41b-209d697bcdaa

Browse Source
This commit is contained in:
sjplimp
2011-05-02 15:02:52 +00:00
parent 2be078632d
commit 5f799182b3
70 changed files with 4489 additions and 2253 deletions
Download Patch File Download Diff File Expand all files Collapse all files

19
lib/gpu/geryon/ucl_nv_kernel.h
View File

@ -13,7 +13,7 @@
copyright : (C) 2010 by W. Michael Brown
email : brownw@ornl.gov
***************************************************************************/
/* -----------------------------------------------------------------------
Copyright (2010) Sandia Corporation. Under the terms of Contract
DE-AC04-94AL85000 with Sandia Corporation, the U.S. Government retains
@ -25,8 +25,18 @@
#ifndef UCL_NV_KERNEL_H
#define UCL_NV_KERNEL_H
#define GLOBAL_ID_X threadIdx.x+__mul24(blockIdx.x,blockDim.x)
#define GLOBAL_ID_Y threadIdx.y+__mul24(blockIdx.y,blockDim.y)
#if (__CUDA_ARCH__ < 200)
#define mul24 __mul24
#define MEM_THREADS 16
#else
#define mul24(X,Y) (X)*(Y)
#define MEM_THREADS 32
#endif
#define GLOBAL_ID_X threadIdx.x+mul24(blockIdx.x,blockDim.x)
#define GLOBAL_ID_Y threadIdx.y+mul24(blockIdx.y,blockDim.y)
#define GLOBAL_SIZE_X mul24(gridDim.x,blockDim.x);
#define GLOBAL_SIZE_Y mul24(gridDim.y,blockDim.y);
#define THREAD_ID_X threadIdx.x
#define THREAD_ID_Y threadIdx.y
#define BLOCK_ID_X blockIdx.x
@ -35,8 +45,9 @@
#define BLOCK_SIZE_Y blockDim.y
#define __kernel extern "C" __global__
#define __local __shared__
#define mul24 __mul24
#define __global
#define __inline static __inline__ __device__
#define atom_add atomicAdd
#endif