cuda实现矩阵相乘实例

CUDA矩阵相乘A*B=C代码，任意输入矩阵A的宽度和矩阵B的宽度（A[wB][wA]*B[wA][wB]），cudaMallocPitch开辟显存空间，cudaMemcpy2D数组复制，Kahan"s Summation Formula提高浮点计算精度。

Win32
Debug
ipch
template_vs2010-6eafb57
Debug
BuildLog.htm
mt.dep
template.cu.obj
template.device-link.options
template.exe.intermediate.manifest
template.pdb
template_gold.obj
vc90.idb
vc90.pdb
template_gold.cpp
template_kernel.cu
template_vs2005.sln
template_vs2005.vcproj
template_vs2008.ncb
template_vs2008.sln
template_vs2008.suo
template_vs2008.vcproj
template_vs2008.vcproj.WIN-LVJ4GKQVHKG.Administrator.user
template_vs2010.sdf
template_vs2010.sln
template_vs2010.suo
template_vs2010.vcxproj
template_vs2010.vcxproj.user
tmpxft_00001350_00000000-8_template.compute_10.cpp3.o
tmpxft_00001350_00000000-12_template.compute_20.cpp3.o
vc90.pdb
vc100.pdb
template.cu