This paper analyses total factor productivity (TFP) of China’s manufacturing sector and its decomposed indexes, i.e., technological progress and technological efficiency by employing Malmquist productivity index based on the Data Envelopment Analysis (DEA) approach. Using panel data of domestic manufacturing sector in China from 2000 to 2005, we estimate the influences of horizontal and vertical linkages on TFP and its decomposition indexes by controlling R&D and exports indexes of domestic manufacturing sector. The empirical results show that R&D and exports have a positive effect on TFP and that FDI inflows lead to positive spillovers significantly in general mainly through technological progress resulting from backward linkages; however, forward linkages have no technology spillover effect while horizontal linkages promote TFP through technological efficiency which has quite different influencing modes compared to that of backward and forward linkages. The grouping regression results also show that FDI technology spillovers have different conductive mechanisms under different technological levels, industry concentrations and export dependency indexes.