In this paper, we describe a data race backend we developed for DIOTA.DIOTA (Dynamic Instrumentation, Optimisation and Transformation ofApplications) is our generic instrumentation tool, and this tool usesso-called backends to use the information gathered using theinstrumentation. Our novel data race backend uses innovative technologieslike multilevel bitmaps, snooped matrix clocks and segment merging in orderto limit the amount of memory used. The tool was implemented for Linuxsystems running on IA32 processors and is fully operational.