The impact of soft errors in logic and its commercialisation in ARM IP

The significance of soft errors in logic has grown because of reduced memory vulnerability and the shrinking dimensions of semiconductor technology coupled with the increasing amount of logic integrated into a chip. Consequently, some of ARM’s customers are concerned about how soft errors on the bus interconnect will affect the dependability of their systems, since the interconnect is a critical hub of communication in a SoC and represents a substantial and growing amount of logic. With the rising complexity of their systems, the interconnect will become larger and more complex in the future, adding to their concern. In this work the impact of soft errors on the bus interconnect logic was investigated and a product was developed to ameliorate the effects of such errors on ARM’s customers’ products. Methods to measure the SER of ARM IP were investigated by focusing on logical masking, which is a component in the calculation of the SER. The effect that the topology of a combinatorial logic circuit has on its logical masking rate was considered by performing gate-level statistical fault injection on different implementations of adder circuits. Significant variation in logical masking was found ranging from a factor of 3.1 at a synthesis frequency of 100 MHz to a factor of 2.1 at 900 MHz. This difference is explained in an original way by correlating logical masking with the circuit’s path length and fan-out. These properties could be used to create a static method of measuring the logical masking rather than the current time-consuming method of dynamic simulation. Additionally, nearly 30% of faults injected cause more than one error, which means that the combinational SER will be underestimated if research does not take gate fan-out into consideration. Using this methodology a circuit designer can now base his choice or development of a circuit on its reliability as well as its performance, power, and area. Studying the variation in the factors that affect the SER is important to ensure accuracy in addressing customer requirements. Although it is important to consider the rate of soft error occurrence, in this work the impact of errors is demonstrated to be critical. Using protocol-level fault injection it is shown that faults on the ARM AXI bus interconnect can have a serious effect on the reliability of the entire SoC such as deadlock, memory corruption, or undefined behaviour. Using a fault-path traversal algorithm, it is demonstrated that traditional error detection codes are not sufficient at preventing these failures when faults occur on certain AXI bus signals. This led to the development of novel fault tolerant methods that provide protection for these identified signals. Based on these developments, a product was proposed for an add-on to the AXI bus interconnect that can detect, correct, and report logic soft errors without changing the AMBA standard or the customer’s connecting IP