Tools and Benchmarks for Real-Time Systems

Dear WATERS Community, due to multiple requests, we enhanced our provided 2017 challenge model with specified communication types. This model is now based on the latest APP4MC https://www.eclipse.org/app4mc/ release 0.8.2. which supports the specification of the used implementation type for each com...

Hi Julio, Michael, Javi, and Juan, the statement that the runnable execution time is without label access and those have to come additionally is valid for all communication semantics. The semantics (from the view of the runnable) only differ in placement of the accessed labels, and thus different ac...

Hi Simon, Arne, We need to ask you a couple of questions/clarifications more: (1) About the LET model for synchronization, you say: "With the LET model, tasks always read data at the beginning of the activation interval and write data at the end of the activation interval." This does not state clea...

Hi Julio,

yes that interpretation is correct. Every label has exactly one writer task, or more precisely one writer-runnable. All other tasks only read that label.
In the provided Amalthea model, the writer or reader role is attached to the label access within a runnable.

Best, Simon

Hi Julio, Javi, Juan, and Michael, thanks for your questions and your participation in the challenge! :-) Please find some answers below: 1 - By looking at the last paragraph in section B. Implicit Communication, which says: "However on the memory storage front, more local storage is required, since...

Hi Junchul, the unit of runnable execution is instructions. Number of instructions divided by the instructions per cycle (IPC) value from the core (HW element) results to the actual taken time in cpu cycles. In our model, the IPC is 1, so number of instructions equals the number of cycles. The 0ns d...

Hi Nacho, preemptive and cooperative tasks can be located on the same core. That's possible. The scheduling policy in this case is still priority based. Preemptive tasks here have a higher priority than cooperative ones. This means that still cooperative tasks preempt each other only at runnable bor...

Hello Juan, effect chains here don't have a blocking/waiting semantic. They are driven by data. In that example Runnable_100ms_7 produces data that is read by Runnable_10ms_19. As you figured out correctly, both runnables have different rates, so here the result of Runnable_100ms_7 is read multiple ...

Hello Juan, please see my answers below. Best, Simon Hello all, I am Juan M Rivas from the University of Cantabria. We are trying to solve the FMTV challenge, and we have came up with some questions. Maybe someone could help us. - The first sub-challenge states that we have to ignore the memory acce...

Title: Real World Automotive Benchmark For Free Authors: Simon Kramer, Dirk Ziegenbein, Arne Hamann (Corporate Research, Robert Bosch GmbH, Renningen, Germany) Abstract: The progress and comparability of real-time analysis methods that are applicable to real-world is slowed by the absence of realis...