There is a new post at my Wind River blog, about how a team of researchers at the University of Nebraska at Lincoln is using Simics to force rare bugs to manifest themselves as errors. They used Simics to control a target system to force it into rare situations much more likely to trigger latent bugs, requiring far fewer test runs compared to just randomly rerunning tests again and again and hoping to see a bug.
Category Archives: Testing
There is a new post at my Wind River blog, about how you actually do fault injection in Simics. This particular post is pretty detailed, showing the actual architecture of a fault injector in Simics, not just “yes you can do it”. It includes actual diagrams of system components and how you can insert fault injection into an existing system, so it is a bit more technical than most my Wind River blog posts that tend to be more conceptual.
Fault Injection is a topic that has fascinated me for a long time. Not just the area of software-to-software fault injection, but more so how you inject faults into hardware using hardware (and how to conveniently approximate this using a simulator). I just stumbled on a short interesting note about such hardware-actuated fault injection in a Fujitsu article.
There is a new post at my Wind River blog, an interview with Dan Poirot at RTI who is using Simics to model and test heterogeneous, distributed, networked systems.
There seems to be no shortage of bugs that “should have been obvious” and subject to the “how can you not check that your own products work together” phenomenon. Just the other day, I stumbled on another one. This time, it was the Microsoft set of applications and operating systems that do not quite work together the way you would expect them to.
We recently got ourselves an iPod Touch, to entertain our oldest child on long trips. It is a brilliant device in many ways, I can understand why people love their iPhones (even though I am very happy with the very different style of the Blackberry phone that I was given by my employer). However, I have found one weird behavior in the music player that leaves me wondering how it got through into the shipping product.
Continuing on the thread from my previous post about the testing of products that fail to find problems that become obvious to (some) users after a very short time, I just read an article (in Swedish) about how the famed Tesla roadster cars behaved when they were confronted with Scandinavian winters.