![]() So a software watchdog is only useful if you have the watchdog running in such a way that it is more reliable than the task in question, for example it may be there to prevent a task from wandering out of its memory space and getting a protection fault (which you could have just watched for in a fault handler rather than setup a watchdog, but anyway). ![]() ![]() There is a problem if the questionable code fails due to hardware which could take down all of the code including the software watchdog. Watchdog organizations and individuals are like sentries. The express purpose of watchdogs the reason they’re called watchdogs is protection. For a software watchdog, one software task/thread/whatever is assumed to be more reliable than another, if the questionable code fails to kick the reliable code then the reliable code kills the questionable code (or whatever the design dictates). The watchdog role overlaps with that of the advocate, but the thrust of most advocacy is the advancement of a cause or the improvement of conditions for a particular population or geographic area. The hardware is assumed to be more reliable than the software. For a hardware watchdog, there is hardware that if it is not kicked often enough will assume the software has hung and will reset the system usually or whatever the design dictates.
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |