Revisiting "too small to fail"

Posted May 22, 2017 15:28 UTC (Mon) by mageta (subscriber, #89696)
In reply to: Revisiting "too small to fail" by zlynx
Parent article: Revisiting "too small to fail"

> Or possibly a simpler idea, take one of the existing automatic unit test generators and modify it for kernel code, although there's always the problem of knowing if the hardware is being simulated correctly.

If there is simulation at all. CPU/MMU alright, even some basic I/O. But I'd wager, for over 90% of the hardware for which the kernel has drivers there is no simulation at all. There is also no trend to make more "simulations", its (understandably) much more interesting to do para-virtualized solutions.