| |||||||||||||
Jones: system call abuseJones: system call abusePosted Nov 10, 2010 11:28 UTC (Wed) by ms (subscriber, #41272)Parent article: Jones: system call abuse
I'm too lazy to read through Dave's posts, but is he aware of the body of work that's been done on this with things like QuickCheck, SmallCheck, LazySmallCheck &c?
(Log in to post comments)
Jones: system call abuse Posted Nov 10, 2010 15:28 UTC (Wed) by adobriyan (guest, #30858) [Link]
How many kernel bugs these foochecks have found?
Jones: system call abuse Posted Nov 10, 2010 15:35 UTC (Wed) by ms (subscriber, #41272) [Link]
Erm, well if you rewrote the kernel in Haskell, no doubt a vast number, but the kernel would be vastly less buggy anyway due to the existence of a decent type system that eliminates one large class of bugs.
(I'm _not_ suggesting that writing a kernel in Haskell is a good idea. I'm merely pointing out that randomised testing based on type analysis of the inputs to functions is a well studied area.)
|
|||||||||||||