Subj : Re: Formal Mutex Semantics To : comp.programming.threads From : Joe Seigh Date : Tue Aug 16 2005 03:45 pm John Hickin wrote: > "Joe Seigh" wrote in message > news:JcqdnebRnZy0b5zeRVn-qA@comcast.com... > >>Maciej Sobczak wrote:> Posix though it best not to confuse you by trying > > to define anything. > >>You supposedly have inate intuitive knowlege of mutex semantics. >> > > > Another way to look at this is that the mutex api augments the C language > that a programmer might use. Considering this augmented thing as a formal > language there are some useful theorems that we want to prove. Then, there > are potentially several models where such proof is possible. So not > specifying the model might be a good thing. If they had different semantics then programs correct using one set may not be correct using another set, which would be a bad thing. If they're equivalent semantics then what does it matter which one is the "official" set of semantics? -- Joe Seigh When you get lemons, you make lemonade. When you get hardware, you make software. .