Re: [moca] definitions of race conditions?

[Martin Berger <martinb@xxxxxxxxxxxxxx>]

> I don't have a citation to back this up, but I always thought of a race
> condition as being any failure of the diamond property,

but that's way too rqestrictive. it would decree just about any
concurrent program i have ever, including

     lock( mutex )     |    lock( mutex )
     x := 2            |    x := 22
     unlock( mutex )   |    unlock( mutex ),

to be suffereing from race conditions.

martin


 
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
The "models for mobility" mailing list     mailto:moca@xxxxxxxxxxxxxxx
http://www-sop.inria.fr/mimosa/personnel/Davide.Sangiorgi/moca.html