>>> On 16.01.14 at 17:04, Julian Stecklina <jsteckli@xxxxxxxxxxxxxxxxxxxx> wrote: > On 01/16/2014 04:04 PM, Jan Beulich wrote: >> I don't think so - this would only be an issue if the conditions used >> | instead of ||. || implies a sequence point between evaluating the >> left and right sides, and the standard says: "The presence of a >> sequence point between the evaluation of expressions A and B >> implies that every value computation and side effect associated >> with A is sequenced before every value computation and side >> effect associated with B." > > This only applies to single-threaded code. Multithreaded code must be > data-race free for that to be true. See > > https://lwn.net/Articles/508991/ > >> And even if there was a problem (i.e. my interpretation of the >> above being incorrect), I don't think you'd need ACCESS_ONCE() >> here: The same local variable can't have two different values in >> two different use sites when there was no intermediate >> assignment to it. > > Same comment as above. One half of this doesn't apply here, due to the explicit barriers that are there. The half about converting local variable accesses back to memory reads (i.e. eliding the local variable), however, is only a theoretical issue afaict: If a compiler really did this, I think there'd be far more places where this would hurt. Jan -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
