[concurrency-interest] Java Memory Model and ParallelStream
luke.hutch at gmail.com
Fri Mar 6 10:16:42 EST 2020
On Fri, Mar 6, 2020, 7:45 AM Andrew Haley <aph at redhat.com> wrote:
> To answer that I'd need to be told what a "computational barrier" is. I
> have some guesses, but I think you should spell it out.
I was referring to a barrier in the standard sense of a synchronization
mechanism that waits for some set of threads to finish some set of tasks
before continuing. I added the word "computational" and "memory" before
"barrier" to disambiguate the "happens-before" of computational work from
the "happens-before" of writing values to memory, viz JMM. I then asked if
these two "happens-before" relationships can be assumed to be exactly
equivalent (i.e. taking into account JMM, and how JMM works with cache
coherency behavior, out-of-order execution, etc.).
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the Concurrency-interest