On Thu, 25 Jun 2015, Steve Reinhardt wrote:
On Thu, Jun 25, 2015 at 12:28 PM Nilay Vaish <[email protected]> wrote:
Since you insist on using it, I'll drop the patch. While I am confident
that this structure is incorrect, I would really love to listen from Brad
or Lisa or you, how you make sense of the order of messages in this
structure. If at any time you realize that your messages are actually not
ordered, use MessageBuffer with ordered=true and connect the controllers
directly.
Thanks, Nilay. Once we post our GPU model (which uses WireBuffers) then we
can look at whether it makes sense to replace them. Replacing the
WireBuffers at this point in time would only add needless complexity to our
already daunting task of cleaning up and merging this code for release.
I forgot to add. I am going to change one thing the WireBuffer. I am
going to drop MessageBufferNode from it. You can look following the link
for the reason: http://reviews.gem5.org/r/2897/.
--
Nilay
_______________________________________________
gem5-dev mailing list
[email protected]
http://m5sim.org/mailman/listinfo/gem5-dev
