Steve Reinhardt wrote: > On Sat, Mar 27, 2010 at 12:20 AM, Gabe Black <[email protected]> wrote: > >> What about one way packets that don't collapse back to the sender? Do we >> have any of those? Or do we always collapse back at least with the ack? >> > > I'm not sure what you're referring to, Gabe... are you talking about > atomic mode (that's what "collapse back" makes me think of), or > packets that don't have responses, or both? > > Generally SenderState isn't used in atomic mode, since your stack > frame does just as good a job. Also it's not useful if there's no > reply. > > We should make sure that we don't assume that SenderState objects are > always dynamically allocated... I don't know if we ever statically > allocate them, but we could. Overwriting the pointer with NULL just > to signify that we're done with it in that case seems OK to me though. > > Steve > _______________________________________________ > m5-dev mailing list > [email protected] > http://m5sim.org/mailman/listinfo/m5-dev > Me either :). I just wasn't sure with my somewhat limited and/or rusty knowledge of how that worked that something like that -couldn't- happen, and I wanted to make sure you guys had considered it.
Gabe _______________________________________________ m5-dev mailing list [email protected] http://m5sim.org/mailman/listinfo/m5-dev
