Re: barriers before {req/rsp}_cons = cons?

[Samuel Thibault <samuel.thibault@eu.citrix.com>]

Keir Fraser, le Mon 21 Jul 2008 09:28:38 +0100, a écrit :
> On 18/7/08 20:33, "Samuel Thibault" <samuel.thibault@eu.citrix.com> wrote:
> 
> >> Could you describe the race you believe is made possible by the absence of
> >> the barrier?
> > 
> > It is very unlikely, but the write into req_cons could happen before
> > actually doing the memcpy, and so another processor could try to reuse
> > the same request slot.
> 
> req_cons is not shared with the frontend driver,

Aow, right, I had forgotten that on the frontend side we are using the
rsp index, so that there is no more in-flight requests than ring slots.

Samuel