[ovs-dev] [PATCH 2/2] poll-loop: Remove static variable n_waiters.
blp at nicira.com
Mon Aug 15 09:56:39 PDT 2011
I have a story there. I had a whole "counted list" implementation
ready to go, and then when I went through the tree to find potential
users I found only two. And then when I looked closer I found that
they didn't really need to count. And that's why there's this patch
series that makes them not count anymore, instead of adding a counted
On Mon, Aug 15, 2011 at 09:51:51AM -0700, Justin Pettit wrote:
> Okay, sounds good. It's not large enough that it makes sense to
> make the lists counting or anything.
> On Aug 15, 2011, at 9:03 AM, Ben Pfaff wrote:
> > When I occasionally look, I usually see about a dozen or so. Not
> > enough to concern me, anyway. Some are probably duplicates (same fd).
> > On Fri, Aug 12, 2011 at 05:06:06PM -0700, Justin Pettit wrote:
> >> It looks fine, but do we have an idea on the number of waiters that typically exist?
> >> --Justin
> >> On Aug 10, 2011, at 3:05 PM, Ben Pfaff wrote:
> >>> It's always a little risky to track the length of a list by hand, because
> >>> it is easy to miss a spot where the length can change. So it seems like
> >>> a small cleanup to just measure the length of the 'waiters' list at the
> >>> point where we need to know it. list_size() is O(n) in the length of the
> >>> list, but the function that calls it is already O(n) in that length so it
> >>> seems like a fair trade-off.
More information about the dev