or maybe fewer frames per ring. The 8MB per ring uniformly set on each port
could be a rigid setting.
Ideally the ring size should be a function of the need, which is the amount
of netlink traffic for the port using that socket.
Kais
On Wed, Dec 4, 2013 at 9:20 AM, Thomas Graf wrote:
> On 12/0
looks good.
Kais
On Wed, Aug 21, 2013 at 1:01 PM, Ben Pfaff wrote:
> Otherwise new Clang complains about this function because it only sometimes
> releases the lock (that is, it only does it when there is a lock to
> release).
>
> I first noticed these warnings with Clang 1:3.4~svn188890-1~exp1
Andy, I didn't see this one assigned to you in Patchwork, so it got two
reviews.
The changes look good to me too.
Kais.
On Fri, Aug 16, 2013 at 2:16 PM, Andy Zhou wrote:
> Looks good.
> Acked-by: Andy Zhou
>
>
> On Fri, Aug 16, 2013 at 1:09 PM, Ben Pfaff wrote:
>
>> htb_parse_qdisc_details_
sounds good.
Kais
On Wed, Aug 7, 2013 at 11:25 PM, Ben Pfaff wrote:
> On Wed, Aug 07, 2013 at 05:02:42PM -0700, Kais Belgaied wrote:
> > proposed changes look good.
> > comment needed regarding the reason for internally managed per-thread
> data
> > (limitation of
proposed changes look good.
comment needed regarding the reason for internally managed per-thread data
(limitation of Posix' PTHREAD_KEYS_MAX)
Kais
On Tue, Aug 6, 2013 at 4:56 PM, Ben Pfaff wrote:
> DEFINE_PER_THREAD_DATA always declared its data item as "static", meaning
> that it was only di
Changes look fine by me.
Kais.
On Tue, Aug 6, 2013 at 4:56 PM, Ben Pfaff wrote:
> Signed-off-by: Ben Pfaff
> ---
> lib/ovs-thread.c | 33 -
> lib/ovs-thread.h | 17 +
> 2 files changed, 49 insertions(+), 1 deletion(-)
>
> diff --git a/lib/o
looks good to me.
Kais.
On Tue, Aug 6, 2013 at 4:56 PM, Ben Pfaff wrote:
> Signed-off-by: Ben Pfaff
> ---
> lib/ovs-thread.c |1 +
> lib/ovs-thread.h |5 +++--
> lib/poll-loop.c |2 +-
> 3 files changed, 5 insertions(+), 3 deletions(-)
>
> diff --git a/lib/ovs-thread.c b/lib/ovs
