Re: divert and deadlock issues

Christian S.J. Peron wrote: Group, I've come up with a basic patch, here are the highlights as per our discussion: - Check for the presence of socket options, if they are present duplicate them using m_dup(9) - Drop the INP/INFO locks after duplication - Activate ip_output() with the cloned m

Re: IPSEC connection drops and doesn't recover

-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Isaac Kohen <[EMAIL PROTECTED]> wrote: > > I get these in dmesg-- does it mean anything? > > IPv4 ESP input: no key association found for spi 94246771 > IPv4 ESP input: no key association found for spi 94246771 > IPv4 ESP input: no key association fou

Re: divert and deadlock issues

Group, I've come up with a basic patch, here are the highlights as per our discussion: - Check for the presence of socket options, if they are present duplicate them using m_dup(9) - Drop the INP/INFO locks after duplication - Activate ip_output() with the cloned mbuf (for socket options). Als

Re: SMPable version of EM driver

On 8/1/07, Vladimir Ivanov <[EMAIL PROTECTED]> wrote: > Hi, > > I've just published revision of EM (mainstream RELENG_6 version w/patch) > driver which is being used in our company to increase network > performance. The main benefit - significantly better SMP utilization. > > http://people.yandex-t

Re: Creating a "non-bridge"

At 12:08 AM 8/1/2007, Julian Elischer wrote: >possibly you could just bridge them together but use ipfw on the bridge to >enforce isolation. Will IPFW block ARP? IPX? Other protocols which may be either demultiplexed or "teed" within the network stack? >OR you could possibly make a netgraph ve

Re: divert and deadlock issues

Christian S.J. Peron wrote: Well, it's still the intent to keep the ability to divert and re-inject multicast packets. This change would basically say: "You cant specify multicast options via the divert socket". Which in practice doesn't happen anyway (where I looked). I dont think we should be

Re: SMPable version of EM driver

Bill Marquette wrote: [skip] What type of performance differences are you seeing with these changes? Is this with FreeBSD acting as a router/firewall, or purely RX queue is being processed w/more than one thread. TX queue thread isn't locked with RX anymore. Extra CPU time can be used by e.

Re: SMPable version of EM driver

Vladimir Ivanov wrote: Hi, I've just published revision of EM (mainstream RELENG_6 version w/patch) driver which is being used in our company to increase network performance. The main benefit - significantly better SMP utilization. http://people.yandex-team.ru/~wawa/em-6.2.9-yandex.tar.gz. T

SMPable version of EM driver

Hi, I've just published revision of EM (mainstream RELENG_6 version w/patch) driver which is being used in our company to increase network performance. The main benefit - significantly better SMP utilization. http://people.yandex-team.ru/~wawa/em-6.2.9-yandex.tar.gz. The driver should be used

Re: IPSEC connection drops and doesn't recover

I get these in dmesg-- does it mean anything? IPv4 ESP input: no key association found for spi 94246771 IPv4 ESP input: no key association found for spi 94246771 IPv4 ESP input: no key association found for spi 94246771 IPv4 ESP input: no key association found for spi 94246771 IPv4 ESP input: no k

IPsec AH tunneling pakcet mis-handling?

Dear all: I do not know the purpose of the following codes in the very beginning in ip6_input(): #ifdef IPSEC /* * should the inner packet be considered authentic? * see comment in ah4_input(). */ if (m) { m->m_flags &= ~M_AUTHIPHDR; m->m_flags &= ~M_AUTHIPDGM;