Re: Enabling VIMAGE by default for FreeBSD 11?

2014-10-14 Thread Craig Rodrigues
On Tue, Oct 14, 2014 at 5:53 AM, Olivier Cochard-Labbé wrote: > > I can use my forwarding/firewalling 10Giga lab for testing VIMAGE impact. > > Thank you for doing the test and providing the results. You put some good work into making those graphs and explaining the results! -- Craig

[Bug 194314] [ixgbe] driver makes some dangerous assumptions with struct mbuf sizing with IXGBE_RX_COPY_LEN

2014-10-14 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=194314 --- Comment #5 from Garrett Cooper --- (In reply to Garrett Cooper from comment #4) > `struct mbuf` as the observation hasn't been violated on FreeBSD CURRENT "as the observation hasn't" -> "; the conditions for the optimization to work ha

[Bug 194314] [ixgbe] driver makes some dangerous assumptions with struct mbuf sizing with IXGBE_RX_COPY_LEN

2014-10-14 Thread bugzilla-noreply
https://bugs.freebsd.org/bugzilla/show_bug.cgi?id=194314 Garrett Cooper changed: What|Removed |Added Priority|--- |Normal --- Comment #4 from Garret

Multicast routing, IGMP, IPTV doubts..

2014-10-14 Thread Marek Salwerowicz
Hi all, My home router is small FreeBSD 10 box, with 3 ethernet ports, I use pf as a firewall. My ISP provides FTTH, that ends at my home with a small box called CPE. The CPE contains 4 ethernet ports. On one port I have Internet access, on the other there is IPTV (that's based on udp/multic

Re: [CFT] alc(4) QAC AR816x/AR817x ethernet controller support

2014-10-14 Thread Andrey Cherkashin
Just wanted to say that with that patch it works on 10.1 with E2200. [  3]  0.0-10.0 sec  1.07 GBytes   923 Mbits/sec --  Andrey Cherkashin Sent with Airmail ___ freebsd-net@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-net

Re: Enabling VIMAGE by default for FreeBSD 11?

2014-10-14 Thread Craig Rodrigues
On Tue, Oct 14, 2014 at 11:17 AM, John D. Hendrickson and Sara Darnell < johnandsa...@cox.net> wrote: > do you mean VIMAGE allows a jail to use an iface device for many IPs or > even MAC? i thought that was already the case all cards can "listen" - > it's only a headers trick per say. > > Search

Re: Enabling VIMAGE by default for FreeBSD 11?

2014-10-14 Thread John D. Hendrickson and Sara Darnell
Alexander V. Chernikov wrote: On 11 Oct 2014, at 21:58, Craig Rodrigues wrote: Hi, What action items are left to enable VIMAGE by default for FreeBSD 11? well the next step is to make it a dependancy so that free bsd won't install without it, and to inject it in many binaries that insure

Re: Enabling VIMAGE by default for FreeBSD 11?

2014-10-14 Thread John D. Hendrickson and Sara Darnell
Alexander V. Chernikov wrote: On 11 Oct 2014, at 21:58, Craig Rodrigues wrote: Hi, What action items are left to enable VIMAGE by default for FreeBSD 11? Are there any tests results showing performance implications on different network-related workloads? Not everyone uses bhyve, so VIMAGE

Re: Enabling VIMAGE by default for FreeBSD 11?

2014-10-14 Thread Benjamin Kaduk
On Tue, 14 Oct 2014, Olivier Cochard-Labbé wrote: > I can use my forwarding/firewalling 10Giga lab for testing VIMAGE impact. > Here are my ministat results (smallest packet size, value in > packet-per-second, about 2000 flows). > => I didn't see lot's of performance impact with VIMAGE option adde

Re: Enabling VIMAGE by default for FreeBSD 11?

2014-10-14 Thread Olivier Cochard-Labbé
On Tue, Oct 14, 2014 at 8:20 AM, Craig Rodrigues wrote: > On Sat, Oct 11, 2014 at 1:20 PM, Alexander V. Chernikov > wrote: > > > > > On 11 Oct 2014, at 21:58, Craig Rodrigues wrote: > > > > > Hi, > > > > > > What action items are left to enable VIMAGE by default for FreeBSD 11? > > Are there an

Re: Enabling VIMAGE by default for FreeBSD 11?

2014-10-14 Thread k simon
在 14-10-13 0:38, Bjoern A. Zeeb 写道: On 12 Oct 2014, at 16:25 , Yamagi Burmeister wrote: Hello, it's been a while since I tested VIMAGE, but at the last time somewhere in 10-CURRENT some UMA memory leaks were left when destroying vnets. They weren't showstoppers for most workloads, but prett