jfv 2008-08-28 22:05:19 UTC
FreeBSD src repository
Modified files:
sys/net if.h if_vlan.c
Log:
SVN rev 182413 on 2008-08-28 22:05:19Z by jfv
Fix to bug kern/126850. Only dispatch event hander if the
interface had a parent (was attached).
Reviewed by:
jhb 2008-08-07 07:23:53 UTC
FreeBSD src repository
Modified files:(Branch: RELENG_6)
sys/net if.h
sys/sys sockio.h
Log:
SVN rev 181359 on 2008-08-07 07:23:53Z by jhb
MFC: Trim some noise from some #ifdef's.
Revision Changes
jhb 2008-08-07 07:25:48 UTC
FreeBSD src repository
Modified files:(Branch: RELENG_7)
sys/net if.h
sys/sys sockio.h
Log:
SVN rev 181360 on 2008-08-07 07:25:48Z by jhb
MFC: Trim some noise from some #ifdef's.
Revision Changes
kmacy 2008-07-30 00:17:40 UTC
FreeBSD src repository
Modified files:(Branch: RELENG_7)
sys/net if.h
Log:
SVN rev 180976 on 2008-07-30 00:17:40Z by kmacy
MFC IFCAP_TOE
Revision ChangesPath
1.108.2.3 +3 -0 src/sys/net/if.h
sam 2008-04-03 23:43:25 UTC
FreeBSD src repository
Modified files:(Branch: RELENG_7)
sys/net if.h
sbin/ifconfigifconfig.8 ifconfig.c
Log:
MFC: WOL infrastructure support
Revision ChangesPath
1.142.2.2 +16 -1
iedowse 2008-04-01 23:40:33 UTC
FreeBSD src repository
Modified files:(Branch: RELENG_7)
sys/net if.h
Log:
MFC 1.113: Add IFF_NEEDSGIANT to IFF_CANTCHANGE to prevent userland
code from clearing IFF_NEEDSGIANT.
Revision ChangesPath
1.108.2.1
iedowse 2008-03-27 18:02:30 UTC
FreeBSD src repository
Modified files:
sys/net if.h
Log:
Add IFF_NEEDSGIANT to IFF_CANTCHANGE, to prevent user-level code
from clearing the IFF_NEEDSGIANT flag on Giant-locked interfaces.
In particular, wpa_supplicant was doing
kmacy 2007-12-15 20:22:09 UTC
FreeBSD src repository
Modified files:
sys/net if.h
Log:
add interface capability for TOE
Revision ChangesPath
1.110 +1 -0 src/sys/net/if.h
___
cvs-all@freebsd.org
kmacy 2007-12-15 21:01:48 UTC
FreeBSD src repository
Modified files:
sys/net if.h
Log:
Create separate capability flags for TCP over IPv4 and TCP over IPv6
Revision ChangesPath
1.111 +3 -1 src/sys/net/if.h
kmacy 2007-12-15 22:06:24 UTC
FreeBSD src repository
Modified files:
sys/net if.h
Log:
fix bonehead cut and paste error in last commit
Revision ChangesPath
1.112 +1 -1 src/sys/net/if.h
___
On Sunday 09 December 2007 09:31:00 pm Sam Leffler wrote:
sam 2007-12-10 02:31:00 UTC
FreeBSD src repository
Modified files:
sys/net if.h
sbin/ifconfigifconfig.8 ifconfig.c
Log:
Wake On Lan (WOL) infrastructure
Submitted by: Stefan
sam 2007-12-10 02:31:00 UTC
FreeBSD src repository
Modified files:
sys/net if.h
sbin/ifconfigifconfig.8 ifconfig.c
Log:
Wake On Lan (WOL) infrastructure
Submitted by: Stefan Sperling [EMAIL PROTECTED]
Reviewed by:brooks
Revision
Sam Leffler wrote:
sam 2007-12-10 02:31:00 UTC
FreeBSD src repository
Modified files:
sys/net if.h
sbin/ifconfigifconfig.8 ifconfig.c
Log:
Wake On Lan (WOL) infrastructure
Submitted by: Stefan Sperling [EMAIL PROTECTED]
Reviewed by:
Kip Macy wrote:
To the best of my knowledge, it is a Solarisism and the 10G cards
from Sun (now licensed to Marvel?) are already delivering this.
Hm. Almost every GigE card out there has it. And yes Marvell is
OEM'ing the card.
The only thing novel about Neptune is the unusually large
Andre Oppermann [EMAIL PROTECTED] wrote:
Sam Leffler wrote:
Andre Oppermann wrote:
andre 2007-06-11 20:08:12 UTC
FreeBSD src repository
Modified files:
sys/net if.h Log:
Add IFCAP_LRO flag for drivers to announce their TCP Large Receive
Offload
* Kip Macy [EMAIL PROTECTED] [070611 15:54] wrote:
I think its a windowism, stands for Receive Side Scaling, and what it means
is multiple receive queues, each with an MSI/X vector so they can interrupt
different CPUs, or they can be tied to virtual guests, or MACs, etc etc...
AFAICT FreeBSD
* Robert Watson [EMAIL PROTECTED] [070611 17:24] wrote:
On Mon, 11 Jun 2007, Kip Macy wrote:
I think its a windowism, stands for Receive Side Scaling, and what it
means is multiple receive queues, each with an MSI/X vector so they can
interrupt different CPUs, or they can be tied to
True, but support it is relatively trivial from the kernel PoV.
Basically you allow the process to fetch which cpu a socket is
bound to, then you can use descriptor passing (processes) or shared
fdtable (threads) and a call to bind to cpu to properly juggle
things around.
Thanks for
* Kip Macy [EMAIL PROTECTED] [070612 11:50] wrote:
True, but support it is relatively trivial from the kernel PoV.
Basically you allow the process to fetch which cpu a socket is
bound to, then you can use descriptor passing (processes) or shared
fdtable (threads) and a call to bind to cpu to
On Tue, 12 Jun 2007, Alfred Perlstein wrote:
AFAICT FreeBSD can't currently benefit from this as there is no cpu
affinity for connections. I may be wrong, but I see lower
single-connection throughput using a receive queue per core than using a
single receive queue. RSS is done by hashing a
Jack Vogel wrote:
On 6/11/07, Andre Oppermann [EMAIL PROTECTED] wrote:
Jack Vogel wrote:
On 6/11/07, Andre Oppermann [EMAIL PROTECTED] wrote:
Jack Vogel wrote:
On 6/11/07, Jack Vogel [EMAIL PROTECTED] wrote:
On 6/11/07, Sam Leffler [EMAIL PROTECTED] wrote:
Andre Oppermann wrote:
To the best of my knowledge, it is a Solarisism and the 10G cards
from Sun (now licensed to Marvel?) are already delivering this.
Hm. Almost every GigE card out there has it. And yes Marvell is
OEM'ing the card.
The only thing novel about Neptune is the unusually large number of
TX/RX dma
andre 2007-06-11 20:08:12 UTC
FreeBSD src repository
Modified files:
sys/net if.h
Log:
Add IFCAP_LRO flag for drivers to announce their TCP Large Receive Offload
capabilities.
Revision ChangesPath
1.108 +1 -0 src/sys/net/if.h
Andre Oppermann wrote:
andre 2007-06-11 20:08:12 UTC
FreeBSD src repository
Modified files:
sys/net if.h
Log:
Add IFCAP_LRO flag for drivers to announce their TCP Large Receive Offload
capabilities.
Revision ChangesPath
1.108 +1 -0
On 6/11/07, Sam Leffler [EMAIL PROTECTED] wrote:
Andre Oppermann wrote:
andre 2007-06-11 20:08:12 UTC
FreeBSD src repository
Modified files:
sys/net if.h
Log:
Add IFCAP_LRO flag for drivers to announce their TCP Large Receive Offload
capabilities.
On 6/11/07, Jack Vogel [EMAIL PROTECTED] wrote:
On 6/11/07, Sam Leffler [EMAIL PROTECTED] wrote:
Andre Oppermann wrote:
andre 2007-06-11 20:08:12 UTC
FreeBSD src repository
Modified files:
sys/net if.h
Log:
Add IFCAP_LRO flag for drivers to
Sam Leffler wrote:
Andre Oppermann wrote:
andre 2007-06-11 20:08:12 UTC
FreeBSD src repository
Modified files:
sys/net if.h Log:
Add IFCAP_LRO flag for drivers to announce their TCP Large Receive
Offload
capabilities.
Revision ChangesPath
1.108
Jack Vogel wrote:
On 6/11/07, Jack Vogel [EMAIL PROTECTED] wrote:
On 6/11/07, Sam Leffler [EMAIL PROTECTED] wrote:
Andre Oppermann wrote:
andre 2007-06-11 20:08:12 UTC
FreeBSD src repository
Modified files:
sys/net if.h
Log:
Add IFCAP_LRO flag for
On 6/11/07, Andre Oppermann [EMAIL PROTECTED] wrote:
Jack Vogel wrote:
On 6/11/07, Jack Vogel [EMAIL PROTECTED] wrote:
On 6/11/07, Sam Leffler [EMAIL PROTECTED] wrote:
Andre Oppermann wrote:
andre 2007-06-11 20:08:12 UTC
FreeBSD src repository
Modified files:
Jack Vogel wrote:
On 6/11/07, Andre Oppermann [EMAIL PROTECTED] wrote:
Jack Vogel wrote:
On 6/11/07, Jack Vogel [EMAIL PROTECTED] wrote:
On 6/11/07, Sam Leffler [EMAIL PROTECTED] wrote:
Andre Oppermann wrote:
andre 2007-06-11 20:08:12 UTC
FreeBSD src repository
On 6/11/07, Andre Oppermann [EMAIL PROTECTED] wrote:
Jack Vogel wrote:
On 6/11/07, Andre Oppermann [EMAIL PROTECTED] wrote:
Jack Vogel wrote:
On 6/11/07, Jack Vogel [EMAIL PROTECTED] wrote:
On 6/11/07, Sam Leffler [EMAIL PROTECTED] wrote:
Andre Oppermann wrote:
andre
I think its a windowism, stands for Receive Side Scaling, and what it means
is multiple receive queues, each with an MSI/X vector so they can interrupt
different CPUs, or they can be tied to virtual guests, or MACs, etc etc...
AFAICT FreeBSD can't currently benefit from this as there is no cpu
On Mon, 11 Jun 2007, Kip Macy wrote:
I think its a windowism, stands for Receive Side Scaling, and what it means
is multiple receive queues, each with an MSI/X vector so they can interrupt
different CPUs, or they can be tied to virtual guests, or MACs, etc etc...
AFAICT FreeBSD can't
brooks 2007-05-16 18:37:37 UTC
FreeBSD src repository
Modified files:
sys/net if.h if_var.h
Log:
The struct if_data members ifi_recvquota and ifi_xmitquota have been
unused for ages. Rename them to ifi_spare_char1 and ifi_spare_char2
respectively to indicate
yar 2007-05-02 11:13:48 UTC
FreeBSD src repository
Modified files:
sys/net if.h
Log:
Fix a couple of typos in a comment.
Revision ChangesPath
1.106 +2 -2 src/sys/net/if.h
___
cvs-all@freebsd.org
Joao Barros wrote this message on Wed, Sep 06, 2006 at 23:25 +0100:
On 9/6/06, Andre Oppermann [EMAIL PROTECTED] wrote:
Log:
First step of TSO (TCP segmentation offload) support in our network
stack.
I have been following the thread on TSO with the Intel boys.
Can you point out
Andre Oppermann [EMAIL PROTECTED] wrote:
Joao Barros wrote:
On 9/6/06, Andre Oppermann [EMAIL PROTECTED] wrote:
Log:
First step of TSO (TCP segmentation offload) support in our network
stack.
Hurray!
I have been following the thread on TSO with the Intel boys.
Can you point out
andre 2006-09-06 18:06:04 UTC
FreeBSD src repository
Modified files:
sys/net if.h if_var.h
sys/sys mbuf.h
Log:
Improve description of if_capabilities, if_capenable and ifi_hwassist.
Sponsored by: TCP/IP Optimization Fundraise 2005
andre 2006-09-06 21:51:59 UTC
FreeBSD src repository
Modified files:
sys/net if.h
sys/netinet ip_output.c tcp_input.c tcp_subr.c
tcp_var.h
sys/netinet6 icmp6.c
sys/sys mbuf.h
Log:
First step of
Andre Oppermann wrote:
andre 2006-09-06 21:51:59 UTC
FreeBSD src repository
Modified files:
sys/net if.h
sys/netinet ip_output.c tcp_input.c tcp_subr.c
tcp_var.h
sys/netinet6 icmp6.c
sys/sys
On 9/6/06, Andre Oppermann [EMAIL PROTECTED] wrote:
Log:
First step of TSO (TCP segmentation offload) support in our network stack.
I have been following the thread on TSO with the Intel boys.
Can you point out which drivers/hardware will support this feature?
--
Joao Barros
Joao Barros wrote:
On 9/6/06, Andre Oppermann [EMAIL PROTECTED] wrote:
Log:
First step of TSO (TCP segmentation offload) support in our network
stack.
I have been following the thread on TSO with the Intel boys.
Can you point out which drivers/hardware will support this feature?
First
42 matches
Mail list logo
