Trace Control for Netem: Emulate network properties such as long range
dependency and self-similarity of cross-traffic.
A new option (trace) has been added to the netem command. If the trace option
is used, the values for packet delay etc. are read from a pregenerated trace
file, afterwards
Trace Control for Netem: Emulate network properties such as long range
dependency and self-similarity of cross-traffic.
user space (iproute2):
The directory tc/netem was split in two parts, one containing the original
distribution tables and the other the tools to generate trace files as well
Trace Control for Netem: Emulate network properties such as long range
dependency and self-similarity of cross-traffic.
kernel space:
The delay, drop, duplication and corruption values are readout in user space
and sent to kernel space via configfs. The userspace process will hang on
write
In article [EMAIL PROTECTED] (at Fri, 22 Sep 2006 02:31:59 -0500), William
Pitcock [EMAIL PROTECTED] says:
This patch allows for a user to disable the requirement to meet the
CAP_NET_BIND_SERVICE capability for a non-superuser. It is toggled by
the net.ipv4.allow_lowport_bind_nonsuperuser
On Fri, 22 Sep 2006 07:27:18 + (GMT)
Holger Kiehl [EMAIL PROTECTED] wrote:
I get some of the page allocation failure errors. My hardware is 4 CPU
Opteron with one quad + one dual intel e1000 cards. Kernel is plain 2.6.18
and for two cards MTU is set to 9000.
Sep 21 21:03:15 athena
From: Jesse Huang [EMAIL PROTECTED]
Change Logs:
- Restore the original TX FIFO overflow process.
Signed-off-by: Jesse Huang [EMAIL PROTECTED]
---
drivers/net/sundance.c | 45 +++--
1 files changed, 27 insertions(+), 18 deletions(-)
Stephen,
you will be glad to hear that the long crashing saga (at least in my
case) turns to the end.
The combination 2.6.18 / skge 1.8 is working well with SMP/IRQ
migration.
The last test which crashed was 2.6.17 / skge 1.6 + PCI posting fixes.
So some fix in 1.7/1.8 seems to make the
On Thu, 2006-09-21 at 23:51, Jiri Benc wrote:
Hi,
sorry for the delay.
On Fri, 01 Sep 2006 11:37:57 +0800, Hong Liu wrote:
@@ -2925,7 +2932,8 @@ static int ieee80211_ioctl_giwencode(str
return 0;
}
- erq-length = 0;
+ erq-length = min((int)erq-length,
On Thu, 2006-09-21 at 16:30 -0400, Dan Williams wrote:
That's kind a hole in the WE API. In this case, I think, SIOCGIWAP
should always return the BSSID of the current association, or none if
there is no association.
I was never sure if it was a hole or just a lack of clearly defined
If there's anything I can do to help isolate this issue please let me
know.
Looking through the logs I'm also intermittently seeing some of this:
sky2 eth1: rx error, status 0x7ffc0001 length 844
I'm not sure if it's related to the TX issue but I thought I'd mention it.
-
To unsubscribe
On 22-09-2006 00:37, Andrew Morton wrote:
Methinks CONFIG_NETFILTER_XT_TARGET_CLASSIFY should depend upon
CONFIG_BRIDGE_NETFILTER. Because brnf_deferred_hooks is defined in
net/bridge/br_netfilter.c and is referred to in net/netfilter/xt_physdev.c.
Or something else ;)
...
I am unsure of
This patch changes (hopefully!) all occurrences in d80211 of
list_for_each to list_for_each_entry (and _safe variants where they were
used before).
Signed-off-by: Johannes Berg [EMAIL PROTECTED]
---
This is the respun patch including no locking changes at all.
---
On Fri, Sep 22, 2006 at 10:41:08AM +0200, Jarek Poplawski wrote:
On 22-09-2006 00:37, Andrew Morton wrote:
Methinks CONFIG_NETFILTER_XT_TARGET_CLASSIFY should depend upon
CONFIG_BRIDGE_NETFILTER. Because brnf_deferred_hooks is defined in
net/bridge/br_netfilter.c and is referred to in
Sorry linux-kernel - it should go to netdev.
On 22-09-2006 10:03, Jarek Poplawski wrote:
On 22-09-2006 00:37, Andrew Morton wrote:
Methinks CONFIG_NETFILTER_XT_TARGET_CLASSIFY should depend upon
CONFIG_BRIDGE_NETFILTER. Because brnf_deferred_hooks is defined in
net/bridge/br_netfilter.c and
On Sep 22, 2006, at 2:41 AM, YOSHIFUJI Hideaki wrote:
In article [EMAIL PROTECTED] (at
Fri, 22 Sep 2006 02:31:59 -0500), William Pitcock
[EMAIL PROTECTED] says:
This patch allows for a user to disable the requirement to meet the
CAP_NET_BIND_SERVICE capability for a non-superuser. It is
[DCCP]: If an application does not want to provide a service code, that's
fine, it will be given 0. Otherwise, service codes can be set
via socket options as before.
This has fixes as below.
| Not sure what happened here but I can't apply this with git-apply. Can
| you
On Fri, Sep 22, 2006 at 11:27:31AM +0200, Jarek Poplawski wrote:
...
There is CONFIG_BRIDGE=m and this is the source of a problem:
config shouldn't allow for:
CONFIG_NETFILTER_BRIDGE=y
CONFIG_NETFILTER_XT_MATCH_PHYSDEV=y
So here is a patch proposal.
Jarek P.
diff -Nurp
Jarek Poplawski wrote:
On Fri, Sep 22, 2006 at 11:27:31AM +0200, Jarek Poplawski wrote:
...
There is CONFIG_BRIDGE=m and this is the source of a problem:
config shouldn't allow for:
CONFIG_NETFILTER_BRIDGE=y
CONFIG_NETFILTER_XT_MATCH_PHYSDEV=y
So here is a patch proposal.
Jarek P.
On Fri, Sep 22, 2006 at 12:15:43PM +0200, Patrick McHardy wrote:
...
We should fix the physdev dependencies since this is what is causing
problems.
...
config NETFILTER_XT_MATCH_PHYSDEV
tristate 'physdev match support'
- depends on NETFILTER_XTABLES BRIDGE_NETFILTER
+ depends
Jarek Poplawski wrote:
On Fri, Sep 22, 2006 at 12:15:43PM +0200, Patrick McHardy wrote:
...
We should fix the physdev dependencies since this is what is causing
problems.
...
config NETFILTER_XT_MATCH_PHYSDEV
tristate 'physdev match support'
- depends on NETFILTER_XTABLES
Thu, 21 Sep 2006 15:39:40 -0700, mabbas pise:
if the stack handle it then no need to know. When I call
ieee80211_register_hw with driver's channel, rate and txpower info. the
txpower values are overwritten by ieee80211_init_client. I get all the
txpower info from EEPROMin the NIC and they
Hello,
I'm having problems with my sky2 NIC hanging under heavy load. This
appears to be an old problem since it happened for me with 2.6.17 as
well. Upgrading the affected systems to 2.6.18 has not solved the
problem. It's easily reproducible for me since I'm running some
application stress
Hello.
I've found strange behaviour of transport mode IPsec in 2.6.18 tree.
After key daemons exchanged keys (I use racoon) I try following command
on 2.6.18 machine: telnet 192.168.4.79 22 (telnet from 2.6.18 to 2.6.17 based
one)
and get very slow response, here is related tcpdump output:
On Fri, Sep 22, 2006 at 03:29:48PM +0400, Evgeniy Polyakov ([EMAIL PROTECTED])
wrote:
Hello.
I've found strange behaviour of transport mode IPsec in 2.6.18 tree.
After key daemons exchanged keys (I use racoon) I try following command
on 2.6.18 machine: telnet 192.168.4.79 22 (telnet from
On Fri, Sep 22, 2006 at 12:54:42PM +0200, Patrick McHardy wrote:
Jarek Poplawski wrote:
On Fri, Sep 22, 2006 at 12:15:43PM +0200, Patrick McHardy wrote:
...
We should fix the physdev dependencies since this is what is causing
problems.
...
config NETFILTER_XT_MATCH_PHYSDEV
Jarek Poplawski wrote:
Anyway this special logic is also something misleading:
a) BRIDGE_NETFILTER=y and BRIDGE_NETFILTER depends on BRIDGE
sugest something which depends on BRIDGE is compiled into
kernel, which is usually impossible if BRIDGE=m
b) let's assume a) is right; shouldn't this
On Fri, 22 Sep 2006 11:04:32 +0200, Johannes Berg wrote:
This patch changes (hopefully!) all occurrences in d80211 of
list_for_each to list_for_each_entry (and _safe variants where they were
used before).
Applied to my tree, thanks for the patch!
Jiri
--
Jiri Benc
SUSE Labs
-
To
On Mon, 21 Aug 2006 09:41:08 +0200, Johannes Berg wrote:
This patch makes d80211 export LED triggers for rx/tx and introduces
functions to allow device drivers to query the trigger names for setting
default triggers. It also cleans up the Makefile LED related stuff.
Applied to my tree with two
On Fri, 22 Sep 2006 15:56:26 +0800, Hong Liu wrote:
Resend.
Applied to my tree, thanks for the patch!
Jiri
--
Jiri Benc
SUSE Labs
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message to [EMAIL PROTECTED]
More majordomo info at
| This creates two new socket options DCCP_SOCKOPT_TX_PACKET_SIZE
| and DCCP_SOCKOPT_RX_PACKET_SIZE. DCCP_SOCKOPT_PACKET_SIZE doesn't
| work and packet size should be set independently on two half
| connections.
I disagree with this solution: it solves one problem by introducing two
new ones:
| This removes DCCP_SOCKOPT_PACKET_SIZE for two reasons:
| * the current code doesn't work
| * tx and rx should be different (introduced in former patch)
Agree that DCCP_SOCKOPT_PACKET_SIZE is redundant, patch looks good and
has been tested.
Acked-by: Gerrit Renker [EMAIL PROTECTED]
|
Quoting Ian McDonald:
| Set initial packet size to defaults as existing code doesn't work
| as set_sockopt occurs after initialisation so dccps_packet_size
| is of no use really.
Please see comments to patch 5/7; rather than setting a default,
ccid3hc{rx,tx}_s should be derived from
On Fri, 22 Sep 2006, Andrew Morton wrote:
On Fri, 22 Sep 2006 07:27:18 + (GMT)
Holger Kiehl [EMAIL PROTECTED] wrote:
I get some of the page allocation failure errors. My hardware is 4 CPU
Opteron with one quad + one dual intel e1000 cards. Kernel is plain 2.6.18
and for two cards MTU is
On Tue, 22 Aug 2006 10:34:19 -0700, David Kimdon wrote:
This ioctl is used when radar is delected on a channel. Data frames must stop
but management frames must be allowed to continue for some time to communicate
the channel switch to stations.
Any progress here? Is what Simon wrote (quoted
On Fri, Sep 22, 2006 at 03:29:48PM +0400, Evgeniy Polyakov ([EMAIL PROTECTED])
wrote:
Hello.
I've found strange behaviour of transport mode IPsec in 2.6.18 tree.
After key daemons exchanged keys (I use racoon) I try following command
on 2.6.18 machine: telnet 192.168.4.79 22 (telnet from
Evgeniy Polyakov wrote:
I started process but if there will be no results in about an hour I
will continue after weekend only if there will be no interesting results
from other developers.
FWIW: I've tried myself and it appears to works fine here. Policy
similar to yours, no netfilter.
-
To
Jiri Benc wrote:
Thu, 21 Sep 2006 15:39:40 -0700, mabbas pise:
if the stack handle it then no need to know. When I call
ieee80211_register_hw with driver's channel, rate and txpower info. the
txpower values are overwritten by ieee80211_init_client. I get all the
txpower info from EEPROMin the
On Fri, 2006-09-22 at 08:24 -0500, Larry Finger wrote:
That project is on hold now, but I'm willing to share the information I have
and to participate in
further developments.
The database seems useful, maybe nl80211 could subsume the userspace
interface?
johannes
-
To unsubscribe from
Hello!
transactions to data segments is fubar. That issue is also why I wonder
about the setting of tcp_abc.
Yes, switching ABC on/off has visible impact on amount of segments.
When ABC is off, amount of segments is almost the same as number of
transactions. When it is on, ~1.5% are merged.
Johannes Berg wrote:
On Fri, 2006-09-22 at 08:24 -0500, Larry Finger wrote:
That project is on hold now, but I'm willing to share the information I have and to participate in
further developments.
The database seems useful, maybe nl80211 could subsume the userspace
interface?
That sounds
On Fri, Sep 22, 2006 at 02:23:17PM +0200, Patrick McHardy ([EMAIL PROTECTED])
wrote:
Evgeniy Polyakov wrote:
I started process but if there will be no results in about an hour I
will continue after weekend only if there will be no interesting results
from other developers.
FWIW: I've
On Fri, 2006-09-22 at 10:32 +0200, Johannes Berg wrote:
On Thu, 2006-09-21 at 16:30 -0400, Dan Williams wrote:
That's kind a hole in the WE API. In this case, I think, SIOCGIWAP
should always return the BSSID of the current association, or none if
there is no association.
I was never
On Fri, Sep 22, 2006 at 12:03:11PM +, Holger Kiehl ([EMAIL PROTECTED])
wrote:
This is going to cause an 9000-byte MTU to use a 16384-byte allocation.
e1000_alloc_rx_buffers() adds two bytes to that, so we do kmalloc(16386),
which causes the slab allocator to request 32768 bytes. All for a
When we found the cause of NETDEV watchdog timeouts in the wireless-2.6 code, I knew that the 2.6.18
release code would cause a serious regression. I immediately prepared and tested a patch to fix
this; however, I was unable to get it pushed into the stable code before release of 2.6.18. The
On Fri, Sep 22, 2006 at 08:20:08AM -0500, Larry Finger wrote:
This patch, which was originally sent to John Linville on 9/14/06,
has been taken against 2.6.18. It changes more lines
than would be absolutely necessary to affect the fix; however, it
ends up with this section looking exactly like
On 9/22/06, Gerrit Renker [EMAIL PROTECTED] wrote:
[DCCP]: If an application does not want to provide a service code, that's
Signed-off-by: Gerrit Renker [EMAIL PROTECTED]
Applied
Thanks,
- Arnaldo
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message
On Fri, Sep 22, 2006 at 02:25:36PM +1200, Ian McDonald wrote:
This change introduces a constant for CCID numbers.
Signed-off-by: Ian McDonald [EMAIL PROTECTED]
Applied
- Arnaldo
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message to [EMAIL PROTECTED]
On Fri, Sep 22, 2006 at 02:26:44PM +1200, Ian McDonald wrote:
With constants for CCID numbers this now uses them in some places.
Signed-off-by: Ian McDonald [EMAIL PROTECTED]
Applied, could have been combined with the previous one tho :)
- Arnaldo
-
To unsubscribe from this list: send the
On Fri, Sep 22, 2006 at 02:28:01PM +1200, Ian McDonald wrote:
This adds DCCP probing shamelessly ripped off from TCP probes by Stephen
Hemminger.
I've put in here support for further CCID3 variables as well.
Andrea/Arnaldo might look to extend for CCID2.
Signed-off-by: Ian McDonald [EMAIL
Erik Mouw wrote:
On Fri, Sep 22, 2006 at 08:20:08AM -0500, Larry Finger wrote:
This patch, which was originally sent to John Linville on 9/14/06,
has been taken against 2.6.18. It changes more lines
than would be absolutely necessary to affect the fix; however, it
ends up with this section
On 9/14/06, Gerrit Renker [EMAIL PROTECTED] wrote:
Sorry kmail garbled this, clean text below.
- Gerrit
Applied, thanks
- Arnaldo
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message to [EMAIL PROTECTED]
More majordomo info at
On 9/22/06, Gerrit Renker [EMAIL PROTECTED] wrote:
[DCCP]: current upper limit for option length is 128 bytes;
much less than actually possible. This patch sets a safe upper
bound on option length of 992, under worst-case assumptions.
This patches against the 2.6.19 tree by David.
Gerrit,
[DCCP]: current upper limit for option length is 128 bytes;
much less than actually possible. This patch sets a safe upper
bound on option length of 992, under worst-case assumptions.
This patches against the 2.6.19 tree by David.
Signed-off-by: Gerrit Renker [EMAIL PROTECTED]
---
dccp.h | 17
Stupid me, it should read 4 * 255 = 1020, not 8 * 255; this is
fixed below.
--
diff --git a/net/dccp/dccp.h b/net/dccp/dccp.h
index 272e858..adabf3e 100644
--- a/net/dccp/dccp.h
+++ b/net/dccp/dccp.h
@@ -40,12 +40,17 @@ extern void dccp_tw_deschedule(struct in
extern void dccp_time_wait(struct
Fix cut/paste error in TCPPROBE help text.
Signed-off-by: Grant Grundler [EMAIL PROTECTED]
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -231,7 +231,7 @@ config NET_TCPPROBE
TCP congestion avoidance modules. If you don't understand
what was just said, you don't need it: say N.
-
On Fri, 22 Sep 2006, Evgeniy Polyakov wrote:
17:45:04.770225 IP 192.168.4.79 192.168.4.78: ESP(spi=0x070635c0,seq=0x1),
length 84
17:45:04.770344 IP 192.168.4.78 192.168.4.79: ESP(spi=0x01f452be,seq=0x2),
length 84
17:45:04.777560 IP 192.168.4.79.ssh 192.168.4.78.56527: P
* Rainer Baumann | 2006-09-22 08:15:13 [+0200]:
Patch for linux kernel 2.6.16.19: http://tcn.hypert.net/tcnKernel_procfs.patch
Coding Style need at least some work ...
Whitespaces around operators and parentheses, useless parentheses, braces for
the else branch, mixes C99/C89 comments,
Andrey Savochkin wrote:
Non-trivial part of socket namespaces: asynchronous events
should be run in proper context.
Signed-off-by: Andrey Savochkin [EMAIL PROTECTED]
---
af_inet.c| 10 ++
inet_timewait_sock.c |8
tcp_timer.c |9 +
3 files
On Fri, 22 Sep 2006, James Morris wrote:
Can you try this with either a fully manual config (setkey only) or
openswan?
Just confirming that everything seems to be working ok with manual config
under 2.6.18 (haven't tried racoon yet).
--
James Morris
[EMAIL PROTECTED]
-
To unsubscribe from
On Thursday 21 September 2006 17:15, Rick Jones wrote:
I was reading David's blog entries on the netdev meeting in Japan, and
have a question about this bit:
Currently, things like Xen have to put the card into promiscuous
mode, accepting all packets, which is quite inefficient.
Is the
On Thu, 14 Sep 2006 12:49:23 +0200, Johannes Berg wrote:
There should be support for notifications, but we need to figure
out if we remove the sysfs based add/remove virtual interface
thing completely or allow the driver to create a notification
through some new API here.
Feel free to remove
On Fri, 22 Sep 2006 13:24:43 +0200
Martin Lucina [EMAIL PROTECTED] wrote:
Hello,
I'm having problems with my sky2 NIC hanging under heavy load. This
appears to be an old problem since it happened for me with 2.6.17 as
well. Upgrading the affected systems to 2.6.18 has not solved the
On Fri, 22 Sep 2006 08:47:10 -0600
Grant Grundler [EMAIL PROTECTED] wrote:
Fix cut/paste error in TCPPROBE help text.
Signed-off-by: Grant Grundler [EMAIL PROTECTED]
--- a/net/Kconfig
+++ b/net/Kconfig
@@ -231,7 +231,7 @@ config NET_TCPPROBE
TCP congestion avoidance modules. If
Alexey Kuznetsov wrote:
Hello!
transactions to data segments is fubar. That issue is also why I wonder
about the setting of tcp_abc.
Yes, switching ABC on/off has visible impact on amount of segments.
When ABC is off, amount of segments is almost the same as number of
transactions. When
On Fri, 22 Sep 2006 08:15:13 +0200
Rainer Baumann [EMAIL PROTECTED] wrote:
Trace Control for Netem: Emulate network properties such as long range
dependency and self-similarity of cross-traffic.
kernel space:
The delay, drop, duplication and corruption values are readout in user space
On Fri, 22 Sep 2006 18:40:00 +1000
Andrew Hall [EMAIL PROTECTED] wrote:
If there's anything I can do to help isolate this issue please let me
know.
Looking through the logs I'm also intermittently seeing some of this:
sky2 eth1: rx error, status 0x7ffc0001 length 844
That means
Hello!
I can't even find a reference to SIOCGSTAMP in the
dhcp-2.0pl5 or dhcp3-3.0.3 sources shipped in Ubuntu.
But I will note that tpacket_rcv() expects to always get
valid timestamps in the SKB, it does a:
It is equally unlikely it uses mmapped packet socket (tpacket_rcv).
I even
On Friday 22 September 2006 17:35, Alexey Kuznetsov wrote:
Hello!
I can't even find a reference to SIOCGSTAMP in the
dhcp-2.0pl5 or dhcp3-3.0.3 sources shipped in Ubuntu.
But I will note that tpacket_rcv() expects to always get
valid timestamps in the SKB, it does a:
It is equally
That came from named. It opens lots of sockets with SIOCGSTAMP.
No idea what it needs that many for.
IIRC ISC BIND named opens a socket for each IP it finds on the system.
Presumeably in this way it knows implicitly the destination IP without
using platform-specific recvfrom/whatever
Andrew Morton wrote:
On Fri, 22 Sep 2006 07:27:18 + (GMT)
Holger Kiehl [EMAIL PROTECTED] wrote:
I get some of the page allocation failure errors. My hardware is 4 CPU
Opteron with one quad + one dual intel e1000 cards. Kernel is plain 2.6.18
and for two cards MTU is set to 9000.
Sep
On 9/23/06, Douglas Leith [EMAIL PROTECTED] wrote:
For those interested in TCP for high-speed environments, and perhaps
also people interested in TCP evaluation generally, I'd like to point
you towards the results of a detailed experimental study which are now
available at:
Stephen,
[EMAIL PROTECTED] said:
You can turn off TX pause and get the same effect.
OK, I'll try that and get back to you.
Is this a dual port on single port card?
Single port, copper media.
-mato
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message to
On Fri, 22 Sep 2006 20:29:25 +0200
Martin Lucina [EMAIL PROTECTED] wrote:
[EMAIL PROTECTED] said:
You can turn off TX pause and get the same effect.
Not sure if I did the right thing, but:
# ifdown eth1
...
# ethtool -A eth1 tx off
# ethtool -a eth1
Pause parameters for eth1:
[EMAIL PROTECTED] said:
You can turn off TX pause and get the same effect.
Not sure if I did the right thing, but:
# ifdown eth1
...
# ethtool -A eth1 tx off
# ethtool -a eth1
Pause parameters for eth1:
Autonegotiate: on
RX: on
TX: off
# ifup eth1
...
sky2 eth1:
[EMAIL PROTECTED] said:
To get tx flow control to turn off. You need a patch (already in netdev-2.6
upstream)
and then you can do:
ethtool -A eth1 autoneg off tx off
Sorry, you've lost me. Which patch? You're saying that turning off tx
flow control will fix the hangs I'm seeing?
On Fri, 22 Sep 2006 20:38:13 +0200
Martin Lucina [EMAIL PROTECTED] wrote:
[EMAIL PROTECTED] said:
To get tx flow control to turn off. You need a patch (already in netdev-2.6
upstream)
and then you can do:
ethtool -A eth1 autoneg off tx off
Sorry, you've lost me. Which patch?
Francois Romieu wrote:
Is there any chance to get linux worked at least as fast, as freebsd-6.1
with small udp packets?
Any help will be more than welcome!
Please, CC: me, due to I'm not subscribed to [EMAIL PROTECTED]
Can you publish dmesg, vmstat 1 and /proc/interrupts somewhere ?
Ian McDonald wrote:
On 9/23/06, Douglas Leith [EMAIL PROTECTED] wrote:
For those interested in TCP for high-speed environments, and perhaps
also people interested in TCP evaluation generally, I'd like to point
you towards the results of a detailed experimental study which are now
available at:
From: Randy Dunlap [EMAIL PROTECTED]
Net devices should depend on NETDEVICES, so revert part of
Paolo's previous patch.
See http://marc.theaimsgroup.com/?l=linux-kernelm=115566326218740w=2
for history.
Signed-off-by: Randy Dunlap [EMAIL PROTECTED]
---
drivers/net/Kconfig |5 +
1 files
I wasn't aware of the planned move to cubic in Linux. Can I ask the
rationale for this ? Cubic is, of course, closely related to HTCP
(borrowing the HTCP idea of using elapsed time since last backoff as the
quantity used to adjust the cwnd increase rate) which *is* tested in the
reported study.
On Sep 22, 2006, at 2:41 AM, YOSHIFUJI Hideaki / 吉藤英明 wrote:
In article [EMAIL PROTECTED] (at
Fri, 22 Sep 2006 02:31:59 -0500), William Pitcock
[EMAIL PROTECTED] says:
This patch allows for a user to disable the requirement to meet the
CAP_NET_BIND_SERVICE capability for a non-superuser.
How is this different than just having the application mmap() the file
and recv() into that buffer?
Ashwini Kulkarni wrote:
My name is Ashwini Kulkarni and I have been working at Intel Corporation for
the past 4 months as an engineering intern. I have been working on the 'TCP
socket splice'
From: William Pitcock [EMAIL PROTECTED]
Date: Fri, 22 Sep 2006 03:27:22 -0500
* The software is untrusted by the end user, in the event that the
software is not trustworthy, the amount of damage it can do running
as a normal user is less than as a superuser. As it is, the bind()
may
On Wed, 20 Sep 2006 13:35:47 +0400
Evgeniy Polyakov [EMAIL PROTECTED] wrote:
Generic event handling mechanism.
Consider for inclusion.
Ulrich's objections sounded substantial, and afaik remain largely
unresolved. How do we sort this out?
-
To unsubscribe from this list: send the line
http://bugzilla.kernel.org/show_bug.cgi?id=7018
Looking at this, it seems that when NFS hands UDP a fragmented (non-linear skb),
it traps because it tries to trim bytes off a non-linear skb. Not sure if
skb_trim
needs fixing, or IP is broken..
---
ssize_t ip_append_page(struct sock *sk, struct
From: Alexey Kuznetsov [EMAIL PROTECTED]
Date: Tue, 5 Sep 2006 13:05:30 +0400
Look into old rfc2401, search for word fragment.
Then search for the same word in new rfc4301. All those 100K of new text
deal with various design bugs in IPsec, mostly with pathologies encountered
in the case of
From: Stephen Hemminger [EMAIL PROTECTED]
Date: Fri, 22 Sep 2006 13:35:12 -0700
http://bugzilla.kernel.org/show_bug.cgi?id=7018
Looking at this, it seems that when NFS hands UDP a fragmented
(non-linear skb), it traps because it tries to trim bytes off a
non-linear skb. Not sure if skb_trim
Boris B. Zhmurov [EMAIL PROTECTED] :
[...]
Thanks Francois, after your suggestions, network performance was twice
increased, but I'm still loosing to freebsd:
Ok, now that the Linux kernel is not in heavy debugging mode, it takes
some ~10 us/(sent packet). It is not _too_ far from the usual
From: YOSHIFUJI Hideaki [EMAIL PROTECTED]
Date: Thu, 21 Sep 2006 15:59:52 +0900 (JST)
[IPV6] NDISC: Handle NDP messages to proxied addresses.
[IPV6]: Don't forward packets to proxied link-local address.
[IPV6] NDISC: Avoid updating neighbor cache for proxied address in
receiving
Linux Kernel Mailing List wrote:
diff --git a/crypto/internal.h b/crypto/internal.h
index b110b97..7dc04ef 100644
--- a/crypto/internal.h
+++ b/crypto/internal.h
@@ -67,9 +67,9 @@ static inline void crypto_kunmap(void *v
kunmap_atomic(vaddr, crypto_kmap_type(out));
}
-static inline
this patch is a replacement for d80211-iwlist-fix.patch. Now we
set bss's phymode and freq using rx_status. The low level driver need to
set these values.
Signed-off-by: Mohamed Abbas [EMAIL PROTECTED]
diff --git a/net/d80211/ieee80211_sta.c b/net/d80211/ieee80211_sta.c
index a933d92..ddbaf14
| +case DCCP_SOCKOPT_TX_PACKET_SIZE:
| +if (dmsk-dccpms_tx_ccid != DCCPC_CCID3)
| +err = -EINVAL;
This is unfortunate: it is in the generic dccp code but
works only for CCID 3. Same for DCCP_SOCKOPT_RX_PACKET_SIZE.
I agree with Gerrit. An average packet
Hi, just a couple clarifications and notes. I haven't deeply thought through
the new socket options, but disagree with your objections :)
Gerrit Renker wrote:
| This creates two new socket options DCCP_SOCKOPT_TX_PACKET_SIZE
| and DCCP_SOCKOPT_RX_PACKET_SIZE. DCCP_SOCKOPT_PACKET_SIZE
Please pull from 'upstream-linus' branch of
master.kernel.org:/pub/scm/linux/kernel/git/jgarzik/netdev-2.6.git
upstream-linus
to receive the following updates:
drivers/net/lp486e.c |6 +++---
drivers/net/mv643xx_eth.c |2 +-
2 files changed, 4 insertions(+), 4 deletions(-)
Jeff
Linus, please pull from
master.kernel.org:/pub/scm/linux/kernel/git/roland/infiniband.git for-linus
This tree is also available from kernel.org mirrors at:
git://git.kernel.org/pub/scm/linux/kernel/git/roland/infiniband.git
for-linus
This tree has:
- Add support for iWARP (RDMA over
This is a minor release update to the bridge utilities to fix some
build issues and remove the usage of libsysfs. libsysfs wasn't making the
code easier and it was a pain for other build environments.
Also some error messages and spelling fixes, etc.
Download from:
ACK, but patch does not apply:
Applying 'Spidernet: add ethtool -S (show statistics)'
error: patch fragment without header at line 14: @@ -472,6 +472,7 @@
spider_net_prepare_rx_descr(struct spide
error: patch fragment without header at line 22: @@ -492,6 +493,7 @@
applied
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
ACK patches 1-3, but does not apply to #upstream. Also, it adds
trailing whitespace.
Applying 'bonding: Validate probe replies in ARP monitor'
Adds trailing whitespace.
.dotest/patch:42:
Adds trailing whitespace.
.dotest/patch:46:
error: patch failed: drivers/net/bonding/bond_main.c:1785
applied
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message to [EMAIL PROTECTED]
More majordomo info at http://vger.kernel.org/majordomo-info.html
1 - 100 of 123 matches
Mail list logo