Sargun Dhillon wrote:
I'm getting an error when I try to compile the netunshare tool.
Do you have any ideas to the error:
netunshare.c:50: error: conflicting types for 'unshare'
/usr/include/bits/sched.h:78: error: previous declaration of 'unshare' was here
are you using the netunshare.c found
Eric W. Biederman wrote:
> Benjamin Thery <[EMAIL PROTECTED]> writes:
>
>> Support for network namespaces in mainline is pretty complete for
>> some time now, but there is still this issue with sysfs that prevents
>> more people to use it easily.
>
return err;
+ if (net == &init_net) {
+ err = device_rename(&dev->dev, dev->name);
+ if (err) {
+ memcpy(dev->name, oldname, IFNAMSIZ);
+ return err;
+ }
}
Otherwise,
Acked-b
Support for network namespaces in mainline is pretty complete for
some time now, but there is still this issue with sysfs that prevents
more people to use it easily.
Reminder for those not aware of the netns/sysfs issue:
With network namespaces, the kernel must be able to support net devices
wi
very well with CONFIG_SYSFS_DEPRECATED=y
The filtering of /sys/class/net with CONFIG_SYSFS_DEPRECATED=y is more
difficult to do because in this case /sys/class/net contains
the actual directories (not symlinks).
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
Tested-by: Serge Hallyn <[EMAIL PRO
This small patch renames a couple of net_device variables from
'net' to 'netdev' to avoid confusion with 'struct net' variables
that will be introduced in the next patch.
'net' is commonly used to designate 'struct net' in the rest of
the ne
This patches exports the new 'struct net' net->id value to /proc/net/nsid
file.
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
net/core/net_namespace.c | 50 +++
1 file changed, 50 insertions(+)
Index: net-next-2.6/net/c
re recycled when IDs are freed (unless you
have thousands of netns running on your host).
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
Acked-by: Serge Hallyn <[EMAIL PROTECTED]>
---
include/net/net_namespace.h |2 ++
net/core/net_namespace.c| 24
Hi Mark,
Indeed booting a distro without sysfs can be tricky because, for
example, initrd might need it (eg. Fedora).
IIRC, to boot without sysfs, in addition to the "root=" option, you
have to create some devices nodes in advance in your /dev tree
(mainly the /dev/sda* ones, plus a bunch of oth
Jiri Slaby wrote:
> Any progress here? Nothing seems reached lkml...
Have a look at Greg's tree. :)
He started to merge the last patches today.
Benjamin
>
> Eric W. Biederman napsal(a):
>> Greg KH <[EMAIL PROTECTED]> writes:
>>
>>> I'll add it to my tree, and get it some testing in the next cyc
here:
git://git.kernel.org/pub/scm/linux/kernel/git/gregkh/patches.git
It contains the patches serie to be applied on top of
linux-2.6 (with 'quilt' for example).
Benjamin
>
> Thanks,
>
> Regards,
> DS
>
>
> On Tue, Sep 2, 2008 at 8:01 PM, Greg KH <[EMAIL
>
> On Wed, Aug 27, 2008 at 6:18 PM, Benjamin Thery <[EMAIL PROTECTED]> wrote:
>> Eric W. Biederman wrote:
>>> The problem. When implementing a network namespace I need to be able
>>> to have multiple network devices with the same name. Currently this
>>
nterfaces to give us anything
> to discriminate on which makes it useless, and there are
> no potential users which makes it an uninteresting problem
> to solve.
>
> Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
> Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
Vivek Goyal wrote:
> Hi All,
>
> I thought of playing with containers but I can't find any instructions
> or any user space bits which tell me how to setup and launch containers.
>
> Any pointers will help...
If you want to play with network namespaces, you can have a look at the
howto there:
ht
Eric W. Biederman wrote:
> Daniel Lezcano <[EMAIL PROTECTED]> writes:
>
>> The kobject events are sent through a netlink message which is not currently
>> per
>> network namespace. Shouldn't be useful to have a way to retrieve from the
>> kobject the network namespace or the uevent socket associa
Quoting Dave Hansen <[EMAIL PROTECTED]>:
> On Wed, 2008-06-18 at 13:12 -0700, Eric W. Biederman wrote:
>>
>> - The data in sysfs fundamentally changes behind the back of the
>> VFS and we need to keep the VFS in sync. Essentially this is the
>> distributed filesystem problem.
>>
>> - In parti
.
Run two instances of sockudp, one server and one client.
Checkpoint the server.
Stop both instances.
Restart server. He will wait for more messages.
Run a new client. The server will continue to sum the data it receives!
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
tests/Ma
Eric W. Biederman wrote:
> Dave Hansen <[EMAIL PROTECTED]> writes:
>
>> On Wed, 2008-06-18 at 19:07 +0200, Benjamin Thery wrote:
>>> To support mounting multiple instances of sysfs occassionally I
>>> need to walk through all of the currently present sysfs
with this
namespace.
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/mount.c | 21 +
include/linux/sysfs.h |8
2 files changed, 29 insertions(+)
Index: linux-mm/fs/sysfs/m
happen on devices whose classes have
tag_ops that they work properly.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
drivers/base/class.c | 30 ++
drivers/base/core
y see a limited set of
the network devices.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/mount.c |8
include/linux/sysfs.h |2 ++
net/Kconfig |2 +-
net/core/ne
/ entries by user
namespace.
Signed-off-by: Serge Hallyn <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/mount.c|3 +++
include/linux/sched.h |1 +
include/linux/sysfs.h |2 ++
kernel/user.c | 21 +
inate on which makes it useless, and there are
no potential users which makes it an uninteresting problem
to solve.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/bin.c|
gned-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/symlink.c| 31 +++
include/linux/sysfs.h | 17 +
2 files changed, 48 insertions(+)
Index: linux-mm/fs/sysfs/symlink.c
sysfs: sysfs_chmod_file handle multiple superblocks
Teach sysfs_chmod_file how to handle multiple sysfs
superblocks.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/f
sysfs: Rename Support multiple superblocks
This patch modifies the sysfs_rename_dir and sysfs_move_dir routines
to support multiple sysfs dentry tries rooted in different
sysfs superblocks.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL
sysfs: Implement __sysfs_get_dentry
This function is similar but much simpler to sysfs_get_dentry
returns a sysfs dentry if one curently exists.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/
sysfs: sysfs_get_dentry add a sb parameter
In preparation for multiple mounts of sysfs add a superblock parameter to
sysfs_get_dentry.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/dir.c | 12 +++-
fs/
surrounded by these no more sysfs super blocks
will be either created or destroyed.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/mount.c | 79 +--
fs/sysfs/sy
Greg, Andrew,
Here is yet another updated version of Eric Biederman's patchset to
implement tagged directories in sysfs ported on top of 2.6.26-rc5-mm3.
This may be easier for you to review as 2.6.26-rc2-mm1 was getting a
bit old now.
There is no major changes since the last version which intro
on
Simple test programs for UDP sockets available upon request :)
Regards,
Benjamin
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
cr.c | 93 ---
1 file changed, 73 insertions(+), 20 deletions(-)
Ind
Dave Hansen <[EMAIL PROTECTED]> a écrit :
> On Mon, 2008-06-16 at 17:38 +0200, Benjamin Thery wrote:
>> Today, I played a bit with inet sockets and cryo. With few modifications
>> I managed to "checkpoint-restart" programs that use UDP sockets to listen
>> and
ve/restore socket type: we can checkpoint UDP sockets too.
Simple test programs for UDP sockets available upon request :)
Regards,
Benjamin
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
cr.c | 63 ++-
1 file changed, 50 i
Benjamin Thery wrote:
> I found the cause of one of the general protection faults I saw with
> my test program and I finally managed to completely restart (a very
> dumb) program for the first time!
>
> My program was failing (GPF) at restart in glibc code. After some
> deb
I found the cause of one of the general protection faults I saw with
my test program and I finally managed to completely restart (a very
dumb) program for the first time!
My program was failing (GPF) at restart in glibc code. After some
debugging I found the failures occur on SINGLE_THREAD_P cal
This is the dumb test program I managed to restart after I re-enabled
the checkpointing of thread area stuff.
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
--
---
tests/Makefile |2 +-
tests/compute.c | 17 +
2 files changed, 18 insertions(+), 1 deletion(-)
This patch re-enable the code that checkpoints (and restore) and thread
area (ldt) using ptrace_get_thread_area(). This is seem to improve the
situation a lot on systems with NPTL: it solved one of the general
protection fault I had when restarting a program.
Signed-off-by: Benjamin Thery
Fix missing initialization in save_msgq_data() that triggers a segfault
in realloc() when cr is compiled with -g.
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
container_ckpt.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Index: cryodev/container_
This tiny patch fixes the handling of option -l.
'l' must be added to the getopt() call and the subsequent switch{] if
we don't want to exit too soon.
This makes the following command works:
$ cr -l -t test.cryo
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
cr
27;s style :) ).
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
cr.c |2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Index: cryodev/cr.c
===
--- cryodev.orig/cr.c
+++ cryodev/cr.c
@@ -1234,7 +1234,7 @@ int mai
Serge,
I've cloned your cryo tree and I'm starting to play with it.
Here are a couple of trivial patches to fix some small issues
with options handling: -l (list) and -f in checkpoint mode.
Still trying to restart the 'sleep' test program.
Benjamin
--
___
y see a limited set of
the network devices.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/mount.c |8
include/linux/sysfs.h |2 ++
net/Kconfig |2 +-
net/core/ne
sysfs: Implement __sysfs_get_dentry
This function is similar but much simpler to sysfs_get_dentry
returns a sysfs dentry if one curently exists.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/
/ entries by user
namespace.
Signed-off-by: Serge Hallyn <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/mount.c |3 +++
include/linux/sched.h |1 +
include/linux/sysfs.h |2 ++
include/linux/user_namespace.h |
sysfs: Rename Support multiple superblocks
This patch modifies the sysfs_rename_dir and sysfs_move_dir routines
to support multiple sysfs dentry tries rooted in different
sysfs superblocks.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL
with this
namespace.
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/mount.c | 21 +
include/linux/sysfs.h |8
2 files changed, 29 insertions(+)
Index: linux-mm/fs/sysfs/m
gned-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/symlink.c| 31 +++
include/linux/sysfs.h | 17 +
2 files changed, 48 insertions(+)
Index: linux-mm/fs/sysfs/symlink.c
happen on devices whose classes have
tag_ops that they work properly.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
drivers/base/class.c | 30 ++
drivers/base/core
sysfs: sysfs_chmod_file handle multiple superblocks
Teach sysfs_chmod_file how to handle multiple sysfs
superblocks.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/f
inate on which makes it useless, and there are
no potential users which makes it an uninteresting problem
to solve.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/bin.c|
sysfs: sysfs_get_dentry add a sb parameter
In preparation for multiple mounts of sysfs add a superblock parameter to
sysfs_get_dentry.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/dir.c | 12 +++-
fs/
surrounded by these no more sysfs super blocks
will be either created or destroyed.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
fs/sysfs/mount.c | 79 +--
fs/sysfs/sy
Greg,
Here is an updated version of the sysfs tagged directories that improves
a bit the situation over the previous one.
I've modified the patch 09 ("Enable tagging for net_class directories in
sysfs") to be a bit less intrusive in sysfs core. I removed the #ifdef'd
parts you didn't like in fs
Serge E. Hallyn wrote:
> Quoting Cedric Le Goater ([EMAIL PROTECTED]):
>> Hello !
>>
>> Serge E. Hallyn wrote:
>>> Hi,
>>>
>>> the proposed agenda for the mini-summit is up (and editable)
>>> at http://wiki.openvz.org/Containers/Mini-summit_2008. Please
>>> take a look and edit or email the list w
Pavel Emelyanov wrote:
> Benjamin Thery wrote:
>> Eric, Pavel,
>>
>> I haven't followed everything about the sysfs/netns issue recently and I
>> think I have missed some of its recent developments in the past weeks.
>> So I'm wondering what is the cu
Eric, Pavel,
I haven't followed everything about the sysfs/netns issue recently and I
think I have missed some of its recent developments in the past weeks.
So I'm wondering what is the current status? Is anyone still working on it?
As several netns patches for IPv4 and IPv6 have already been m
Daniel Lezcano wrote:
> Rick Jones wrote:
>> Daniel Lezcano wrote:
>>> Hi,
>>>
>>> Some performance tests was made by Benjamin to watch out the impact
>>> of the network namespace. The good news is there is no impact when
>>> used with or without namespaces. That has been checked using a real
>>
Benjamin Thery wrote:
> On Mon, Mar 3, 2008 at 3:55 PM, Pavel Emelyanov <[EMAIL PROTECTED]> wrote:
>> Benjamin Thery wrote:
>> > Daniel Lezcano wrote:
>> >> Hi,
>> >>
>> >> Some performance tests was made by Benjamin to watch out
On Mon, Mar 3, 2008 at 3:55 PM, Pavel Emelyanov <[EMAIL PROTECTED]> wrote:
> Benjamin Thery wrote:
> > Daniel Lezcano wrote:
> >> Hi,
> >>
> >> Some performance tests was made by Benjamin to watch out the impact of
> >> the network names
One more thing about these results: the kernel.
The version used to run these tests was 2.6.25-rc1 from Dave Miller's
net-2.6 tree.
(and I included results from a vanilla 2.6.23.16 as reference)
Benjamin
Daniel Lezcano wrote:
> Hi,
>
> Some performance tests was made by Benjamin to watch out th
Daniel Lezcano wrote:
> Hi,
>
> Some performance tests was made by Benjamin to watch out the impact of
> the network namespace. The good news is there is no impact when used
> with or without namespaces. That has been checked using a real network
> device inside a network namespace.
>
> These
On Jan 31, 2008 3:58 PM, Daniel Lezcano <[EMAIL PROTECTED]> wrote:
> Denis V. Lunev wrote:
> > Here are some preparations and cleanups to enable network device/inet
> > address notifiers inside a namespace.
> >
> > This set of patches has been originally sent last Friday. One cleanup
> > patch fro
Pavel Machek wrote:
Hi!
* Patches 9 to 15 propose to add some functionalities, and thus are
submitted here for RFC, about both the interest and their implementation.
These functionalities are:
- Two new control-commands:
. IPC_SETID: to change an IPC's id.
. IPC_SETALL: beha
Eric W. Biederman wrote:
> Daniel Lezcano <[EMAIL PROTECTED]> writes:
>
>>> Could you please place the struct net *net pointer up by the
>>> network device pointer.
};
>>> I know we need a net pointer in struct rt_table, because it
>>> is a hash table that we can't dynamically allocate so we
Daniel Lezcano wrote:
> Denis V. Lunev wrote:
>> Hello, All!
>>
>> We are completely bite to ground with the current Eric's patchset today
>> by Dave Miller. flowi tagging considered wrong. The same opinion has
>> been received from Alexey Kuznetsov :(
>>
>> So, it seems that we can't push this app
Denis V. Lunev wrote:
> Daniel Lezcano wrote:
>> Denis V. Lunev wrote:
>>> Daniel Lezcano wrote:
Hi all,
while doing ipv6 namespace, we were faced to a problem with the loopback
and the dst_destroy function.
When the network namespace exits, the cleanup function is cal
Pavel Emelyanov wrote:
> Daniel Lezcano wrote:
>> Denis V. Lunev wrote:
>> > Daniel Lezcano wrote:
>> >> Denis V. Lunev wrote:
>> >>> Daniel Lezcano wrote:
>> >>>
>> * the first one is the locking of the network namespace list by
>> rtnl_lock, so from the timer callback we can not
hen initialized the
module tries to register its pernet operations and add them in
the pernet_list. Unfortunately this list is corrupted as its
first entries have been freed at the end of the boot sequence.
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
drivers/net/loopback.c
Eric W. Biederman wrote:
> Benjamin Thery <[EMAIL PROTECTED]> writes:
>
>> Denis V. Lunev wrote:
>>> The patch attached should help. The idea is simple. The "init" should be
>>> called only once without NETNS. Period. No need for any lists.
>>
send it to Dave after the ACK.
Tested on x86_64 with CONFIG_NET_NS=n and y.
It fixes the issue we observed.
Acked-by: Benjamin Thery <[EMAIL PROTECTED]>
> Regards,
> Den
>
> Benjamin Thery wrote:
>> Hello Pavel,
>>
>> I've found a problem with o
Hello Pavel,
I've found a problem with one of your patch related to netns:
* [NETNS] Move some code into __init section when CONFIG_NET_NS=n (v2)
http://www.spinics.net/lists/netdev/msg43310.html
This patch introduces the __net_init/__net_exit/__net_initdata
defines to save some memory when C
Denis V. Lunev wrote:
> Benjamin Thery wrote:
>> Denis V. Lunev wrote:
>>> Daniel Lezcano wrote:
>>>> Oh, by the way, I forgot something important you spotted with the list
>>>> protected by the mutex.
>>>>
>>>> When looking at
Denis V. Lunev wrote:
> Daniel Lezcano wrote:
>> Oh, by the way, I forgot something important you spotted with the list
>> protected by the mutex.
>>
>> When looking at ipv6/fib_hash.c with Benjamin, we need to browse the
>> network namespaces list for the garbage collecting, but we are in an
>> in
Add CLONE_NEWNET flag in sys_unshare() test.
This patch should be merged with commit:
"net: Add network namespace clone & unshare support."
Signed-off-by: Benjamin Thery <[EMAIL PROTECTED]>
---
kernel/fork.c |3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
In
The patchset looks fine to me.
This looks like a nice and simple way to obtain basic IP isolation.
Benjamin
[EMAIL PROTECTED] wrote:
Paul Menage mentionned, a few weeks ago, he wanted a bind filtering
for containers. Here it is :)
The following patches are a proposition to bring IP isolatio
[EMAIL PROTECTED] wrote:
From: Daniel Lezcano <[EMAIL PROTECTED]>
This patch creates the network container subsystem.
It consists for the moment on a single file "network.ipv4".
The interface is pretty simple:
To add an IP address to the container:
echo add AB12 > network.ipv4
Kirill Korotaev wrote:
Acked-By: Kirill Korotaev <[EMAIL PROTECTED]>
Acked-by: Benjamin Thery <[EMAIL PROTECTED]>
Looks good to me.
Compiled and tested on i386.
[EMAIL PROTECTED] wrote:
From: Daniel Lezcano <[EMAIL PROTECTED]>
Doing this makes loopback.c a better exam
Hi Eric,
I've cloned the repository and built the kernel.
Unshare support is broken. It misses a small change in sys_unshare()
to accept the CLONE_NEWNET flag.
I guess it should be added in commit "net: Add network namespace
clone & unshare support." (See attached patch below)
Also, have you re
Following a discussion we had at OLS concerning L2 network namespace
performances and how the new macvlan driver could potentially improve
them, I've ported the macvlan patchset on top of Eric's net namespace
patchset on 2.6.22-rc4-mm2.
A little bit of history:
Some months ago, when we ran some
Eric,
Thanks a lot for the updated patchset.
I haven't look at the details yet, but I've already integrated it to
your network namespaces patchset and ported the whole thing to
2.6.22-rc4-mm2 today.
It is far easier to do when you don't have to think about the sysfs stuff :)
I only performed
Last one, that contains the dirty hack that "fixes" symlinks through
shadow directories in my pacthset.
If you have time to help me port/fix these patches that would be great.
If you already have an updated version pf the patchset that would be
even better.
Thanks for your help.
Benjamin
Inde
Third one.
Benjamin
>From 2ccb66d6173de22e96615f2ca5c99940e3e0 Mon Sep 17 00:00:00 2001
From: Eric W. Biederman <[EMAIL PROTECTED]>
Date: Thu, 5 Apr 2007 11:05:34 -0600
Subject: [PATCH] sysfs: Implement sysfs_delete_link and sysfs_rename_link
When removing a symlink sysfs_remove_link does no
seless, and there are
not potential users which makes it an uniteresting problem
to solve.
Changelog:
- Port to 2.6.21-mm2 on top of GregKH's sysfs patches - Benjamin Thery <[EMAIL PROTECTED]>
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
===
Here are the patches for the sysfs shadow directories ported to
2.6.21-mm2 in case you want to look at them to spot the issues.
They are extracted from the full netns patchset.
Benjamin
--
B e n j a m i n T h e r y - BULL/DT/Open Software R&D
http://www.bull.com
>From c25e55e214c208c2820
Hi Eric,
For the past few weeks, I've been trying to port your netns patchset
on top of 2.6.21-mm2. It took me a lot more time than I first expected
to have something working.
I started the port based on your latest public git repository tag:
"netns/v2.6.21-rc6-netns17".
I met a few diffic
David Miller wrote:
From: Kirill Korotaev <[EMAIL PROTECTED]>
Date: Thu, 07 Jun 2007 12:14:29 +0400
David Miller wrote:
From: Pavel Emelianov <[EMAIL PROTECTED]>
Date: Wed, 06 Jun 2007 19:11:38 +0400
Veth stands for Virtual ETHernet. It is a simple tunnel driver
that works at the link layer
Eric W. Biederman wrote:
> Daniel Lezcano <[EMAIL PROTECTED]> writes:
>
>> Hi,
>>
>> as suggested Rick, I added the Service Demand results to the matrix.
>
> A couple of random thoughts in trying to understand the numbers you are
> seeing.
>
> - Checksum offloading?
>
> You have noted that wi
Eric W. Biederman wrote:
> Daniel Lezcano <[EMAIL PROTECTED]> writes:
[...]
>> * When do you expect to have the network namespace into mainline ?
> My current goal is to finish my rebase against 2.6.linus_lastest in
> the next couple of days after having figured out how to deal with sysfs.
Great
Eric W. Biederman wrote:
> Kirill Korotaev <[EMAIL PROTECTED]> writes:
>
>> Benjamin,
>>
>> checksumming can be optimized out as well.
>> We had an experimental patch for OpenVZ venet device, which adds
>> NETIF_F_LLTX | NETIF_F_HW_CSUM | NETIF_F_SG | NETIF_F_HIGHDMA
>> features to venet device an
Hi,
Yesterday, I applied a patch similar to Kirill's one that skip
skb_cow() in ip_forward when the device is a etun, and it does help a lot.
With the patch the cpu load increase is reduced by 50%. Part of the
problem is "solved".
Here are the figures for netperf:
(Host A -> Host B
Host A i
[EMAIL PROTECTED] (Eric W. Biederman)22/03/2007 14:02 CST Pour : Benjamin Thery <[EMAIL PROTECTED]> cc : "Denis V. Lunev" <[EMAIL PROTECTED]>, Linux Containers <[EMAIL PROTECTED]>, Dave Hansen <[EMAIL PROTECTED]> ccc : Objet : Re: [Devel] Re: [PATCHSET]
Eric W. Biederman wrote:
> "Denis V. Lunev" <[EMAIL PROTECTED]> writes:
>
>> Kirill Korotaev wrote:
>>> if network device inside container has MTU higher then eth0 outside the
>> container,
>>> then packets will get fragmented.
>>> First time to MTU1 inside container and refragmented to MTU2
>>> o
Hello Eric,
See comments about __dev_alloc_name() below.
Regards,
Benjamin
Eric W. Biederman wrote:
> From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
>
> When forcibly changing the network namespace of a device
> I need something that can generate a name for the device
> in the new names
93 matches
Mail list logo
