From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
While C99 converting the ctl_table initializers I realized
that the binary sysctl numbers were in conflict with the binary
values under CTL_KERN. Including CTL_KERN KERN_VERSION as used
by glibc. So I just removed the sysctl
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
With this change the sysctl inodes can be cached and
nothing needs to be done when removing a sysctl table.
For a costk of 2K code we will save about 4K of static tables
(when we remove de from ctl_table) and 70K in proc_dir_entries
t
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
fs/xfs/linux-2.6/xfs_sysctl.c | 258
1 files changed, 180 insertions(+), 78 deletions(-)
diff --git a/fs/xfs/linux-2.6/xfs_sysct
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
The semantic effect of insert_at_head is that it would allow
new registered sysctl entries to override existing sysctl entries
of the same name. Which is pain for caching and the proc interface
never implemented.
I have done an aud
Kirill Korotaev <[EMAIL PROTECTED]> writes:
> Eric, really good job!
>
> Patches: 1-13, 15-24, 26-32, 34-44, 46-49, 52-55, 57 (all except below)
> Acked-By: Kirill Korotaev <[EMAIL PROTECTED]>
>
> 14/59 - minor (extra space)
> 25/59 - minor note
> 33/59 - not sorted sysctl IDs
> 45/59 - typo
>
Kirill Korotaev <[EMAIL PROTECTED]> writes:
> Eric, though I personally don't care much:
> 1. I ask for not setting your authorship/copyright on the code which you just
> copied
> from other places. Just doesn't look polite IMHO.
I can't claim complete ownership of the code, there was plenty of
sion, here is a simple fix for
the fib proc iterator so it works even when there is only a single route
in a routing table.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
net/ipv4/fib_trie.c | 21 -
1 files changed, 16 insertions(+), 5 deletions(-)
diff
While enhancing the neighbour code to handle multiple network
namespaces I noticed that decnet is assuming neigh_parms_alloc
will allways succeed, which is clearly wrong. So handle the
failure.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
net/decnet/dn_dev.c |5 +
1
David Miller <[EMAIL PROTECTED]> writes:
> From: Robert Olsson <[EMAIL PROTECTED]>
> Date: Wed, 24 Jan 2007 22:43:30 +0100
>
>> Yes the case when the trie is just a single leaf got wrong with the
>> iterator and your patchs cures it. I think we have a similar problem
>> with /proc/net/fib_trie
>
>
While enhancing the neighbour code to handle multiple network
namespaces I noticed that decnet is assuming neigh_parms_alloc
will allways succeed, which is clearly wrong. So handle the
failure.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
net/decnet/dn_dev.c | 11
Dmitry Mishin <[EMAIL PROTECTED]> writes:
> On Wednesday 17 January 2007 23:16, Eric W. Biederman wrote:
>> Dmitry Mishin <[EMAIL PROTECTED]> writes:
>>
>> > Added functions and macros required to operate with network namespaces.
>> > They are requ
Ben Greear <[EMAIL PROTECTED]> writes:
> Eric W. Biederman wrote:
>> From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
>>
>> etun is a simple two headed tunnel driver that at the link layer
>> looks like ethernet. It's target audience is communica
Stephen Hemminger <[EMAIL PROTECTED]> writes:
>> +
>> +#define __per_net_start ((char *)0)
>> +#define __per_net_end ((char *)0)
>
> Don't use these use NULL
NULL has the wrong data type. These are compiled out character array
normally generated by the linker script. I'm not even certa
The idea of a network namespace is fundamentally quite simple. We create
a mechanism that from the users perspective allows creation of separate
instances of the network stack. When combined with mechanism like chroot
this results in a much more complete isolation. When seen in the context
of a
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
This is the network namespace from which all which all sockets
and anything else under user control ultimately get their network
namespace parameters.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
include/linux/nspr
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
In combination with the sysfs support I am in the process of merging
with gregkh, creates a separate instance of the /sys/class/net directory
for each network namespace so two devices with the same name do not conflict.
Then a n
er cpu variables for having the linker build and dynamically change
structures for us appears applicable here and a lot less nuisance then
what I did before so I am implementing a tailored version of that
technique as well, and again this makes it very simple to compile the
code out.
Signed-off-b
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Because of the global nature of garbage collection, and
because of the cost of per namespace hash tables
unix_socket_table has been kept global. With a filter
added on lookups so we don't see sockets from the wrong
namespace.
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
etun is a simple two headed tunnel driver that at the link layer
looks like ethernet. It's target audience is communicating
between network namespaces but it is general enough it may
have other uses as well.
Signed-off-by: Eric W
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Before I can enable rtnetlink to work in all network namespaces
I need to be certain that something won't break. So this
patch deliberately disables all of the methods and when they
are audited this extra check can be disabled.
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Every user of the network device notifiers is either a protocol
stack or a pseudo device. If a protocol stack that does not have
support for multiple network namespaces receives an event for a
device that is not in the initial n
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
This patch makes the loopback_dev per network namespace.
The loopback device registers itself as a pernet_device so
we can register the new loopback_dev instance when we add
a new network namespace and so we can unregister the
loopback
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
I haven't a clue if this interface will meet with widespread approval but
at this point it is simple, and very useful.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
net/core/n
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Please note that network devices do not increase the count
count on the network namespace. The are inside the network
namespace and so the network namespace tag is in the nature
of a back pointer and so getting and putting the n
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Sockets need to get a reference to their network namespace,
or possibly a simple hold if someone registers on the network
namespace notifier and will free the sockets when the namespace
is going to be destroyed.
Signed-off-by:
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
The initial network namespace is special and we need to use it for various
things. Probably the biggest initial use will be to ensure code that
can't cope with multiple namespaces only sees the initial network namespace.
For that
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
It turns out after a quick audit that except for removing the checks
there is really nothing to do here.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
net/core/rtnetlink.c | 21 +++--
1 file
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
This patch allows you to create a new network namespace
using sys_clone(...).
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
include/linux/sched.h|1 +
kernel/nsproxy.c | 11 +++
net/core/net
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
In particalure I moved:
/proc/sys/net/core/xfrm_aevent_etime
/proc/sys/net/core/xfrm_aevent_rseqth
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
include/net/xfrm.h |4 ++--
net/core/sysctl_net_c
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
include/linux/socket.h |3 ++-
net/core/sysctl_net_core.c | 16
net/socket.c |7 ---
3 files changed, 14 insertions(+),
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Except for carefully selected pseudo devices all network
interfaces should start out in the initial network namespace.
Ultimately it will be register_netdev that examines what
dev->nd_net is set to and places a device in a network
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
After this patch none of the netlink callback support anything
except the initial network namespace but the rtnetlink infrastructure
now handles multiple network namespaces.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
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 namespace without overwriting the old name.
__dev_alloc_name provides me that functionality.
Signed-off-by:
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Add the config option to enable multiple network namespaces.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
net/Kconfig |7 +++
1 files changed, 7 insertions(+), 0 deletions(-)
diff --git a/net/Kconfig b/net/K
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
net/core/wireless.c | 15 ++-
1 files changed, 10 insertions(+), 5 deletions(-)
diff --git a/net/core/wireless.c b/net/core/wireless.c
index 9036359..d5
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
This is done by making all of the relevant global variables
per network namespace.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
net/packet/af_packet.c | 125 +++-
1 files
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
The user interface is: register_net_sysctl_table and
unregister_net_sysctl_table. Very much like the current
interface except there is an network namespace parameter.
This this any sysctl in the net_root_table and it's
subdire
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
This patch introduces NETIF_F_NETNS_LOCAL a flag to indicate
a network device is local to a single network namespace and
should never be moved. Useful for pseudo devices that we
need an instance in each network namespace (like the lo
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Many of the changes to the network stack will simply be adding a
network namespace parameter to function calls or moving variables
from globals to being per network namespace. When those variables
have initializers that cannot stat
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Support is added for the .data.pernet section where all of
the variables who have a single instance in each network
namespace will live. Every architectures linker script
is modified so is should work.
Summarizing the functions:
net_n
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
Each netlink socket will live in exactly one network namespace,
this includes the controlling kernel sockets.
This patch updates all of the existing netlink protocols
to only support the initial network namespace. Request
by clie
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
This patch modifies every packet receive function
registered with dev_add_pack() to drop packets if they
are not from the initial network namespace, in addition
to ensure consistency of argument passing the unnecessary
device parame
From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
This patch passes in the namespace a new socket should be created in
and has the socket code do the appropriate reference counting. By
virtue of this all socket create methods are touched. In addition
the socket create methods are mo
Stephen Hemminger <[EMAIL PROTECTED]> writes:
> Can all this be a nop if a CONFIG option is not selected?
That is exactly what this infrastructure supports.
What you see is the version that comes into effect when
the CONFIG option is not selected.
>From using an empty structure to replace a poin
Robert Olsson <[EMAIL PROTECTED]> writes:
> David Miller writes:
>
> > > Yes the case when the trie is just a single leaf got wrong with the
> > > iterator and your patchs cures it. I think we have a similar problem
> > > with /proc/net/fib_trie
> >
> > I'm happy to review a fix for that :-)
Auke Kok <[EMAIL PROTECTED]> writes:
> Adrian Bunk wrote:
>> This email lists some known regressions in 2.6.20-rc7 compared to 2.6.19
>> that are not yet fixed in Linus' tree.
>>
>> If you find your name in the Cc header, you are either submitter of one
>> of the bugs, maintainer of an affectected
Auke Kok <[EMAIL PROTECTED]> writes:
> maybe I've been unclear, but here's how e1000 detects link changes:
>
> 1) by checking every 2 seconds in the watchdog by reading PHY registers
> 2) by receiving an interrupt from the NIC with the LSI bit in the interrupt
> control register
>
> if the link is
Auke Kok <[EMAIL PROTECTED]> writes:
> that's explained by a driver change that did that. Since at initialization
> we're
> basically waiting for a link change to tell the stack that we're up, we
> decided
> to change the order to have the hardware fire an LSI interrupt to trigger a
> watchdog r
"Adam Kropelin" <[EMAIL PROTECTED]> writes:
>> Can I get the corresponding lspci -xxx output. I suspect the BIOS
>> did not program the hypertransport MSI mapping capabilities correctly.
>> All it has to do is set the enable but still, occasionally BIOS
>> writers miss the most amazing things.
>
"Adam Kropelin" <[EMAIL PROTECTED]> writes:
> Naive question... Can the pci layer (or e1000) detect that MSI is not enabled
> in
> the hardware and avoid using it in that case? With the number of MSI problems
> showing up it seems risky to assume it's usable on any given platform without
> some s
Jeremy Fitzhardinge <[EMAIL PROTECTED]> writes:
> +++ b/drivers/xen/Kconfig.net
> @@ -0,0 +1,14 @@
> +menu "Xen network device drivers"
> +depends on NETDEVICES && XEN
> +
> +config XEN_NETDEV_FRONTEND
> + tristate "Network-device frontend driver"
> + depends on XEN
> + default
I believe the culprit is 43cb76d91ee85f579a69d42bc8efc08bac560278.
For some reason network devices are now showing up under the pci
device tree, in directories that have something other than network
devices.
# ls -l /sys/class/net/eth0
lrwxrwxrwx 1 root root 0 Feb 17 23:19 /sys/class/net/eth0 ->
Greg KH <[EMAIL PROTECTED]> writes:
> On Sun, Feb 18, 2007 at 08:55:20AM -0700, Eric W. Biederman wrote:
>> User space is allowed to rename network devices to anything any name
>> not currently taken by another network device.
>>
>> However when I now do someth
Greg KH <[EMAIL PROTECTED]> writes:
> We need our own namespace for these devices, and we have it today
> already. Look if you enable CONFIG_SYSFS_DEPRECATED, or on a pre-2.6.19
> machine at what shows up in the pci device directories:
> -r--r--r-- 1 root root 4096 2007-02-18 13:06 vendor
Inte
jamal <[EMAIL PROTECTED]> writes:
> Ok, its a little more than grat arp (which i assumed it was). However,
> the cause-fatale given in section 4 for why you wanna broadcast does
> sound very lame. To quote: You do this in case "two disjoint network
> links are joined".
It isn't quite a gratuitous
Khalid Aziz <[EMAIL PROTECTED]> writes:
> On Thu, 2005-06-30 at 17:52 -0600, Eric W. Biederman wrote:
>
> I have found that I can not walk reboot_notifier list in all cases
> before kexec'ing a new kernel. For instance, when handling INIT on ia64,
> we are running in int
Daniel Lezcano <[EMAIL PROTECTED]> writes:
>> +
>> +mutex_lock(&net_mutex);
>> +err = setup_net(new_net);
>> +if (err)
>> +goto out_unlock;
>>
> Should we "net_free" in case of error ?
Oops. Yes we should.
Thanks.
>> +net_lock();
>> +net_list_append(new_net);
>>
Daniel Lezcano <[EMAIL PROTECTED]> writes:
> Eric W. Biederman wrote:
>> From: Eric W. Biederman <[EMAIL PROTECTED]> - unquoted
>>
>> This patch introduces NETIF_F_NETNS_LOCAL a flag to indicate
>> a network device is local to a single network namespace an
"Michael S. Tsirkin" <[EMAIL PROTECTED]> writes:
>> > Why is neighbour->dev changed here?
>>
>> It holds reference to device and prevents its destruction.
>> If dst is held somewhere, we cannot destroy the device and deadlock
>> while unregister.
>
> BTW, can this ever happen for the loopback dev
David Miller <[EMAIL PROTECTED]> writes:
> From: "Michael S. Tsirkin" <[EMAIL PROTECTED]>
> Date: Mon, 19 Mar 2007 00:42:34 +0200
>> > Hmm. Then the code moving dst->dev to point to the loopback
>> > device will have to be fixed too. I'll post a patch a bit later.
>>
>> Does this look sane (unte
Rusty Russell <[EMAIL PROTECTED]> writes:
> On Sun, 2007-03-18 at 13:08 +0100, Andi Kleen wrote:
>> > The idea is _NOT_ that you go look for references to the paravirt_ops
>> > members structure, that would be stupid and you wouldn't be able to
>> > use the most efficient addressing mode on a give
David Miller <[EMAIL PROTECTED]> writes:
> From: Linus Torvalds <[EMAIL PROTECTED]>
> Date: Mon, 19 Mar 2007 20:18:14 -0700 (PDT)
>
>> > > Please don't subject us to another couple months of hair-pulling only
>> > > to have Linus yank the thing out again, there are certainly more
>> > > useful thi
Linus Torvalds <[EMAIL PROTECTED]> writes:
> On Mon, 19 Mar 2007, Jeremy Fitzhardinge wrote:
>> Actually, it still does need a temp register. The sequence for cli is:
>>
>> mov %fs:xen_vcpu, %eax
>> movb $1,1(%eax)
>
> We should just do this natively. There's been several tests over the
David Miller <[EMAIL PROTECTED]> writes:
> I think the thing to do is to just leave the loopback references
> in place, try to unregister the per-namespace loopback device,
> and that will safely wait for all the references to go away.
Right. The only thing I have found that needs to be changed
Andi Kleen <[EMAIL PROTECTED]> writes:
>> I'm conflicted about the dwarf unwinder. I was off doing other things
>> at the time so I missed the pain, but I do have a distinct recollection of
>> the back traces on x86_64 being distinctly worse the on i386.
>
> The only case were i386 was better wa
Jay Cliburn <[EMAIL PROTECTED]> writes:
> We're trying to track down the source of a problem that occurs whenever
> the atl1 network driver is activated on a 32-bit 2.6.21-rc4 kernel. We
> need help from netdev.
>
> We can load the driver just fine, but whenever we activate the
> network, we see
Jay Cliburn <[EMAIL PROTECTED]> writes:
> On Tue, 27 Mar 2007 14:42:20 -0600
> [EMAIL PROTECTED] (Eric W. Biederman) wrote:
>
> Thanks for replying, Eric. I've added atl1-devel to the cc list.
>
>> Do you have msi's working in 2.6.21-rc4 in the x86_64 ker
Daniel Lezcano <[EMAIL PROTECTED]> writes:
>
> 3. General observations
> ---
>
> The objective to have no performances degrations, when the network
> namespace is off in the kernel, is reached in both solutions.
>
> When the network is used outside the container and the network
Daniel Lezcano <[EMAIL PROTECTED]> writes:
> Eric W. Biederman wrote:
>> Daniel Lezcano <[EMAIL PROTECTED]> writes:
>>
>>> 3. General observations
>>> ---
>>>
>>> The objective to have no performances degrations,
Kirill Korotaev <[EMAIL PROTECTED]> writes:
>> Ideally we can optimize the bridge code or something equivalent to
>> it so that we can take one look at the destination mac address and
>> know which network namespace we should be in. Potentially moving this
>> work to hardware when the hardware s
Benjamin Thery <[EMAIL PROTECTED]> writes:
> 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.
Daniel Lezcano <[EMAIL PROTECTED]> writes:
> Hi,
>
> as suggested Rick, I added the Service Demand results to the matrix.
Thanks. The latency number is interesting and it confirms what we were
seeing looking at cpu usage.
We don't have an inexpesive way to get a packet from the outside world
t
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 with the bridge netfilter support disabled you
Benjamin Thery <[EMAIL PROTECTED]> writes:
> Eric W. Biederman wrote:
>> A couple of random thoughts in trying to understand the numbers you are
>> seeing.
>>
>> - Checksum offloading?
>>
>> You have noted that with the bridge netfilter support
/module/etun/parameters/newif
To destroy a pair of devices:
echo -n 'veth0' > /sys/module/etun/parameters/delif
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
drivers/net/Kconfig | 14 ++
drivers/net/Makefile |1 +
drivers/net/etun.c | 486 +
Daniel Lezcano <[EMAIL PROTECTED]> writes:
> Hi,
>
> Eric Biederman has posted a few weeks ago a RFC-patchset concerning the
> network namespace.
> I ported it to the 2.6.20 kernel and uploaded the patchset to
>
> http://lxc.sourceforge.net/network.php
>
> For the part I had to used (TCP/UDP-IPV4
Ben Greear <[EMAIL PROTECTED]> writes:
> Is there any way to tell for certain if an interface is a etun or not? Maybe
> a file could be found (or not) in sysfs somewhere?
Link for any decent network driver ethtool -i
> Also, how do you find the peer device from user-space? This would be very
Roland Dreier <[EMAIL PROTECTED]> writes:
> > +/*
> > + * The higher levels take care of making this non-reentrant (it's
> > + * called with bh's disabled).
> > + */
> > +static int etun_xmit(struct sk_buff *skb, struct net_device *tx_dev)
>
> You have this comment, but then...
>
> > + dev-
Stephen Hemminger <[EMAIL PROTECTED]> writes:
> Why not implement a true virtual network rather than simple
> tunnel pairs?
Mostly I don't even need etun. It is a cheap way to make up for
the lack of sufficient physical network adapters on the machine.
If you plug in enough network cards into th
Ben Greear <[EMAIL PROTECTED]> writes:
> I guess that will do, but then if you ever change the strings, any user-space
> that is
> depending on this will break or have to be modified with additional cruft. It
> seems
> cleaner to me to have an ioctl or a specific place in /proc or some other
> vi
David Miller <[EMAIL PROTECTED]> writes:
> From: Patrick McHardy <[EMAIL PROTECTED]>
> Date: Mon, 09 Apr 2007 18:58:13 +0200
>
>> It would be nice if someone would finally come up with a generic
>> interface based on netlink (RTM_NEWLINK) instead of adding yet
>> another couple of homegrown interf
Johannes Berg <[EMAIL PROTECTED]> writes:
> Our virtual devices are always associated with a piece of hardware, and
> we really want them to be associated with that at all times, even when
> not UP. Everything else seems like a huge complication if only because
> then we can't have whoever will be
"David S. Miller" <[EMAIL PROTECTED]> writes:
> From: [EMAIL PROTECTED] (Eric W. Biederman)
> Date: Wed, 25 Jan 2006 13:08:53 -0700
>
>> Analysis:
>> Currently tcp_v6_send_synack is never called with a dst entry
>> so dst always comes in as NULL.
>>
Jeff Garzik <[EMAIL PROTECTED]> writes:
> Andi Kleen wrote:
>> There was already talk some time ago to make NAPI drivers use
>> the hardware mitigation again. The reason is when you have
>
>
> This was discussed on the netdev list, and the conclusion was that you want
> both
> NAPI and hw mitigat
Herbert Xu <[EMAIL PROTECTED]> writes:
> On Fri, Jan 27, 2006 at 01:00:49AM -0700, Eric W. Biederman wrote:
>>
>> However I do know I have correctly found the leak.
>
> Yes you are right. The locking/refcounting in addrconf.c is such
> a mess. I've asked a nu
"Christopher Friesen" <[EMAIL PROTECTED]> writes:
> Eric W. Biederman wrote:
>> Jeff Garzik <[EMAIL PROTECTED]> writes:
>
>>> This was discussed on the netdev list, and the conclusion was that
>>> you want both NAPI and hw mitigation. This was
"Leonid Grossman" <[EMAIL PROTECTED]> writes:
> There two facilities (at least, in our ASIC, but there is no reason this
> can't be part of the generic multi-channel driver interface that I will
> get to shortly) to deal with it.
>
> - hardware supports more than one utilization-based interrupt ra
"Ayaz Abdulla" <[EMAIL PROTECTED]> writes:
> Can we remove "Reverse Engineered" and "(EXPERIMENTAL)" now?
Can we get better bit names in the driver?
Having unknown xyz in half of the names is scare?
Can we have some help fixing the problem where the on some boots
the driver cannot receive packets
rate down
to 1 node in 16 but that is still horribly bad.
> I will work on clearing up some of those names.
Thanks.
> -Original Message-
> From: Eric W. Biederman [mailto:[EMAIL PROTECTED]
> Sent: Thursday, March 16, 2006 9:58 AM
> To: Ayaz Abdulla
> Cc: Stephen Hemm
"Ayaz Abdulla" <[EMAIL PROTECTED]> writes:
> Its hard for me to believe that you would think that there is no
> indication of the driver being actively supported. Take a look at the
> revision history in the forcedeth.c file. There have been 18 revisions
> in less then a year timeframe.
There is
Jeff Garzik <[EMAIL PROTECTED]> writes:
> Eric W. Biederman wrote:
>> There is no maintainer contact in the driver nor in the maintainers file.
>
>> It has clearly been bug fixed but simply didn't look like there
>> was a maintainer.
>
> Then you clearly
"Jesse Brandeburg" <[EMAIL PROTECTED]> writes:
> On 9/28/06, Sukadev Bhattiprolu <[EMAIL PROTECTED]> wrote:
>> $ cat /proc/interrupts
>>
>>CPU0 CPU1
>> 28: 0 0 IO-APIC-fasteoi eth0
>> NMI: 96 35
>> LOC: 18251 18524
>> ERR:
tcp_make_synack consume the dst entry so we still have
a dst_entry with a bumped refrence count at the end of
this function.
Therefore we need to call dst_release just before we return
just like tcp_v4_send_synack does.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
net/ipv6/tcp_ipv6.c
still elevated.
Fix this by simply moving rt6_ifdown where we flush the routes after
we have removed all of the addresses from the interface.
Signed-off-by: Eric W. Biederman <[EMAIL PROTECTED]>
---
net/ipv6/addrconf.c |7 ++-
1 files changed, 6 insertions(+),
Herbert Xu <[EMAIL PROTECTED]> writes:
> On Wed, Jan 25, 2006 at 08:12:02PM +0000, Eric W. Biederman wrote:
>>
>> Unfortunately because we have already call rt6_ifdown() the route is
>> not found in the routing table, the dst_free does not decrement the
>> coun
in and those 100 mount namespaces that share with each other
are your real concern. But I don't know.
> On Tue, Jul 05, 2016 at 09:44:00AM -0500, Eric W. Biederman wrote:
>> Phil Sutter writes:
>>
>> > Stress-testing OpenStack Neutron revealed poor performance of
Stephen Hemminger writes:
> On Tue, 5 Jul 2016 16:51:18 +0200
> Phil Sutter wrote:
>
>> Stress-testing OpenStack Neutron revealed poor performance of 'ip netns'
>> when dealing with a high amount of namespaces. The cause of this lies in
>> the combination of how iproute2 mounts NETNS_RUN_DIR an
Rick Jones writes:
> On 07/07/2016 08:48 AM, Phil Sutter wrote:
>> On Thu, Jul 07, 2016 at 02:59:48PM +0200, Nicolas Dichtel wrote:
>>> Le 07/07/2016 13:17, Phil Sutter a écrit :
>>> [snip]
The issue came up during OpenStack Neutron testing, see this ticket for
reference:
http
Rick Jones writes:
> On 07/07/2016 09:34 AM, Eric W. Biederman wrote:
>> Rick Jones writes:
>>> 300 routers is far from the upper limit/goal. Back in HP Public
>>> Cloud, we were running as many as 700 routers per network node (*),
>>> and more than four n
patches are also available at:
git://git.kernel.org/pub/scm/linux/kernel/git/ebiederm/user-namespace.git
for-testing
Changes since v1:
- Compilation issues with !CONFIG_SYSCTL and !CONFIG_USER_NS have been addressed
- Comment improvements.
- A few names have been changed to be clearer.
Eric W
601 - 700 of 1164 matches
Mail list logo