David Miller wrote:
From: Bodo Eggert [EMAIL PROTECTED]
Date: Wed, 16 Aug 2006 02:02:03 +0200
Stephen Hemminger [EMAIL PROTECTED] wrote:
IMHO idiots who put space's in filenames should be ignored. As long as the
bonding code doesn't throw a fatal error, it has every right to return
No
On Tue, 2006-08-15 at 12:29 -0400, Dan Williams wrote:
We might want to take the time to fix up a few of the ambiguities of
WEXT that we've encountered over the past few years:
Yes, I definitely agree.
o Separate attributes for signal strength units; signed integer type for
dBm, unsigned
On Tue, 2006-08-15 at 15:59 -0400, Dan Williams wrote:
Ok, so if somebody magically opens up new unlicensed ISM spectrum
around, say, 7GHz, does that space get broken into channels and assigned
specific numbers by the IEEE?
I know there are stable channel #s for abg range. What about the
On Tue, 2006-08-15 at 12:14 -0400, Luis R. Rodriguez wrote:
Basically redo WE completely from scratch using netlink.
Not quite, I hope! As Dan mentioned, for example all the key management
stuff ought to be consolidated. Same for some other things.
For per packet this makes sense, for
Herbert Xu wrote:
kiran kandi [EMAIL PROTECTED] wrote:
In what context hard_start_xmit function is called. Is it called in soft
irq or a processes context.
softirq
Also can you call kfree_skb in soft irq context.
Yes. Don't do it in hard irq context though.
FWIW there is also
UDPv4 and UDPv6 use an almost identical version of the get_port function,
which is unnecessary since the (long) code differs in only one if-statement.
By disentangling the if statement and adding v4/v6 checks appropriately, this
code duplication can be removed and further
* udp_port_rover can
Hello.
Network tree allocator can be used to allocate memory for all network
operations from any context.
Changes from previous release:
* added dynamically grown cache
* changed some inline issues
* reduced code size
* removed AVL tree implementation from the sources
* changed minimum
Jeff Garzik wrote:
Herbert Xu wrote:
kiran kandi [EMAIL PROTECTED] wrote:
In what context hard_start_xmit function is called. Is it called in
soft irq or a processes context.
softirq
It can be process too...doesn't pktgen call it directly?
Ben
-
To unsubscribe from this list: send
On Wed, Aug 16, 2006 at 09:35:46AM +0400, Evgeniy Polyakov wrote:
On Tue, Aug 15, 2006 at 10:21:22PM +0200, Arnd Bergmann ([EMAIL PROTECTED])
wrote:
Am Monday 14 August 2006 13:04 schrieb Evgeniy Polyakov:
?* full per CPU allocation and freeing (objects are never freed on
Hello.
In article [EMAIL PROTECTED] (at Wed, 16 Aug 2006 08:46:48 +0100), [EMAIL
PROTECTED] says:
UDPv4 and UDPv6 use an almost identical version of the get_port function,
which is unnecessary since the (long) code differs in only one if-statement.
:
:
+#if defined(CONFIG_IPV6) ||
On Wed, Aug 16, 2006 at 09:48:08AM +0100, Christoph Hellwig ([EMAIL PROTECTED])
wrote:
Doesn't alloc_pages() automatically switches to alloc_pages_node() or
alloc_pages_current()?
That's not what's wanted. If you have a slow interconnect you always want
to allocate memory on the node the
From: Evgeniy Polyakov [EMAIL PROTECTED]
Date: Wed, 16 Aug 2006 13:00:31 +0400
So I would like to know how to determine which node should be used for
allocation. Changes of __get_user_pages() to alloc_pages_node() are
trivial.
netdev_alloc_skb() knows the netdevice, and therefore you can
On Wed, Aug 16, 2006 at 02:05:03AM -0700, David Miller wrote:
From: Evgeniy Polyakov [EMAIL PROTECTED]
Date: Wed, 16 Aug 2006 13:00:31 +0400
So I would like to know how to determine which node should be used for
allocation. Changes of __get_user_pages() to alloc_pages_node() are
trivial.
On Wed, Aug 16, 2006 at 10:10:29AM +0100, Christoph Hellwig ([EMAIL PROTECTED])
wrote:
On Wed, Aug 16, 2006 at 02:05:03AM -0700, David Miller wrote:
From: Evgeniy Polyakov [EMAIL PROTECTED]
Date: Wed, 16 Aug 2006 13:00:31 +0400
So I would like to know how to determine which node should
On Wed, Aug 16, 2006 at 01:32:02PM +0400, Evgeniy Polyakov wrote:
On Wed, Aug 16, 2006 at 10:10:29AM +0100, Christoph Hellwig ([EMAIL
PROTECTED]) wrote:
On Wed, Aug 16, 2006 at 02:05:03AM -0700, David Miller wrote:
From: Evgeniy Polyakov [EMAIL PROTECTED]
Date: Wed, 16 Aug 2006 13:00:31
On Wed, Aug 16, 2006 at 01:00:31PM +0400, Evgeniy Polyakov wrote:
On Wed, Aug 16, 2006 at 09:48:08AM +0100, Christoph Hellwig ([EMAIL
PROTECTED]) wrote:
Doesn't alloc_pages() automatically switches to alloc_pages_node() or
alloc_pages_current()?
That's not what's wanted. If you have
From: Christoph Hellwig [EMAIL PROTECTED]
Date: Wed, 16 Aug 2006 10:38:37 +0100
We could, but I'd rather waste 4 bytes in struct net_device than
having such ugly warts in common code.
Why not instead have struct device store some default node value?
The node decision will be sub-optimal on
On Wed, Aug 16, 2006 at 02:40:08AM -0700, David Miller wrote:
From: Christoph Hellwig [EMAIL PROTECTED]
Date: Wed, 16 Aug 2006 10:38:37 +0100
We could, but I'd rather waste 4 bytes in struct net_device than
having such ugly warts in common code.
Why not instead have struct device store
On Wed, Aug 16, 2006 at 01:48:56AM -0700, Ben Greear wrote:
softirq
It can be process too...doesn't pktgen call it directly?
Only with BH disabled.
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmVHI~} [EMAIL PROTECTED]
Home Page: http://gondor.apana.org.au/~herbert/
PGP
Library function to create a seq_file in proc filesystem,
showing some information for each netdevice.
This code is present in the kernel in about 10 instances, and
all of them can be converted to using introduced library function.
Signed-off-by: Andrey Savochkin [EMAIL PROTECTED]
---
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 changed, 27 insertions(+)
Hi All,
I'd like to resurrect our discussion about network namespaces.
In our previous discussions it appeared that we have rather polar concepts
which seemed hard to reconcile.
Now I have an idea how to look at all discussed concepts to enable everyone's
usage scenario.
1. The most
Structures related to IPv4 rounting (FIB and routing cache)
are made per-namespace.
Signed-off-by: Andrey Savochkin [EMAIL PROTECTED]
---
include/linux/net_ns.h | 10 +++
include/net/flow.h |3 +
include/net/ip_fib.h | 46
net/core/dev.c |8 ++
CONFIG_NET_NS and net_namespace structure are introduced.
List of network devices is made per-namespace.
Each namespace gets its own loopback device.
Signed-off-by: Andrey Savochkin [EMAIL PROTECTED]
---
drivers/net/loopback.c| 69 -
include/linux/init_task.h |9 ++
Temporary code to play with network namespaces in the simplest way.
Do
exec 7 /proc/net/net_ns
in your bash shell and you'll get a brand new network namespace.
There you can, for example, do
ip link set lo up
ip addr list
ip addr add 1.2.3.4 dev lo
ping -n
Socket hash lookups are made within namespace.
Hash tables are common for all namespaces, with
additional permutation of indexes.
Signed-off-by: Andrey Savochkin [EMAIL PROTECTED]
---
include/linux/ipv6.h |3 ++-
include/net/inet6_hashtables.h |6 --
Destructor field added proc_dir_entries,
standard destructor kfree'ing data introduced.
Signed-off-by: Andrey Savochkin [EMAIL PROTECTED]
---
fs/proc/generic.c | 10 --
fs/proc/root.c |1 +
include/linux/proc_fs.h |4
3 files changed, 13 insertions(+), 2
A simple device to pass packets between a namespace and its child.
Signed-off-by: Andrey Savochkin [EMAIL PROTECTED]
---
Makefile |3
veth.c | 327 +++
2 files changed, 330 insertions(+)
--- ./drivers/net/Makefile.veveth
Temporary code to debug and play with pass-through device.
Create device pair by
modprobe veth
echo 'add veth1 0:1:2:3:4:1 eth0 0:1:2:3:4:2' /proc/net/veth_ctl
and your shell will appear into a new namespace with `eth0' device.
Configure device in this namespace
ip l s eth0
* Herbert Xu [EMAIL PROTECTED] 2006-08-16 12:58
I'm not comfortable with that change since it implies the message
originated from a user-space process.
The netlink header pid is really akin to sadb_msg_pid from RFC 2367.
IMHO it should always be zero if the kernel is the originator of the
On Wed, 2006-16-08 at 12:58 +0200, Thomas Graf wrote:
* Herbert Xu [EMAIL PROTECTED] 2006-08-16 12:58
I'm not comfortable with that change since it implies the message
originated from a user-space process.
The netlink header pid is really akin to sadb_msg_pid from RFC 2367.
IMHO it
Hi Thomas:
On Wed, Aug 16, 2006 at 12:58:56PM +0200, Thomas Graf wrote:
All route and tc notifications already use the pid so applications
can decide whether the event was caused by them. A notification
is a reply to a request so it doesn't even violate RFC 2367.
Actually most IPv4
* Herbert Xu [EMAIL PROTECTED] 2006-08-16 21:12
On Wed, Aug 16, 2006 at 12:58:56PM +0200, Thomas Graf wrote:
All route and tc notifications already use the pid so applications
can decide whether the event was caused by them. A notification
is a reply to a request so it doesn't even
On Wed, Aug 16, 2006 at 09:12:40PM +1000, herbert wrote:
Any notification that sets the netlink pid to current-pid is
*completely* bogus. Let me repeat this, the netlink pid is not
a process ID.
BTW, I'm not having a go at either Thomas or Jamal. You guys
are oo the same side for once :).
On Wednesday 16 August 2006 11:00, Evgeniy Polyakov wrote:
There is drawback here - if data was allocated on CPU wheere NIC is
closer and then processed on different CPU it will cost more than
in case where buffer was allocated on CPU where it will be processed.
But from other point of
Hi John,
Please pull the patch to add Larry as bcm43xx-softmac
maintainer into wireless-dev and _after_ that please apply
the following patch to mark the d80211 branch explicitely.
--
Add MAINTAINERS for bcm43xx-d80211
Signed-off-by: Michael Buesch [EMAIL PROTECTED]
Index:
Quoting Andrey Savochkin ([EMAIL PROTECTED]):
Hi All,
I'd like to resurrect our discussion about network namespaces.
In our previous discussions it appeared that we have rather polar concepts
which seemed hard to reconcile.
Now I have an idea how to look at all discussed concepts to enable
On Wed, Aug 16, 2006 at 01:40:03PM +0200, Thomas Graf wrote:
It was added to help quagga identify which route modifications
are self caused. It's not possible to use rtm_protocol for this
purpose as other applications can delete routes added by quagga.
Actually it's not that bad. I just
On Wed, 2006-16-08 at 21:39 +1000, Herbert Xu wrote:
So let's step back a bit and think about where does this pid really
come from. The field in question is nlmsg_pid. Its primary purpose
is to identify unicast transactions along with the field nlmsg_seq.
It was not designed to identify the
On Wed, 2006-16-08 at 14:05 +0200, Thomas Graf wrote:
Right, but he forgot the bits in IPv6 which I now fixed. The
changeset introducing those current-pid uses was definitely
simply wrong. I'm not questioning that :)
Herbert, if you look at the thread as well I am no longer questioning
that
* Herbert Xu [EMAIL PROTECTED] 2006-08-16 21:57
On Wed, Aug 16, 2006 at 01:40:03PM +0200, Thomas Graf wrote:
It was added to help quagga identify which route modifications
are self caused. It's not possible to use rtm_protocol for this
purpose as other applications can delete routes
* jamal [EMAIL PROTECTED] 2006-08-16 08:04
current-pid i think is coming out to be a bad idea. Thomas' patches
revert it out. Again this has everything to do with the original idea
what maps to pid now changing to socketid.
It probably developed from autobind using current-tid.
-
To
On Wed, Aug 16, 2006 at 08:04:24AM -0400, jamal wrote:
What do you think of the idea of infact rewriting the pid to be that of
the socket id?
Rewriting it with the netlink socket address? That's fine by me as
long as there is a clear 1-to-1 relationship between the request
and the
* Herbert Xu [EMAIL PROTECTED] 2006-08-16 21:39
For a broadcast notification, the nlmsg_pid field is meaningless
because the nlmsg_seq field is also meaningless. I'm not denying
that it wouldn't be useful to have the originator's socket address
in there. What I'm saying is that it's the
On Wed, 2006-16-08 at 14:08 +0200, Thomas Graf wrote:
* jamal [EMAIL PROTECTED] 2006-08-16 08:04
current-pid i think is coming out to be a bad idea. Thomas' patches
revert it out. Again this has everything to do with the original idea
what maps to pid now changing to socketid.
It
Hello!
The netlink header pid is really akin to sadb_msg_pid from RFC 2367.
IMHO it should always be zero if the kernel is the originator of the
message.
No. Analogue of sadb_msg_pid is nladdr.nl_pid.
Netlink header pid is not originator of the message, but author of
the change. The notion
Hey,
I just noticed that my rt2500usb device turns on the link LED when I
just add an active monitor interface. I can't imagine that being on
purpose, but I'm not sure based on what it is controlled.
johannes
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a
Hello!
In one conversation with Alexey he told me there was some inspiration
from pfkey in the semantics of it i.e processid.
Inspiration, but not a copy. :-)
Unlike pfkeyv2 it uses addressing usual for networking i.e.
struct sockaddr_nl.
Alexey
-
To unsubscribe from this list: send the line
On Wed, 2006-16-08 at 17:04 +0400, Alexey Kuznetsov wrote:
Hello!
In one conversation with Alexey he told me there was some inspiration
from pfkey in the semantics of it i.e processid.
Inspiration, but not a copy. :-)
Oh, absolutely. Netlink is way superior. I should have said
On Tue, 2006-15-08 at 22:59 -0400, jamal wrote:
How about moving
it to linux/jiffies.h and rewrite in the style of msec_to_jiffies?
Is there something other than the boundary check already done you
foresee being made? If yes, do you wanna take a crack at it?
Herbert, I actually dont
poll/select() notifications. Timer notifications.
This patch includes generic poll/select and timer notifications.
kevent_poll works simialr to epoll and has the same issues (callback
is invoked not from internal state machine of the caller, but through
process awake).
Timer notifications can
Core files.
This patch includes core kevent files:
- userspace controlling
- kernelspace interfaces
- initialization
- notification state machines
Signed-off-by: Evgeniy Polyakov [EMAIL PROTECTED]
diff --git a/arch/i386/kernel/syscall_table.S b/arch/i386/kernel/syscall_table.S
index
Hi Yoshifuji,
| +#if defined(CONFIG_IPV6) || defined(CONFIG_IPV6_MODULE)
| + else if(sk-sk_family == PF_INET6
| + ipv6_rcv_saddr_equal(sk, sk2) )
| + goto fail;
| + }
|
On Mon, Aug 14, 2006 at 10:21:36AM +0400, Evgeniy Polyakov wrote:
Generic event handling mechanism.
Hi, I've just started looking into this, so some comments here first
on the submission process:
- could you send new revisions of the patches in a new thread so one can
easily find them?
-
On Mon, Aug 14, 2006 at 10:21:36AM +0400, Evgeniy Polyakov wrote:
poll/select() notifications. Timer notifications.
This patch includes generic poll/select and timer notifications.
kevent_poll works simialr to epoll and has the same issues (callback
is invoked not from internal state
Giacomo A. Catenazzi ([EMAIL PROTECTED]) said:
Are you willing to work to add the special case code necessary to
handle whitespace characters in the device name over all of the kernel
code and also all of the userland tools too?
But if you don't handle spaces in userspace, you handle *,
On Wed, Aug 16, 2006 at 02:26:31PM +0100, Christoph Hellwig ([EMAIL PROTECTED])
wrote:
On Mon, Aug 14, 2006 at 10:21:36AM +0400, Evgeniy Polyakov wrote:
Generic event handling mechanism.
Hi, I've just started looking into this, so some comments here first
on the submission process:
On Wed, Aug 16, 2006 at 02:30:14PM +0100, Christoph Hellwig ([EMAIL PROTECTED])
wrote:
On Mon, Aug 14, 2006 at 10:21:36AM +0400, Evgeniy Polyakov wrote:
poll/select() notifications. Timer notifications.
This patch includes generic poll/select and timer notifications.
kevent_poll
diff --git a/include/linux/kevent.h b/include/linux/kevent.h
new file mode 100644
index 000..03a
--- /dev/null
+++ b/include/linux/kevent.h
@@ -0,0 +1,310 @@
+/*
+ * kevent.h
Please don't put filenames in the top of file block comments. They're
redudant and as history shows out
The sysfs attributes add_iface and remove_iface both check for
CAP_NET_ADMIN whenever something is written. Hence, permissions for the
files should be relaxed so that someone who is not root but happens to
have CAP_NET_ADMIN can do things.
Signed-off-by: Johannes Berg [EMAIL PROTECTED]
---
Hi
I'm not sure the second one is quite right. The case of concern
is where an interface is deleted. If you joined (or left) the group by
address and then deleted the interface, then you wouldn't match the
index (which wouldn't be set) so the leave wouldn't work, still.
That's right I
On Tue, 2006-08-15 at 18:48 +0400, Andrey Savochkin wrote:
/* Can survive without statistics */
stats = kmalloc(sizeof(struct net_device_stats), GFP_KERNEL);
if (stats) {
memset(stats, 0, sizeof(struct net_device_stats));
-
The same patch as in previous e-mail with a few typos in comments corrected:
Signed-off-by: Michal Ruzicka [EMAIL PROTECTED]
--- linux-2.6.17.8/net/ipv4/igmp.c.orig 2006-08-11 11:50:46.0 +0200
+++ linux-2.6.17.8/net/ipv4/igmp.c 2006-08-16 16:53:08.0 +0200
@@ -1369,13
I Recently built a 2.6.17.7 and wanted to turn off CONFIG_NET_ESTIMATOR
but can't using menuconfig.
Is it on by default now, or is it a config issue?
I wanted it off to play with chains of policers and unless I
misunderstand it uses Hz, and is inaccurate when Hz=250 with its'
minimum time
Hello!
(application) containers. Performance aside, are there any reasons why
this approach would be problematic for c/r?
This approach is just perfect for c/r.
Probably, this is the only approach when migration can be done
in a clean and self-consistent way.
Alexey
-
To unsubscribe from
John,
Please pull this patch for the wireless-2.6 tree.
This patch depends on the 64bit DMA patch, which is already
submitted for inclusion.
Convert the bitfields in the bcm43xx DMA code to properly
aligned u8 booleans. These flags are accessed in the DMA
hotpath, so it's a good idea to waste a
On Wed, 2006-08-16 at 10:36 -0500, Larry Finger wrote:
+ /* Boolean. Is this a TX ring? */
+ u8 tx
+ /* Boolean. 64bit DMA if true, 32bit DMA otherwise. */
+ u8 dma64;
does that compile?
johannes
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of
Make functions and constants generic, add support for two more
SMSC PHY models with identical interrupt source and mask registers
Signed-off-by: Steve Glendinning [EMAIL PROTECTED]
---
drivers/net/phy/smsc.c | 112 ++--
1 files changed, 89
Kok, Auke-jan H wrote:
Auke Kok wrote:
Jay Vosburgh wrote:
Running both 2.6.17.6 plus the e1000 7.2.7 from sourceforge, or
the e1000 in netdev-2.6#upstream (7.1.9-k4).
Starting up ethtool -p ethX then unplugging the cable
connected to the identified port is causing my system to
Trivial style fixes to match other PHY drivers
Signed-off-by: Steve Glendinning [EMAIL PROTECTED]
---
drivers/net/phy/smsc.c | 15 +++
1 files changed, 7 insertions(+), 8 deletions(-)
diff --git a/drivers/net/phy/smsc.c b/drivers/net/phy/smsc.c
index 25e31fb..2119bd7 100644
---
On Wed, 16 Aug 2006 07:46:43 -0700
Dave Hansen [EMAIL PROTECTED] wrote:
On Tue, 2006-08-15 at 18:48 +0400, Andrey Savochkin wrote:
/* Can survive without statistics */
stats = kmalloc(sizeof(struct net_device_stats), GFP_KERNEL);
if (stats) {
On Tue, 15 Aug 2006 18:48:43 +0400
Andrey Savochkin [EMAIL PROTECTED] wrote:
Temporary code to play with network namespaces in the simplest way.
Do
exec 7 /proc/net/net_ns
in your bash shell and you'll get a brand new network namespace.
There you can, for example, do
ip link
On Mon, 14 Aug 2006 10:12:01 +0200, Johannes Berg wrote:
I'd like to see a link from the wiphy to the master interface that
belongs to it so one can tell this easily on systems that have multiple
wireless devices.
As wiphy and master interface are closely bind to each other, this makes
On Tue, 8 Aug 2006 16:30:12 +0200, Ivo van Doorn wrote:
This will add the rfkill driver to the input/misc section of the kernel.
rfkill is usefull for newtwork devices that contain a hardware button
to enable or disable the radio.
With rfkill a generic interface is created for the network
Stephen Hemminger [EMAIL PROTECTED] writes:
On Tue, 15 Aug 2006 18:48:43 +0400
Andrey Savochkin [EMAIL PROTECTED] wrote:
Temporary code to play with network namespaces in the simplest way.
Do
exec 7 /proc/net/net_ns
in your bash shell and you'll get a brand new network namespace.
Alexey Kuznetsov [EMAIL PROTECTED] writes:
Hello!
(application) containers. Performance aside, are there any reasons why
this approach would be problematic for c/r?
This approach is just perfect for c/r.
Yes. For c/r you need to take your state with you.
Probably, this is the only
Stephen,
the reproducible crashes with all skge versions (where sk98lin works
fine) on my box are SMP related.
I booted with maxcpus=1 and the box survived my usual crash test, I will
keep an eye.
Daniel
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message
I'd suggest that the new signal strength measure should be defined as
'RCPI' - the 'Received Channel Power Indicator' - which is defined in
IEEE 802.11k (the Radio Measurements amendment to 802.11). Here is the
full text of the definition from 802.11k draft 5.0:
received channel power indicator
On Wed, 16 Aug 2006 19:47:08 +0200
Beschorner Daniel [EMAIL PROTECTED] wrote:
Stephen,
the reproducible crashes with all skge versions (where sk98lin works
fine) on my box are SMP related.
I booted with maxcpus=1 and the box survived my usual crash test, I will
keep an eye.
Daniel
Is
On Tue, 15 Aug 2006 18:23:19 -0500
Jay Cliburn [EMAIL PROTECTED] wrote:
Stephen Hemminger wrote:
On Sun, 13 Aug 2006 19:11:42 -0500
Jay Cliburn [EMAIL PROTECTED] wrote:
...snip...
I've read the LKML FAQ regarding new driver submissions, but it implies
that the submitter be willing to
On Wed, 16 Aug 2006 11:29:00 +1000
Herbert Xu [EMAIL PROTECTED] wrote:
On Tue, Aug 15, 2006 at 11:29:59AM -0700, [EMAIL PROTECTED] wrote:
http://bugzilla.kernel.org/show_bug.cgi?id=6936
It's actually a bug in the bridging code :)
[BRIDGE]: Disable SG/GSO if TX checksum is off
When the
Brandeburg, Jesse [EMAIL PROTECTED] wrote:
Kok, Auke-jan H wrote:
Auke Kok wrote:
Jay Vosburgh wrote:
Running both 2.6.17.6 plus the e1000 7.2.7 from sourceforge, or
the e1000 in netdev-2.6#upstream (7.1.9-k4).
Starting up ethtool -p ethX then unplugging the cable
connected to the
Stephen Hemminger wrote:
On Tue, 15 Aug 2006 18:23:19 -0500
Jay Cliburn [EMAIL PROTECTED] wrote:
Stephen Hemminger wrote:
On Sun, 13 Aug 2006 19:11:42 -0500
Jay Cliburn [EMAIL PROTECTED] wrote:
...snip...
I've read the LKML FAQ regarding new driver submissions, but it implies
that the
Michal,
I believe the patch I submitted yesterday fixes this
problem, and in a simpler way.
+-DLS
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message to [EMAIL PROTECTED]
More majordomo info at
On Mon, 14 Aug 2006, David Miller wrote:
From: Jay Vosburgh [EMAIL PROTECTED]
Date: Thu, 03 Aug 2006 18:01:35 -0700
In this case (bond0.555 above bond0 above eth0,eth1,etc),
skb_bond doesn't suppress duplicates because skb_bond is called with the
skb-dev set to the bond0.555 dev,
On Mon, 14 Aug 2006, [EMAIL PROTECTED] wrote:
From: Russell King [EMAIL PROTECTED]
When booting using root-nfs, I'm seeing (independently) two lockdep dumps
in the smc91x driver. The patch below fixes both. Both dumps look like
real locking issues.
Nico - please review and ack if you
Nicolas Pitre wrote:
On Mon, 14 Aug 2006, [EMAIL PROTECTED] wrote:
From: Russell King [EMAIL PROTECTED]
When booting using root-nfs, I'm seeing (independently) two lockdep dumps
in the smc91x driver. The patch below fixes both. Both dumps look like
real locking issues.
Nico - please review
On Wed, Aug 16, 2006 at 12:30:29PM -0400, Jeff Garzik wrote:
Linas Vepstas wrote:
The recent set of low-waterark patches for the spider result in a
Let's not reinvented NAPI, shall we...
??
I was under the impression that NAPI was for the receive side only.
This round of patches were for
Linas Vepstas wrote:
I was under the impression that NAPI was for the receive side only.
That depends on the driver implementation.
Jeff
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message to [EMAIL PROTECTED]
More majordomo info at
From: Jeff Garzik [EMAIL PROTECTED]
Date: Wed, 16 Aug 2006 16:34:31 -0400
Linas Vepstas wrote:
I was under the impression that NAPI was for the receive side only.
That depends on the driver implementation.
What Jeff is trying to say is that TX reclaim can occur in
the NAPI poll routine,
Hello folks,
In looking at a few benchmarks (especially netperf) run locally, it seems
that tcp is unable to make full use of available CPU cycles as the sender
is throttled waiting for ACKs to arrive. The problem is exacerbated when
the sender is using a small send buffer -- running netperf
Last year I reported to the linux-hams list and to Tom Sailer
that I could not get AX25 to work using a serial baycom modem with
baycom_ser_fdx under vanilla kernel 2.6.11.6 although it ran fine
under kernel 2.4.30:
http://he.fi/archive/linux-hams/200505/0021.html
I got no answer, but other
On Wed, 16 Aug 2006 16:55:32 -0400
Benjamin LaHaise [EMAIL PROTECTED] wrote:
Hello folks,
In looking at a few benchmarks (especially netperf) run locally, it seems
that tcp is unable to make full use of available CPU cycles as the sender
is throttled waiting for ACKs to arrive. The
From: Stephen Hemminger [EMAIL PROTECTED]
Date: Wed, 16 Aug 2006 12:11:12 -0700
What ethernet hardware? The defaults are often not big enough
for full speed on gigabit hardware. I need increase rmem/wmem to allow
for more buffering.
Current kernels allow the TCP send and receive socket
On Wed, 16 Aug 2006 13:44:43 -0500
John Haller [EMAIL PROTECTED] wrote:
Stephen Hemminger wrote:
On Tue, 15 Aug 2006 18:23:19 -0500
Jay Cliburn [EMAIL PROTECTED] wrote:
Stephen Hemminger wrote:
On Sun, 13 Aug 2006 19:11:42 -0500
Jay Cliburn [EMAIL PROTECTED] wrote:
...snip...
The point of delayed ack's was to merge the response and the ack on
request/response
protocols like NFS or telnet. It does make sense to get it out sooner though.
Well, to a point at least - I wouldn't go so far as to suggest immediate
ACKs.
However, I was always under the impression that
On Wed, Aug 16, 2006 at 12:11:12PM -0700, Stephen Hemminger wrote:
is throttled waiting for ACKs to arrive. The problem is exacerbated when
the sender is using a small send buffer -- running netperf -C -c -- -s 1024
show a miserable 420Kbit/s at essentially 0% CPU usage. Tests over gige
Allow NVM to setup LPLU for IGP2 and IGP3. Only IGP needs LPLU D3
disabled during init here.
Signed-off-by: Jeff Kirsher [EMAIL PROTECTED]
Signed-off-by: Auke Kok [EMAIL PROTECTED]
---
drivers/net/e1000/e1000_hw.c | 13 -
1 files changed, 8 insertions(+), 5 deletions(-)
diff
Signed-off-by: Auke Kok [EMAIL PROTECTED]
---
drivers/net/e1000/e1000_main.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/drivers/net/e1000/e1000_main.c b/drivers/net/e1000/e1000_main.c
index 627f224..ea3d504 100644
--- a/drivers/net/e1000/e1000_main.c
+++
MDIO/MDIO-X was broken due to a wrong errata. Removing the workaround
code fixes for affected NICs.
Signed-off-by: Jeff Kirsher [EMAIL PROTECTED]
Signed-off-by: Auke Kok [EMAIL PROTECTED]
---
drivers/net/e100.c | 14 +-
1 files changed, 5 insertions(+), 9 deletions(-)
diff --git
1 - 100 of 162 matches
Mail list logo