| Julia Lawall | [PATCH 1/3] drivers/net/smsc911x.c: Fix resource size off by...
From: Julia Lawall <julia@diku.dk>
The call resource_size(res) returns res->end - res->start + 1 and thus the
second change is semantics-preserving. res_size is then used as the second
argument of a call to request_mem_region, and the memory allocated by this
call appears to be the same as what is released in the two calls to
release_mem_region. So the size argument for those calls should be
resource_size(size) as well. Alternatively, in the second call to
release_mem_region, the sec...
| Jul 4, 5:31 pm 2009 |
| egcdcgis | JACKPOT JOY
You have won congratulation,Please Verify this mail by sending your name, address, age, phone number Occuption etc to cgnlcenter@9.cn
--
| Jul 4, 1:52 pm 2009 |
| Jeff Garzik | Re: >10% performance degradation since 2.6.18
In the pre-MSI-X days, you'd have cachelines bouncing all over the place
if you distributed networking interrupts across CPUs, particularly given
that NAPI would run some things on a single CPU anyway.
Today, machines are faster, we have multiple interrupts per device, and
we have multiple RX/TX queues. I would be interested to see hard
numbers (as opposed to guesses) about various new ways to distributed
interrupts across CPUs.
What's the best setup for power usage?
What's the best setu...
| Jul 4, 5:19 am 2009 |
| Alex Dubov |
| Jul 4, 3:58 am 2009 |
| Gerrit Renker | Re: [Bugme-new] [Bug 13692] New: dccp_ipv4 has >1 usage c...
|
| Well, I tried "rmmod -f" first thing after noticing this behavior
| and got a kernel crash dump immediately.
|
| I'm not near that machine currently, but I'll post the crash dump on
| Monday.
|
What is most important to know: did you do the "rmmod -f" while any
sockets were active? Did you have DCCP traffic going on before (not
assuming while) this happened?
(It is possible to see the current socket state using "ss -dane".)
--
| Jul 4, 2:45 pm 2009 |
| Cyrill Gorcunov | [patch 1/2] net, bridge: align br_nf_ops assignment
No functional change -- just for easier reading.
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
---
net/bridge/br_netfilter.c | 96 ++++++++++++++++++++++++++--------------------
1 file changed, 56 insertions(+), 40 deletions(-)
Index: linux-2.6.git/net/bridge/br_netfilter.c
=====================================================================
--- linux-2.6.git.orig/net/bridge/br_netfilter.c
+++ linux-2.6.git/net/bridge/br_netfilter.c
@@ -905,46 +905,62 @@ static unsigned int ip_...
| Jul 4, 2:11 am 2009 |
| Cyrill Gorcunov | [patch 2/2] net, netns_xt: shrink netns_xt members
In case if kernel was compiled without ebtables support
there is no need to keep ebt_table pointers in netns_xt
structure.
Make it config dependent.
Signed-off-by: Cyrill Gorcunov <gorcunov@openvz.org>
---
Probably GCC will authomatically shrink the structure
if the members are not in use. Please review.
include/net/netns/x_tables.h | 3 +++
1 file changed, 3 insertions(+)
Index: linux-2.6.git/include/net/netns/x_tables.h
=======================================================...
| Jul 4, 2:11 am 2009 |
| Cyrill Gorcunov | [patch 0/2] a few bridge code cleanups
Hi,
here is a few bridge cleanups - code alignment
and conditional compilation for structure members.
Please review. Hope I didn't miss anything.
Anyway if it breaks someone patch queue --
just drop this series then.
Cyrill
--
| Jul 4, 2:11 am 2009 |
| David Miller |
| Jul 3, 10:10 pm 2009 |
| David Miller |
| Jul 3, 10:11 pm 2009 |
| Jarek Poplawski | Re: Soft-Lockup/Race in networking in 2.6.31-rc1+195 ( possi...
Since there are still no 100% proofs nor suspects, here are some
suggestions of additional checking. This bisected commit could
probably be additionally verified by applying to 2.6.30 with a
preceding one; I attach both of them below.
Another suggestion is to try this without lockdep e.g. by setting
/sys/module/lockdep/parameters/lock_stat and prove_locking to 0.
Thanks,
Jarek P.
--------------->
commit 597d0275736dad9c3bda6f0a00a1c477dc0f37b1
Author: Arun R Bharadwaj <arun@linux.vne...
| Jul 4, 11:18 am 2009 |
| David Miller | Re: Soft-Lockup/Race in networking in 2.6.31-rc1+195 ( possi...
From: Jarek Poplawski <jarkao2@gmail.com>
Let's say that a particular networking timer always gets
re-added on the cpu where the timer fires.
In that case, beforehand, no inter-cpu races could possibly
be tested. But with the new timer code, such races could
now be potentially triggered.
--
| Jul 3, 9:55 pm 2009 |
| Jarek Poplawski |
| Jul 4, 2:36 am 2009 |
| David Miller |
| Jul 3, 10:12 pm 2009 |
| David Miller |
| Jul 3, 10:12 pm 2009 |
| David Miller |
| Jul 3, 10:12 pm 2009 |
| David Miller |
| Jul 3, 10:12 pm 2009 |
| David Miller |
| Jul 3, 10:11 pm 2009 |
| David Miller |
| Jul 3, 10:13 pm 2009 |
| David Miller |
| Jul 3, 11:14 pm 2009 |
| David Miller |
| Jul 3, 11:14 pm 2009 |
| David Miller | Re: [PATCH 1/4] net: skb_orphan on dev_hard_start_xmit
From: Herbert Xu <herbert@gondor.apana.org.au>
Agreed, but alas we are foaming at the mouth until we have a truly
usable alternative.
In particular the case of handling a device without usable TX
completion event indications is still quite troublesome.
--
| Jul 3, 11:02 pm 2009 |
| Herbert Xu |
| Jul 3, 11:08 pm 2009 |
| David Miller | Re: [PATCH 1/4] net: skb_orphan on dev_hard_start_xmit
From: Herbert Xu <herbert@gondor.apana.org.au>
NIU
I basically can't defer interrupts because the chip supports
per-TX-desc interrupt indications but it lacks an "all TX queue sent"
event. So if, say, tell it to interrupt every 1/4 of the TX queue
then up to 1/4 of the queue can have packets "stuck" in there
if TX activity all of a sudden ceases.
The only thing I've come up with to be able to mitigate interrupts is
to use an hrtimer of some sort. But that's going to be hard to get
rig...
| Jul 3, 11:13 pm 2009 |
| Herbert Xu | Re: [PATCH 1/4] net: skb_orphan on dev_hard_start_xmit
Here's an idea: We let the sender decide whether we need to enable
notification. This decision would be carried as a flag in the skb.
For example, UDP would set this flag when its socket buffer is close
to capacity. Routing would set this flag per NAPI run, etc.
Of course you'd ignore this flag completely if the qdisc queue is
non-empty.
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au>
Home Page: [ message continues ] " title="http://gondor.apana.... ">http://gondor.apana....
| Jul 4, 3:42 am 2009 |
| Herbert Xu | Re: [PATCH 1/4] net: skb_orphan on dev_hard_start_xmit
Actually it doesn't even matter for routing because only those
that are charged by the skb's or the pages care and they're the
only ones that would need to set this.
One potential problem is if the socket is constantly running
close to capacity, but that should only happen if the device
TX queue is also close to capacity which means that the qdisc
queue should be non-empty.
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email: Herbert Xu ~{PmV>HI~} <herbert@gondor.apana.org.au&...
| Jul 4, 5:09 am 2009 |
| Eric W. Biederman |
| Jul 3, 8:46 pm 2009 |
| Denys Fedoryschenko | Re: [RFC] arp announce, arp_proxy and windows ip conflict ve...
ARP proxy generally was supposed to not export default route(by RFC and by
logic) it should be maybe other name, arp_proxy_no_default_gateway_export -
but i guess it is too long. With exporting default gateway i have two
choices - eat it and have answer to almost any ARP request, or loose for
proxy_arp completely this interface (by setting same medium id).
Btw usually proxy_arp "answers" only for directly attached networks, e.g. link
layer routes (no gateway), but sure this is not a rule, and s...
| Jul 4, 3:55 am 2009 |
| Eric W. Biederman |
| Jul 4, 11:00 am 2009 |
| Denys Fedoryschenko |
| Jul 4, 11:03 am 2009 |
| Eric W. Biederman |
| Jul 4, 5:57 pm 2009 |
| Denys Fedoryschenko |
| Jul 4, 6:00 pm 2009 |
| Eric W. Biederman |
| Jul 4, 7:47 pm 2009 |
| Mark Smith | Re: [RFC] arp announce, arp_proxy and windows ip conflict ve...
On Sun, 5 Jul 2009 01:00:08 +0300
I'm pretty sure it doesn't.
It's a common enough situation. A second subnet (or more) is assigned
to a link because the first isn't large enough, and renumbering the
hosts into a larger subnet is not practical at the time. A
'one-armed router' is used up stream to have traffic go between the
different subnets, at the cost of traffic double traversing the link.
(The worst example I've seen is 25 subnets operating this way!)
Fundamentally it is no different to r...
| Jul 4, 7:22 pm 2009 |
