Signed-off-by: Ilpo Järvinen [EMAIL PROTECTED]
---
net/ipv4/tcp_input.c |2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 9fc9096..84bcdc9 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -1400,7 +1400,7 @@
Hi,
Here are some changes that should go to net-2.6. Boot simple
transfer tested. I'll do future cleanups that are now possible
in a different series to a next kernel version.
Reordering detection fails to take account that the reordered
skb may have pcount larger than 1. In such case the lowest of
them had the largest reordering, the old formula used the
highest of them which is pcount - 1 packets less reordered.
Signed-off-by: Ilpo Järvinen [EMAIL PROTECTED]
---
Fixes subtle bug like the one with fastpath_cnt_hint happening
due to the way the GSO and hints interact. Because hints are not
reset when just a GSOed skb is partially ACKed, there's no
guarantee that the relevant part of the write queue is going to
be processed in sacktag at all (skbs below
In case we run out of mem when fragmenting, the clearing of
FLAG_ONLY_ORIG_SACKED might get missed which then feeds FRTO
with false information. Move clearing outside skb processing
loop so that it will get executed even if the skb loop
terminates prematurely due to out-of-mem.
Besides, now the
Hi
Got two setups: with an ARM-based router and with a PC, connecting to
a remote site using IPSec in tunnel mode over ppp. Setup is identical,
PC works, ARM router doesn't. Kernel 2.6.23.1.
The setup looks like
PCa --- routerA ==== routerB --- PCb
where routerA is our problematic
Hi Dave:
It looks like the addition of UDP-Lite has upset netstat:
$ netstat -s
Ip:
1257344 total packets received
6 with invalid addresses
0 forwarded
0 incoming packets discarded
1257338 incoming packets delivered
1257151 requests sent out
Icmp:
0 ICMP messages
In article [EMAIL PROTECTED] (at Sat, 10 Nov 2007 21:14:29 +0800), Herbert Xu
[EMAIL PROTECTED] says:
It looks like the addition of UDP-Lite has upset netstat:
$ netstat -s
Ip:
:
Udp:
30 packets received
0 packets to unknown port received.
0 packet receive errors
17
YOSHIFUJI Hideaki / 吉藤英明 a écrit :
In article [EMAIL PROTECTED] (at Sat, 10 Nov 2007 21:14:29 +0800), Herbert Xu
[EMAIL PROTECTED] says:
It looks like the addition of UDP-Lite has upset netstat:
$ netstat -s
Ip:
:
Udp:
30 packets received
0 packets to unknown port received.
0
Both places look like
if (err == XXX)
goto yyy;
done:
while both yyy targets look like
err = XXX;
goto done;
so this is ok to remove the above if-s.
yyy labels are used in other places and are not removed.
Signed-off-by: Pavel Emelyanov [EMAIL
There are many places that get the dst entry, increase the
__use counter and set the lastuse time stamp.
Make a helper for this.
Signed-off-by: Pavel Emelyanov [EMAIL PROTECTED]
---
diff --git a/include/net/dst.h b/include/net/dst.h
index e9ff4a4..2f65e89 100644
--- a/include/net/dst.h
+++
The inetpeer.c tracks the LRU list of inet_perr-s, but makes
it by hands. Use the list_head-s for this.
Signed-off-by: Pavel Emelyanov [EMAIL PROTECTED]
---
diff --git a/include/net/inetpeer.h b/include/net/inetpeer.h
index aa10a81..ad8404b 100644
--- a/include/net/inetpeer.h
+++
On Sat, Nov 10, 2007 at 10:32:43PM +0900, YOSHIFUJI Hideaki / 吉藤英明 wrote:
Hmm? netstat 1.42 (net-tools 1.60) seems fine.
Which netstat are you using?
The one from Debian etch:
$ netstat -V
net-tools 1.60
netstat 1.42 (2001-04-15)
Fred Baumgarten, Alan Cox, Bernd Eckenfels, Phil Blundell,
In article [EMAIL PROTECTED] (at Sat, 10 Nov 2007 22:33:25 +0800), Herbert Xu
[EMAIL PROTECTED] says:
On Sat, Nov 10, 2007 at 10:32:43PM +0900, YOSHIFUJI Hideaki / 吉藤英明 wrote:
Hmm? netstat 1.42 (net-tools 1.60) seems fine.
Which netstat are you using?
The one from Debian etch:
On Sat, Nov 10, 2007 at 11:48:54PM +0900, YOSHIFUJI Hideaki / 吉藤英明 wrote:
Hmm. netstat -s from etch (i386) and etch (x86_64) work
fine for me. Same version, same architecture. Strange...
Interesting. What does your /proc/net/snmp file look like?
And what does netstat -s actually
Herbert Xu a écrit :
On Sat, Nov 10, 2007 at 11:48:54PM +0900, YOSHIFUJI Hideaki / 吉藤英明 wrote:
Hmm. netstat -s from etch (i386) and etch (x86_64) work
fine for me. Same version, same architecture. Strange...
Interesting. What does your /proc/net/snmp file look like?
And what does
Eric Dumazet a écrit :
Herbert Xu a écrit :
On Sat, Nov 10, 2007 at 11:48:54PM +0900, YOSHIFUJI Hideaki / 吉藤英明
wrote:
Hmm. netstat -s from etch (i386) and etch (x86_64) work
fine for me. Same version, same architecture. Strange...
Interesting. What does your /proc/net/snmp file look
On Sat, Nov 10, 2007 at 04:16:41PM +0100, Eric Dumazet wrote:
It fails to parse /proc/net/netstat , because TcpExt line is bigger than
1024 chars.
Good catch. Perhaps it's time someone rewrote this in netlink
or add this to ss.
Cheers,
--
Visit Openswan at http://www.openswan.org/
Email:
On Fri, 9 Nov 2007 16:40:05 -0600
[EMAIL PROTECTED] (Linas Vepstas) wrote:
On Fri, Nov 09, 2007 at 06:02:34PM +0100, Ingo Oeser wrote:
Linas Vepstas schrieb:
+ * napi_enabled_p - return non-zero if napi enabled
+ *
+ * Mnemonic: _p stands for predicate, returning a yes/no
+
This patch adds support for the RDC R6040 MAC we can find in the RDC R-321x
System-on-chips and some other devices.
You will need the RDC PCI identifiers if you want to test this driver :
RDC_PCI_VENDOR_ID = 0x17f3
RDC_PCI_DEVICE_ID_RDC_R6040 = 0x6040
Changes from first patch :
- use
+static int __devinit r6040_init_one(struct pci_dev *pdev,
+ const struct pci_device_id *ent)
+{
+ struct net_device *dev;
+ struct r6040_private *lp;
+ void __iomem *ioaddr;
+ int err, io_size = R6040_IO_SIZE;
+ static int card_idx
Eric Dumazet [EMAIL PROTECTED] writes:
I meant a netstat bug of course, sorry :(
It fails to parse /proc/net/netstat , because TcpExt line is bigger
than 1024 chars.
guilty -- i wrote that code a long time ago.
To correct it, we might enlarge buf1[] and buf2[] from 1024 to 2048 in
Templin, Fred L [EMAIL PROTECTED] writes:
+#if defined(CONFIG_IPV6_ISATAP)
+ /* ISATAP (RFC4214) - router address in daddr */
+ if (!strncmp(parms-name, isatap, 6)) {
Modern distributions tend to have daemons to automatically rename
network interfaces using SIOCSIFNAME. Not sure
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:
MAINTAINERS | 10 ++-
drivers/net/Kconfig |2 +-
On Sat, 10 Nov 2007, Guillaume Chazarain wrote:
Doing some bittorrent with linux-2.6.24-rc2, my box crashed with this
in the log:
4WARNING: at net/ipv4/tcp_input.c:1571 tcp_remove_reno_sacks()
This gets triggered when SACKED + LOST marked are more than packets_out.
sacked_out is dealt
IMHO linux should not drop unclassified vlan packets to pass these to sniffers.
isn't it?
But, since the __vlan_hwaccel_rx drops unclassified vlan packets, sniffers
cannot see them.
This make the __vlan_hwaccel_rx receive and process an unclassified vlan packet
as a PACKET_OTHERHOST.
Any check,
IMHO even though netdevice is in the promiscuous mode, we should receive all of
ingress packets.
This disable the vlan filtering feature when a vlan hw accel configured e1000
device goes into promiscuous mode.
This make packets visible to sniffers though it's not vlan id of itself.
Any check,
Hello Ilpo, thanks a lot for your investigation
Do you have GSO enabled?
According to ethtool -k, no.
Is this reproducable?
Unfortunately not, I saw it only once.
You can try to provoke it by setting tcp_sack
sysctl
to 0 as this seems to be non-SACK related... If so, you could try the
Fred L. Templin wrote:
What do you suggest?
To remove them unless there's some justification
to keep them.
-
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
From: Ilpo_Järvinen [EMAIL PROTECTED]
Date: Sat, 10 Nov 2007 12:55:59 +0200
Reordering detection fails to take account that the reordered
skb may have pcount larger than 1. In such case the lowest of
them had the largest reordering, the old formula used the
highest of them which is pcount - 1
From: Ilpo_Järvinen [EMAIL PROTECTED]
Date: Sat, 10 Nov 2007 12:56:00 +0200
Fixes subtle bug like the one with fastpath_cnt_hint happening
due to the way the GSO and hints interact. Because hints are not
reset when just a GSOed skb is partially ACKed, there's no
guarantee that the relevant
From: Ilpo_Järvinen [EMAIL PROTECTED]
Date: Sat, 10 Nov 2007 12:56:01 +0200
Signed-off-by: Ilpo Järvinen [EMAIL PROTECTED]
Applied.
-
To unsubscribe from this list: send the line unsubscribe netdev in
the body of a message to [EMAIL PROTECTED]
More majordomo info at
From: Ilpo_Järvinen [EMAIL PROTECTED]
Date: Sat, 10 Nov 2007 12:56:02 +0200
In case we run out of mem when fragmenting, the clearing of
FLAG_ONLY_ORIG_SACKED might get missed which then feeds FRTO
with false information. Move clearing outside skb processing
loop so that it will get executed
From: Pavel Emelyanov [EMAIL PROTECTED]
Date: Sat, 10 Nov 2007 17:27:50 +0300
Both places look like
if (err == XXX)
goto yyy;
done:
while both yyy targets look like
err = XXX;
goto done;
so this is ok to remove the above if-s.
yyy
From: Pavel Emelyanov [EMAIL PROTECTED]
Date: Sat, 10 Nov 2007 17:29:52 +0300
There are many places that get the dst entry, increase the
__use counter and set the lastuse time stamp.
Make a helper for this.
Signed-off-by: Pavel Emelyanov [EMAIL PROTECTED]
Nice cleanup, applied.
Thanks!
From: Pavel Emelyanov [EMAIL PROTECTED]
Date: Sat, 10 Nov 2007 17:32:58 +0300
The inetpeer.c tracks the LRU list of inet_perr-s, but makes
it by hands. Use the list_head-s for this.
Signed-off-by: Pavel Emelyanov [EMAIL PROTECTED]
This makes every inetpeer struct consume 8 more bytes, and
From: Patrick McHardy [EMAIL PROTECTED]
Date: Sat, 10 Nov 2007 01:13:42 +0100
Joe Perches wrote:
On Sat, 2007-11-10 at 00:12 +0100, Patrick McHardy wrote:
This may cause a use-after-free since __dev_addr_delete frees the address
when all references are gone.
How about a
From: Arnaldo Carvalho de Melo [EMAIL PROTECTED]
Date: Fri, 9 Nov 2007 12:37:42 -0200
Em Fri, Nov 09, 2007 at 04:39:41PM +0300, Pavel Emelyanov escreveu:
The packet_socks_nr variable is used purely for debugging
the number of sockets.
As Arnaldo pointed out, there's already an
From: Arnaldo Carvalho de Melo [EMAIL PROTECTED]
Date: Fri, 9 Nov 2007 12:39:08 -0200
Em Fri, Nov 09, 2007 at 04:42:05PM +0300, Pavel Emelyanov escreveu:
Just like in the af_packet.c, the ipx_sock_nr variable is used
for debugging purposes.
Switch to using existing infrastructure.
From: Pavel Emelyanov [EMAIL PROTECTED]
Date: Fri, 09 Nov 2007 16:09:58 +0300
This one is used in all the smth-to-ip tunnels we have and
looks ... not very good. Make this a regular function in the
tunnel4.ko module.
Signed-off-by: Pavel Emelyanov [EMAIL PROTECTED]
This breaks the build if
From: Pavel Emelyanov [EMAIL PROTECTED]
Date: Fri, 09 Nov 2007 16:14:12 +0300
The tunnel64_protocol uses the tunnel4_protocol's err_handler and
thus calls the tunnel4_protocol's handlers.
This is not very good, as in case of (icmp) error the wrong error
handlers will be called (e.g. ipip
From: Pavel Emelyanov [EMAIL PROTECTED]
Date: Fri, 09 Nov 2007 16:16:23 +0300
Both check for the family to select an appropriate tunnel list.
Consolidate this check and make the for() loop more readable.
Signed-off-by: Pavel Emelyanov [EMAIL PROTECTED]
Applied, thanks.
-
To unsubscribe from
From: Pavel Emelyanov [EMAIL PROTECTED]
Date: Fri, 09 Nov 2007 16:18:48 +0300
Two sets - tunnel[6]4_rcv() and tunnel[6]4_err - do the same
thing, but scan for different lists of tunnels, so this code
is easily consolidated.
Signed-off-by: Pavel Emelyanov [EMAIL PROTECTED]
This doesn't
From: Patrick McHardy [EMAIL PROTECTED]
Date: Thu, 08 Nov 2007 18:19:53 +0100
[VLAN]: Don't synchronize addresses while the vlan device is down
While the VLAN device is down, the unicast addresses are not configured
on the underlying device, so we shouldn't attempt to sync them.
Noticed
From: Patrick McHardy [EMAIL PROTECTED]
Date: Thu, 08 Nov 2007 18:20:19 +0100
This patch allows to change a VLAN device's MAC address while
the device is up. Doesn't seem worth to hold back until
2.6.25-rc since it can't break anything without actually
getting used (and it does work :))
From: Chuck Lever [EMAIL PROTECTED]
Date: Thu, 08 Nov 2007 11:59:19 -0500
The intent of the assertion in skb_truesize_check() is to check
for skb-truesize being decremented too much by other code,
resulting in a wraparound below zero.
The type of the right side of the comparison causes the
From: jamal [EMAIL PROTECTED]
Date: Thu, 08 Nov 2007 08:45:31 -0500
On Thu, 2007-08-11 at 13:07 +0200, Radu Rendec wrote:
Computing the rank of the first set bit in the hash mask (for using later
in u32_hash_fold()) was done with plain C code. Using ffs() instead makes
the code more
From: John W. Linville [EMAIL PROTECTED]
Date: Wed, 7 Nov 2007 21:11:19 -0500
On Wed, Nov 07, 2007 at 04:32:00PM -0800, David Miller wrote:
From: John W. Linville [EMAIL PROTECTED]
Date: Wed, 7 Nov 2007 13:51:54 -0500
Hold-off on this one for now if -- clearly Johannes and I need to
From: PJ Waskiewicz [EMAIL PROTECTED]
Date: Tue, 06 Nov 2007 07:25:47 -0800
The socket option for packet sockets to return the original ifindex instead
of the bonded ifindex will not match multicast traffic. Since this socket
option is the most useful for layer 2 traffic and multicast
From: Pavel Emelyanov [EMAIL PROTECTED]
Date: Wed, 07 Nov 2007 16:52:36 +0300
This counter is _always_ modified under the unix_gc_lock spinlock,
so its atomicity can be provided w/o additional efforts.
Signed-off-by: Pavel Emelyanov [EMAIL PROTECTED]
Applied.
-
To unsubscribe from this
From: Pavel Emelyanov [EMAIL PROTECTED]
Date: Wed, 07 Nov 2007 16:56:21 +0300
The scan_inflight() routine scans through the unix sockets and calls
some passed callback. The fact is that all these callbacks work with
the unix_sock objects, not the sock ones, so make this conversion in
the
From: Pavel Emelyanov [EMAIL PROTECTED]
Date: Wed, 07 Nov 2007 17:01:17 +0300
The unix_nr_socks value is limited with the 2 * get_max_files() value,
as seen from the unix_create1(). However, the check and the actual
increment are separated with the GFP_KERNEL allocation, so this limit
can be
From: Alexey Dobriyan [EMAIL PROTECTED]
Date: Wed, 7 Nov 2007 15:51:33 +0300
* it already statically initialized
* reinitializing live global spinlock every time netns is
setup is also wrong
Signed-off-by: Alexey Dobriyan [EMAIL PROTECTED]
Good catch.
Applied, thanks Alexey.
-
To
From: Denis V. Lunev [EMAIL PROTECTED]
Date: Wed, 7 Nov 2007 16:36:36 +0300
This patch fixes a small memory leak. Default fib rules can be deleted by
the user if the rule does not carry FIB_RULE_PERMANENT flag, f.e. by
ip rule flush
Such a rule will not be freed as the ref-counter has
sysctl_{r,w}mem_max can now be unexported.
Signed-off-by: Adrian Bunk [EMAIL PROTECTED]
---
d49d51e5f46615c30e0f96c5333733e0ab1c85e6
diff --git a/net/core/sock.c b/net/core/sock.c
index 8fc2f84..c519b43 100644
--- a/net/core/sock.c
+++ b/net/core/sock.c
@@ -2097,7 +2097,3 @@
This patch removes the following unused EXPORT_SYMBOL's:
- ip_vs_try_bind_dest
- ip_vs_find_dest
Signed-off-by: Adrian Bunk [EMAIL PROTECTED]
---
net/ipv4/ipvs/ip_vs_conn.c |1 -
net/ipv4/ipvs/ip_vs_ctl.c |1 -
2 files changed, 2 deletions(-)
ad9f400d4f66ea3423f97e609d6ef2572055c603
56 matches
Mail list logo
