* Re: [PATCH] 3c515: Write outside array bounds
From: David Miller @ 2009-07-30 20:28 UTC (permalink / raw)
To: roel.kluin; +Cc: jarkao2, netdev, akpm
In-Reply-To: <4A717558.2050407@gmail.com>
From: Roel Kluin <roel.kluin@gmail.com>
Date: Thu, 30 Jul 2009 12:26:32 +0200
> if dev_alloc_skb() fails on the first iteration, a write to
> cp->rx_ring[-1] occurs.
>
> Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Applied.
^ permalink raw reply
* Re: Bonding ALB sends bogus packets
From: Jay Vosburgh @ 2009-07-30 20:28 UTC (permalink / raw)
To: Stephen Hemminger; +Cc: Ben Greear, bonding-devel, netdev
In-Reply-To: <20090730130656.680e9152@nehalam>
Stephen Hemminger <shemminger@vyatta.com> wrote:
>The bonding ALB mode builds packets with type, ETH_P_LOOP.
>
>Well ETH_P_LOOP is defined as 0x0060 which looks completely bogus.
>All Ethernet types less than 1536 are interpreted as 802.2 frames.
>The result is that the resulting packet looks like a bogus 802.2
>frame to the other host (or switch). I have no idea what the
>initial design was or what the purpose of this code is, but it
>should either send a real packet or nothing at all.
I've looked at that code before, too, although I hadn't noticed
that ETH_P_LOOP is under the limit for interpretation as a length
instead of a type.
I believe the purpose of the code is to update the switch's MAC
address table for the port and insure it's correct, since the alb mode
can move MAC addresses around amongst the set of slaves.
-J
---
-Jay Vosburgh, IBM Linux Technology Center, fubar@us.ibm.com
^ permalink raw reply
* Re: [PATCH] eexpress: Read buffer overflow
From: David Miller @ 2009-07-30 20:28 UTC (permalink / raw)
To: jarkao2; +Cc: roel.kluin, netdev, akpm
In-Reply-To: <20090729200517.GB3058@ami.dom.local>
From: Jarek Poplawski <jarkao2@gmail.com>
Date: Wed, 29 Jul 2009 22:05:17 +0200
> On Wed, Jul 29, 2009 at 03:18:56PM +0200, Roel Kluin wrote:
>> start_code is 69 words, but the code always writes a multiple of 16 words,
>> so the last 11 words written are outside the array.
>>
>> Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
>> ---
>> > Now you seem to make my previous math working :-)
>> >
>> >>> (max) i = 64, (max) j = 14, (64+14+16)/2 = 47 < 69, so it seems to copy
>> >>> less than its size?
>> >
>> > Jarek P.
>> >
>>
>> You're right, thanks for reviewing, this one should be correct.
>
> Looks OK to me.
Applied.
^ permalink raw reply
* Re: [PATCH] [IPV4]: Buffer overflow
From: David Miller @ 2009-07-30 20:28 UTC (permalink / raw)
To: roel.kluin; +Cc: jarkao2, netdev, akpm
In-Reply-To: <4A716C13.6030208@gmail.com>
From: Roel Kluin <roel.kluin@gmail.com>
Date: Thu, 30 Jul 2009 11:46:59 +0200
> If arp_format_neigh_entry() can be called with n->dev->addr_len == 0, then a
> write to hbuffer[-1] occurs.
>
> Signed-off-by: Roel Kluin <roel.kluin@gmail.com>
Applied.
^ permalink raw reply
* Re: [PATCH V3] CAN: Add Flexcan CAN controller driver
From: Wolfgang Grandegger @ 2009-07-30 20:22 UTC (permalink / raw)
To: Sascha Hauer; +Cc: Socketcan-core, Linux Netdev List
In-Reply-To: <4A719463.2060201@grandegger.com>
Wolfgang Grandegger wrote:
> Sascha Hauer wrote:
[...]
>> errcnt = readl(®s->errcnt);
>> rxerr = (errcnt >> 8) & 0xff;
>> txerr = errcnt & 0xff;
>> if ((rxerr >= 96 || txerr >= 96) && state == CAN_STATE_ERROR_ACTIVE)
>> state = CAN_STATE_ERROR_WARNING;
>
> Yes, or do not handle the state in the "case 0" above appropriately. The
> CAN spec only specifies the bus error states error active, error passive
> and bus off. The warning is somehow optional but available on most CAN
> controllers. Nevertheless, we should create a message if the ISR
> realizes a state changes.
FYI, I will be on holiday for the next 1.5 weeks.
Wolfgang.
^ permalink raw reply
* Re: [PATCH 0/4] au1000_eth platform device/driver conversion
From: David Miller @ 2009-07-30 20:17 UTC (permalink / raw)
To: florian; +Cc: sshtylyov, ralf, manuel.lauss, linux-mips, netdev
In-Reply-To: <200907302214.00525.florian@openwrt.org>
From: Florian Fainelli <florian@openwrt.org>
Date: Thu, 30 Jul 2009 22:13:59 +0200
> Ok, I got to respin them to address some comments on the MIPS-side
> anyway and will not forget to CC netdev and you this time.
Thank you.
^ permalink raw reply
* Re: [PATCH] vlan: allow creating vlan when real device is not up
From: David Miller @ 2009-07-30 20:17 UTC (permalink / raw)
To: shemminger; +Cc: kaber, netdev
In-Reply-To: <20090730112321.35753739@nehalam>
From: Stephen Hemminger <shemminger@vyatta.com>
Date: Thu, 30 Jul 2009 11:23:21 -0700
> There is no reason for the arbitrary restriction that device must be
> up to create a vlan. This patch was added to Vyatta kernel to resolve startup
> ordering issues where vlan's are created but real device was disabled.
>
> Note: the vlan already correctly inherits the operstate from real device; so
> if vlan is created and real device is marked down, the vlan is marked
> down.
>
> Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
Looks fine, added to net-next-2.6
We can revert if Patrick or someone else finds a potential problem
with allowing this.
Thanks.
^ permalink raw reply
* Re: [PATCH 0/4] au1000_eth platform device/driver conversion
From: Florian Fainelli @ 2009-07-30 20:13 UTC (permalink / raw)
To: David Miller; +Cc: sshtylyov, ralf, manuel.lauss, linux-mips, netdev
In-Reply-To: <20090730.131000.259116391.davem@davemloft.net>
Hi,
Le Thursday 30 July 2009 22:10:00 David Miller, vous avez écrit :
> From: Sergei Shtylyov <sshtylyov@ru.mvista.com>
> Date: Thu, 30 Jul 2009 23:48:13 +0400
>
> > Because they were only posted to linux-mips. And we didn't see #2 and
> > #4 in linux-mips either.
>
> That's not going to work.
>
> You can't apply some of these patches to one tree, and some to
> the other. That leaves the driver in a half-baked state in BOTH
> trees during the entire development period.
Sorry about that I did not want to "spam" netdev with MIPS specific bits as I
know those specific changes would lead to a discussion for sure.
>
> So one of the two trees has to be chosen and all 4 patches added
> there, and there only.
Ok, I got to respin them to address some comments on the MIPS-side anyway and
will not forget to CC netdev and you this time.
--
Best regards, Florian Fainelli
Email: florian@openwrt.org
Web: http://openwrt.org
IRC: [florian] on irc.freenode.net
-------------------------------
^ permalink raw reply
* Re: [PATCH v2] docbook: fix printk of ip address
From: David Miller @ 2009-07-30 20:10 UTC (permalink / raw)
To: randy.dunlap; +Cc: netdev, tklauser
In-Reply-To: <20090730123746.1fe58340.randy.dunlap@oracle.com>
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Thu, 30 Jul 2009 12:37:46 -0700
> From: Tobias Klauser <tklauser@distanz.ch>
>
> Use the %pI4 format string instead of %d.%d.%d.%d and NIPQUAD.
>
> Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
> Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
Applied, thanks.
^ permalink raw reply
* Re: [PATCH 0/4] au1000_eth platform device/driver conversion
From: David Miller @ 2009-07-30 20:10 UTC (permalink / raw)
To: sshtylyov; +Cc: florian, ralf, manuel.lauss, linux-mips, netdev
In-Reply-To: <4A71F8FD.3060002@ru.mvista.com>
From: Sergei Shtylyov <sshtylyov@ru.mvista.com>
Date: Thu, 30 Jul 2009 23:48:13 +0400
> Because they were only posted to linux-mips. And we didn't see #2 and
> #4 in linux-mips either.
That's not going to work.
You can't apply some of these patches to one tree, and some to
the other. That leaves the driver in a half-baked state in BOTH
trees during the entire development period.
So one of the two trees has to be chosen and all 4 patches added
there, and there only.
^ permalink raw reply
* Re: IPv6 default routes timing out?
From: John Dykstra @ 2009-07-30 20:09 UTC (permalink / raw)
To: Roland Dreier; +Cc: Roland Dreier, netdev
In-Reply-To: <aday6q6qatu.fsf@cisco.com>
On Thu, 2009-07-30 at 11:34 -0700, Roland Dreier wrote:
> Well, I see router advertisements (and pings to ff02::1) from the router
> on both of the two laptops I have handy, and then after a while the
> openbsd system seems to stop broadcasting -- I no longer see the
> broadcast ipv6 packets on either laptop, but if I just do "ifconfig ral0
> down; ifconfig ral0 up" on the openbsd system then everything seems to
> come back.
Note that these are actually multicasts, not broadcasts. There's not
much difference on the sending end, but on the receiving end there's
usually filtering at the NIC level that doesn't apply to broadcasts.
That's why I'm a little bit suspicious of the laptop end of your
connection. It's not uncommon for multicast to be broken in various
ways without people noticing, because they don't use it very much on
IPv4. However, it's fundamental to IPv6.
-- John
^ permalink raw reply
* Bonding ALB sends bogus packets
From: Stephen Hemminger @ 2009-07-30 20:06 UTC (permalink / raw)
To: Jay Vosburgh, Ben Greear; +Cc: bonding-devel, netdev
The bonding ALB mode builds packets with type, ETH_P_LOOP.
Well ETH_P_LOOP is defined as 0x0060 which looks completely bogus.
All Ethernet types less than 1536 are interpreted as 802.2 frames.
The result is that the resulting packet looks like a bogus 802.2
frame to the other host (or switch). I have no idea what the
initial design was or what the purpose of this code is, but it
should either send a real packet or nothing at all.
--
^ permalink raw reply
* Re: [PATCH 0/4] au1000_eth platform device/driver conversion
From: Sergei Shtylyov @ 2009-07-30 19:48 UTC (permalink / raw)
To: David Miller; +Cc: florian, ralf, manuel.lauss, linux-mips, netdev
In-Reply-To: <20090730.123450.224830320.davem@davemloft.net>
Hello.
David Miller wrote:
> From: Florian Fainelli <florian@openwrt.org>
> Date: Tue, 28 Jul 2009 23:00:04 +0200
>
>> Hi Ralf, David, Manuel,
>>
>> The following patches convert the AMD Alchemy au1000_eth driver
>> to become a platform device/driver. The patches are splitted that way:
>>
>
> I was only able to see patch #2 and #4 in this series and
> checking the netdev patch postings that made it to:
>
> http://patchwork.ozlabs.org/project/netdev/list/
>
> confirms that patches #1 and #3 didn't make it to the list
> either.
Because they were only posted to linux-mips. And we didn't see #2 and
#4 in linux-mips either.
WBR, Sergei
^ permalink raw reply
* [PATCH v2] docbook: fix printk of ip address
From: Randy Dunlap @ 2009-07-30 19:37 UTC (permalink / raw)
To: David Miller; +Cc: netdev, tklauser
In-Reply-To: <20090730.123151.191002348.davem@davemloft.net>
From: Tobias Klauser <tklauser@distanz.ch>
Use the %pI4 format string instead of %d.%d.%d.%d and NIPQUAD.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
---
Documentation/DocBook/kernel-hacking.tmpl | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
--- linux-2.6.31-rc4-git3.orig/Documentation/DocBook/kernel-hacking.tmpl
+++ linux-2.6.31-rc4-git3/Documentation/DocBook/kernel-hacking.tmpl
@@ -449,8 +449,8 @@ printk(KERN_INFO "i = %u\n", i);
</para>
<programlisting>
-__u32 ipaddress;
-printk(KERN_INFO "my ip: %d.%d.%d.%d\n", NIPQUAD(ipaddress));
+__be32 ipaddress;
+printk(KERN_INFO "my ip: %pI4\n", &ipaddress);
</programlisting>
<para>
---
~Randy
LPC 2009, Sept. 23-25, Portland, Oregon
http://linuxplumbersconf.org/2009/
^ permalink raw reply
* Re: [PATCH 0/4] au1000_eth platform device/driver conversion
From: David Miller @ 2009-07-30 19:34 UTC (permalink / raw)
To: florian; +Cc: ralf, manuel.lauss, linux-mips, netdev
In-Reply-To: <200907282300.07144.florian@openwrt.org>
From: Florian Fainelli <florian@openwrt.org>
Date: Tue, 28 Jul 2009 23:00:04 +0200
> Hi Ralf, David, Manuel,
>
> The following patches convert the AMD Alchemy au1000_eth driver
> to become a platform device/driver. The patches are splitted that way:
I was only able to see patch #2 and #4 in this series and
checking the netdev patch postings that made it to:
http://patchwork.ozlabs.org/project/netdev/list/
confirms that patches #1 and #3 didn't make it to the list
either.
^ permalink raw reply
* Re: [PATCH] docbook: fix printk of ip address
From: David Miller @ 2009-07-30 19:31 UTC (permalink / raw)
To: randy.dunlap; +Cc: netdev, tklauser
In-Reply-To: <20090730122518.995a83ba.randy.dunlap@oracle.com>
From: Randy Dunlap <randy.dunlap@oracle.com>
Date: Thu, 30 Jul 2009 12:25:18 -0700
> Does this need to be changed to __be32?
Yes, the value passed in must be in network byte order
when using %pI4.
^ permalink raw reply
* Re: [net-next 4/4] bnx2x: Version 1.48.114-1
From: David Miller @ 2009-07-30 19:29 UTC (permalink / raw)
To: eilong; +Cc: netdev
In-Reply-To: <1248862811.27379.18.camel@lb-tlvb-eilong>
From: "Eilon Greenstein" <eilong@broadcom.com>
Date: Wed, 29 Jul 2009 13:20:11 +0300
>
> Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Applied.
^ permalink raw reply
* Re: [net-next 3/4] bnx2x: WoL only with current MAC address
From: David Miller @ 2009-07-30 19:29 UTC (permalink / raw)
To: eilong; +Cc: netdev
In-Reply-To: <1248862810.27379.17.camel@lb-tlvb-eilong>
From: "Eilon Greenstein" <eilong@broadcom.com>
Date: Wed, 29 Jul 2009 13:20:10 +0300
> In some cases with 57711E, depending on the functions unload sequence, other
> functions MAC address could have been used to wake the system as well. Make sure
> to block all but the current function if WoL is required by changing the mode
> to single function WoL.
>
> Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Applied.
^ permalink raw reply
* Re: [net-next 2/4] bnx2x: Self-test kills management traffic
From: David Miller @ 2009-07-30 19:29 UTC (permalink / raw)
To: eilong; +Cc: netdev
In-Reply-To: <1248862808.27379.16.camel@lb-tlvb-eilong>
From: "Eilon Greenstein" <eilong@broadcom.com>
Date: Wed, 29 Jul 2009 13:20:08 +0300
> Self test used to play with the management FIFO possibly while management was
> running...
>
> Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Applied.
^ permalink raw reply
* [PATCH] docbook: fix printk of ip address
From: Randy Dunlap @ 2009-07-30 19:25 UTC (permalink / raw)
To: netdev; +Cc: davem, Tobias Klauser
Does this need to be changed to __be32?
---
From: Tobias Klauser <tklauser@distanz.ch>
Use the %pI4 format string instead of %d.%d.%d.%d and NIPQUAD.
Signed-off-by: Tobias Klauser <tklauser@distanz.ch>
Signed-off-by: Randy Dunlap <randy.dunlap@oracle.com>
---
Documentation/DocBook/kernel-hacking.tmpl | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
--- linux-2.6.31-rc4-git3.orig/Documentation/DocBook/kernel-hacking.tmpl
+++ linux-2.6.31-rc4-git3/Documentation/DocBook/kernel-hacking.tmpl
@@ -450,7 +450,7 @@ printk(KERN_INFO "i = %u\n", i);
<programlisting>
__u32 ipaddress;
-printk(KERN_INFO "my ip: %d.%d.%d.%d\n", NIPQUAD(ipaddress));
+printk(KERN_INFO "my ip: %pI4\n", &ipaddress);
</programlisting>
<para>
---
~Randy
LPC 2009, Sept. 23-25, Portland, Oregon
http://linuxplumbersconf.org/2009/
^ permalink raw reply
* Re: [net-next 1/4] bnx2x: Receive traffic that maches management filtering rules
From: David Miller @ 2009-07-30 19:20 UTC (permalink / raw)
To: eilong; +Cc: netdev
In-Reply-To: <1248862804.27379.15.camel@lb-tlvb-eilong>
From: "Eilon Greenstein" <eilong@broadcom.com>
Date: Wed, 29 Jul 2009 13:20:04 +0300
> Due to lack of configuration, if the BMC configures the chip to pass all
> broadcast/multicast traffic to it, the host will not receive it. On top of
> fixing it, also make sure that in promiscuous mode, the host will receive the
> management traffic as well.
>
> Signed-off-by: Eilon Greenstein <eilong@broadcom.com>
Applied.
^ permalink raw reply
* Re: IPv6 default routes timing out?
From: Roland Dreier @ 2009-07-30 18:34 UTC (permalink / raw)
To: John Dykstra; +Cc: Roland Dreier, netdev
In-Reply-To: <1248975058.7167.28.camel@Maple>
> > Thanks, I did some more debugging it seems the issue is in the
> > router's (openbsd) wireless driver -- it thinks it is broadcasting
> > router advertisements but they aren't making it to the laptop.
> Could it be that there's something buggy in the laptop wireless driver
> or hardware re filtering of multicast packets at the L2 level?
Well, I see router advertisements (and pings to ff02::1) from the router
on both of the two laptops I have handy, and then after a while the
openbsd system seems to stop broadcasting -- I no longer see the
broadcast ipv6 packets on either laptop, but if I just do "ifconfig ral0
down; ifconfig ral0 up" on the openbsd system then everything seems to
come back. So it seems unlikely that the problem is anywhere except on
the openbsd system sending.
It's interesting to note that the only broadcasts that the openbsd
router is going to send that matter in my setup are ipv6 router
advertisements -- before I started using ipv4, all the broadcasts would
have been arps from the laptops to the router, so I would have never
noticed that the router stopped being able to broadcast.
- R.
^ permalink raw reply
* [PATCH] vlan: allow creating vlan when real device is not up
From: Stephen Hemminger @ 2009-07-30 18:23 UTC (permalink / raw)
To: Patrick McHardy, David Miller; +Cc: netdev
There is no reason for the arbitrary restriction that device must be
up to create a vlan. This patch was added to Vyatta kernel to resolve startup
ordering issues where vlan's are created but real device was disabled.
Note: the vlan already correctly inherits the operstate from real device; so
if vlan is created and real device is marked down, the vlan is marked
down.
Signed-off-by: Stephen Hemminger <shemminger@vyatta.com>
--- a/net/8021q/vlan.c 2009-07-30 11:15:34.983190412 -0700
+++ b/net/8021q/vlan.c 2009-07-30 11:16:03.590055942 -0700
@@ -225,12 +225,6 @@ int vlan_check_real_dev(struct net_devic
return -EOPNOTSUPP;
}
- /* The real device must be up and operating in order to
- * assosciate a VLAN device with it.
- */
- if (!(real_dev->flags & IFF_UP))
- return -ENETDOWN;
-
if (__find_vlan_dev(real_dev, vlan_id) != NULL)
return -EEXIST;
^ permalink raw reply
* Re: [PATCH] iscsi: Use GFP_ATOMIC in iscsi_offload_mesg().
From: David Miller @ 2009-07-30 18:04 UTC (permalink / raw)
To: michaelc; +Cc: open-iscsi, James.Bottomley, netdev, linux-scsi
In-Reply-To: <4A70FA22.7010706@cs.wisc.edu>
From: Mike Christie <michaelc@cs.wisc.edu>
Date: Wed, 29 Jul 2009 20:40:50 -0500
> On 07/29/2009 01:49 PM, Michael Chan wrote:
>> Changing to GFP_ATOMIC because the only caller in cnic/bnx2i may
>> be calling this function while holding spin_lock.
>>
>> This problem was discovered by Mike Christie.
>>
>> Signed-off-by: Michael Chan<mchan@broadcom.com>
...
> Acked-by: Mike Christie <michaelc@cs.wisc.edu>
>
> Dave Miller probably wants to take this in his tree since it fixes a
> bug with this patch
> http://git.kernel.org/?p=linux/kernel/git/davem/net-2.6.git;a=commit;h=6d7760a88c25057c2c2243e5dfe2d731064bd31d
Yep I'll take it, thanks everyone.
^ permalink raw reply
* Re: [PATCH 4/5] net/netlabel: Add kmalloc NULL tests
From: David Miller @ 2009-07-30 17:59 UTC (permalink / raw)
To: paul.moore; +Cc: julia, netdev, linux-kernel, kernel-janitors
In-Reply-To: <200907301053.37416.paul.moore@hp.com>
From: Paul Moore <paul.moore@hp.com>
Date: Thu, 30 Jul 2009 10:53:37 -0400
> On Thursday 30 July 2009 10:38:19 am Julia Lawall wrote:
>> From: Julia Lawall <julia@diku.dk>
>>
>> The test on map4 should be a test on map6.
>>
>> The semantic match that finds this problem is as follows:
>> (http://www.emn.fr/x-info/coccinelle/)
...
>> Signed-off-by: Julia Lawall <julia@diku.dk>
>
> Much better, thank you.
>
> Acked-by: Paul Moore <paul.moore@hp.com>
Applied, thanks.
^ permalink raw reply
page: next (older) | prev (newer) | latest
- recent:[subjects (threaded)|topics (new)|topics (active)]
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox