Bug#511963: Bug #511963 linux-image-2.6.26-1-xen-amd64: bridge + ipv6 = WARNING: at net/core/dev.c:1457 skb_gso_segment+0x88/0x1af()

[Hans van Kranenburg]

retitle 511963 linux-image-2.6.26-1-xen-amd64: bridge + ipv6 = WARNING: 
at net/core/dev.c:1457 skb_gso_segment+0x88/0x1af()

found 511963 2.6.26-13
tag 511963 ipv6
thank you

I found #506419 (cloned to 509156, set to wishlist) which is about 
exactly the same issue:

http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=506419

Strange enough that bug did not show up in my search. Anyway, #506419 
contains extra information, as well as another link to kerneltrap:

http://kerneltrap.org/mailarchive/linux-netdev/2008/10/16/3674714

It seems this issue is only Bridge + IPv6 related and it not related to 
do with the specific network card/driver. At least, I can reproduce 
immediately with latest lenny 2.6.26-13 on different hardware. The first 
box using tg3, this one is a SuperMicro system with "nVidia Corporation 
MCP55" chipset, like the system #506419 was originally about. Hm it 
could be SuperMicro related, I'll try to reproduce on some other system 
tomorrow.


Also in this case, issuing `ethtool -K eth0 gso off` is a workaround 
now, but probably not a solution. ;]


Hans van Kranenburg

[134650.920807] WARNING: at net/core/dev.c:1451 skb_gso_segment+0x88/0x1af()
[134651.001951] Modules linked in: bridge ipv6 loop ide_cd_mod cdrom 
snd_pcm snd_timer snd soundcore snd_page_alloc floppy k8temp psmouse 
serio_raw shpchp pcspkr pci_hotplug forcedeth amd74xx ohci_hcd 
i2c_nforce2 ehci_hcd button i2c_core evdev reiserfs dm_mirror dm_log 
dm_snapshot dm_mod raid1 md_mod sd_mod thermal processor fan thermal_sys 
ide_pci_generic ide_core ata_generic sata_nv libata scsi_mod dock

[134651.348821] Pid: 0, comm: swapper Not tainted 2.6.26-1-amd64 #1
[134651.348821]
[134651.348821] Call Trace:
[134651.348821][] warn_on_slowpath+0x51/0x7a
[134651.348821]  [] dev_queue_xmit+0x273/0x2b1
[134651.348821]  [] 
:bridge:br_dev_queue_push_xmit+0x75/0x79

[134651.348821]  [] :bridge:br_dev_xmit+0x60/0x74
[134651.348821]  [] dev_queue_xmit+0x273/0x2b1
[134651.348821]  [] :ipv6:ip6_output+0xc11/0xc37
[134651.348821]  [] skb_gso_segment+0x88/0x1af
[134651.348821]  [] dev_hard_start_xmit+0x17f/0x24e
[134651.348821]  [] dev_queue_xmit+0x1f0/0x2b1
[134651.348821]  [] :ipv6:ip6_output+0xc11/0xc37
[134651.348821]  [] :ipv6:ip6_xmit+0x35c/0x44e
[134651.348821]  [] neigh_update+0x3b9/0x3e7
[134651.348821]  [] :ipv6:inet6_csk_xmit+0x219/0x235
[134651.348821]  [] tcp_transmit_skb+0x731/0x76e
[134651.348821]  [] tso_fragment+0xef/0x13b
[134651.348821]  [] __tcp_push_pending_frames+0x6d6/0x7a9
[134651.348821]  [] tcp_rcv_established+0x10c/0x8b4
[134651.348821]  [] :ipv6:tcp_v6_do_rcv+0x626/0x8ff
[134651.348821]  [] enqueue_task+0x56/0x61
[134651.348821]  [] :ipv6:ip6_pol_route_input+0x0/0x11
[134651.348821]  [] :ipv6:fib6_rule_action+0x79/0x196
[134651.348821]  [] _write_lock_bh+0x9/0x1c
[134651.348821]  [] __inet6_lookup_established+0xed/0x258
[134651.348821]  [] :ipv6:tcp_v6_rcv+0x638/0x685
[134651.348821]  [] :ipv6:ip6_input_finish+0x1d5/0x34c
[134651.348821]  [] :ipv6:ipv6_rcv+0x32d/0x38e
[134651.348821]  [] 
:bridge:br_handle_frame_finish+0x107/0x13e

[134651.348821]  [] :bridge:br_handle_frame+0x1b0/0x1c9
[134651.348821]  [] netif_receive_skb+0x29f/0x3b8
[134651.348821]  [] __alloc_skb+0x7f/0x12d
[134651.348821]  [] process_backlog+0x86/0xef
[134651.348821]  [] dev_alloc_skb+0x16/0x2d
[134651.348821]  [] net_rx_action+0xab/0x1da
[134651.348821]  [] __do_softirq+0x5c/0xd1
[134651.348821]  [] call_softirq+0x1c/0x28
[134651.348821]  [] do_softirq+0x3c/0x81
[134651.348821]  [] irq_exit+0x3f/0x83
[134651.348821]  [] do_IRQ+0xb9/0xd9
[134651.348821]  [] default_idle+0x0/0x49
[134651.348821]  [] ret_from_intr+0x0/0x19
[134651.348821][] lapic_next_event+0x0/0x13
[134651.348821]  [] native_safe_halt+0x2/0x3
[134651.348821]  [] default_idle+0x2a/0x49
[134651.348821]  [] cpu_idle+0x89/0xb3
[134651.348821]
[134651.348821] ---[ end trace 0af407c51eea6e77 ]---



--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Processed (with 1 errors): Re: Bug #511963 linux-image-2.6.26-1-xen-amd64: bridge + ipv6 = WARNING: at net/core/dev.c:1457 skb_gso_segment+0x88/0x1af()

[Debian Bug Tracking System]

Processing commands for cont...@bugs.debian.org:

> retitle 511963 linux-image-2.6.26-1-xen-amd64: bridge + ipv6 = WARNING:
Bug#511963: linux-image-2.6.26-1-xen-amd64: xen + tg3 + ipv6 = WARNING: at 
net/core/dev.c:1457 skb_gso_segment+0x88/0x1af()
Changed Bug title to `linux-image-2.6.26-1-xen-amd64: bridge + ipv6 = WARNING:' 
from `linux-image-2.6.26-1-xen-amd64: xen + tg3 + ipv6 = WARNING: at 
net/core/dev.c:1457 skb_gso_segment+0x88/0x1af()'.

> at net/core/dev.c:1457 skb_gso_segment+0x88/0x1af()
Unknown command or malformed arguments to command.

> found 511963 2.6.26-13
Bug#511963: linux-image-2.6.26-1-xen-amd64: bridge + ipv6 = WARNING:
Bug marked as found in version 2.6.26-13.

> tag 511963 ipv6
Bug#511963: linux-image-2.6.26-1-xen-amd64: bridge + ipv6 = WARNING:
There were no tags set.
Tags added: ipv6

> thank you
Stopping processing here.

Please contact me if you need assistance.

Debian bug tracking system administrator
(administrator, Debian Bugs database)


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: Complete source for vserver image

[Ben Hutchings]

On Mon, 2009-01-19 at 22:05 +1100, Declan Mullen wrote:
> Hi
> 
> I would like to recompile Lenny's "linux-image-2.6.26-1-vserver-686"
> kernel. I have installed "linux-source-2.6.26" but it seems to be
> missing the VServer bits.
> 
> I have used the default "/boot/config-2.6.26-1-vserver-686" file for the
> new ".config". I do a "make menuconfig" and do not change anything but
> instead just save the config. Yet the newly saved ".config" does not
> contain any VServer configuration options.
> 
> Do I have to patch this kernel source ? If so, what is the Lenny package
> I install to get it ? If not, what should I be doing ?

The extra patches for vserver can be found in linux-patch-debian-2.6.26.
To apply them, set your current directory to the Linux source tree and
run
"/usr/src/kernel-patches/all/2.6.26/apply/debian -a i386 -f vserver".

Ben.



signature.asc
Description: This is a digitally signed message part

Bug#512453: initramfs-tools.8.gz: Please document environment variables for hook scripts

[Arnaud Fontaine]

Package: initramfs-tools
Version: 0.92o
Severity: wishlist

Hello,

When  trying  to write  an  initramfs hook  script,  I  didn't find  any
explanation in initramfs-tools.8.gz about environment variables exported
for  hook scripts,  except  $DESTDIR.   I was  actually  trying to  find
something  such  as  $MODULESDIR.Could  you  please  document  these
variables?

Cheers,
Arnaud Fontaine


pgpdhizLkUARu.pgp
Description: PGP signature

Bug#502326: Current status?

[Noah Meyerhans]

On Tue, Jan 13, 2009 at 02:03:10PM +0100, Oliver Bock wrote:
> I'd like to know the current status of this bug as the root cause got 
> fixed upstream apparently. Is the fixed microcode already scheduled for 
> integration into "firmware-iwlwifi" already?

The bug still occurs in linux-image-2.6.26-1-686 (2.6.26-13) even with
the new firmware.  I believe we also need
http://git.kernel.org/?p=linux/kernel/git/stable/linux-2.6.27.y.git;a=commitdiff;h=55d6a3cd0cc85ed90c39cf32e16f622bd003117b;hp=47cbb1107e4172f3632713d74dc8651a32ceb294
though I'm not completely sure why.  It seems to me that, if the
firmware fixes the problem, the above patch shouldn't be needed.

In any case, I'm going to build 2.6.26 with the above patch for testing.
If it helps, it'd be really helpful to get it included with an upcoming
lenny revision.  The lenny kernel is pretty much unusable to me with
this bug present...

noah



signature.asc
Description: Digital signature

Re: Kernel 2.6.27.x (is a good idea to Lenny?)

[maximilian attems]

On Tue, Jan 20, 2009 at 01:13:12PM -0200, Renato S. Yamane wrote:
> OK, I know that Lenny is froozen, but Kernel 2.6.27 will be supported by
> a very long time as we can see in:
> 
> 
> I think that is a good idea use this release in Lenny, because it will
> be up-to-date by a long time (by Kernel developers).
> 
> 2.6.27 make easy maintain it.
> 2.6.26 is not so easy.
> 

once lenny release, we'll work on lenny+1/2 kernel, presumably 2.6.30

kind regards

-- 
maks


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Kernel 2.6.27.x (is a good idea to Lenny?)

[Renato S. Yamane]

OK, I know that Lenny is froozen, but Kernel 2.6.27 will be supported by
a very long time as we can see in:


I think that is a good idea use this release in Lenny, because it will
be up-to-date by a long time (by Kernel developers).

2.6.27 make easy maintain it.
2.6.26 is not so easy.

Best regards,
Renato S. Yamane
Brazil


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

linux-headers-2.6.28-1-686 ininstallable

[Rainer Dorsch]

Hello,
Hello,

I tried to install linux-headers-2.6.28-1-686 from 
http://kernel-archive.buildserver.net but failed with

blackbox:~# aptitude install linux-headers-2.6.28-1-686
Reading package lists... Done
Building dependency tree
Reading state information... Done
Reading extended state information
Initializing package states... Done
Reading task descriptions... Done
The following packages are BROKEN:
  linux-headers-2.6.28-1-686
The following NEW packages will be installed:
  linux-headers-2.6.28-1-common{a}
0 packages upgraded, 2 newly installed, 0 to remove and 6 not upgraded.
Need to get 4448kB of archives. After unpacking 35.3MB will be used.
The following packages have unmet dependencies:
  linux-headers-2.6.28-1-686: Depends: linux-kbuild-2.6.28 which is a virtual 
package.
The following actions will resolve these dependencies:

Keep the following packages at their current version:
linux-headers-2.6.28-1-686 [Not Installed]

Score is -9881

Accept this solution? [Y/n/q/?] q
Abandoning all efforts to resolve these dependencies.
Abort.
blackbox:~# tail -1 /etc/apt/sources.list
deb http://kernel-archive.buildserver.net/debian-kernel/ trunk main
blackbox:~#


Is that package missing intentionally?

Thanks,
Rainer

-- 
Rainer Dorsch
Lärchenstr. 6
D-72135 Dettenhausen
07157-734133
email: rdor...@web.de
jabber: rdor...@jabber.org
GPG Fingerprint: 5966 C54C 2B3C 42CC 1F4F  8F59 E3A8 C538 7519 141E
Full GPG key: http://pgp.mit.edu/


--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Processed: Re: Bug#512337: linux-2.6: One GNU libc test fails on a Xen kernel

[Debian Bug Tracking System]

Processing commands for cont...@bugs.debian.org:

> reassign 512337 xen-hypervisor-3.0.3-1-i386-pae
Bug#512337: linux-2.6: One GNU libc test fails on a Xen kernel
Bug reassigned from package `linux-2.6' to `xen-hypervisor-3.0.3-1-i386-pae'.

> notfound 512337 2.6.18.dfsg.1-23etch1
Bug#512337: linux-2.6: One GNU libc test fails on a Xen kernel
Bug no longer marked as found in version 2.6.18.dfsg.1-23etch1.

> found 512337 3.0.3-0-4
Bug#512337: linux-2.6: One GNU libc test fails on a Xen kernel
Bug marked as found in version 3.0.3-0-4.

> retitle 512337 xen-hypervisor: One GNU libc test fails in a Xen environment
Bug#512337: linux-2.6: One GNU libc test fails on a Xen kernel
Changed Bug title to `xen-hypervisor: One GNU libc test fails in a Xen 
environment' from `linux-2.6: One GNU libc test fails on a Xen kernel'.

> thanks
Stopping processing here.

Please contact me if you need assistance.

Debian bug tracking system administrator
(administrator, Debian Bugs database)


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#512337: linux-2.6: One GNU libc test fails on a Xen kernel

[Aurelien Jarno]

reassign 512337 xen-hypervisor-3.0.3-1-i386-pae
notfound 512337 2.6.18.dfsg.1-23etch1
found 512337 3.0.3-0-4
retitle 512337 xen-hypervisor: One GNU libc test fails in a Xen environment
thanks

On Tue, Jan 20, 2009 at 08:21:54AM +, Ian Campbell wrote:
> 
> Thinking about this I now remember that the -ve segment fixups are
> actually done by the hypervisor not the kernel. Could you reassign this
> bug to the correct version of the hypervisor please.

It's the Etch hypervisor.

> Is the dom0 running Etch also? It would be worth trying the Lenny
> hypervisor.

Actually I have done all my tests directly in the dom0, running Etch. I
guess the build daemon is also using Etch for dom0.

> I can't reproduce any issue running nptl/tst-tls4 from the 2.7-18 glibc
> tree in a guest running the Etch kernel and Sid userspace on a Sid
> hypervisor. Oh wait, I'm running a 64 bit hypervisor which won't need
> these fixups, nevermind... I'll try and dig up another box to test on
> but that might me a bit.

I have just done the test with and Etch kernel and the hypervisor from
Lenny, and the bug is not reproducible on this version.

I doubt that this bug will be fixed in Etch, so are you still interested
by a reduced testcase?

-- 
  .''`.  Aurelien Jarno | GPG: 1024D/F1BCDB73
 : :' :  Debian developer   | Electrical Engineer
 `. `'   aure...@debian.org | aurel...@aurel32.net
   `-people.debian.org/~aurel32 | www.aurel32.net



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#512337: linux-2.6: One GNU libc test fails on a Xen kernel

[Aurelien Jarno]

On Mon, Jan 19, 2009 at 10:35:09PM +, Ian Campbell wrote:
> On Mon, 2009-01-19 at 23:26 +0100, Aurelien Jarno wrote:
> > On Mon, Jan 19, 2009 at 09:58:41PM +, Ian Campbell wrote:
> > > On Mon, 2009-01-19 at 20:26 +0100, Aurelien Jarno wrote:
> > > > Package: linux-2.6
> > > > Version: 2.6.18.dfsg.1-23etch1
> > > > Severity: important
> > > > 
> > > > When using a Xen kernel (in a Dom0 or a DomU), the following test fails:
> > > > - tst-tls4.out, Error 1
> > > 
> > > Which one?
> > > 
> > > glibc-2.3.6$ find -name tst-tls4.c
> > > ./elf/tst-tls4.c
> > > ./nptl/tst-tls4.c
> > 
> > It's the NPTL version. I haven't tried to reproduce it with the glibc 
> > version
> > 2.3.6, but at least is is reproducible with version 2.7.
> 
> OK, so just to be clear -- the experimental buildd is running an Etch
> kernel? It doesn't matter I would just have expected an experimental
> buildd to be running sid or at least testing. Which userspace is it
> using? Would I expect to be able to build glibc 2.7 on Etch usersapce
> (because I can't ;-))

Yes, it is running the Etch kernel, and the build happens in a 
sid/experimental chroot.

Note that while the build is failing with glibc 2.9, it does not fail
with glibc 2.7. But in both case this single test fails with a
segmentation fault.

-- 
  .''`.  Aurelien Jarno | GPG: 1024D/F1BCDB73
 : :' :  Debian developer   | Electrical Engineer
 `. `'   aure...@debian.org | aurel...@aurel32.net
   `-people.debian.org/~aurel32 | www.aurel32.net



-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: ext4 in 2.6.28

[Tshepang Lekhonkhobe]

On Tue, Jan 20, 2009 at 12:04 PM, maximilian attems  wrote:
> On Tue, Jan 20, 2009 at 10:52:28AM +0200, Tshepang Lekhonkhobe wrote:
>>
>> Thanks for providing 2.6.28, but it is nearly unusable for me because
>> I saw ext4, which my /home partition lives on, is disabled. Any reason
>> why?
>
> was due to the renaming of ext4dev to ext4, next build should
> it have enabled.

cool

> kind regards
>
> maks
>
> ps if you'd looked up the linux-2.6 bug reports you would see that one
>   pending! ;)

That didn't cross my mind. Thanks for encouragement.


-- 
my place on the web:
floss-and-misc.blogspot.com


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: ext4 in 2.6.28

[maximilian attems]

On Tue, Jan 20, 2009 at 10:52:28AM +0200, Tshepang Lekhonkhobe wrote:
> 
> Thanks for providing 2.6.28, but it is nearly unusable for me because
> I saw ext4, which my /home partition lives on, is disabled. Any reason
> why?

was due to the renaming of ext4dev to ext4, next build should
it have enabled.

kind regards

maks

ps if you'd looked up the linux-2.6 bug reports you would see that one
   pending! ;)


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Re: Complete source for vserver image

[Bastian Blank]

On Tue, Jan 20, 2009 at 07:59:05PM +1100, Declan Mullen wrote:
> Do I have to patch this kernel source ? If so, what is the Lenny package
> I install to get it ? If not, what should I be doing ?

linux-patch-debian-2.6.26. However this is not really documented.

Bastian

-- 
Insults are effective only where emotion is present.
-- Spock, "Who Mourns for Adonais?"  stardate 3468.1


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Complete source for vserver image

[Declan Mullen]

Hi

I would like to recompile Lenny's "linux-image-2.6.26-1-vserver-686"
kernel. I have installed "linux-source-2.6.26" but it seems to be
missing the VServer bits.

I have used the default "/boot/config-2.6.26-1-vserver-686" file for the
new ".config". I do a "make menuconfig" and do not change anything but
instead just save the config. Yet the newly saved ".config" does not
contain any VServer configuration options.

Do I have to patch this kernel source ? If so, what is the Lenny package
I install to get it ? If not, what should I be doing ?

Thanks,
Declan




--
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

ext4 in 2.6.28

[Tshepang Lekhonkhobe]

Hi,

Thanks for providing 2.6.28, but it is nearly unusable for me because
I saw ext4, which my /home partition lives on, is disabled. Any reason
why?

-- 
my place on the web:
floss-and-misc.blogspot.com


-- 
To UNSUBSCRIBE, email to debian-kernel-requ...@lists.debian.org
with a subject of "unsubscribe". Trouble? Contact listmas...@lists.debian.org

Bug#512337: linux-2.6: One GNU libc test fails on a Xen kernel

[Ian Campbell]

Thinking about this I now remember that the -ve segment fixups are
actually done by the hypervisor not the kernel. Could you reassign this
bug to the correct version of the hypervisor please.

Is the dom0 running Etch also? It would be worth trying the Lenny
hypervisor.

I can't reproduce any issue running nptl/tst-tls4 from the 2.7-18 glibc
tree in a guest running the Etch kernel and Sid userspace on a Sid
hypervisor. Oh wait, I'm running a 64 bit hypervisor which won't need
these fixups, nevermind... I'll try and dig up another box to test on
but that might me a bit.

Ian.

-- 
Ian Campbell

QOTD:
"Who?  Me?  No, no, NO!!  But I do sell rugs."


signature.asc
Description: This is a digitally signed message part