* [Qemu-devel] BUG - qdev - partial loss of network connectivity
@ 2010-09-22 17:18 Leszek Urbanski
2010-09-22 17:41 ` Anthony Liguori
0 siblings, 1 reply; 18+ messages in thread
From: Leszek Urbanski @ 2010-09-22 17:18 UTC (permalink / raw)
To: qemu-devel; +Cc: tygrys
Hi,
This is a qemu-kvm (i.e. not qemu) bug report. I've been told on IRC (#kvm)
that this bug report should go to this list anyway.
host and guest kernel: 2.6.32.22
arch: amd64
qemu-kvm: 0.12.5 and 0.13.0-rc1
How to reproduce: copy a large (few hundred MB) file to an NFS mount
(guest is the client), using qdev syntax for virtio:
-device virtio-net-pci,netdev=tap1,mac=ab:cd:ef:01:23:45 \
-netdev type=tap,id=tap1,ifname=tap1
all NFS mounts immediately stall and processes go into D state and are stuck
on nfs_wait_bit_uninterruptible. There is no further communication between
the client and the server. It occurs with UDP and TCP mounts.
That same guest, when switched to the old -net/-net syntax:
-net nic,vlan=0,model=virtio,name=eth0,macaddr=ab:cd:ef:01:23:45 \
-net tap,vlan=0,ifname=tap1
works fine. It also works fine on bare metal.
I can reproduce the problem every time within 10 minutes with qdev.
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: [Qemu-devel] BUG - qdev - partial loss of network connectivity
2010-09-22 17:18 [Qemu-devel] BUG - qdev - partial loss of network connectivity Leszek Urbanski
@ 2010-09-22 17:41 ` Anthony Liguori
2010-09-22 18:20 ` Leszek Urbanski
0 siblings, 1 reply; 18+ messages in thread
From: Anthony Liguori @ 2010-09-22 17:41 UTC (permalink / raw)
To: Leszek Urbanski; +Cc: qemu-devel
On 09/22/2010 12:18 PM, Leszek Urbanski wrote:
> Hi,
>
> This is a qemu-kvm (i.e. not qemu) bug report. I've been told on IRC (#kvm)
> that this bug report should go to this list anyway.
>
>
> host and guest kernel: 2.6.32.22
> arch: amd64
> qemu-kvm: 0.12.5 and 0.13.0-rc1
>
> How to reproduce: copy a large (few hundred MB) file to an NFS mount
> (guest is the client), using qdev syntax for virtio:
>
> -device virtio-net-pci,netdev=tap1,mac=ab:cd:ef:01:23:45 \
> -netdev type=tap,id=tap1,ifname=tap1
>
> all NFS mounts immediately stall and processes go into D state and are stuck
> on nfs_wait_bit_uninterruptible. There is no further communication between
> the client and the server. It occurs with UDP and TCP mounts.
>
> That same guest, when switched to the old -net/-net syntax:
>
> -net nic,vlan=0,model=virtio,name=eth0,macaddr=ab:cd:ef:01:23:45 \
> -net tap,vlan=0,ifname=tap1
>
> works fine. It also works fine on bare metal.
>
> I can reproduce the problem every time within 10 minutes with qdev.
>
Is the guest kernel vanilla 2.6.32.22 or is it a distro kernel? If the
later, what distro?
The difference in the two invocations is that with the -device syntax,
you're getting offload features enabled.
Regards,
Anthony Liguori
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: [Qemu-devel] BUG - qdev - partial loss of network connectivity
2010-09-22 17:41 ` Anthony Liguori
@ 2010-09-22 18:20 ` Leszek Urbanski
2010-09-22 18:35 ` Anthony Liguori
0 siblings, 1 reply; 18+ messages in thread
From: Leszek Urbanski @ 2010-09-22 18:20 UTC (permalink / raw)
To: Anthony Liguori; +Cc: qemu-devel
<4C9A3FAF.9090503@codemonkey.ws>; from Anthony Liguori on Wed, Sep 22, 2010 at 12:41:03 -0500
> Is the guest kernel vanilla 2.6.32.22 or is it a distro kernel? If the
> later, what distro?
>
> The difference in the two invocations is that with the -device syntax,
> you're getting offload features enabled.
It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
(based on 2.6.32.21).
If offload is the only difference, I'll play with different offload
options and check which one causes it.
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: [Qemu-devel] BUG - qdev - partial loss of network connectivity
2010-09-22 18:20 ` Leszek Urbanski
@ 2010-09-22 18:35 ` Anthony Liguori
2010-09-23 14:04 ` Leszek Urbanski
0 siblings, 1 reply; 18+ messages in thread
From: Anthony Liguori @ 2010-09-22 18:35 UTC (permalink / raw)
To: Leszek Urbanski; +Cc: qemu-devel
On 09/22/2010 01:20 PM, Leszek Urbanski wrote:
> <4C9A3FAF.9090503@codemonkey.ws>; from Anthony Liguori on Wed, Sep 22, 2010 at 12:41:03 -0500
>
>
>> Is the guest kernel vanilla 2.6.32.22 or is it a distro kernel? If the
>> later, what distro?
>>
>> The difference in the two invocations is that with the -device syntax,
>> you're getting offload features enabled.
>>
> It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> (based on 2.6.32.21).
>
> If offload is the only difference, I'll play with different offload
> options and check which one causes it.
>
It's not technically the only difference but it's the most likely
culprit IMHO.
Regards,
Anthony Liguori
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: [Qemu-devel] BUG - qdev - partial loss of network connectivity
2010-09-22 18:35 ` Anthony Liguori
@ 2010-09-23 14:04 ` Leszek Urbanski
2010-09-26 15:43 ` [Qemu-devel] " Michael S. Tsirkin
0 siblings, 1 reply; 18+ messages in thread
From: Leszek Urbanski @ 2010-09-23 14:04 UTC (permalink / raw)
To: Anthony Liguori; +Cc: qemu-devel
<4C9A4C77.2080806@codemonkey.ws>; from Anthony Liguori on Wed, Sep 22, 2010 at 13:35:35 -0500
> >>Is the guest kernel vanilla 2.6.32.22 or is it a distro kernel? If the
> >>later, what distro?
> >>
> >>The difference in the two invocations is that with the -device syntax,
> >>you're getting offload features enabled.
> >>
> >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> >(based on 2.6.32.21).
> >
> >If offload is the only difference, I'll play with different offload
> >options and check which one causes it.
> >
>
> It's not technically the only difference but it's the most likely
> culprit IMHO.
udp fragmentation offload is definitely the culprit.
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
* [Qemu-devel] Re: BUG - qdev - partial loss of network connectivity
2010-09-23 14:04 ` Leszek Urbanski
@ 2010-09-26 15:43 ` Michael S. Tsirkin
2010-09-27 21:32 ` Leszek Urbanski
2010-09-27 21:32 ` Leszek Urbanski
0 siblings, 2 replies; 18+ messages in thread
From: Michael S. Tsirkin @ 2010-09-26 15:43 UTC (permalink / raw)
To: Leszek Urbanski; +Cc: qemu-devel
On Thu, Sep 23, 2010 at 04:04:37PM +0200, Leszek Urbanski wrote:
> <4C9A4C77.2080806@codemonkey.ws>; from Anthony Liguori on Wed, Sep 22, 2010 at 13:35:35 -0500
>
> > >>Is the guest kernel vanilla 2.6.32.22 or is it a distro kernel? If the
> > >>later, what distro?
> > >>
> > >>The difference in the two invocations is that with the -device syntax,
> > >>you're getting offload features enabled.
> > >>
> > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > >(based on 2.6.32.21).
> > >
> > >If offload is the only difference, I'll play with different offload
> > >options and check which one causes it.
> > >
> >
> > It's not technically the only difference but it's the most likely
> > culprit IMHO.
>
> udp fragmentation offload is definitely the culprit.
I see. Most likely guest bug - won't be the first bug around UFO.
If so pls copy netdev linux-nfs and virtualization.
Do you see anything in dmesg? Can try 2.6.36-rc5?
> --
> Leszek "Tygrys" Urbanski, SCSA, SCNA
> "Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
> wretched hive of bugs and flamers. We must be cautious." -- DECWARS
> http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: BUG - qdev - partial loss of network connectivity
2010-09-26 15:43 ` [Qemu-devel] " Michael S. Tsirkin
2010-09-27 21:32 ` Leszek Urbanski
@ 2010-09-27 21:32 ` Leszek Urbanski
1 sibling, 0 replies; 18+ messages in thread
From: Leszek Urbanski @ 2010-09-27 21:32 UTC (permalink / raw)
To: Michael S. Tsirkin; +Cc: netdev, linux-nfs, qemu-devel, virtualization
<20100926154324.GD21843@redhat.com>; from Michael S. Tsirkin on Sun, Sep 26, 2010 at 17:43:24 +0200
> > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > >(based on 2.6.32.21).
> > > >
> > > >If offload is the only difference, I'll play with different offload
> > > >options and check which one causes it.
> > > >
> > >
> > > It's not technically the only difference but it's the most likely
> > > culprit IMHO.
> >
> > udp fragmentation offload is definitely the culprit.
>
> I see. Most likely guest bug - won't be the first bug around UFO.
> If so pls copy netdev linux-nfs and virtualization.
> Do you see anything in dmesg? Can try 2.6.36-rc5?
(for reference: first post is at:
http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
fixed it, or should I dissect?
2.6.32.x - there's nothing interesting in dmesg, apart from traces related
to tasks in D state waiting on the NFS mounts:
[ 84.396127] nfs: server 10.0.0.1 not responding, still trying
[ 240.568162] INFO: task cp:1838 blocked for more than 120 seconds.
[ 240.569715] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 240.571486] cp D 0000000000000002 0 1838 1831 0x00000000
[ 240.573340] ffff88011fa5b880 0000000000000082 0000000000000000 ffff88011e45bb44
[ 240.575508] ffff88011e45bcc8 ffffffff8102cdac 000000000000f9e0 ffff88011e45bfd8
[ 240.578827] 0000000000015780 0000000000015780 ffff88011c7ce2e0 ffff88011c7ce5d8
[ 240.580502] Call Trace:
[ 240.581132] [<ffffffff8102cdac>] ? pvclock_clocksource_read+0x3a/0x8b
[ 240.582427] [<ffffffff8102cdac>] ? pvclock_clocksource_read+0x3a/0x8b
[ 240.583869] [<ffffffff810b3bdd>] ? sync_page+0x0/0x46
[ 240.585034] [<ffffffff810b3bdd>] ? sync_page+0x0/0x46
[ 240.586087] [<ffffffff812f9939>] ? io_schedule+0x73/0xb7
[ 240.587287] [<ffffffff810b3c1e>] ? sync_page+0x41/0x46
[ 240.588202] [<ffffffff812f9e46>] ? __wait_on_bit+0x41/0x70
[ 240.589314] [<ffffffff810b3da2>] ? wait_on_page_bit+0x6b/0x71
[ 240.590630] [<ffffffff81064a1c>] ? wake_bit_function+0x0/0x23
[ 240.591906] [<ffffffff810bb9ea>] ? pagevec_lookup_tag+0x1a/0x21
[ 240.592954] [<ffffffff810b4577>] ? wait_on_page_writeback_range+0x69/0x11b
[ 240.594403] [<ffffffff810b536e>] ? filemap_write_and_wait+0x26/0x32
[ 240.595563] [<ffffffffa02c0d35>] ? nfs_setattr+0xb9/0x117 [nfs]
[ 240.596670] [<ffffffff810b3a0b>] ? find_get_page+0x1a/0x77
[ 240.598012] [<ffffffff810b3bb9>] ? lock_page+0x9/0x1f
[ 240.598878] [<ffffffff810b41ee>] ? filemap_fault+0xb9/0x2f6
[ 240.599839] [<ffffffff810ca3c2>] ? __do_fault+0x38c/0x3c3
[ 240.601003] [<ffffffff810ee1ce>] ? do_sync_write+0xce/0x113
[ 240.602082] [<ffffffff81051e75>] ? current_fs_time+0x1e/0x24
[ 240.602968] [<ffffffff811009b7>] ? notify_change+0x180/0x2c5
[ 240.604245] [<ffffffff8110b7b5>] ? utimes_common+0x12d/0x14d
[ 240.605355] [<ffffffff8110b856>] ? do_utimes+0x81/0xca
[ 240.606558] [<ffffffff8110b9ab>] ? sys_utimensat+0x5b/0x6a
[ 240.607817] [<ffffffff81010b42>] ? system_call_fastpath+0x16/0x1b
[ 240.609124] INFO: task find:1866 blocked for more than 120 seconds.
[ 240.610409] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 240.612066] find D 0000000000000000 0 1866 1863 0x00000000
[ 240.613490] ffffffff8145d1f0 0000000000000086 0000000000000000 ffff88011e2d2350
[ 240.615188] 00000022b63d07c7 ffff88011c55e000 000000000000f9e0 ffff8800c78a5fd8
[ 240.616576] 0000000000015780 0000000000015780 ffff88011e2d2350 ffff88011e2d2648
[ 240.618297] Call Trace:
[ 240.618777] [<ffffffff810e5369>] ? virt_to_head_page+0x9/0x2a
[ 240.619906] [<ffffffff812fa07a>] ? __mutex_lock_common+0x122/0x192
[ 240.621324] [<ffffffff812fa1a2>] ? mutex_lock+0x1a/0x31
[ 240.622543] [<ffffffff81102c11>] ? mntput_no_expire+0x23/0xee
[ 240.623860] [<ffffffffa02c0b03>] ? nfs_getattr+0x3b/0xda [nfs]
[ 240.625219] [<ffffffff810f1839>] ? vfs_fstatat+0x43/0x57
[ 240.626290] [<ffffffff810f185e>] ? sys_newfstatat+0x11/0x30
[ 240.627594] [<ffffffff81102c11>] ? mntput_no_expire+0x23/0xee
[ 240.628768] [<ffffffff8101195b>] ? device_not_available+0x1b/0x20
[ 240.629644] [<ffffffff81010b42>] ? system_call_fastpath+0x16/0x1b
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: BUG - qdev - partial loss of network connectivity
@ 2010-09-27 21:32 ` Leszek Urbanski
0 siblings, 0 replies; 18+ messages in thread
From: Leszek Urbanski @ 2010-09-27 21:32 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: Anthony Liguori, qemu-devel, netdev, linux-nfs, virtualization
<20100926154324.GD21843@redhat.com>; from Michael S. Tsirkin on Sun, Sep 26, 2010 at 17:43:24 +0200
> > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > >(based on 2.6.32.21).
> > > >
> > > >If offload is the only difference, I'll play with different offload
> > > >options and check which one causes it.
> > > >
> > >
> > > It's not technically the only difference but it's the most likely
> > > culprit IMHO.
> >
> > udp fragmentation offload is definitely the culprit.
>
> I see. Most likely guest bug - won't be the first bug around UFO.
> If so pls copy netdev linux-nfs and virtualization.
> Do you see anything in dmesg? Can try 2.6.36-rc5?
(for reference: first post is at:
http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
fixed it, or should I dissect?
2.6.32.x - there's nothing interesting in dmesg, apart from traces related
to tasks in D state waiting on the NFS mounts:
[ 84.396127] nfs: server 10.0.0.1 not responding, still trying
[ 240.568162] INFO: task cp:1838 blocked for more than 120 seconds.
[ 240.569715] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 240.571486] cp D 0000000000000002 0 1838 1831 0x00000000
[ 240.573340] ffff88011fa5b880 0000000000000082 0000000000000000 ffff88011e45bb44
[ 240.575508] ffff88011e45bcc8 ffffffff8102cdac 000000000000f9e0 ffff88011e45bfd8
[ 240.578827] 0000000000015780 0000000000015780 ffff88011c7ce2e0 ffff88011c7ce5d8
[ 240.580502] Call Trace:
[ 240.581132] [<ffffffff8102cdac>] ? pvclock_clocksource_read+0x3a/0x8b
[ 240.582427] [<ffffffff8102cdac>] ? pvclock_clocksource_read+0x3a/0x8b
[ 240.583869] [<ffffffff810b3bdd>] ? sync_page+0x0/0x46
[ 240.585034] [<ffffffff810b3bdd>] ? sync_page+0x0/0x46
[ 240.586087] [<ffffffff812f9939>] ? io_schedule+0x73/0xb7
[ 240.587287] [<ffffffff810b3c1e>] ? sync_page+0x41/0x46
[ 240.588202] [<ffffffff812f9e46>] ? __wait_on_bit+0x41/0x70
[ 240.589314] [<ffffffff810b3da2>] ? wait_on_page_bit+0x6b/0x71
[ 240.590630] [<ffffffff81064a1c>] ? wake_bit_function+0x0/0x23
[ 240.591906] [<ffffffff810bb9ea>] ? pagevec_lookup_tag+0x1a/0x21
[ 240.592954] [<ffffffff810b4577>] ? wait_on_page_writeback_range+0x69/0x11b
[ 240.594403] [<ffffffff810b536e>] ? filemap_write_and_wait+0x26/0x32
[ 240.595563] [<ffffffffa02c0d35>] ? nfs_setattr+0xb9/0x117 [nfs]
[ 240.596670] [<ffffffff810b3a0b>] ? find_get_page+0x1a/0x77
[ 240.598012] [<ffffffff810b3bb9>] ? lock_page+0x9/0x1f
[ 240.598878] [<ffffffff810b41ee>] ? filemap_fault+0xb9/0x2f6
[ 240.599839] [<ffffffff810ca3c2>] ? __do_fault+0x38c/0x3c3
[ 240.601003] [<ffffffff810ee1ce>] ? do_sync_write+0xce/0x113
[ 240.602082] [<ffffffff81051e75>] ? current_fs_time+0x1e/0x24
[ 240.602968] [<ffffffff811009b7>] ? notify_change+0x180/0x2c5
[ 240.604245] [<ffffffff8110b7b5>] ? utimes_common+0x12d/0x14d
[ 240.605355] [<ffffffff8110b856>] ? do_utimes+0x81/0xca
[ 240.606558] [<ffffffff8110b9ab>] ? sys_utimensat+0x5b/0x6a
[ 240.607817] [<ffffffff81010b42>] ? system_call_fastpath+0x16/0x1b
[ 240.609124] INFO: task find:1866 blocked for more than 120 seconds.
[ 240.610409] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 240.612066] find D 0000000000000000 0 1866 1863 0x00000000
[ 240.613490] ffffffff8145d1f0 0000000000000086 0000000000000000 ffff88011e2d2350
[ 240.615188] 00000022b63d07c7 ffff88011c55e000 000000000000f9e0 ffff8800c78a5fd8
[ 240.616576] 0000000000015780 0000000000015780 ffff88011e2d2350 ffff88011e2d2648
[ 240.618297] Call Trace:
[ 240.618777] [<ffffffff810e5369>] ? virt_to_head_page+0x9/0x2a
[ 240.619906] [<ffffffff812fa07a>] ? __mutex_lock_common+0x122/0x192
[ 240.621324] [<ffffffff812fa1a2>] ? mutex_lock+0x1a/0x31
[ 240.622543] [<ffffffff81102c11>] ? mntput_no_expire+0x23/0xee
[ 240.623860] [<ffffffffa02c0b03>] ? nfs_getattr+0x3b/0xda [nfs]
[ 240.625219] [<ffffffff810f1839>] ? vfs_fstatat+0x43/0x57
[ 240.626290] [<ffffffff810f185e>] ? sys_newfstatat+0x11/0x30
[ 240.627594] [<ffffffff81102c11>] ? mntput_no_expire+0x23/0xee
[ 240.628768] [<ffffffff8101195b>] ? device_not_available+0x1b/0x20
[ 240.629644] [<ffffffff81010b42>] ? system_call_fastpath+0x16/0x1b
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
* [Qemu-devel] Re: BUG - qdev - partial loss of network connectivity
@ 2010-09-27 21:32 ` Leszek Urbanski
0 siblings, 0 replies; 18+ messages in thread
From: Leszek Urbanski @ 2010-09-27 21:32 UTC (permalink / raw)
To: Michael S. Tsirkin; +Cc: netdev, linux-nfs, qemu-devel, virtualization
<20100926154324.GD21843@redhat.com>; from Michael S. Tsirkin on Sun, Sep 26, 2010 at 17:43:24 +0200
> > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > >(based on 2.6.32.21).
> > > >
> > > >If offload is the only difference, I'll play with different offload
> > > >options and check which one causes it.
> > > >
> > >
> > > It's not technically the only difference but it's the most likely
> > > culprit IMHO.
> >
> > udp fragmentation offload is definitely the culprit.
>
> I see. Most likely guest bug - won't be the first bug around UFO.
> If so pls copy netdev linux-nfs and virtualization.
> Do you see anything in dmesg? Can try 2.6.36-rc5?
(for reference: first post is at:
http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
fixed it, or should I dissect?
2.6.32.x - there's nothing interesting in dmesg, apart from traces related
to tasks in D state waiting on the NFS mounts:
[ 84.396127] nfs: server 10.0.0.1 not responding, still trying
[ 240.568162] INFO: task cp:1838 blocked for more than 120 seconds.
[ 240.569715] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 240.571486] cp D 0000000000000002 0 1838 1831 0x00000000
[ 240.573340] ffff88011fa5b880 0000000000000082 0000000000000000 ffff88011e45bb44
[ 240.575508] ffff88011e45bcc8 ffffffff8102cdac 000000000000f9e0 ffff88011e45bfd8
[ 240.578827] 0000000000015780 0000000000015780 ffff88011c7ce2e0 ffff88011c7ce5d8
[ 240.580502] Call Trace:
[ 240.581132] [<ffffffff8102cdac>] ? pvclock_clocksource_read+0x3a/0x8b
[ 240.582427] [<ffffffff8102cdac>] ? pvclock_clocksource_read+0x3a/0x8b
[ 240.583869] [<ffffffff810b3bdd>] ? sync_page+0x0/0x46
[ 240.585034] [<ffffffff810b3bdd>] ? sync_page+0x0/0x46
[ 240.586087] [<ffffffff812f9939>] ? io_schedule+0x73/0xb7
[ 240.587287] [<ffffffff810b3c1e>] ? sync_page+0x41/0x46
[ 240.588202] [<ffffffff812f9e46>] ? __wait_on_bit+0x41/0x70
[ 240.589314] [<ffffffff810b3da2>] ? wait_on_page_bit+0x6b/0x71
[ 240.590630] [<ffffffff81064a1c>] ? wake_bit_function+0x0/0x23
[ 240.591906] [<ffffffff810bb9ea>] ? pagevec_lookup_tag+0x1a/0x21
[ 240.592954] [<ffffffff810b4577>] ? wait_on_page_writeback_range+0x69/0x11b
[ 240.594403] [<ffffffff810b536e>] ? filemap_write_and_wait+0x26/0x32
[ 240.595563] [<ffffffffa02c0d35>] ? nfs_setattr+0xb9/0x117 [nfs]
[ 240.596670] [<ffffffff810b3a0b>] ? find_get_page+0x1a/0x77
[ 240.598012] [<ffffffff810b3bb9>] ? lock_page+0x9/0x1f
[ 240.598878] [<ffffffff810b41ee>] ? filemap_fault+0xb9/0x2f6
[ 240.599839] [<ffffffff810ca3c2>] ? __do_fault+0x38c/0x3c3
[ 240.601003] [<ffffffff810ee1ce>] ? do_sync_write+0xce/0x113
[ 240.602082] [<ffffffff81051e75>] ? current_fs_time+0x1e/0x24
[ 240.602968] [<ffffffff811009b7>] ? notify_change+0x180/0x2c5
[ 240.604245] [<ffffffff8110b7b5>] ? utimes_common+0x12d/0x14d
[ 240.605355] [<ffffffff8110b856>] ? do_utimes+0x81/0xca
[ 240.606558] [<ffffffff8110b9ab>] ? sys_utimensat+0x5b/0x6a
[ 240.607817] [<ffffffff81010b42>] ? system_call_fastpath+0x16/0x1b
[ 240.609124] INFO: task find:1866 blocked for more than 120 seconds.
[ 240.610409] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 240.612066] find D 0000000000000000 0 1866 1863 0x00000000
[ 240.613490] ffffffff8145d1f0 0000000000000086 0000000000000000 ffff88011e2d2350
[ 240.615188] 00000022b63d07c7 ffff88011c55e000 000000000000f9e0 ffff8800c78a5fd8
[ 240.616576] 0000000000015780 0000000000015780 ffff88011e2d2350 ffff88011e2d2648
[ 240.618297] Call Trace:
[ 240.618777] [<ffffffff810e5369>] ? virt_to_head_page+0x9/0x2a
[ 240.619906] [<ffffffff812fa07a>] ? __mutex_lock_common+0x122/0x192
[ 240.621324] [<ffffffff812fa1a2>] ? mutex_lock+0x1a/0x31
[ 240.622543] [<ffffffff81102c11>] ? mntput_no_expire+0x23/0xee
[ 240.623860] [<ffffffffa02c0b03>] ? nfs_getattr+0x3b/0xda [nfs]
[ 240.625219] [<ffffffff810f1839>] ? vfs_fstatat+0x43/0x57
[ 240.626290] [<ffffffff810f185e>] ? sys_newfstatat+0x11/0x30
[ 240.627594] [<ffffffff81102c11>] ? mntput_no_expire+0x23/0xee
[ 240.628768] [<ffffffff8101195b>] ? device_not_available+0x1b/0x20
[ 240.629644] [<ffffffff81010b42>] ? system_call_fastpath+0x16/0x1b
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: BUG - qdev - partial loss of network connectivity
2010-09-26 15:43 ` [Qemu-devel] " Michael S. Tsirkin
2010-09-27 21:32 ` Leszek Urbanski
@ 2010-09-27 21:32 ` Leszek Urbanski
1 sibling, 0 replies; 18+ messages in thread
From: Leszek Urbanski @ 2010-09-27 21:32 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: netdev, linux-nfs, qemu-devel, Anthony Liguori, virtualization
<20100926154324.GD21843@redhat.com>; from Michael S. Tsirkin on Sun, Sep 26, 2010 at 17:43:24 +0200
> > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > >(based on 2.6.32.21).
> > > >
> > > >If offload is the only difference, I'll play with different offload
> > > >options and check which one causes it.
> > > >
> > >
> > > It's not technically the only difference but it's the most likely
> > > culprit IMHO.
> >
> > udp fragmentation offload is definitely the culprit.
>
> I see. Most likely guest bug - won't be the first bug around UFO.
> If so pls copy netdev linux-nfs and virtualization.
> Do you see anything in dmesg? Can try 2.6.36-rc5?
(for reference: first post is at:
http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
fixed it, or should I dissect?
2.6.32.x - there's nothing interesting in dmesg, apart from traces related
to tasks in D state waiting on the NFS mounts:
[ 84.396127] nfs: server 10.0.0.1 not responding, still trying
[ 240.568162] INFO: task cp:1838 blocked for more than 120 seconds.
[ 240.569715] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 240.571486] cp D 0000000000000002 0 1838 1831 0x00000000
[ 240.573340] ffff88011fa5b880 0000000000000082 0000000000000000 ffff88011e45bb44
[ 240.575508] ffff88011e45bcc8 ffffffff8102cdac 000000000000f9e0 ffff88011e45bfd8
[ 240.578827] 0000000000015780 0000000000015780 ffff88011c7ce2e0 ffff88011c7ce5d8
[ 240.580502] Call Trace:
[ 240.581132] [<ffffffff8102cdac>] ? pvclock_clocksource_read+0x3a/0x8b
[ 240.582427] [<ffffffff8102cdac>] ? pvclock_clocksource_read+0x3a/0x8b
[ 240.583869] [<ffffffff810b3bdd>] ? sync_page+0x0/0x46
[ 240.585034] [<ffffffff810b3bdd>] ? sync_page+0x0/0x46
[ 240.586087] [<ffffffff812f9939>] ? io_schedule+0x73/0xb7
[ 240.587287] [<ffffffff810b3c1e>] ? sync_page+0x41/0x46
[ 240.588202] [<ffffffff812f9e46>] ? __wait_on_bit+0x41/0x70
[ 240.589314] [<ffffffff810b3da2>] ? wait_on_page_bit+0x6b/0x71
[ 240.590630] [<ffffffff81064a1c>] ? wake_bit_function+0x0/0x23
[ 240.591906] [<ffffffff810bb9ea>] ? pagevec_lookup_tag+0x1a/0x21
[ 240.592954] [<ffffffff810b4577>] ? wait_on_page_writeback_range+0x69/0x11b
[ 240.594403] [<ffffffff810b536e>] ? filemap_write_and_wait+0x26/0x32
[ 240.595563] [<ffffffffa02c0d35>] ? nfs_setattr+0xb9/0x117 [nfs]
[ 240.596670] [<ffffffff810b3a0b>] ? find_get_page+0x1a/0x77
[ 240.598012] [<ffffffff810b3bb9>] ? lock_page+0x9/0x1f
[ 240.598878] [<ffffffff810b41ee>] ? filemap_fault+0xb9/0x2f6
[ 240.599839] [<ffffffff810ca3c2>] ? __do_fault+0x38c/0x3c3
[ 240.601003] [<ffffffff810ee1ce>] ? do_sync_write+0xce/0x113
[ 240.602082] [<ffffffff81051e75>] ? current_fs_time+0x1e/0x24
[ 240.602968] [<ffffffff811009b7>] ? notify_change+0x180/0x2c5
[ 240.604245] [<ffffffff8110b7b5>] ? utimes_common+0x12d/0x14d
[ 240.605355] [<ffffffff8110b856>] ? do_utimes+0x81/0xca
[ 240.606558] [<ffffffff8110b9ab>] ? sys_utimensat+0x5b/0x6a
[ 240.607817] [<ffffffff81010b42>] ? system_call_fastpath+0x16/0x1b
[ 240.609124] INFO: task find:1866 blocked for more than 120 seconds.
[ 240.610409] "echo 0 > /proc/sys/kernel/hung_task_timeout_secs" disables this message.
[ 240.612066] find D 0000000000000000 0 1866 1863 0x00000000
[ 240.613490] ffffffff8145d1f0 0000000000000086 0000000000000000 ffff88011e2d2350
[ 240.615188] 00000022b63d07c7 ffff88011c55e000 000000000000f9e0 ffff8800c78a5fd8
[ 240.616576] 0000000000015780 0000000000015780 ffff88011e2d2350 ffff88011e2d2648
[ 240.618297] Call Trace:
[ 240.618777] [<ffffffff810e5369>] ? virt_to_head_page+0x9/0x2a
[ 240.619906] [<ffffffff812fa07a>] ? __mutex_lock_common+0x122/0x192
[ 240.621324] [<ffffffff812fa1a2>] ? mutex_lock+0x1a/0x31
[ 240.622543] [<ffffffff81102c11>] ? mntput_no_expire+0x23/0xee
[ 240.623860] [<ffffffffa02c0b03>] ? nfs_getattr+0x3b/0xda [nfs]
[ 240.625219] [<ffffffff810f1839>] ? vfs_fstatat+0x43/0x57
[ 240.626290] [<ffffffff810f185e>] ? sys_newfstatat+0x11/0x30
[ 240.627594] [<ffffffff81102c11>] ? mntput_no_expire+0x23/0xee
[ 240.628768] [<ffffffff8101195b>] ? device_not_available+0x1b/0x20
[ 240.629644] [<ffffffff81010b42>] ? system_call_fastpath+0x16/0x1b
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: BUG - qdev - partial loss of network connectivity
2010-09-27 21:32 ` Leszek Urbanski
(?)
@ 2010-09-28 9:50 ` Michael S. Tsirkin
-1 siblings, 0 replies; 18+ messages in thread
From: Michael S. Tsirkin @ 2010-09-28 9:50 UTC (permalink / raw)
To: Leszek Urbanski; +Cc: netdev, linux-nfs, qemu-devel, virtualization
On Mon, Sep 27, 2010 at 11:32:03PM +0200, Leszek Urbanski wrote:
> <20100926154324.GD21843@redhat.com>; from Michael S. Tsirkin on Sun, Sep 26, 2010 at 17:43:24 +0200
>
> > > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > > >(based on 2.6.32.21).
> > > > >
> > > > >If offload is the only difference, I'll play with different offload
> > > > >options and check which one causes it.
> > > > >
> > > >
> > > > It's not technically the only difference but it's the most likely
> > > > culprit IMHO.
> > >
> > > udp fragmentation offload is definitely the culprit.
> >
> > I see. Most likely guest bug - won't be the first bug around UFO.
> > If so pls copy netdev linux-nfs and virtualization.
> > Do you see anything in dmesg? Can try 2.6.36-rc5?
>
> (for reference: first post is at:
> http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
>
> I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
> fixed it, or should I dissect?
bisect, yes: there were many UFO related patches since 2.6.32.
--
MST
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: BUG - qdev - partial loss of network connectivity
@ 2010-09-28 9:50 ` Michael S. Tsirkin
0 siblings, 0 replies; 18+ messages in thread
From: Michael S. Tsirkin @ 2010-09-28 9:50 UTC (permalink / raw)
To: Leszek Urbanski
Cc: Anthony Liguori, qemu-devel, netdev, linux-nfs, virtualization
On Mon, Sep 27, 2010 at 11:32:03PM +0200, Leszek Urbanski wrote:
> <20100926154324.GD21843@redhat.com>; from Michael S. Tsirkin on Sun, Sep 26, 2010 at 17:43:24 +0200
>
> > > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > > >(based on 2.6.32.21).
> > > > >
> > > > >If offload is the only difference, I'll play with different offload
> > > > >options and check which one causes it.
> > > > >
> > > >
> > > > It's not technically the only difference but it's the most likely
> > > > culprit IMHO.
> > >
> > > udp fragmentation offload is definitely the culprit.
> >
> > I see. Most likely guest bug - won't be the first bug around UFO.
> > If so pls copy netdev linux-nfs and virtualization.
> > Do you see anything in dmesg? Can try 2.6.36-rc5?
>
> (for reference: first post is at:
> http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
>
> I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
> fixed it, or should I dissect?
bisect, yes: there were many UFO related patches since 2.6.32.
--
MST
^ permalink raw reply [flat|nested] 18+ messages in thread
* [Qemu-devel] Re: BUG - qdev - partial loss of network connectivity
@ 2010-09-28 9:50 ` Michael S. Tsirkin
0 siblings, 0 replies; 18+ messages in thread
From: Michael S. Tsirkin @ 2010-09-28 9:50 UTC (permalink / raw)
To: Leszek Urbanski; +Cc: netdev, linux-nfs, qemu-devel, virtualization
On Mon, Sep 27, 2010 at 11:32:03PM +0200, Leszek Urbanski wrote:
> <20100926154324.GD21843@redhat.com>; from Michael S. Tsirkin on Sun, Sep 26, 2010 at 17:43:24 +0200
>
> > > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > > >(based on 2.6.32.21).
> > > > >
> > > > >If offload is the only difference, I'll play with different offload
> > > > >options and check which one causes it.
> > > > >
> > > >
> > > > It's not technically the only difference but it's the most likely
> > > > culprit IMHO.
> > >
> > > udp fragmentation offload is definitely the culprit.
> >
> > I see. Most likely guest bug - won't be the first bug around UFO.
> > If so pls copy netdev linux-nfs and virtualization.
> > Do you see anything in dmesg? Can try 2.6.36-rc5?
>
> (for reference: first post is at:
> http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
>
> I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
> fixed it, or should I dissect?
bisect, yes: there were many UFO related patches since 2.6.32.
--
MST
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: BUG - qdev - partial loss of network connectivity
2010-09-27 21:32 ` Leszek Urbanski
(?)
(?)
@ 2010-09-28 9:50 ` Michael S. Tsirkin
-1 siblings, 0 replies; 18+ messages in thread
From: Michael S. Tsirkin @ 2010-09-28 9:50 UTC (permalink / raw)
To: Leszek Urbanski
Cc: netdev, linux-nfs, qemu-devel, Anthony Liguori, virtualization
On Mon, Sep 27, 2010 at 11:32:03PM +0200, Leszek Urbanski wrote:
> <20100926154324.GD21843@redhat.com>; from Michael S. Tsirkin on Sun, Sep 26, 2010 at 17:43:24 +0200
>
> > > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > > >(based on 2.6.32.21).
> > > > >
> > > > >If offload is the only difference, I'll play with different offload
> > > > >options and check which one causes it.
> > > > >
> > > >
> > > > It's not technically the only difference but it's the most likely
> > > > culprit IMHO.
> > >
> > > udp fragmentation offload is definitely the culprit.
> >
> > I see. Most likely guest bug - won't be the first bug around UFO.
> > If so pls copy netdev linux-nfs and virtualization.
> > Do you see anything in dmesg? Can try 2.6.36-rc5?
>
> (for reference: first post is at:
> http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
>
> I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
> fixed it, or should I dissect?
bisect, yes: there were many UFO related patches since 2.6.32.
--
MST
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: BUG - qdev - partial loss of network connectivity
2010-09-27 21:32 ` Leszek Urbanski
(?)
@ 2010-10-05 21:29 ` Leszek Urbanski
-1 siblings, 0 replies; 18+ messages in thread
From: Leszek Urbanski @ 2010-10-05 21:29 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: Anthony Liguori, qemu-devel-qX2TKyscuCcdnm+yROfE0A,
netdev-u79uwXL29TY76Z2rM5mHXA, linux-nfs-u79uwXL29TY76Z2rM5mHXA,
virtualization-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA
<20100927213203.GA28089-4cL5GMJPxME@public.gmane.org>; from Leszek Urbanski on Mon, Sep 27, 2010 at 23:32:03 +0200
> > > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > > >(based on 2.6.32.21).
> > > > >
> > > > >If offload is the only difference, I'll play with different offload
> > > > >options and check which one causes it.
> > > > >
> > > >
> > > > It's not technically the only difference but it's the most likely
> > > > culprit IMHO.
> > >
> > > udp fragmentation offload is definitely the culprit.
> >
> > I see. Most likely guest bug - won't be the first bug around UFO.
> > If so pls copy netdev linux-nfs and virtualization.
> > Do you see anything in dmesg? Can try 2.6.36-rc5?
>
> (for reference: first post is at:
> http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
>
> I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
This one definitely fixes it: http://patchwork.ozlabs.org/patch/55643/
(Herbert Xu: udp: Fix bogus UFO packet generation)
The patch works with 2.6.32.24 too - it should probably be backported to
2.6.32.x.
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
--
To unsubscribe from this list: send the line "unsubscribe linux-nfs" in
the body of a message to majordomo-u79uwXL29TY76Z2rM5mHXA@public.gmane.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: BUG - qdev - partial loss of network connectivity
@ 2010-10-05 21:29 ` Leszek Urbanski
0 siblings, 0 replies; 18+ messages in thread
From: Leszek Urbanski @ 2010-10-05 21:29 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: Anthony Liguori, qemu-devel, netdev, linux-nfs, virtualization
<20100927213203.GA28089@moo.pl>; from Leszek Urbanski on Mon, Sep 27, 2010 at 23:32:03 +0200
> > > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > > >(based on 2.6.32.21).
> > > > >
> > > > >If offload is the only difference, I'll play with different offload
> > > > >options and check which one causes it.
> > > > >
> > > >
> > > > It's not technically the only difference but it's the most likely
> > > > culprit IMHO.
> > >
> > > udp fragmentation offload is definitely the culprit.
> >
> > I see. Most likely guest bug - won't be the first bug around UFO.
> > If so pls copy netdev linux-nfs and virtualization.
> > Do you see anything in dmesg? Can try 2.6.36-rc5?
>
> (for reference: first post is at:
> http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
>
> I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
This one definitely fixes it: http://patchwork.ozlabs.org/patch/55643/
(Herbert Xu: udp: Fix bogus UFO packet generation)
The patch works with 2.6.32.24 too - it should probably be backported to
2.6.32.x.
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
* [Qemu-devel] Re: BUG - qdev - partial loss of network connectivity
@ 2010-10-05 21:29 ` Leszek Urbanski
0 siblings, 0 replies; 18+ messages in thread
From: Leszek Urbanski @ 2010-10-05 21:29 UTC (permalink / raw)
To: Michael S. Tsirkin; +Cc: netdev, linux-nfs, qemu-devel, virtualization
<20100927213203.GA28089@moo.pl>; from Leszek Urbanski on Mon, Sep 27, 2010 at 23:32:03 +0200
> > > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > > >(based on 2.6.32.21).
> > > > >
> > > > >If offload is the only difference, I'll play with different offload
> > > > >options and check which one causes it.
> > > > >
> > > >
> > > > It's not technically the only difference but it's the most likely
> > > > culprit IMHO.
> > >
> > > udp fragmentation offload is definitely the culprit.
> >
> > I see. Most likely guest bug - won't be the first bug around UFO.
> > If so pls copy netdev linux-nfs and virtualization.
> > Do you see anything in dmesg? Can try 2.6.36-rc5?
>
> (for reference: first post is at:
> http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
>
> I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
This one definitely fixes it: http://patchwork.ozlabs.org/patch/55643/
(Herbert Xu: udp: Fix bogus UFO packet generation)
The patch works with 2.6.32.24 too - it should probably be backported to
2.6.32.x.
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
* Re: BUG - qdev - partial loss of network connectivity
2010-09-27 21:32 ` Leszek Urbanski
` (4 preceding siblings ...)
(?)
@ 2010-10-05 21:29 ` Leszek Urbanski
-1 siblings, 0 replies; 18+ messages in thread
From: Leszek Urbanski @ 2010-10-05 21:29 UTC (permalink / raw)
To: Michael S. Tsirkin
Cc: netdev, linux-nfs, qemu-devel, Anthony Liguori, virtualization
<20100927213203.GA28089@moo.pl>; from Leszek Urbanski on Mon, Sep 27, 2010 at 23:32:03 +0200
> > > > >It's vanilla 2.6.32.22, but I also reproduced this on Debian's 2.6.32-23
> > > > >(based on 2.6.32.21).
> > > > >
> > > > >If offload is the only difference, I'll play with different offload
> > > > >options and check which one causes it.
> > > > >
> > > >
> > > > It's not technically the only difference but it's the most likely
> > > > culprit IMHO.
> > >
> > > udp fragmentation offload is definitely the culprit.
> >
> > I see. Most likely guest bug - won't be the first bug around UFO.
> > If so pls copy netdev linux-nfs and virtualization.
> > Do you see anything in dmesg? Can try 2.6.36-rc5?
>
> (for reference: first post is at:
> http://lists.nongnu.org/archive/html/qemu-devel/2010-09/msg01685.html )
>
> I can't reproduce it on 2.6.36-rc5. Do you have an idea which patch may have
This one definitely fixes it: http://patchwork.ozlabs.org/patch/55643/
(Herbert Xu: udp: Fix bogus UFO packet generation)
The patch works with 2.6.32.24 too - it should probably be backported to
2.6.32.x.
--
Leszek "Tygrys" Urbanski, SCSA, SCNA
"Unix-to-Unix Copy Program;" said PDP-1. "You will never find a more
wretched hive of bugs and flamers. We must be cautious." -- DECWARS
http://cygnus.moo.pl/ -- Cygnus High Altitude Balloon
^ permalink raw reply [flat|nested] 18+ messages in thread
end of thread, other threads:[~2010-10-05 21:29 UTC | newest]
Thread overview: 18+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2010-09-22 17:18 [Qemu-devel] BUG - qdev - partial loss of network connectivity Leszek Urbanski
2010-09-22 17:41 ` Anthony Liguori
2010-09-22 18:20 ` Leszek Urbanski
2010-09-22 18:35 ` Anthony Liguori
2010-09-23 14:04 ` Leszek Urbanski
2010-09-26 15:43 ` [Qemu-devel] " Michael S. Tsirkin
2010-09-27 21:32 ` Leszek Urbanski
2010-09-27 21:32 ` [Qemu-devel] " Leszek Urbanski
2010-09-27 21:32 ` Leszek Urbanski
2010-09-28 9:50 ` Michael S. Tsirkin
2010-09-28 9:50 ` Michael S. Tsirkin
2010-09-28 9:50 ` [Qemu-devel] " Michael S. Tsirkin
2010-09-28 9:50 ` Michael S. Tsirkin
[not found] ` <20100927213203.GA28089-4cL5GMJPxME@public.gmane.org>
2010-10-05 21:29 ` Leszek Urbanski
2010-10-05 21:29 ` [Qemu-devel] " Leszek Urbanski
2010-10-05 21:29 ` Leszek Urbanski
2010-10-05 21:29 ` Leszek Urbanski
2010-09-27 21:32 ` Leszek Urbanski
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.