ASIX USB-Ethernet driver noticably slow, dropping packets and reporting RX errors

[Adam Gandelman]

After updating a natty system to precise beta-1, the asix USB ethernet device began flooding console and kernel log with messages:

[68246.610332] asix 1-5:1.0: eth1: asix_rx_fixup() Bad RX Length 1416

ifconfig showed RX errors incrementing as traffic passed through the interface, and throughput dwindled as low as 100Kb/s on a 100Mb network.

Compiling the vendor's driver seems to remedy this. It seems there is a history of bugs related to this specific driver since 2.6.36 -- see https://bugzilla.kernel.org/show_bug.cgi?id=16831
---
AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.24.
ApportVersion: 1.94-0ubuntu1
Architecture: i386
ArecordDevices:
 Home directory /home/adam not ours.
 **** List of CAPTURE Hardware Devices ****
 card 0: NVidia [HDA NVidia], device 0: ALC662 rev1 Analog [ALC662 rev1 Analog]
   Subdevices: 1/1
   Subdevice #0: subdevice #0
AudioDevicesInUse: Error: command ['fuser', '-v', '/dev/snd/by-path', '/dev/snd/controlC0', '/dev/snd/hwC0D0', '/dev/snd/hwC0D3', '/dev/snd/pcmC0D0c', '/dev/snd/pcmC0D0p', '/dev/snd/pcmC0D3p', '/dev/snd/seq', '/dev/snd/timer'] failed with exit code 1:
CRDA: Error: command ['iw', 'reg', 'get'] failed with exit code 1: nl80211 not found.
Card0.Amixer.info:
 Card hw:0 'NVidia'/'HDA NVidia at 0xfae78000 irq 22'
   Mixer name : 'Nvidia MCP79/7A HDMI'
   Components : 'HDA:10ec0662,10250222,00100101 HDA:10de0007,10de0101,00100100'
   Controls : 19
   Simple ctrls : 9
DistroRelease: Ubuntu 12.04
HibernationDevice: RESUME=UUID=766017f1-32ff-4cb5-ae6a-4663fd5c0b13
IwConfig: Error: [Errno 2] No such file or directory
MachineType: Acer Aspire R1600
NonfreeKernelModules: nvidia
Package: linux (not installed)
ProcEnviron:
 TERM=xterm
 LANG=en_US
 SHELL=/bin/bash
ProcFB:

ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.2.0-17-generic root=UUID=3f712443-bb8a-49cc-8a3d-939a13400643 ro quit nosplash loglevel=0 vt.handoff=7
ProcVersionSignature: Ubuntu 3.2.0-17.27-generic 3.2.6
PulseSinks:
 Error: command ['pacmd', 'list-sinks'] failed with exit code 1: Home directory /home/adam not ours.
 No PulseAudio daemon running, or not running as session daemon.
PulseSources:
 Error: command ['pacmd', 'list-sources'] failed with exit code 1: Home directory /home/adam not ours.
 No PulseAudio daemon running, or not running as session daemon.
RelatedPackageVersions:
 linux-restricted-modules-3.2.0-17-generic N/A
 linux-backports-modules-3.2.0-17-generic N/A
 linux-firmware 1.71
RfKill: Error: [Errno 2] No such file or directory
Tags: precise
Uname: Linux 3.2.0-17-generic i686
UpgradeStatus: No upgrade log present (probably fresh install)
UserGroups:

dmi.bios.date: 06/17/2009
dmi.bios.vendor: American Megatrends Inc.
dmi.bios.version: P01-A1
dmi.board.asset.tag: To Be Filled By O.E.M.
dmi.board.name: FMCP7A-ION
dmi.board.vendor: Acer
dmi.chassis.type: 3
dmi.chassis.vendor: Acer
dmi.modalias: dmi:bvnAmericanMegatrendsInc.:bvrP01-A1:bd06/17/2009:svnAcer:pnAspireR1600:pvr:rvnAcer:rnFMCP7A-ION:rvr:cvnAcer:ct3:cvr:
dmi.product.name: Aspire R1600
dmi.sys.vendor: Acer

[Adam Gandelman (gandelman-a)]

The driver from the vendors site [1] works as expected compiled against 3.2.0-17-generic with the following patch: http://paste.ubuntu.com/870961/

[Brad Figg (brad-figg)]

This bug is missing log files that will aid in diagnosing the problem. From a terminal window please run:

apport-collect 947723

and then change the status of the bug to 'Confirmed'.

If, due to the nature of the issue you have encountered, you are unable to run this command, please add a comment stating that fact and change the bug status to 'Confirmed'.

This change has been made by an automated script, maintained by the Ubuntu Kernel Team.

[Adam Gandelman (gandelman-a)]

URL referenced above: http://www.asix.com.tw/FrootAttach/driver/AX88772B_772A_760_772_178_LINUX_Driver_v4.2.0_Source.zip

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Adam Gandelman (gandelman-a)]

apport information

[Joseph Salisbury (jsalisbury)]

There is a newer version of the kernel than the one you tested when this issue was found. Please test again with the newer kernel and indicate in the bug if this issue still exists or not. You can upgrade just your kernel by running the following in a terminal:

sudo apt-get update
sudo apt-get install linux

If the bug still exists in the latest kernel, please change the status back to "Confirmed". If the bug is fixed, please change the status to "Fix Released".

Thanks in advance

[Launchpad Janitor (janitor)]

[Expired for linux (Ubuntu) because there has been no activity for 60 days.]

[Diego Carrera Gallego (diegocarrera2000)]

I got same problem on ux31e and ubuntu 12.04 & kernel 3.2.0-22.

On syslog i found a lot of logs about Bad Header Length...
asix 2-1:1.0: eth0: asix_rx_fixup() Bad Header Length
asix 2-1:1.0: eth0: asix_rx_fixup() Bad RX Length 1716

So, when i copy a big file (size ~ 3,7 Gb ) from home network with 100MBps of capacity, my ux31e only transfers 300kB/s.
I checked the network hardware with "sudo lshw -c network" and its says that support 100Mbits.

[Diego Carrera Gallego (diegocarrera2000)]

I found a solution on wiki about AsusZenbook.
https://help.ubuntu.com/community/AsusZenbook

You have to download driver from
http://www.asix.com.tw/download.php?sub=driverdetail&PItemID=105

For 3.2 kernels, the following patch has to be applied on this drivers:

--- a/asix.c 2011-03-10 13:46:05.000000000 +0100
+++ b/asix.c 2011-11-26 09:39:48.000000000 +0100
@@ -829,7 +829,7 @@
        .ndo_do_ioctl = ax8817x_ioctl,
        .ndo_set_mac_address = ax8817x_set_mac_addr,
        .ndo_validate_addr = eth_validate_addr,
- .ndo_set_multicast_list = ax8817x_set_multicast,
+ .ndo_set_rx_mode = ax8817x_set_multicast,
 };
 #endif

@@ -1522,7 +1522,7 @@
        .ndo_get_stats = axusbnet_get_stats,
        .ndo_set_mac_address = ax8817x_set_mac_addr,
        .ndo_validate_addr = eth_validate_addr,
- .ndo_set_multicast_list = ax88772b_set_multicast,
+ .ndo_set_rx_mode = ax88772b_set_multicast,
 };
 #endif

Make & make install

and it is fixed .

I hope that Ubuntu kernel team include this patch.

[Chris (gentlec)]

I can confirm this on Ubuntu Server 12.04 running kernel 3.2.0-25-generic-pae. It's surprising that this bug isn't getting more attention since the majority of USB NICs seem to have Asix chipsets. I'm using Ubuntu Server on an Atom box as a router and unfortunately am having to use a USB NIC as the public ethernet interface going to my cable modem. Because of this bug, my throughput is cut down from 40Mbps to around 10Mbps.

[Ming Lei (tom-leiming)]

Could you guys provide some products(brand, type) which is using Asix
chipsets?

I'd like to port the vendor specific driver into upstream if I have a NICs to
test.

On Thu, Jun 21, 2012 at 2:06 AM, mr_echo <email address hidden> wrote:
> I can confirm this on Ubuntu Server 12.04 running kernel 3.2.0-25
> -generic-pae.  It's surprising that this bug isn't getting more
> attention since the majority of USB NICs seem to have Asix chipsets.
> I'm using Ubuntu Server on an Atom box as a router and unfortunately am
> having to use a USB NIC as the public ethernet interface going to my
> cable modem.  Because of this bug, my throughput is cut down from 40Mbps
> to around 10Mbps.
>
> --
> You received this bug notification because you are subscribed to linux
> in Ubuntu.
> https://bugs.launchpad.net/bugs/947723
>
> Title:
>  ASIX USB-Ethernet driver noticably slow, dropping packets and
>  reporting RX errors
>
> Status in “linux” package in Ubuntu:
>  Confirmed
>
> Bug description:
>  After updating a natty system to precise beta-1, the asix USB ethernet
>  device began flooding console and kernel log with messages:
>
>  [68246.610332] asix 1-5:1.0: eth1: asix_rx_fixup() Bad RX Length 1416
>
>  ifconfig showed RX errors incrementing as traffic passed through the
>  interface, and throughput dwindled as low as 100Kb/s on a 100Mb
>  network.
>
>  Compiling the vendor's driver seems to remedy this.  It seems there is a history of bugs related to this specific driver since 2.6.36 -- see https://bugzilla.kernel.org/show_bug.cgi?id=16831
>  ---
>  AlsaVersion: Advanced Linux Sound Architecture Driver Version 1.0.24.
>  ApportVersion: 1.94-0ubuntu1
>  Architecture: i386
>  ArecordDevices:
>   Home directory /home/adam not ours.
>   **** List of CAPTURE Hardware Devices ****
>   card 0: NVidia [HDA NVidia], device 0: ALC662 rev1 Analog [ALC662 rev1 Analog]
>     Subdevices: 1/1
>     Subdevice #0: subdevice #0
>  AudioDevicesInUse: Error: command ['fuser', '-v', '/dev/snd/by-path', '/dev/snd/controlC0', '/dev/snd/hwC0D0', '/dev/snd/hwC0D3', '/dev/snd/pcmC0D0c', '/dev/snd/pcmC0D0p', '/dev/snd/pcmC0D3p', '/dev/snd/seq', '/dev/snd/timer'] failed with exit code 1:
>  CRDA: Error: command ['iw', 'reg', 'get'] failed with exit code 1: nl80211 not found.
>  Card0.Amixer.info:
>   Card hw:0 'NVidia'/'HDA NVidia at 0xfae78000 irq 22'
>     Mixer name : 'Nvidia MCP79/7A HDMI'
>     Components : 'HDA:10ec0662,10250222,00100101 HDA:10de0007,10de0101,00100100'
>     Controls      : 19
>     Simple ctrls  : 9
>  DistroRelease: Ubuntu 12.04
>  HibernationDevice: RESUME=UUID=766017f1-32ff-4cb5-ae6a-4663fd5c0b13
>  IwConfig: Error: [Errno 2] No such file or directory
>  MachineType: Acer Aspire R1600
>  NonfreeKernelModules: nvidia
>  Package: linux (not installed)
>  ProcEnviron:
>   TERM=xterm
>   LANG=en_US
>   SHELL=/bin/bash
>  ProcFB:
>
>  ProcKernelCmdLine: BOOT_IMAGE=/boot/vmlinuz-3.2.0-17-generic root=UUID=3f712443-bb8a-49cc-8a3d-939a13400643 ro quit nosplash loglevel=0 vt.handoff=7
>  ProcVersionSignature: Ubuntu 3.2.0-17.27-generic 3.2.6
>  PulseSinks:
>   Error: command ['pacmd', 'list-sinks'] failed with exit code 1: Home directory /home/adam not ours.
> ...

[Gino (ginoputrino)]

Hi,
  for me it is the usb to ethernet adapter that came with my asus zenbook U31E. (USB ID is 0b95:7e2b).

You should be able to buy these individually, as you probably don't want to buy a whole laptop to get one to test.

[Ming Lei (tom-leiming)]

I have integrated two recent related fix on rx path, could anyone test
the images
in below link?

http://kernel.ubuntu.com/~ming/bugs/947723/

BTW: the transfer is ongoing, please wait 10 minutes to download the images.

[Chris (gentlec)]

The adapter I'm using is the D-Link DUB-E100 USB 2.0 Fast Ethernet Adapter.

[Chris (gentlec)]

By the way, the D-Link DUB-E100 USB 2.0 Fast Ethernet Adapter uses the Asix AX88772 chip.

[Chris (gentlec)]

I am unable to test your kernel packages. I'm on a 32 bit Atom and your packages are 64 bit. Did you do a 32 bit build?

[Ming Lei (tom-leiming)]

On Fri, Jun 22, 2012 at 8:05 AM, mr_echo <email address hidden> wrote:
> I am unable to test your kernel packages.  I'm on a 32 bit Atom and your
> packages are 64 bit.  Did you do a 32 bit build?

I just built 32bit images, see below link:

          http://kernel.ubuntu.com/~ming/bugs/947723/i386/

Thanks,

[Gino (ginoputrino)]

Hi Ming,
     I've given the 64-bit 3.2.0-26-generic kernel you supplied a try:
Generally performance seems better. Hard to tell because the rest of the network is busy right now.
My dmesg is still filling up with these:

[ 872.587526] asix 2-1:1.0: eth0: asix_rx_fixup() Bad RX Length 1434
[ 872.587537] asix 2-1:1.0: eth0: asix_rx_fixup() Bad Header Length
[ 872.628519] asix 2-1:1.0: eth0: asix_rx_fixup() Bad RX Length 1434
[ 872.628529] asix 2-1:1.0: eth0: asix_rx_fixup() Bad Header Length
[ 873.376564] asix 2-1:1.0: eth0: asix_rx_fixup() Bad RX Length 1416
[ 873.376577] asix 2-1:1.0: eth0: asix_rx_fixup() Bad Header Length

I will keep testing and see how it goes.

[Ming Lei (tom-leiming)]

On Fri, Jun 22, 2012 at 12:46 PM, Gino <email address hidden> wrote:
> Hi Ming,
>     I've given the 64-bit 3.2.0-26-generic kernel you supplied a try:
> Generally performance seems better. Hard to tell because the rest of the network is busy right now.

If you have two boxes, you can try iperf to test performance. Generally,
TCP Rx or Tx can reach at least 90Mbps on a 100Mbps NIC.

> My dmesg is still filling up with these:
>
> [  872.587526] asix 2-1:1.0: eth0: asix_rx_fixup() Bad RX Length 1434
> [  872.587537] asix 2-1:1.0: eth0: asix_rx_fixup() Bad Header Length
> [  872.628519] asix 2-1:1.0: eth0: asix_rx_fixup() Bad RX Length 1434
> [  872.628529] asix 2-1:1.0: eth0: asix_rx_fixup() Bad Header Length
> [  873.376564] asix 2-1:1.0: eth0: asix_rx_fixup() Bad RX Length 1416
> [  873.376577] asix 2-1:1.0: eth0: asix_rx_fixup() Bad Header Length
>
> I will keep testing and see how it goes.

[Gino (ginoputrino)]

Hi Ming,
     no this kernel does not resolve the issue.
It starts off with ok speed, but over time degrades in exactly the same way. Using the vendor driver as recommended in the op definitely rectifies the issue.

Is there any debug info you want me to grab while running your module?

[Ming Lei (tom-leiming)]

On Mon, Jun 25, 2012 at 2:47 PM, Gino <email address hidden> wrote:
> Hi Ming,
>     no this kernel does not resolve the issue.
> It starts off with ok speed, but over time degrades in exactly the same way. Using the vendor driver as recommended in the op definitely rectifies the issue.
>
> Is there any debug info you want me to grab while running your module?

Could you test upstream kernel to see if the same problem can be reproduced?

            http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.5-rc4-quantal/

Thanks,

[Ming Lei (tom-leiming)]

After some discussion in upstream maillist, looks the problem is caused
by below commit:

commit a9e0aca4b37885b5599e52211f098bd7f565e749
Author: Eric Dumazet <email address hidden>
Date: Wed Mar 14 20:18:32 2012 +0000

    asix: asix_rx_fixup surgery to reduce skb truesizes

    asix_rx_fixup() is complex, and does some unnecessary memory copies (at
    least on x86 where NET_IP_ALIGN is 0)

    Also, it tends to provide skbs with a big truesize (4096+256 with
    MTU=1500) to upper stack, so incoming trafic consume a lot of memory and
    I noticed early packet drops because we hit socket rcvbuf too fast.

    Switch to a different strategy, using copybreak so that we provide nice
    skbs to upper stack (including the NET_SKB_PAD to avoid future head
    reallocations in some paths)

    With this patch, I no longer see packets drops or tcp collapses on
    various tcp workload with a AX88772 adapter.

so I built kernel image with the patch, anyone can have a test?

          http://kernel.ubuntu.com/~ming/bugs/947723/i386/2/

thanks

[Adam Gandelman (gandelman-a)]

Hi Ming-

Sorry for the delay on testing. I just managed to test your latest 32 bit kernel and it seems to be working great! No spam in my kernel log, ifconfig is reporting no errors and I am able to sustain good throughput with my asix device.

+1 to the proposed patch. Thanks, Ming.

[Ming Lei (tom-leiming)]

On Sat, Jul 14, 2012 at 1:46 PM, Adam Gandelman
<email address hidden> wrote:
> Hi Ming-
>
> Sorry for the delay on testing. I just managed to test your latest 32
> bit kernel and it seems to be working great! No spam in my kernel log,
> ifconfig is reporting no errors and I am able to sustain good throughput
> with my asix device.
>
>
> +1 to the proposed patch. Thanks, Ming.

Great, thanks for your test, and I will push the patch to precise and
other release.

Thanks

[Luis Henriques (henrix)]

This bug is awaiting verification that the kernel for Precise in -proposed solves the problem (3.2.0-29.46). Please test the kernel and update this bug with the results. If the problem is solved, change the tag 'verification-needed-precise' to 'verification-done-precise'.

If verification is not done by one week from today, this fix will be dropped from the source code, and this bug will be closed.

See https://wiki.ubuntu.com/Testing/EnableProposed for documentation how to enable and use -proposed. Thank you!

[David (launux)]

Hello,

I have this bug since yesterday, after the kernel proposed :

3.2.0-29-generic-pae #46-Ubuntu SMP Fri Jul 27 17:25:43 UTC 2012 i686 i686 i386 GNU/Linux

Jul 30 20:47:01 kerbex kernel: [ 203.590907] asix 1-6:1.0: eth1: asix_rx_fixup() Bad Header Length
Jul 30 20:47:02 kerbex kernel: [ 203.730211] asix 1-6:1.0: eth1: asix_rx_fixup() Bad RX Length 1386
Jul 30 20:47:02 kerbex kernel: [ 203.731536] asix 1-6:1.0: eth1: asix_rx_fixup() Bad Header Length
Jul 30 20:47:02 kerbex kernel: [ 203.829838] asix 1-6:1.0: eth1: asix_rx_fixup() Bad RX Length 1386
Jul 30 20:47:02 kerbex kernel: [ 203.831164] asix 1-6:1.0: eth1: asix_rx_fixup() Bad Header Length
Jul 30 20:47:02 kerbex kernel: [ 203.849838] asix 1-6:1.0: eth1: asix_rx_fixup() Bad RX Length 1386
Jul 30 20:47:02 kerbex kernel: [ 203.851158] asix 1-6:1.0: eth1: asix_rx_fixup() Bad Header Length
Jul 30 20:47:02 kerbex kernel: [ 203.869836] asix 1-6:1.0: eth1: asix_rx_fixup() Bad RX Length 1386
Jul 30 20:47:02 kerbex kernel: [ 203.871139] asix 1-6:1.0: eth1: asix_rx_fixup() Bad Header Length
Jul 30 20:47:02 kerbex kernel: [ 203.930213] asix 1-6:1.0: eth1: asix_rx_fixup() Bad RX Length 1386
Jul 30 20:47:02 kerbex kernel: [ 203.931518] asix 1-6:1.0: eth1: asix_rx_fixup() Bad Header Length
Jul 30 20:47:03 kerbex kernel: [ 204.872998] asix 1-6:1.0: eth1: asix_rx_fixup() Bad RX Length 1386

The frequency seems less important than before and the length (1386) is always the same unlike the old kernel

Thanks

David

[Ming Lei (tom-leiming)]

On Tue, Jul 31, 2012 at 2:57 AM, David <email address hidden> wrote:

>
> The frequency seems less important than before and the length (1386) is
> always the same unlike the old kernel

Could you test throughput to see if the speed is slowed down when the log
is printed?

Thanks

[David (launux)]

I do not know how to measure throughput, but is more slowly. The computer is used to make a bridge with vlan 802.1q for my Freebox. My TV flow is not fluid since the new kernel.

[Ming Lei (tom-leiming)]

On Tue, Jul 31, 2012 at 4:16 PM, David <email address hidden> wrote:
> I do not know how to measure throughput, but is more slowly. The
> computer is used to make a bridge with vlan 802.1q for my Freebox. My TV

Could you just test the common ethernet interface function instead of vlan
802.1q to see if it is OK? Wrt. 802.1q, it may be another problem.

> flow is not fluid since the new kernel.

IMO, this should be a way to test throughput.

Thanks,

[David (launux)]

I don't know IMO ! Without VLAN, nautilus transfert file via NFS with throughput of 11mb/s. However, there are always some error messages in the syslog.

After a Rollback of the Kernel, I have also the problem. I had not sufficiently paid attention before.

I think that there is no regression, but all is not corrected.

[Luis Henriques (henrix)]

So, there are still some issues with the asix driver but it looks like these are different from the original bug report. Do we all agree with this? If so, I would propose to open a new bug report. Can anyone (the original reporter, maybe) confirm the original issue is solve with the kernel currently on -proposed? (See comment #40 for details on how to test -proposed.)

[Ming Lei (tom-leiming)]

On Wed, Aug 1, 2012 at 7:59 PM, David <email address hidden> wrote:
> I don't know IMO ! Without VLAN, nautilus transfert file via NFS with
> throughput of 11mb/s. However, there are always some error messages in
> the syslog.
>
> After a Rollback of the Kernel, I have also the problem. I had not
> sufficiently paid attention before.
>
> I think that there is no regression, but all is not corrected.

OK, could you test the upstream kernel in the below link to see
if your problem can be fixed?

        http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.5-quantal/

Thanks,

[Adam Gandelman (gandelman-a)]

I'm happy to re-test the new package in proposed, but I've been watching http://archive.ubuntu.com/ubuntu/dists/precise-proposed/main/binary-i386/Packages.gz since it was accepted and the latest kernel I find is 3.2.0.29.31, not 3.2.0-29.46. Am I looking in the wrong place? Do proposed kernel updates go somewhere else?

[Luis Henriques (henrix)]

Adam, you can check https://wiki.ubuntu.com/Testing/EnableProposed for instructions on how to use the -proposed pocket. After enabling the -proposed pocket, you don't need to manually download the kernel -- just run apt-get in a console, for example.

[Adam Gandelman (gandelman-a)]

I can confirm that the package in -proposed fixes my issue. I can saturate my connection for prolonged periods, and the only msg related to the asix driver that hits my kernel log is from the module's initialization.

[David (launux)]

Hi Ming,
The 3.5 kernel seems to block the USB ports. The network and keyboard not working.

[Ming Lei (tom-leiming)]

On Fri, Aug 3, 2012 at 12:27 PM, David <email address hidden> wrote:
> Hi Ming,
> The 3.5 kernel seems to block the USB ports. The network and keyboard not working.

It is weird, could you post the output of 'dmesg'?

Thanks,

[Luis Henriques (henrix)]

Adam, thanks a lot for testing and verifying that the new kernel solves the original issue.

David, Ming: do we all agree that we're now dealing with a different issue? If so, could you please open a new bug?

Thank you all!

[Ming Lei (tom-leiming)]

On Fri, Aug 3, 2012 at 4:29 PM, Luis Henriques
<email address hidden> wrote:
> David, Ming: do we all agree that we're now dealing with a different
> issue? If so, could you please open a new bug?

I Agree.

David, could you open a new bug about your specific vlan issue since
you can reproduce the bug? Also I suggest you to try the latest
3.6-rc1 kernel first if the usb devices of your box can't be brought up
on 3.5 kernel.

  http://kernel.ubuntu.com/~kernel-ppa/mainline/v3.6-rc1-quantal/

Thanks,

[David (launux)]

I think that enclose this bug is not a good idea.
During my tests without VLAN by downloading a 1gb file, I still had errors (not much).
With VLAN Bridge with my errors have significantly decreased. I think the bug is the same but it has not been fully corrected.
Enclosed is my kern.log file with Kernel 3.5 (no network and no keyboard after boot).
With kernel 3.6, I still have the same problem (network and keyboard are ok).

[Ming Lei (tom-leiming)]

On Sat, Aug 4, 2012 at 3:08 AM, David <email address hidden> wrote:
> With kernel 3.6, I still have the same problem (network and keyboard are ok).

If so, it is surely a new problem on your case, could you open a new bug entry?
Also there is the same problem on upstream kernel.

Thanks,

[Adam Conrad (adconrad)]

The verification of this Stable Release Update has completed successfully and the package has now been released to -updates. Subsequently, the Ubuntu Stable Release Updates Team is being unsubscribed and will not receive messages about this bug report. In the event that you encounter a regression using the package from -updates please report a new bug using ubuntu-bug and tag the bug report regression-update so we can easily find any regresssions.

[Petr E. Antonov (antpeter)]

Still have this bug.

uname -a

Linux eee-server 3.6.3-030603-generic #201210211349 SMP Sun Oct 21 17:57:45 UTC 2012 i686 i686 i686 GNU/Linux

tailf /var/log/kern.log
Oct 28 21:20:47 eee-server kernel: [278761.126491] asix 1-2:1.0: eth1: asix_rx_fixup() Bad Header Length
Oct 28 21:20:51 eee-server kernel: [278764.611074] asix 1-2:1.0: eth1: asix_rx_fixup() Bad RX Length 527
Oct 28 21:20:51 eee-server kernel: [278764.613722] asix 1-2:1.0: eth1: asix_rx_fixup() Bad Header Length
Oct 28 21:20:52 eee-server kernel: [278765.475047] asix 1-2:1.0: eth1: asix_rx_fixup() Bad RX Length 637
Oct 28 21:20:52 eee-server kernel: [278765.477655] asix 1-2:1.0: eth1: asix_rx_fixup() Bad Header Length
Oct 28 21:21:25 eee-server kernel: [278798.301808] asix 1-2:1.0: eth1: asix_rx_fixup() Bad RX Length 119
Oct 28 21:21:25 eee-server kernel: [278798.304409] asix 1-2:1.0: eth1: asix_rx_fixup() Bad Header Length
Oct 28 21:23:47 eee-server kernel: [278940.601259] asix 1-2:1.0: eth1: asix_rx_fixup() Bad RX Length 528
Oct 28 21:29:24 eee-server kernel: [279277.612150] asix 1-2:1.0: eth1: asix_rx_fixup() Bad RX Length 666
Oct 28 21:29:24 eee-server kernel: [279277.614717] asix 1-2:1.0: eth1: asix_rx_fixup() Bad Header Length

[Gino (ginoputrino)]

Yes, this bug still occurs for me.

Using the manufacturer's source from

http://www.asix.com.tw/FrootAttach/driver/AX88772B_772A_760_772_178_LINUX_Driver_v4.4.0_Source.zip

to build a kernel module solves the problem though.

What is ubuntu patching the working manufacturer's driver with that stops it from working?

[Petr E. Antonov (antpeter)]

Still have this bug.

$ uname -a
Linux eee-server 3.6.5-030605-generic #201211011211 SMP Thu Nov 1 16:20:33 UTC 2012 i686 i686 i686 GNU/Linux

I don't want to patch the kernel because this should be done by Kernel Team.
Please pay more attention to this problem

[Sean DS (se4n-1)]

I have this on Raring.

$ uname -a
Linux xxxxxxxxxxxxxxxx 3.8.0-27-generic #40-Ubuntu SMP Tue Jul 9 00:19:35 UTC 2013 i686 athlon i686 GNU/Linux

[Art (wildart)]

Same thing.

asix 2-1:1.0 eth0: asix_rx_fixup() Bad RX Length 813
asix 2-1:1.0 eth0: asix_rx_fixup() Bad Header Length
asix 2-1:1.0 eth0: asix_rx_fixup() Bad RX Length 952
asix 2-1:1.0 eth0: asix_rx_fixup() Bad Header Length

and so on.

$ uname -a
Linux AD-UX31E 3.8.0-27-generic #40-Ubuntu SMP Tue Jul 9 00:17:05 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

[Art (wildart)]

Last update did not help.

$ uname -a
Linux xxxxxxxx 3.8.0-29-generic #42-Ubuntu SMP Tue Aug 13 19:40:39 UTC 2013 x86_64 x86_64 x86_64 GNU/Linux

Please, look at this.

[Shehavtsov Nikolay (ayanami-dev)]

Art (wildart), try just install the latest driver from http://www.asix.com.tw/search.php , compile it and install, that is easy. This driver fixes this problem, no more errors.