[gutsy] orinoco_cs.ko missing

Ditto. This is still present in 2.6.22-8-generic. Would be nice to see it included as orinoco cards are quite common and generally always 'just work'.

lontra@kubuntu:~$ modprobe orinoco_cs
FATAL: Module orinoco_cs not found.

And it's still missing in the 2.6.22-9-generic image that hit the repos yesterday.

Just FYI, this bug is related to bug 57146 which concerns the incompatability between network-manager and recent hostap_* drivers.

I don't know much about kernel drivers, especially wireless one, but I do see that orinoco module is in 2.6.22-9, is this really different than orinoco_cs?

Yes it is

On 8/9/07, Paul Dufresne <email address hidden> wrote:
> I don't know much about kernel drivers, especially wireless one, but I
> do see that orinoco module is in 2.6.22-9, is this really different than
> orinoco_cs?
>
> ** Changed in: linux-source-2.6.22 (Ubuntu)
> Assignee: (unassigned) => Paul Dufresne
> Status: New => Incomplete
>
> --
> [gutsy] orinoco_cs.ko missing
> https://bugs.launchpad.net/bugs/125832
> You received this bug notification because you are a direct subscriber
> of the bug.
>

Normally, all bugs on kernels must have some information described at:
https://wiki.ubuntu.com/KernelTeamBugPolicies

But because of the nature of the bug report, I am confirming without it.

orinoco_* were disabled intentionally by Ben Collins back in late April under some incorrect assumptions. While it is true that the hostap driver is often a better choice for anyone with a Prism 2/2.5/3 -based card the orinoco drivers support two other chipsets (from Symbol and Lucent) which are not supported by hostap - they are not the same chipset. This is explained in greater detail on this page: http://www.hpl.hp.com/personal/Jean_Tourrilhes/Linux/Orinoco.html

The ideal solution would be to give priority (if that's possible!) to hostap but supply orinoco as well for cards not supported by hostap.

I should clarify that the hostap situation is improving but there are still some bugs on that front. orinoco is still a better choice at this point even for many Prism 2/2.5/3 users.

I have a couple cards here that won't even get an interface device associated with them using hostap_cs. This occurs even in feisty, so they may actually require orinoco_cs to work at all. (see bug 131926)

I have an "Orinoco Gold" card that is also affected by the lack of the orinoco_cs. It is produced by Lucent Technologies, and so is not supported by hostap. While hostap may be the better choice for some Prism chipsets, removing orinoco_cs clearly breaks things for a lot of others.

By the way, where can one find the rationale for the removal of orinoco_cs? If it is that hostap works better for Prism chipsets, wouldn't removing the PCI IDs for the prism chipsets from the orinoco_cs driver be enough to cause hostap to be loaded for them and still allow those of us with Lucent or Symbol chipsets to use orinoco_cs? If that would be considered, I can (attempt to) prepare a patch to do so and re-enable the orinoco driver.

It's still missing in 2.6.22-10. Removing this driver makes no sense to me -- it's fully functional and supports some hardware that is not supported by hostap. The changelog of 27 April (https://lists.ubuntu.com/archives/gutsy-changes/2007-May/001245.html) indicates it's being disabled "in favor of hostap_cs", but this makes little sense as hostap_cs does not support all hardware in orinoco_cs. The 2.6.22 kernels are useless for me at present.

I agree. I have an orinoco card by Lucent and it won't work with the hostap_cs. This is the first I have ever run a distribution (either stable or in development) where my orinoco card wasn't detected automatically and supported. Hopefully this can be fixed before Gutsy becomes final. Otherwise I know Debian isn't affected by this and I know where I'll have to go.

I have a compaq wl100. hostap does not work for me so orinoco_cs would be good the have. hostap does suggest that I update the firmware, but 2.6.20-16-generic with orinoco_cs works. 2.6.22-10-generic with hostap fails.

I was wondering if there was any sort of update to this... seems like a relatively easy fix, if nothing else than to revert to the Feisty behavior. Right now, this is a fairly major regression for anyone with one of these chipsets.

I am having the same problem with the new kernel as those above. I have an Orinoco Silver card from Lucent Technologies, and my card will not work properly with the hostap_cs driver. This card is obviously still used by a good number of Ubuntu users and I'm sure we would all appreciate having this driver included in gutsy again. Is there really any harm in having both drivers included in the kernel?

orinoco_cs is missing from 2.6.22-11 as well.

I compiled a 2.6.22-11 kernel (not an expert at this) and copied the orinoco_cs.ko module into /lib/modules/2.6.22-11-generic/drivers/net/wireless (from memory).

I did a depmod -a and modprobe orinoco_cs.

My wireless link is now up. Not sure if it will automatically come up on boot though.

To get the module to compile I just enabled CONFIG_PCMCIA_HERMES=m.

So, yes, it should be an easy fix.

This is my vague attempt at a patch to the ubuntu sources for 2.6.22 to re-enable the orinoco_cs module. This should fix behavior for the most users.

Hey, thanks for the patch. I'll give it a go tomorrow.

I'm a little irritated. Another new kernel has been released, and this bug is still unresolved. Hell, a developer hasn't even looked at this bug report, as far as I can tell. I've resorted to using Debian's stock 2.6.22 kernel and it works flawlessly. The whole "incompatible with network-manger" thing is just smoke. A card that that has worked out of the box for the past three releases, and probably more, is now causing hard locks, kernel panics, or is simply unrecognised. Why isn't someone working on this?

Thanks for the patch David. It would be great is someone could test it as well.

Setting priority High and milestoning as this seems to be low hanging fruit.

The attached patch not only re-enables orinoco_cs, but also tells orinoco_cs not to take control of devices also supported by hostap_cs. This has the effect of deprecating orinoco_cs in favor of hostap_cs while still allowing orinoco_cs to be used by the devices that need it. I don't have any hostap-supported hardware to check, but it seems like it should work. Please test and advise (or optionally include in linux-image-2.6.22-12 or whatever).

compaq wl-100 is (if I read it correctly) removed from the orinoco_cs support list.

This would be a bad thing as the compaq wl-100 only works with the orinoco_cs driver. the hostap* set does not work (at present).

Do your pci ids match those in the patch? The only ones I removed were those listed specifically in hostap_cs.c, which is not to say I could not have made a mistake (or they may have pci ids for which they do not have proper support).

This is the compaq wl100. it matches my pcmcia card.
- PCMCIA_DEVICE_MANF_CARD(0x0138, 0x0002), /* Compaq WL100 11 Mbps Wireless Adapter */

I think the problem is that hostap is supposed to work for this and other cards, but it does not seem to for many of us blessed with this or similar cards.

I also have to blacklist hostap_cs and hostap so it does not take control.

I should add that orinoco_cs in 2.6.20-16 (the last kernel it seemed to be in from ubuntu) does not work as well as it once did, but this could just be me. I can make it work even if I have to ifdown and ifup it occasionally.

I have a Toshiba A20 S207 and my wireless don't work at gutsy.
# cardctl ident
Socket 0:
  product info: "TOSHIBA", "Wireless LAN Card", "Version 01.01", ""
  manfid: 0x0156, 0x0002
  function: 6 (network)

With ubuntu 7.04 my wireless work perfect, with orinoco_cs, i remember.
The module hostap_cs don't work to.
I think my wirless card is a prism2.

Thanks,

I don't see the milestone marker on here, and it still hasn't made it into 2.6.22-12. Is there any plan to fix this horribly obvious regression?

My dell latitude c800 which has out of the box network support for its Dell/Intel truemobile prism2 based wireless card in feisty, but is not recognized at all in gutsy. Additionally, my prism2 based wireless card in my desktop pc doesn't work in gutsy either. This should be a very critical priority! Many Many linux compatible wireless cards are prism2!

Prism 2 cards should be supported by hostap_cs, and so should still work fine.

They should work with hostap, but frequently don't.

This driver fiasco caused me to go out and upgrade my infrastructure to 802.11g with atheros cards, since none of my 5 cards, including the one built into my T30 would work in gutsy.

hostap_cs failing to work with network-manager is bug 57146, and we have code to fix that. If anyone has a card that fulfils all the following criteria:

1) Is detected by hostap_cs, and has an interface created
2) Doesn't work with manual configuration (ifconfig, iwconfig and dhclient - we already know that network-manager is broken)
3) Works fine with orinoco_cs and manual configuration

then please attach the output of

pccardctl info

and explain the problems you see.

Matthew... Is there any word on whether the kernel team will be re-enabling orinoco_cs? Is there anyone I could discuss the issue with? At a minimum, I'd like to understand the rationale for its removal (and lack of replacement) when it clearly breaks some cards that are supported by NO other drivers. I don't mind being patient, but this is a fairly simple bug for something that's been outstanding for more than 2 months.

Yes, orinoco_cs will be reenabled.

Gutsy kernel commit 0fd77efbd8364d6a2625bbc4b84c1f3c26387655. The orinoco_cs is enabled to automatically install only for those PCMCIA IDs that are not also defined in hostap_cs. For those IDs that exist in both drivers hostap_cs takes precedence. To force orinoco_cs in that case you must blacklist hostap_cs and insmod orinoco_cs.

Finally got to test Gutsy 32-bit Beta (20070925.2) on Sony Vaio SRX51P since the unionfs bug is (almost) sorted - the PC started albeit a few problems still reported for unionfs.

However, the Lucent WaveLan/IEEE version 01.01 mini-pci card that is interfaced via the pccard interface and works fine in Feisty is not detected. It looks like the lack of orinoco_cs affects this PC too.

Now I can boot the PC I can work on getting the wlags49 driver (supports WPA) to work with kernel 2.6.22 which will help some users.

Details of the device:

udi = '/org/freedesktop/Hal/devices/pcmcia__1__1'
  info.bus = 'pcmcia' (string)
  info.parent = '/org/freedesktop/Hal/devices/pci_104c_ac50' (string)
  info.product = 'WaveLAN/IEEE' (string)
  info.subsystem = 'pcmcia' (string)
  info.udi = '/org/freedesktop/Hal/devices/pcmcia__1__1' (string)
  info.vendor = 'Lucent Technologies' (string)
  linux.hotplug_type = 2 (0x2) (int)
  linux.subsystem = 'pcmcia' (string)
  linux.sysfs_path = '/sys/devices/pci0000:00/0000:00:1e.0/0000:01:05.0/1.0' (string)
  pcmcia.card_id = 2 (0x2) (int)
  pcmcia.func_id = 6 (0x6) (int)
  pcmcia.manf_id = 342 (0x156) (int)
  pcmcia.prod_id1 = 'Lucent Technologies' (string)
  pcmcia.prod_id2 = 'WaveLAN/IEEE' (string)
  pcmcia.prod_id3 = 'Version 01.01' (string)
  pcmcia.product = 'Lucent Technologies' (string)
  pcmcia.socket_number = 1 (0x1) (int)
  pcmcia.vendor = 'Lucent Technologies' (string)

$ pccardctl ls
Socket 0 Bridge: [yenta_cardbus] (bus ID: 0000:01:02.0)
Socket 1 Bridge: [yenta_cardbus] (bus ID: 0000:01:05.0)
Socket 1 Device 0: [-- no driver --] (bus ID: 1.0)

Just tested three cards:

1) Belkin F5D6020 (unknown revision). This is an older orinoco device (primary firmware 0.3.0, if I'm not mistaken). The hostap driver doesn't load. Once orinoco_cs is reenabled (post beta), this card should start to work again, because hostap disavows knowledge of this card.

2) Netgear MA401 (unknown revision). Same exact situation as #1

3) Netgear MA401 (Rev. D). This is a much newer card (supporting the most recent primary and secondary firmwares from http://linux.junsun.net/intersil-prism/ ). It is supported by the hostap driver. It _does_ work with the command line tools, but fails to work with network manager. Therefore, it is affected by bug 57146.

I bricked my internal mini-PCI wireless NIC device a while back trying to update the firmware, so I can't test it.

As far as I can tell, once bug 57146 is fixed and the new kernel is released including orinoco_*, all my cards should be working.

Thanks!
mike

There is no lum dependency.

my prism2 card:

1) Is detected by hostap_cs, and has an interface created: eth1 as
well as wlan0_ren (something like that)
2) Doesn't work with manual configuration (ifconfig, iwconfig and dhclient)
3) Works fine with orinoco_cs and manual configuration (I'm using it
now with a hacked 2.6.20 kernel)

PRODID_1="Compaq"
PRODID_2="WL100_11Mbps_Wireless_PC_Card"
PRODID_3="Version 01.00"
PRODID_4=""
MANFID=0138,0002
FUNCID=6
PRODID_1=""
PRODID_2=""
PRODID_3=""
PRODID_4=""
MANFID=0000,0000
FUNCID=255

The interface eth1 seems configured and associated to the AP but it
does not get and IP address via DHCP.

In response to Matthew's request for pccardctl data:
My toshiba Satellite Pro 4600 laptop (affected by the bug) generates the following:
PRODID_1="TOSHIBA"
PRODID_2="Wireless LAN Card"
PRODID_3="Version 01.01"
PRODID_4=""
MANFID=0156,0002
FUNCID=6
PRODID_1=""
PRODID_2=""
PRODID_3=""
PRODID_4=""
MANFID=0000,0000
FUNCID=255
PRODID_1=""
PRODID_2=""
PRODID_3=""
PRODID_4=""
MANFID=0000,0000
FUNCID=255

I hope there is a patch available very soon, as my primary motivation for upgrading it was the kernel enhancements to improve battery life. (And the new OpenOffice version)

My Toshiba Satellite Pro 6100 has the same info as earthforce_1's

Bill and earthforce_1: Could you please open a bug describing the issues you have with hostap_cs?

The problem I has is what I described in the other bug: wireless no longer works after upgrade to Gusty Beta.

Reading this ticket sounded like the same symptoms I saw. And reverting to an earlier kernel (2.6.20) solved the problem. What can I do to confirm that it is a hostap_cs issue?

Also, upgrading to today's 2.6.22.12.4 kernel does not help.

here are the 4 files requested

Bill,

What kind of wireless card do you have? If it's not being seen at all, it might be affected by orinoco_cs. On the 2.6.20 kernel, can you check if hostap_cs or orinoco_cs is loaded? (Or similar) Is your card internal or external?

Matthew:
I already opened a bug, it was flagged as a duplicate of this one. Basically the symptoms are the wireless interface disappears as if it never existed. The network configuration icon on the toolbar reports dial up and wired ethernet interfaces only, the wireless interface simply vanished without a trace. Any more tests you want me do to with the laptop, or if you want to test a workaround let me know and I will post the results here. This is the sole remaining gutsy showstopper I am facing.

Socket 0:
  product info: "TOSHIBA", "Wireless LAN Card", "Version 01.01", ""
  manfid: 0x0156, 0x0002
  function: 6 (network)

and the card is internal

David,

Looks like it loads orinoco with the 2.6.20 kernel.

bill@bill-laptop:~$ lsmod |grep ori
orinoco_cs 16900 1
orinoco 43028 1 orinoco_cs
hermes 8448 2 orinoco_cs,orinoco
pcmcia 39212 1 orinoco_cs
pcmcia_core 40852 4 orinoco_cs,pcmcia,yenta_socket,rsrc_nonstatic

bill@bill-laptop:~$ lsmod |grep host

bill@bill-laptop:~$ uname -a
Linux bill-laptop 2.6.20-16-generic #2 SMP Sun Sep 23 19:50:39 UTC 2007 i686 GNU/Linux

2.6.20 - orinoco_cs loaded - everything works fine (toshiba portege 2000)
2.6.22 - orinoco_cs not loaded - I can't see my wifi card. The wifi card is the same as Bill Chriss has.

linux-source-2.6.22 (2.6.22-13.40) gutsy; urgency=low

  [Amit Kucheria]

  * Enable CONFIG_VM86 for LPIA
    - LP: #146311
  * Update configuration files
  * Disable MSI by default
  * Add mmconf documentation
  * Update configuration files

  [Bartlomiej Zolnierkiewicz]

  * ide-disk: workaround for buggy HPA support on ST340823A (take 3)
    - LP: #26119

  [Ben Collins]

  * ubuntu/cell: Fixup ps3 related modules for d-i, enable RTAS console
  * ubuntu/cell: Enable CELLEB and related modules (pata_scc)
  * ubuntu/cell: Move ps3rom to storage-core. Also use spidernet, not
    spider_net.
  * ubuntu/cell: Set PS3_MANAGER=y
  * ubuntu: Set NR_CPUS=256 for sparc64-smp

  [Chuck Short]

  * [USB] USB] Support for MediaTek MT6227 in cdc-acm.
    - LP: #134123
  * [XEN] Fix xen vif create with more than 14 guests.
    - LP: #14486

  [Jorge Juan Chico]

  * ide: ST320413A has the same problem as ST340823A
    - LP: #26119

  [Kyle McMartin]

  * fix -rt build
  * fix ia32entry-xen.S for CVE-2007-4573
  * fix build when CONFIG_PCI_MSI is not set

  [Matthew Garrett]

  * hostap: send events on data interface as well as master interface
    - LP: #57146
  * A malformed _GTF object should not prevent ATA device recovery
    - LP: #139079
  * hostap: send events on data interface as well as master interface
    - LP: #57146
  * A malformed _GTF object should not prevent ATA device recovery
    - LP: #139079
  * Don't lose appletouch button release events
  * Fix build with appletouch change
  * Disable Thinkpad backlight support on machines with ACPI video
    - LP: #148055
  * Don't attempt to register a callback if there is no CMOS object
    - LP: #145857
  * Update ACPI bay hotswap code to support locking
    - LP: #148219
  * Update ACPI bay hotswap code to support locking
    - LP: #148219
  * Don't attempt to register a callback if there is no CMOS object
    - LP: #145857
  * Disable Thinkpad backlight support on machines with ACPI video
    - LP: #148055

  [Steffen Klassert]

  * 3c59x: fix duplex configuration
    - LP: #94186

  [Thomas Gleixner]

  * clockevents: remove the suspend/resume workaround^Wthinko

  [Tim Gardner]

  * orinoco_cs.ko missing
    - LP: #125832
  * Marvell Technology ethernet card not recognized and not operational
    - LP: #135316
  * Marvell Technology ethernet card not recognized and not operational
    - LP: #135316
  * acpi_scan_rsdp() breaks some PCs by not honouring ACPI specification
    - LP: #144336
  * VIA southbridge Intel id missing
    - LP: #128289
  * Add T-Sinus 111card to hostap_cs driver to be able to upload firmware
    - LP: #132466
  * RTL8111 PCI Express Gigabit driver r8169 big files produce slow file
    transfer
    - LP: #114171
  * Guest OS does not recognize a lun with non zero target id on Vmware ESX
    Server
    - LP: #140761
  * Modualrize vesafb
    - LP: #139505
  * Nikon cameras need support in unusual_devs.h
    - LP: #134477
  * agp for i830m broken in gutsy
    - LP: #139767
  * hdaps: Added support for Thinkpad T61
    - LP: #147383
  * xen: Update config for i386
    - LP: #139047
  * xen: resync for amd64
    - LP: #139047
  * ide-disk: workaround for buggy HPA support ...

I booted the new 2.6.22-13.40 kernel. It still hard locks on boot when my Netgear MA401 is plugged in. Come on guys, Debian's 2.6.22 kernel boots fine, which means one of the Ubuntu patches to Debian's kernel is causing hard locks on some machines.

Tell me what logs you need (if you can get any from a hard lock at boot), and I will post them.

I can confirm this behavior on a Dell Inspiron with an Orinoco PCMCIA card. Debian kernel is fine

Ben,

If you're having kernel hard locks, that's a completely different bug from this. I recommend opening a new bug, and providing the standard kernel information (lspci -vvnn, dmesg, etc.) Boot in recovery mode and see if you can see anything when it locks.

trshemanske,

Have you tried 2.6.22-13? orinoco_cs is in that kernel, so this bug is essentially fixed.

Kernel freezes 1/3 through boot on my Toshiba Satellite 4600 with unlucky kernel -13 Ok with the previous kernel, except for the missing wireless driver. Booting the -13 rescue from the grub menu freezes with a register dump - would this be in a log file anywhere so I could post it?

Also freezes on my generic old P3-800/512 Ram system with a linksys USB wireless card. Two other PCs (one P3, one P4 and Amd-64) are ok with the new kernel. The need for a wireless driver seems to be the common factor between the laptop and the other PC that borked in exactly the same manner. Confirmed that both hung PCs boot ok when I step back to the last working kernel.

Looks like anybody else hitting this bug can recover by hitting <ESC> quickly and booting the previous working kernel from the grub menu. The 2.6.22-13 kernel barfs on on affected systems even in rescue mode, you have to step back to the last kernel.

David:
It may be a new bug, but it certainly seems to be related to this one, as the only systems affected by the kernel lockup are the ones that needed the wireless drivers. I have installed the new kernel on a variety of machines and I don't think it is just a coincidence.

After I've upgraded my notebook to latest guisty, it says "Process udevd (pid: 3648, ti=da144000 task=da464530 task.ti=da144000)" "Stack bla bla". It has toshiba + internal wifi PCMCIA card which uses orinoco_cs driver

Yes, from what I remember ( the Toshiba laptop is not with me at the moment) that was the very same message I got when I booted the -13 kernel in rescue mode, after it hung 1/3 of the way through the normal reboot.

This bug is fixed. It seems that orinoco_cs is causing hangs on boot on some hardware - please open another bug about that. Ideally, press escape at startup, press e, select the line starting "kernel", press e again, delete the words "quiet" and "splash", press enter and then press b. Include the last of the kernel output when filing the new bug. Please don't follow up here - it's unrelated to this issue.

I've created bug #149997 for the hang on boot (with a digital camera screenshot of the kernel output).

2.6.22-13-generic does not work for me with a COMPAQ WL100 PCMCIA.

hostap* loads but does not work.
orinoco_cs will not load since (I guess) it does not support my card's ID 0138 0002 (if that is what all the stuff in modinfo orinoco_cs shows).

To get my older prism2 card working under Gutsy I had to use the older prism2 drivers. I did get finally get it to work.
http://devinvenable.blogspot.com/2007/11/wireless-problems-upgrading-ubuntu-to.html

This problem seems to have been solved in Gutsy. My Dell Truemobile 1150 (orinoco_cs) card now is detected and works fine. I did not have to do anything to fix it except to install the latest release.

However, this problem still exists in Ubuntu Studio (2.6.22-14-rt) which was apparently compiled using a version of original Gutsy kernel.

The problem also occurs if I upgrade from Gutsy to Studio by installing the Studio packages. The kernel upgrade renders my wireless card unusable.

Should this be opened as a new bug for Ubuntu Studio?

Rejecting task on Ubuntu, as a more specific on kernel exist (and is even Fix Released).

Confirming the ubuntustudio task of this bug, as I believe:
https://bugs.launchpad.net/~jason-mancine plus the info in this bug
is enough for Ubuntustudio developer to investigate this bug.

Marking as fix released to match ubuntu