ssb error while loading wifi module b43

[Nicolò Chieffo]

Binary package hint: linux-source-2.6.24

Hello I have a broadcom 43 wireless card in my ho notebook, and I'm using Hardy with kernel 2.6.24-4.
I cannot use wifi at all, because when I try to load the module b43 I get this error in /var/log/messages

Jan 20 17:12:03 babbo-laptop kernel: [ 146.928719] ACPI: PCI Interrupt 0000:02:00.0[A] -> GSI 16 (level, low) -> IRQ 17
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947353] WARNING: at /build/buildd/linux-2.6.24/drivers/ssb/main.c:883 ssb_tmslow_reject_bitmask()
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947366] Pid: 6816, comm: modprobe Not tainted 2.6.24-4-generic #1
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947391] [<f893687b>] ssb_tmslow_reject_bitmask+0x4b/0x60 [ssb]
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947414] [<f8937161>] ssb_device_is_enabled+0x11/0x40 [ssb]
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947430] [<f8939299>] ssb_pcicore_init+0x19/0x60 [ssb]
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947446] [<f89366a3>] ssb_attach_queued_buses+0xd3/0x260 [ssb]
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947463] [<f89386ef>] ssb_pci_xtal+0x17f/0x230 [ssb]
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947487] [<f8936c36>] ssb_bus_register+0x156/0x1d0 [ssb]
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947503] [<f8937f50>] ssb_pci_get_invariants+0x0/0x2c0 [ssb]
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947544] [<f8936d2a>] ssb_bus_pcibus_register+0x2a/0x60 [ssb]
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947556] [r8169:pci_set_master+0x54/0x60] pci_set_master+0x54/0x60
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947572] [<f893890c>] ssb_pcihost_probe+0x6c/0xb0 [ssb]
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947589] [pci_device_probe+0x56/0x80] pci_device_probe+0x56/0x80
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947601] [driver_probe_device+0x88/0x190] driver_probe_device+0x88/0x190
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947609] [scsi_mod:kobject_uevent_env+0xf0/0x2610] kobject_uevent_env+0xf0/0x3d0
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947624] [__driver_attach+0x9e/0xa0] __driver_attach+0x9e/0xa0
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947636] [scsi_mod:bus_for_each_dev+0x3b/0xe0] bus_for_each_dev+0x3b/0x60
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947651] [usbcore:driver_attach+0x16/0x2b0] driver_attach+0x16/0x20
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947657] [__driver_attach+0x0/0xa0] __driver_attach+0x0/0xa0
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947663] [bus_add_driver+0x8a/0x1e0] bus_add_driver+0x8a/0x1e0
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947679] [parport_pc:__pci_register_driver+0x56/0x630] __pci_register_driver+0x56/0x90
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947690] [<f887404a>] ssb_modinit+0x4a/0x80 [ssb]
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947702] [usbcore:blocking_notifier_call_chain+0x17/0x20] blocking_notifier_call_chain+0x17/0x20
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947714] [sys_init_module+0x126/0x19c0] sys_init_module+0x126/0x19c0
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947758] [<c0278d30>] bus_register+0x0/0x230
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947779] [sysenter_past_esp+0x6b/0xa9] sysenter_past_esp+0x6b/0xa9
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947801] =======================
Jan 20 17:12:03 babbo-laptop kernel: [ 146.947897] ssb: Sonics Silicon Backplane found on PCI device 0000:02:00.0

tell me which other information you need to fix this bug. Thanks!

[Leann Ogasawara (leannogasawara)]

Thank you for taking the time to report this bug and helping to make Ubuntu better. Per the kernel team's bug policy, can you please attach the following information. Please be sure to attach each file as a separate attachment.

* uname -a > uname-a.log
* cat /proc/version_signature > version.log
* dmesg > dmesg.log
* sudo lspci -vvnn > lspci-vvnn.log

Also, can you be sure to capture the entire dmesg output only after having attempted to load the module so that it also includes the warning you are seeing.

For more information regarding the kernel team bug policy, please refer to https://wiki.ubuntu.com/KernelTeamBugPolicies . Thanks again and we appreciate your help and feedback.

[Nicolò Chieffo (yelo3)]

I'm sorry I haven't answered for some days, but I have this laptop
only on weekends.
I will explain better the problem:
the module "ssb" is automatically loaded at boot with the module
"ohci_hcd". I blacklisted those modules because I needed to work with
wifi and ndiswrapper is the only way to have it running, and
ndiswrapper didn't work with those loaded.

I wanted to try the new b43 driver, and I realized that when I
modprobe it, it also loads ssb, so I realized that there was the error
message I posted on the description.

[Nicolò Chieffo (yelo3)]

After loading the module b43 this is the list of added modules:
+b43
+cfg80211
+input_polldev
+mac80211
+rfkill
+ssb

the dmesg output exactly AFTER the module load:

[ 91.085949] ACPI: PCI Interrupt 0000:02:00.0[A] -> GSI 16 (level,
low) -> IRQ 17
[ 91.085969] PCI: Setting latency timer of device 0000:02:00.0 to 64
[ 91.100919] WARNING: at
/build/buildd/linux-2.6.24/drivers/ssb/main.c:883
ssb_tmslow_reject_bitmask()
[ 91.100931] Pid: 6017, comm: modprobe Not tainted 2.6.24-5-generic #1
[ 91.100956] [<f8de187b>] ssb_tmslow_reject_bitmask+0x4b/0x60 [ssb]
[ 91.100978] [<f8de2161>] ssb_device_is_enabled+0x11/0x40 [ssb]
[ 91.100992] [<f8de42d9>] ssb_pcicore_init+0x19/0x60 [ssb]
[ 91.101006] [<f8de16a3>] ssb_attach_queued_buses+0xd3/0x260 [ssb]
[ 91.101022] [<f8de372f>] ssb_pci_xtal+0x17f/0x230 [ssb]
[ 91.101042] [<f8de1c36>] ssb_bus_register+0x156/0x1d0 [ssb]
[ 91.101056] [<f8de2f90>] ssb_pci_get_invariants+0x0/0x2c0 [ssb]
[ 91.101091] [<f8de1d2a>] ssb_bus_pcibus_register+0x2a/0x60 [ssb]
[ 91.101103] [<c0221324>] pci_set_master+0x54/0x60
[ 91.101118] [<f8de394c>] ssb_pcihost_probe+0x6c/0xb0 [ssb]
[ 91.101133] [<c0223256>] pci_device_probe+0x56/0x80
[ 91.101143] [<c027e908>] driver_probe_device+0x88/0x190
[ 91.101152] [<c0212040>] kobject_uevent_env+0xf0/0x3d0
[ 91.101166] [<c027eb7e>] __driver_attach+0x9e/0xa0
[ 91.101176] [<c027dd3b>] bus_for_each_dev+0x3b/0x60
[ 91.101188] [<c027e786>] driver_attach+0x16/0x20
[ 91.101194] [<c027eae0>] __driver_attach+0x0/0xa0
[ 91.101201] [<c027e0ba>] bus_add_driver+0x8a/0x1e0
[ 91.101213] [<c0223406>] __pci_register_driver+0x56/0x90
[ 91.101222] [<f8bc304a>] ssb_modinit+0x4a/0x80 [ssb]
[ 91.101233] [<c0146477>] blocking_notifier_call_chain+0x17/0x20
[ 91.101245] [<c01515e6>] sys_init_module+0x126/0x19c0
[ 91.101282] [<c027e2b0>] bus_register+0x0/0x230
[ 91.101299] [<c01053c2>] sysenter_past_esp+0x6b/0xa9
[ 91.101318] =======================
[ 91.101412] ssb: Sonics Silicon Backplane found on PCI device 0000:02:00.0

[Nicolò Chieffo (yelo3)]

problem still present in 2.6.24-7-generic but the dmesg log is a bit different:

[ 17.720342] PCI: Setting latency timer of device 0000:02:00.0 to 64
[ 17.779030] WARNING: at
/build/buildd/linux-2.6.24/drivers/ssb/main.c:883
ssb_tmslow_reject_bitmask()
[ 17.779036] Pid: 1528, comm: modprobe Not tainted 2.6.24-7-generic #1
[ 17.779041] [<f886e87b>] ssb_tmslow_reject_bitmask+0x4b/0x60 [ssb]
[ 17.779052] [<f886f161>] ssb_device_is_enabled+0x11/0x40 [ssb]
[ 17.779059] [<f88712d9>] ssb_pcicore_init+0x19/0x60 [ssb]
[ 17.779067] [<f886e6a3>] ssb_attach_queued_buses+0xd3/0x260 [ssb]
[ 17.779074] [<f887072f>] ssb_pci_xtal+0x17f/0x230 [ssb]
[ 17.779082] [<f886ec36>] ssb_bus_register+0x156/0x1d0 [ssb]
[ 17.779088] [<f886ff90>] ssb_pci_get_invariants+0x0/0x2c0 [ssb]
[ 17.779099] [<f886ed2a>] ssb_bus_pcibus_register+0x2a/0x60 [ssb]
[ 17.779105] [<c0221374>] pci_set_master+0x54/0x60
[ 17.779112] [<f887094c>] ssb_pcihost_probe+0x6c/0xb0 [ssb]
[ 17.779119] [<c02232a6>] pci_device_probe+0x56/0x80
[ 17.779123] [<c027ea18>] driver_probe_device+0x88/0x190
[ 17.779129] [<c027ec8e>] __driver_attach+0x9e/0xa0
[ 17.779132] [<c027de4b>] bus_for_each_dev+0x3b/0x60
[ 17.779137] [<c027e896>] driver_attach+0x16/0x20
[ 17.779140] [<c027ebf0>] __driver_attach+0x0/0xa0
[ 17.779144] [<c027e1ca>] bus_add_driver+0x8a/0x1e0
[ 17.779148] [<c0223456>] __pci_register_driver+0x56/0x90
[ 17.779152] [<f885804a>] ssb_modinit+0x4a/0x80 [ssb]
[ 17.779158] [<c0146477>] blocking_notifier_call_chain+0x17/0x20
[ 17.779163] [<c01515e6>] sys_init_module+0x126/0x19c0
[ 17.779170] [<f887f000>] ahci_init+0x0/0x14 [ahci]
[ 17.779177] [<c027e3c0>] bus_register+0x0/0x230
[ 17.779182] [<c0105442>] syscall_call+0x7/0xb
[ 17.779186] [<c0310000>] vcc_ioctl+0x140/0x2d0
[ 17.779191] =======================
[ 17.779248] ssb: Sonics Silicon Backplane found on PCI device 0000:02:00.0

[Leann Ogasawara (leannogasawara)]

Care to attach your full dmesg output? Thanks.

[Nicolò Chieffo (yelo3)]

this is the full dmesg

[Colin Ian King (colin-king)]

Your card is a BCM4311 revision 02. I've looked at the driver code and the error you are seeing occurs with revision 02 versions of the BCM4311 with this particular driver. Apparently your revision of the device will not be supported in mainline kernels until 2.6.25-rc1.

However, there is a 2.6.24 patch available which I shall explore (c.f. http://linuxwireless.org/en/users/Drivers/b43#FAQ-Frequentlyaskedquestion )

Perhaps using NDIS wrappers is a way forward for the moment, I refer you to https://help.ubuntu.com/7.10/internet/C/ndiswrapper.html

[Nicolò Chieffo (yelo3)]

Yes, I'm using ndiswrapper now. Is it possible to have this patch
included in the main kernel? Lots of people who buy new laptops have
this card installed, and Hardy will be a long term support release, so
I think that this patch should be evaluated...
what do you think?

[Colin Ian King (colin-king)]

Backporting the fixed driver code to 2.6.24 is too intrusive, so we nominating this to be including in Intrepid (Hardy+1).

[Nicolò Chieffo (yelo3)]

I think this decision will affect s very big amount of users (broadcom
cards are inserted in most low end, but new laptops priced <900€),
which will not be able to use wireless in an LTSP release of ubuntu.
Ndiswrapper is not a solution since it works well only in non
protected access points. Is there another solution we can try to have
a working wireless device?

[Colin Ian King (colin-king)]

Larry Finger patch applied to support revision 2 after revisiting your comments.

[Nicolò Chieffo (yelo3)]

Thank you very much Colin! I really appreciate your decision change!

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux - 2.6.24-11.17

---------------
linux (2.6.24-11.17) hardy; urgency=low

  [Alan Cox]

  * Pull in fixes for pata_it821x.
    - LP: #106931

  [Alessio Igor Bogani]

  * rt: Synchronized with upstream (2.6.24.3-rt3)
  * rt: Updated configuration files

  [Amit Kucheria]

  * Add AGP support for Radeon Mobility 9000 chipset
    - LP: #178634
  * Bluetooth: SCO flow control to enable bluetooth headsets

  [Ben Collins]

  * binary: Include vesafs in initrd subdir, should fix vga= usage

  [Colin Ian King]

  * AMD SB700 south bridge support patches
    - LP: #195354
  * BCM4311 Revision 2 fix
    - LP: #184600

  [Mauro Carvalho Chehab]

  * V4L/DVB (6753): Fix vivi to support non-zero minor node

  [Tim Gardner]

  * Merged 2.6.24.3
  * Add atl1 to d-i bits.
    - LP: #159561
  * SAUCE: Add xpad support for RedOctane Guitar Hero
    - LP: #196745

  [Upstream Kernel Changes]

  * DVB: cx23885: add missing subsystem ID for Hauppauge HVR1800 Retail
  * slab: fix bootstrap on memoryless node
  * vm audit: add VM_DONTEXPAND to mmap for drivers that need it
    (CVE-2008-0007)
  * USB: keyspan: Fix oops
  * usb gadget: fix fsl_usb2_udc potential OOPS
  * USB: CP2101 New Device IDs
  * USB: add support for 4348:5523 WinChipHead USB->RS 232 adapter
  * USB: Sierra - Add support for Aircard 881U
  * USB: Adding YC Cable USB Serial device to pl2303
  * USB: sierra driver - add devices
  * USB: ftdi_sio - enabling multiple ELV devices, adding EM1010PC
  * USB: ftdi-sio: Patch to add vendor/device id for ATK_16IC CCD
  * USB: sierra: add support for Onda H600/Zte MF330 datacard to USB Driver
    for Sierra Wireless
  * USB: remove duplicate entry in Option driver and Pl2303 driver for
    Huawei modem
  * USB: pl2303: add support for RATOC REX-USB60F
  * USB: ftdi driver - add support for optical probe device
  * USB: use GFP_NOIO in reset path
  * USB: Variant of the Dell Wireless 5520 driver
  * USB: storage: Add unusual_dev for HP r707
  * USB: fix usbtest halt check on big endian systems
  * USB: handle idVendor of 0x0000
  * USB: Fix usb_serial_driver structure for Kobil cardreader driver.
  * forcedeth: mac address mcp77/79
  * lockdep: annotate epoll
  * sys_remap_file_pages: fix ->vm_file accounting
  * PCI: Fix fakephp deadlock
  * ACPI: update ACPI blacklist
  * x86: restore correct module name for apm
  * sky2: restore multicast addresses after recovery
  * sky2: fix for WOL on some devices
  * b43: Fix suspend/resume
  * b43: Drop packets we are not able to encrypt
  * b43: Fix dma-slot resource leakage
  * b43legacy: fix PIO crash
  * b43legacy: fix suspend/resume
  * b43legacy: drop packets we are not able to encrypt
  * b43legacy: fix DMA slot resource leakage
  * selinux: fix labeling of /proc/net inodes
  * b43: Reject new firmware early
  * sched: let +nice tasks have smaller impact
  * sched: fix high wake up latencies with FAIR_USER_SCHED
  * fix writev regression: pan hanging unkillable and un-straceable
  * Driver core: Revert "Fix Firmware class name collision"
  * drm: the drm really should call pci_set_master..
  * splice: missing user pointer access verification (CVE-2008-0009/10)
  * Linu...

[der_vegi (m-may)]

This patch breaks BCM4312 cards, see bug 197959 .

[Nicolò Chieffo (yelo3)]

re-confirmig due to the patch removal. can we do something? I know that linux 2.6.25 works great with my bcm43xx rev02. Instead with the fix introduced by this bug it did not work, maybe the fix was incomplete! Is it possible?

[Florian Zeitz (florian-zeitz)]

Could anybody actually shed some light on this for me?
Somehow it all is very confusing.
The patch seems to be meant to make bcm4311 cards work, but at the same time breaks bcm4312 cards (that is what I was able to get from the bugs. some places also say different things I guess it is very easy to accidentally type 2 instead of 1).
But for some reason my experience is the exact opposite. I have a bcm4312 (rev 02) chip in my laptop (or at least lspci says so) that stopped working when the patch was removed and worked while it was there.

[Nicolò Chieffo (yelo3)]

Wait. The patch also breaks bcm4311 rev02! ssb loads, the card is
detected but it does not work (it works rarely). So the solution is
hidden in the 2.6.25 kernel, in whitch my bcm4311 rev02 works without
problems

[Colin Ian King (colin-king)]

Hi Nicolo,

I had to revert the patch for rev 02 chipsets because it was a significant patch that broke the broadcom driver for BCM4312 chipsets to name just a few. Apologies for this, note that I understand that 2.6.25 fixes your problem but to isolate the fix and backport it to 2.6.24 so close to the Hardy release without breaking the driver for other BCM chipsets is not possible. I am therefore marking this as "Won't Fix" for Hardy, and keeping it open for Hardy + 1.

Colin

[Nicolò Chieffo (yelo3)]

Nevermind for me, I can wait. But I think that this patch has to be
backported to a long term support distribution, as I've written in my
previous posts.

[gururise (gururise)]

I agree, a fix needs to be backported. This problem affects the majority of HP/Compaq Laptop users who have an AMD Processor.

[Martin Wimmer (martin-wimmer)]

Actually it currently doesn't work for me with bcm4312 (rev02), while it was recognized in previous kernel versions.

b43 isn't loaded at startup. A modprobe doesn't return errors, but the card just isn't recognized.
Is there a chance this will get fixed in Hardy? Or will I have to use ndiswrapper until Hardy +1?

 dmesg | grep 'ssb'
[ 25.544026] WARNING: at /build/buildd/linux-2.6.24/drivers/ssb/main.c:883 ssb_tmslow_reject_bitmask()
[ 25.544048] [<ffffffff881128ca>] :ssb:ssb_tmslow_reject_bitmask+0x4a/0x60
[ 25.544055] [<ffffffff88113316>] :ssb:ssb_device_is_enabled+0x16/0x50
[ 25.544060] [<ffffffff88115731>] :ssb:ssb_pcicore_init+0x21/0x70
[ 25.544066] [<ffffffff88112656>] :ssb:ssb_attach_queued_buses+0x106/0x2d0
[ 25.544072] [<ffffffff88114250>] :ssb:ssb_pci_get_invariants+0x0/0x2d0
[ 25.544078] [<ffffffff88112d0e>] :ssb:ssb_bus_register+0x17e/0x200
[ 25.544086] [<ffffffff88112e22>] :ssb:ssb_bus_pcibus_register+0x32/0x60
[ 25.544092] [<ffffffff88114cdd>] :ssb:ssb_pcihost_probe+0x7d/0xc0
[ 25.544127] [<ffffffff8811d052>] :ssb:ssb_modinit+0x52/0x80
[ 25.544204] ssb: Sonics Silicon Backplane found on PCI device 0000:30:00.0

lsmod | grep 'b43'
b43 126760 0
rfkill 10128 1 b43
mac80211 192532 1 b43
led_class 7176 1 b43
input_polldev 6928 1 b43
ssb 37252 1 b43

[Colin Ian King (colin-king)]

Just to re-iterate:

The patch to fix rev 02 chipsets was reverted because it was a significant patch that broke the broadcom driver for BCM4312 chipsets to name just a few. Apologies for this, note that I understand that 2.6.25 fixes your problem but to isolate the fix and backport it to 2.6.24 so close to the Hardy release without breaking the driver for other BCM chipsets is not possible. I am therefore marking this as "Won't Fix" for Hardy, and keeping it open for Hardy + 1.

[Nicolò Chieffo (yelo3)]

I really don't understand what's happening, but I now tried to boot
ubuntu original kernel 2.6.24-16-generic, and ssb load without
errors... also b43 loads, and the wlan card seems not to have problems
(for now, I'm only at the first boot)

[Nicolò Chieffo (yelo3)]

I'm sorry, I was wrong. I had ndiswrapper installed and I forgot! ssb
loads when ndiswrapper is loaded.

[Nicolò Chieffo (yelo3)]

Now I removed ndiswrapper and compiled compat-wireless-2.6 from
http://linuxwireless.org/download/compat-wireless-2.6/compat-wireless-2.6.tar.bz2
I have to say that this is also an option, instead of using
linux-2.6.25 you keep the ubuntu kernel, and only update the wireless
stack. I have to say that now it works really well.

[Miłosz Kosobucki (mikom)]

So, as I can see the bug isn't going to be resolved until hardy+1? Great :/

[Johan Zaxmy (johan-zaxmy)]

Miłosz Kosobucki wrote:
> So, as I can see the bug isn't going to be resolved until hardy+1? Great
> :/
>
>
Yup really great, this means that all our notebook models at the
university doesn't work. Since HP 2510p, 6910p and 8510p is delivered
with "Broadcom Corporation BCM94311MCG wlan mini-PCI (rev 02)".
:(

-Johan Zaxmy

[Wolfgang (this-ece)]

"Now I removed ndiswrapper and compiled compat-wireless-2.6 from
http://linuxwireless.org/download/compat-wireless-2.6/compat-wireless-2.6.tar.bz2
I have to say that this is also an option, instead of using
linux-2.6.25 you keep the ubuntu kernel, and only update the wireless
stack. I have to say that now it works really well."

this solution was proposed in my report (https://bugs.launchpad.net/ubuntu/+source/linux/+bug/211235).

Now i can see networks, but i can't connect to them, with network-admin or networkmanager. Someone had this bug too ?

[Zandre Bran (zandrebran)]

On Tue, Apr 29, 2008 at 8:42 PM, Wolfgang <email address hidden> wrote:

    Hi. Please, sorry my english, from Brazil.

> "Now I removed ndiswrapper and compiled compat-wireless-2.6 from
> http://linuxwireless.org/download/compat-wireless-2.6/compat-wireless-2.6.tar.bz2
> I have to say that this is also an option, instead of using
> linux-2.6.25 you keep the ubuntu kernel, and only update the wireless
> stack. I have to say that now it works really well."
>
> this solution was proposed in my report
> (https://bugs.launchpad.net/ubuntu/+source/linux/+bug/211235).
>
> Now i can see networks, but i can't connect to them, with network-admin
> or networkmanager. Someone had this bug too ?

     No, my problem is dont work blacklist ssb for load ndiswrapper. See:

# cat /etc/modprobe.d/blacklist
...
blacklist ssb
blacklist b43
...

# lsmod | grep ssb
ssb 32260 0

   But, for work my wire add in rc.local:

# cat /etc/rc.local
...
modprobe -r ssb
modprobe -r ndiswrapper
modprobe ndiswrapper
...

    I'm report new bug #227158

[]s
- Zandre.

[Nicolò Chieffo (yelo3)]

to make ssb not load you have to blacklist it and to execute

sudo update-initramfs -k all -u

I don't know why, but it is loaded in initrd.img during the boot.

[Zandre Bran (zandrebran)]

On Tue, May 6, 2008 at 7:41 AM, Nicolò Chieffo <email address hidden> wrote:

     Thank you for responding.

> to make ssb not load you have to blacklist it and to execute
>
> sudo update-initramfs -k all -u
>
> I don't know why, but it is loaded in initrd.img during the boot.

     Ok, was resolved with an update today (05-may). but there is a
list that is not USN. I suggest the idea:

http://brainstorm.ubuntu.com/idea/1615/

- []s
- Zandre.

[Nicolò Chieffo (yelo3)]

correctly works in Intrepid

[Nicolò Chieffo (yelo3)]

correctly works in Intrepid. reopen if not for you

[Chuck Renner (chuckrenner)]

I have found only two things that work correctly for my BCM4311/12 (rev 02). One is ndiswrapper 0.52/0.53 (newer than the 0.50 available in Ubuntu packages) with the drivers found from HP/Compaq as SP36684A. The second is the wireless-compat-2.6.26 patch.

I am interested in learning how to take the work done in the wireless-compat-2.6.26 (replacement modules for all the wireless drivers they support), and create dpkgs available to the Ubuntu universe (community) repositories. Effectively, there could be a wireless-compat-2.6.26-modules_<arch>_2.6.24-<abi#>-ubuntu<rev#> package for each architecture, kernel, and abi containing the compiled replacement driver modules, as well as a meta-package with a simple name like wireless-compat, that always keeps the correct replacement/compatibility modules installed in case of a kernel upgrade or an abi bump. Since the compatibility patch works for kernels as old as 2.6.20, older versions of Ubuntu could even be supported (at least Gutsy, maybe Fiesty, Edgy, or Dapper).

I'm more interested in learning how to create the packages, dependencies, documentation, etc., and learning how to upload them to the repositories and become a package maintainer, then I am somebody else doing it. Especially since the modules would have built in obsolescence (they would no longer be needed for any version of Ubuntu with the 2.6.26 kernel or higher).

If I am able to pull that off, it might even make sense to make packages with the required (extracted) firmwares for Broadcom cards, and get it available in the Medibuntu repositories, but that is a lot less important, since it is not that painful to do.

But back to the point. This is a confirmation that the new wireless-compat patch that came out on 06-25 works well on Hardy Heron (no problems so far anyway).

I removed my ndiswrapper drivers (moved /etc/ndiswrapper and /etc/modprobe.d/ndiswrapper to folder in my home directory so they could ndiswrapper would not be able to load, but I preserving my settings because I can always move them back). I unblacklisted ssb and b43, and ran all of the following from the extracted archive:
sudo make # builds the compatibility modules
sudo make install
sudo make load
sudo update-initramfs -u -k all -v

Of course, the extracted firmwares available by running b43-fwcutter are still required to be installed, build-essentials are needed, as well as the linux-headers package, but all that is par for the course by now.

Can somebody point me in the right direction as far as learning the required steps to create these repository based packages? This way, anyone using a broadcom and the compat-wireless pack, can be assured that they will have the new modules with each kernel and abi bump. :)

[Chuck Renner (chuckrenner)]

Broken for all linux kernels < 2.6.26. Compatibility patch is available (it replaces the affected modules in the kernel), and should ultimately be available as repository-based meta-package with compiled replacement modules for either universe or main. Affects too many users not to be supported on Hardy. I would be willing to consider becoming package maintainer if someone could point me in the right direction for learning how to do so.

[Tim Gardner (timg-tpi)]

Chuck - you can run a packaged version of compat-wireless by installing linux-backports-modules-2.6.24 from my PPA at http://ppa.launchpad.net/timg-tpi/ubuntu. This LBM package will be uploaded sometime in early July 2008.

[Nicolò Chieffo (yelo3)]

Sorry, for intrepid it is fixed, you should change the Hardy "Affects" key

[Chuck Renner (chuckrenner)]

Thanks guys. It's great to know that this patch will be in this backports module available for Hardy. It does solve the problem for me personally, since I'll add the ppa to my repositories list, and will install the new package from Tim as soon as it is available. I'm still concerned about the sheer number of laptop users that have broadcom chipsets that are affected by this. Adding unsigned code from a ppa repository is not a new trick for me, but it certainly is less friendly than desired for new Ubuntu users. Since Intrepid is obviously using 2.6.26 (or greater), this only affects Hardy or earlier distributions. Officially, I believe Dapper Dan (which was a LTS release), Fiesty Fawn, and Gutsy Gibbon are still supported, so that means they are affected as well. Does this mean we should add these to the affects list? I believe the compat-wireless pack is for all kernels >= 2.6.22 or above, and that the developers plan on supporting all the way back to 2.6.18. I kind of skipped Edgy Eft and Fiesty Fawn, so I dont' know which distributions use which kernels. I did use Gutsy Gibbon, and it was 2.6.24, so it falls into this category. It becomes a moot point if there aren't any affected users.

It seems to me that almost every current AMD-based laptop on the market is using broadcom wireless chipsets, and the Intel based ones are using Intel wireless chipsets to get the logo compliance they need for those nifty little stickers like "Centrino inside". So one of my concerns is that the solution still seems difficult for new linux/Ubuntu users, and that it can hurt in winning over market share for linux/debian/Ubuntu, etc.

Still having it in a maintained repository (even a PPA one), is better than requiring people to compile, and takes away the work for most of us when we the kernel abi gets bumped for a security fix. So thanks again, guys!!!

[Nicolò Chieffo (yelo3)]

For me this works with 8.04.1
feel free to reopen it if this does not work for you!

[Launchpad Janitor (janitor)]

Per a decision made by the Ubuntu Kernel Team, bugs will longer be assigned to the ubuntu-kernel-team in Launchpad as part of the bug triage process. The ubuntu-kernel-team is being unassigned from this bug report. Refer to https://wiki.ubuntu.com/KernelTeamBugPolicies for more information. Thanks.