BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts

I have exactly the same problem. Any solution!?

Nothing yet. :-(

You should say in the bug that you reproduced it.

2006/9/6, chevrier <email address hidden>:
>
> I have exactly the same problem. Any solution!?
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

--
Martin Hamel
(418)261-2222
<email address hidden>

here is the relevant part of my dmesg, but it's really the same, everything is the same. I googled like crazy, but to no avail - no one seems to have this fiigured out at all. The only solution I have found so far is to disable the nvidia driver - note that I had the nvidia linux driver installed. I hope it'll work!

Thanks for keeping me updated on this!

[17179589.584000] ACPI: AC Adapter [AC] (on-line)
[17179589.604000] ACPI: Battery Slot [BAT0] (battery present)
[17179589.688000] ACPI: Lid Switch [LID]
[17179589.688000] ACPI: Power Button (CM) [PBTN]
[17179589.688000] ACPI: Sleep Button (CM) [SBTN]
[17179589.772000] ibm_acpi: ec object not found
[17179589.808000] pcc_acpi: loading...
[17179589.892000] ACPI: Video Device [VID] (multi-head: yes rom: no post: no)
[17179594.204000] ppdev: user-space parallel port driver
[17179594.832000] apm: BIOS not found.
[17179598.064000] NET: Registered protocol family 10
[17179598.064000] lo: Disabled Privacy Extensions
[17179598.064000] ADDRCONF(NETDEV_UP): eth0: link is not ready
[17179598.064000] ADDRCONF(NETDEV_UP): wlan0: link is not ready
[17179598.064000] IPv6 over IPv4 tunneling driver
[17179599.516000] Bluetooth: Core ver 2.8
[17179599.516000] NET: Registered protocol family 31
[17179599.516000] Bluetooth: HCI device and connection manager initialized
[17179599.516000] Bluetooth: HCI socket layer initialized
[17179599.584000] Bluetooth: L2CAP ver 2.8
[17179599.584000] Bluetooth: L2CAP socket layer initialized
[17179599.680000] Bluetooth: RFCOMM socket layer initialized
[17179599.680000] Bluetooth: RFCOMM TTY layer initialized
[17179599.680000] Bluetooth: RFCOMM ver 1.7
[17179636.992000] ADDRCONF(NETDEV_UP): wlan0: link is not ready
[17179644.104000] ADDRCONF(NETDEV_UP): wlan0: link is not ready
[17179644.708000] ADDRCONF(NETDEV_CHANGE): wlan0: link becomes ready
[17179655.588000] wlan0: no IPv6 routers present
[17179666.392000] irq 177: nobody cared (try booting with the "irqpoll" option)
[17179666.392000] [<c014ea3a>] __report_bad_irq+0x2a/0xa0
[17179666.392000] [<c014e1fd>] handle_IRQ_event+0x3d/0x70
[17179666.392000] [<c014eb57>] note_interrupt+0x87/0xf0
[17179666.392000] [<c014e32d>] __do_IRQ+0xfd/0x110
[17179666.392000] [<c0105ce9>] do_IRQ+0x19/0x30
[17179666.392000] [<c0103f26>] common_interrupt+0x1a/0x20
[17179666.392000] [<f8872f35>] acpi_processor_idle+0x21e/0x32d [processor]
[17179666.392000] [<c010111f>] cpu_idle+0x6f/0xc0
[17179666.392000] [<c03aca3f>] start_kernel+0x19f/0x200
[17179666.392000] [<c03ac3c0>] unknown_bootoption+0x0/0x1f0
[17179666.392000] handlers:
[17179666.392000] [<f8c319a0>] (ndis_isr+0x0/0xd0 [ndiswrapper])
[17179666.392000] Disabling IRQ #177

and bye bye my wireless connection...

I'm seeing the same issue-wireless works for a while, then shuts down. I have a Dell D620 with the Broadcom 4311, using the 686 kernel and nvidia driver. I also added the line:

blacklist bcm43xx

to /etc/modprobe.d/blacklist in case there was a conflict with the native Broadcom driver, which doesn't yet work with this chip, but that didn't help. TIA for any help!

Here is the dmesg output:

[17179588.252000] ndiswrapper version 1.23 loaded (preempt=yes,smp=yes)
[17179588.320000] ndiswrapper: driver bcmwl5 (Broadcom,11/02/2005, 4.10.40.0) loaded
[17179588.320000] ACPI: PCI Interrupt 0000:0c:00.0[A] -> GSI 17 (level, low) -> IRQ 177
[17179588.320000] PCI: Setting latency timer of device 0000:0c:00.0 to 64
[17179588.324000] ndiswrapper: using irq 177
[17179588.980000] wlan0: vendor: ''
[17179588.980000] wlan0: ethernet device 00:16:ce:8e:c7:47 using NDIS driver bcmwl5, 14E4:4311.5.conf
[17179588.980000] wlan0: encryption modes supported: WEP; TKIP with WPA, WPA2, WPA2PSK; AES/CCMP with WPA, WPA2, WPA2PSK
[17179594.152000] NET: Registered protocol family 10
[17179594.152000] lo: Disabled Privacy Extensions
[17179594.152000] IPv6 over IPv4 tunneling driver
[17179596.224000] ACPI: AC Adapter [AC] (on-line)
[17179596.248000] ACPI: Battery Slot [BAT0] (battery present)
[17179596.248000] ACPI: Battery Slot [BAT1] (battery absent)
[17179596.336000] ACPI: Lid Switch [LID]
[17179596.336000] ACPI: Power Button (CM) [PBTN]
[17179596.336000] ACPI: Sleep Button (CM) [SBTN]
[17179596.428000] ibm_acpi: ec object not found
[17179596.472000] pcc_acpi: loading...
[17179596.560000] ACPI: Video Device [VID] (multi-head: yes rom: no post: no)
[17179596.560000] ACPI: Video Device [VID] (multi-head: yes rom: no post: no)
[17179596.560000] ACPI: Video Device [VID2] (multi-head: yes rom: no post: no)[17179602.680000] ppdev: user-space parallel port driver
[17179603.212000] apm: BIOS not found.
[17179604.228000] eth0: no IPv6 routers present
[17179604.828000] wlan0: no IPv6 routers present
[17179607.160000] Bluetooth: Core ver 2.8
[17179607.160000] NET: Registered protocol family 31
[17179607.160000] Bluetooth: HCI device and connection manager initialized
[17179607.160000] Bluetooth: HCI socket layer initialized
[17179607.216000] Bluetooth: L2CAP ver 2.8
[17179607.216000] Bluetooth: L2CAP socket layer initialized
[17179607.240000] Bluetooth: RFCOMM socket layer initialized
[17179607.240000] Bluetooth: RFCOMM TTY layer initialized
[17179607.240000] Bluetooth: RFCOMM ver 1.7
[17179624.524000] eth0: no IPv6 routers present
[17179913.992000] ibm_acpi: ec object not found
[17180234.540000] irq 177: nobody cared (try booting with the "irqpoll" option)
[17180234.540000] [<c014fc8a>] __report_bad_irq+0x2a/0xa0
[17180234.540000] [<c014f44d>] handle_IRQ_event+0x3d/0x70
[17180234.540000] [<c014fda7>] note_interrupt+0x87/0xf0
[17180234.540000] [<c014f57d>] __do_IRQ+0xfd/0x110
[17180234.540000] [<c0105c79>] do_IRQ+0x19/0x30
[17180234.540000] [<c0103eb6>] common_interrupt+0x1a/0x20
[17180234.540000] [<f8872f35>] acpi_processor_idle+0x21e/0x32d [processor]
[17180234.540000] [<c010111f>] cpu_idle+0x6f/0xc0
[17180234.540000] [<c03aca3f>] start_kernel+0x19f/0x200
[171802...

I may have found at least a workaround. After posting my earlier comment, I found the Ndiswrapper FAQ (http://ndiswrapper.sourceforge.net/mediawiki/index.php/FAQ) which suggested disabling ACPI irq routing with the boot commandline parameter acpi=noirq. I did that and my wireless has now stayed up for over 2 hours!

I haven't see any ill effects, but I haven't looked hard either.

Here is my dmesg output with acpi=noirq:

[17179588.328000] ndiswrapper version 1.23 loaded (preempt=yes,smp=yes)
[17179588.396000] ndiswrapper: driver bcmwl5 (Broadcom,11/02/2005, 4.10.40.0) loaded
[17179588.396000] PCI: Assigned IRQ 3 for device 0000:0c:00.0
[17179588.396000] PCI: Sharing IRQ 3 with 0000:00:1f.2
[17179588.396000] PCI: Sharing IRQ 3 with 0000:00:1f.3
[17179588.396000] PCI: Setting latency timer of device 0000:0c:00.0 to 64
[17179588.404000] ndiswrapper: using irq 3
[17179589.056000] wlan0: vendor: ''
[17179589.056000] wlan0: ethernet device 00:16:ce:8e:c7:47 using NDIS driver bcmwl5, 14E4:4311.5.conf
[17179589.056000] wlan0: encryption modes supported: WEP; TKIP with WPA, WPA2, WPA2PSK; AES/CCMP with WPA, WPA2, WPA2PSK
[17179592.980000] NET: Registered protocol family 10
[17179592.980000] lo: Disabled Privacy Extensions
[17179592.980000] ADDRCONF(NETDEV_UP): eth0: link is not ready
[17179592.980000] IPv6 over IPv4 tunneling driver
[17179594.124000] ACPI: AC Adapter [AC] (on-line)
[17179594.148000] ACPI: Battery Slot [BAT0] (battery present)
[17179594.148000] ACPI: Battery Slot [BAT1] (battery absent)
[17179594.228000] ACPI: Lid Switch [LID]
[17179594.228000] ACPI: Power Button (CM) [PBTN]
[17179594.228000] ACPI: Sleep Button (CM) [SBTN]
[17179594.324000] ibm_acpi: ec object not found
[17179594.356000] pcc_acpi: loading...
[17179594.448000] ACPI: Video Device [VID] (multi-head: yes rom: no post: no)
[17179594.448000] ACPI: Video Device [VID] (multi-head: yes rom: no post: no)
[17179594.448000] ACPI: Video Device [VID2] (multi-head: yes rom: no post: no)
[17179600.400000] ppdev: user-space parallel port driver
[17179600.892000] apm: BIOS not found.
[17179604.760000] wlan0: no IPv6 routers present
[17179605.144000] Bluetooth: Core ver 2.8
[17179605.144000] NET: Registered protocol family 31
[17179605.144000] Bluetooth: HCI device and connection manager initialized
[17179605.144000] Bluetooth: HCI socket layer initialized
[17179605.196000] Bluetooth: L2CAP ver 2.8
[17179605.196000] Bluetooth: L2CAP socket layer initialized
[17179605.216000] Bluetooth: RFCOMM socket layer initialized
[17179605.216000] Bluetooth: RFCOMM TTY layer initialized
[17179605.216000] Bluetooth: RFCOMM ver 1.7
[17180802.816000] spurious 8259A interrupt: IRQ7.
j
Note the last line, which occurred a while after boot--the same time when I was losing wireless connectivity before.

Following up further, it appears that the acpi=noirq option is NOT a workaround. My connection stayed up the first few times I tried it, but since it seems to fail after about the same period of time.

Sorry for the false lead!

Just a note, this is still an issue with the nvidia beta driver, dunno if anyone expected it to be fixed, but its still an issue there.

Please let me know if theres any way I can help. I am more than willing to arrange a time to sit down on IRC or something and hammer this out the manual way with someone ;). Just get in touch with me through any one of the ways on my launchpad page.

Where do I meet you on IRC?

2006/10/4, Kevin Kubasik <email address hidden>:
>
> Just a note, this is still an issue with the nvidia beta driver, dunno
> if anyone expected it to be fixed, but its still an issue there.
>
> Please let me know if theres any way I can help. I am more than willing
> to arrange a time to sit down on IRC or something and hammer this out
> the manual way with someone ;). Just get in touch with me through any
> one of the ways on my launchpad page.
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

--
Martin Hamel
(418)261-2222
<email address hidden>

kkubasik on irc.gnome.org
or
kkubasik/kkubasi1 on irc.freenode.net

or qub333 on gtalk/jabber or qub333 on aim, any form of contact is fine.

Just a note, I mean this as a user, not someone who can fix it, I have been trying for about a month now with no success, but if someone with more experience with ndiswrapper/the linux kernel is willing to talk/work with me we can try to figure it out.

ok ;-) I thought you where a kernel hacher or something. On my side, I think
more and more about buying a new wifi card.

good luck!

2006/10/4, Kevin Kubasik <email address hidden>:
>
> Just a note, I mean this as a user, not someone who can fix it, I have
> been trying for about a month now with no success, but if someone with
> more experience with ndiswrapper/the linux kernel is willing to
> talk/work with me we can try to figure it out.
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

--
Martin Hamel
(418)261-2222
<email address hidden>

Ok, well best of luck with the new card, I'm gonna figure this out eventually ;)

Could you provide me with the output of 'cat /proc/' ?

Mine looks like this.

           CPU0 CPU1
  0: 7171886 0 IO-APIC-edge timer
  1: 24263 0 IO-APIC-edge i8042
  8: 3 0 IO-APIC-edge rtc
  9: 4 0 IO-APIC-level acpi
 12: 207 0 IO-APIC-edge i8042
 14: 255552 0 IO-APIC-edge libata
 15: 200969 0 IO-APIC-edge libata
 58: 3290794 0 IO-APIC-level uhci_hcd:usb1, ehci_hcd:usb5
 66: 368541 0 IO-APIC-level uhci_hcd:usb2, HDA Intel
 74: 0 0 IO-APIC-level uhci_hcd:usb3
 82: 0 0 IO-APIC-level uhci_hcd:usb4
 90: 10761725 0 PCI-MSI eth0
169: 2008836 0 IO-APIC-level nvidia
177: 5835 0 IO-APIC-level eth1
193: 10 0 IO-APIC-level ohci1394, yenta
NMI: 0 0
LOC: 7171673 7171672
ERR: 0
MIS: 0

           CPU0 CPU1
  0: 2147788 0 IO-APIC-edge timer
  1: 8286 0 IO-APIC-edge i8042
  8: 3 0 IO-APIC-edge rtc
  9: 5 0 IO-APIC-level acpi
 12: 139668 0 IO-APIC-edge i8042
 14: 22116 0 IO-APIC-edge libata
 15: 59541 0 IO-APIC-edge libata
 50: 3889 0 IO-APIC-level uhci_hcd:usb2, HDA Intel
 58: 0 0 IO-APIC-level uhci_hcd:usb3
 66: 0 0 IO-APIC-level uhci_hcd:usb4
177: 470543 0 IO-APIC-level ndiswrapper
185: 7 0 IO-APIC-level yenta
233: 159 0 IO-APIC-level uhci_hcd:usb1, ehci_hcd:usb5
NMI: 0 0
LOC: 2147382 2147381
ERR: 0
MIS: 0

2006/10/4, Kevin Kubasik <email address hidden>:
>
> Ok, well best of luck with the new card, I'm gonna figure this out
> eventually ;)
>
> Could you provide me with the output of 'cat /proc/' ?
>
> Mine looks like this.
>
> CPU0 CPU1
> 0: 7171886 0 IO-APIC-edge timer
> 1: 24263 0 IO-APIC-edge i8042
> 8: 3 0 IO-APIC-edge rtc
> 9: 4 0 IO-APIC-level acpi
> 12: 207 0 IO-APIC-edge i8042
> 14: 255552 0 IO-APIC-edge libata
> 15: 200969 0 IO-APIC-edge libata
> 58: 3290794 0 IO-APIC-level uhci_hcd:usb1, ehci_hcd:usb5
> 66: 368541 0 IO-APIC-level uhci_hcd:usb2, HDA Intel
> 74: 0 0 IO-APIC-level uhci_hcd:usb3
> 82: 0 0 IO-APIC-level uhci_hcd:usb4
> 90: 10761725 0 PCI-MSI eth0
> 169: 2008836 0 IO-APIC-level nvidia
> 177: 5835 0 IO-APIC-level eth1
> 193: 10 0 IO-APIC-level ohci1394, yenta
> NMI: 0 0
> LOC: 7171673 7171672
> ERR: 0
> MIS: 0
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

--
Martin Hamel
(418)261-2222
<email address hidden>

Have you tried using the proper Linux driver? If you're having crashes while

a) running code that we don't have the source for (nvidia), and
b) running code that we don't have the source for and which is written for a different OS (ndiswrapper/broadcom driver)

it's sadly quite difficult for us to do anything.
https://help.ubuntu.com/community/WifiDocs/Driver/bcm43xx/Dapper has information on how to set it up.

Unfortunetly, I have tried bcm43xx and have had no luck getting it to work. Part of the issue might be me trying it under edgy. But NetworkManager has no luck with getting any response from the network interface when it is enabled using bcm43xx.

If there is a more explicit howto I should follow, please let me know, I am more than willing to try most anything.

Ok, when booting with irqpoll I get this error when cpu usage is high, which results in the death of wireless, is this something we could hypothetically fix?

[17179775.704000] BUG: soft lockup detected on CPU#0!
[17179775.704000] <c01493df> softlockup_tick+0x9f/0xf0 <c012c0e1> update_process_times+0x31/0x80
[17179775.704000] <c0114d13> smp_apic_timer_interrupt+0x53/0x60 <c010413c> apic_timer_interrupt+0x1c/0x30
[17179775.704000] <c02d6a99> _spin_unlock_irqrestore+0x9/0x10 <c025ae33> i8042_interrupt+0x1d3/0x220
[17179775.704000] <c010408a> common_interrupt+0x1a/0x20 <c0149533> handle_IRQ_event+0x33/0x60
[17179775.704000] <c01495fd> __do_IRQ+0x9d/0x110 <c0105c89> do_IRQ+0x19/0x30
[17179775.704000] <c010408a> common_interrupt+0x1a/0x20
[17179790.656000] irq 177: nobody cared (try booting with the "irqpoll" option)
[17179790.656000] <c0149bb4> __report_bad_irq+0x24/0x80 <c0149cad> note_interrupt+0x9d/0x270
[17179790.656000] <f8b34f82> ndis_isr+0x52/0xc0 [ndiswrapper] <c0149658> __do_IRQ+0xf8/0x110
[17179790.656000] <c0105c89> do_IRQ+0x19/0x30 <c010408a> common_interrupt+0x1a/0x20
[17179790.656000] <c0149517> handle_IRQ_event+0x17/0x60 <c01495fd> __do_IRQ+0x9d/0x110
[17179790.656000] <c0105c89> do_IRQ+0x19/0x30 <c010408a> common_interrupt+0x1a/0x20
[17179790.656000] handlers:
[17179790.656000] [<f8b34f30>] (ndis_isr+0x0/0xc0 [ndiswrapper])
[17179790.656000] Disabling IRQ #177

The BCM43XX does not support the 4311 yet. Actually, they are working on it
and I eard that it was partially working now. But that version is not in the
2.6.17 kernel. And I don't know what "partially working mean". Well, I guess
we'll have a working driver in edgyh+1. But that leave's us like that for
another 6 months. :-(

2006/10/4, Matthew Garrett <email address hidden>:
>
> Have you tried using the proper Linux driver? If you're having crashes
> while
>
> a) running code that we don't have the source for (nvidia), and
> b) running code that we don't have the source for and which is written for
> a different OS (ndiswrapper/broadcom driver)
>
> it's sadly quite difficult for us to do anything.
> https://help.ubuntu.com/community/WifiDocs/Driver/bcm43xx/Dapper has
> information on how to set it up.
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

--
Martin Hamel
(418)261-2222
<email address hidden>

See that web page for 4311 driver status:
http://bcm43xx.berlios.de/?go=devices

I guess that would be our best hope. Right now, the driver is listed has
unstable for the 4311. We could have something working soon. But not for
edgy. It will mean recompiling our kernel our wait for edgy+1.

2006/10/5, Martin Hamel <email address hidden>:
>
> The BCM43XX does not support the 4311 yet. Actually, they are working on
> it and I eard that it was partially working now. But that version is not in
> the 2.6.17 kernel. And I don't know what "partially working mean". Well, I
> guess we'll have a working driver in edgyh+1. But that leave's us like that
> for another 6 months. :-(
>
> 2006/10/4, Matthew Garrett <email address hidden>:
> >
> > Have you tried using the proper Linux driver? If you're having crashes
> > while
> >
> > a) running code that we don't have the source for (nvidia), and
> > b) running code that we don't have the source for and which is written
> > for a different OS (ndiswrapper/broadcom driver)
> >
> > it's sadly quite difficult for us to do anything.
> > https://help.ubuntu.com/community/WifiDocs/Driver/bcm43xx/Dapper has
> > information on how to set it up.
> >
> > --
> > BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> > https://launchpad.net/bugs/57355
> >
>
>
>
> --
> Martin Hamel
> (418)261-2222
> <email address hidden>
>

--
Martin Hamel
(418)261-2222
<email address hidden>

The 4311 should be somewhat supported by the driver in edgy. Have you tried it?

Yes.. and I do not see any network with the kde wireless manager.

2006/10/9, Matthew Garrett <email address hidden>:
>
> The 4311 should be somewhat supported by the driver in edgy. Have you
> tried it?
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

--
Martin Hamel
(418)261-2222
<email address hidden>

I just wanted too report that I too have this bug with the broadcom BCM4310 card in a Dell D820.

lspci -v output:
0c:00.0 Network controller: Broadcom Corporation BCM4310 UART (rev 01)
        Subsystem: Dell Unknown device 0007
        Flags: bus master, fast devsel, latency 0, IRQ 177
        Memory at dcffc000 (32-bit, non-prefetchable) [size=16K]
        Capabilities: <access denied>

I'm no "kernel hacker" as mentioned above but please let me know if I can help with this one.

Using ndiswrapper with a broadcom card will work, only up till around the first 15 minutes, then it crashes. It seems that I can get a 4306 card to work with the native driver.

How about 64 bits? Does anyone have the driver? I looked around, and could not find anything, except for a few incompatible ones. Dell has not posted any 64 bits drivers (it already took them over a month to post the bios update to get the new merom processor recognized!). I am hoping that going 64 bits might help - who knows!?
thanks and regards, for any help you may provide.

I confirm the same problem using Dell Latitude D820 w/ Dell Wireless 1490 Minicard using ndiswrapper.

My kernel: 2.6.17-10-generic #2 SMP Fri Oct 13 18:45:35 UTC 2006 i686 GNU/Linux

Any progress on this issue?

I have read on other forums, it is a clash of nvidia driver and wifi. I do not want to use the nv driver as the performance of the graphic card with this driver is soooo bad....

Matthew wrote:

> Have you tried using the proper Linux driver? If you're having crashes while

>a) running code that we don't have the source for (nvidia), and
>b) running code that we don't have the source for and which is written for a different OS (ndiswrapper/broadcom driver)

>it's sadly quite difficult for us to do anything.
>https://help.ubuntu.com/community/WifiDocs/Driver/bcm43xx/Dapper has information on how to set it up.

I've got a Dell D620 with this card, and I'm unclear as to what the "Ubuntu-endorsed" way is to get the card to work with Edgy (the instructions above are for Dapper). Are the same? I've been through what feels like hundreds of "How-Tos" at this point, and some have the prerequisite of "start with a fresh installation, since other competing how tos may have hosed your configuration." I'm going to clone my partitions, but reloading them before each "how to" is a nightmare, so I'm looking for the how to with Ubuntu's stamp of approval.

After much trial and error, I have my 4311 working with ndiswrapper, no problems. (I don't have any nvidia, but thought I'd still post a bit here.)

I am running the 2.6.17-10-generic x86_64 kernel, ndiswrapper version 1.27 (downloaded from Sourceforge, and newer than the version in Edgy's repositories), and a driver downloaded from HP. My Dell 1390 wireless card now works on my E1405 laptop. A meandering HOWTO is here: http://wiki.waningsun.net/index.php?title=Dell_E1405_%26_Linux_HOWTO#ndiswrapper.2C_fixed.21

Jamie: You should be able to load and unload modules without re-installing your system, so don't worry about re-installing every time you change something!

Original bug report talks about conflicts between two proprietary drivers (windows ndis driver, and nvidia driver). I cannot fix this.

TO L. Kevin

Yes, with the original graphic card driver, things work fine. the problem occurs with the proprietary nvidia driver...

I wonder why this bug has status REJECTED??

It's rejected because we can't fix bugs related to proprietary drivers. One being a windows NDIS driver, and the other being a proprietary nvidia driver.

But there is a broadcom free driver and they are working on the cards which
are interesting us. So there should be a solution pretty soon. Certainly at
the next kernel. It is not true that nothing is going to be done about it.

2006/12/2, Ben Collins <email address hidden>:
>
> It's rejected because we can't fix bugs related to proprietary drivers.
> One being a windows NDIS driver, and the other being a proprietary
> nvidia driver.
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

--
Martin Hamel
(418)261-2222
<email address hidden>

It's a said story.... This is the face of linux that I find very tricky. I am using dual boot, having xp and ubuntu. Everything is ok in xp. Ubuntu has problems with irq conflict for wifi and nvidia card (rejected bug), screensaver (confirmed bug), sound setting. I like it very much but this just makes work not comfortable.

Can you try booting with "pci=nomsi" on the kernel command line?

I just tried booting with "pci=nomsi" and still the same story - same error.

I have tried "pci=nomsi" too. Same problem...

Ok, it is the conflict of two drivers. I guess in principle any two drivers can have the same irq#. Shouldn't there be a way how to change it? There are devices that can have irq# changed...

Who is responsible for it? Shall we better contact nVidia or Broadcom or guys developing ndiswrapper?

How come that nvidia driver and the same broadcom driver in XP do not have this problem?

according to nvidia, the problem is in the BIOS or kernel. Look at posts from "netllama" here:

http://www.nvnews.net/vbulletin/showthread.php?t=48327&page=2

They do not seem to be interested in doing anything to work around it or fix the problem.

I can confirm this bug with my HP dv6000 series computer. Note that this is not ubuntu specific. Some people argue it is on the bios and/or kernel level. And some other people the issue (specifically for ndiswrapper and nvidia) has been / will be resolved in 2.16.19 linux kernel.

Note that this issue is not a bug in 64 bit kernels. With 64 bit kernel on AMD you can make ndiswrapper and nvidia working together without a workaround. But this does not elminates the probem for 32 bits.

While rejecting this bug is sensible (since it is not ubuntu specific) I would expect developers to take the initiative and offer a solution for users. Or contacting with nvidia , linux kernel, or ndiswrappers and acknowleding about the issue would be a great step. Because wide range of laptop users are affected from this issue.

I realize this is weird, and I'm not sure why this works, but it does, at least for me on my HP Pavillion dv6000z laptop. I'm posting this in hopes that it helps others...doing exactly this, in this order, has worked for me.
1) Stop gdm from starting on boot:
update-rc.d -f gdm remove
2) Make your kernel line in /boot/grub/menu.lst look like this "kernel /boot/vmlinuz-2.6.19 root=/dev/sda2 irqpoll pci=routeirq ro"
3) Once your computer boots, login (remember, gdm isnt going to load)
4) sudo rmmod bcm43xx ehci_hcd ohci_hcd sdhci mmc_core ieee80211softmac ieee80211 ieee80211_crypt
5) modprobe nvidia
6) modprobe ndiswrapper
7) use iwconfig and ifconfig to get your network up
8) Start up gdm "/etc/init.d/gdm start"
9) Have some fun. Yes, this leaves you without USB support, so that might not work for some people.

This works for me, using an hp pavilion dv6000 (dv1650us), which is turion dual 64bit processors). It may work for other systems.

1) Install latest nvidia drivers from their website in linux.
1.1) Download the drivers for linux from the nvidia website.
1.2) right-click on the downloaded file: go to properties: Permissions tab: make executable.
1.3) You may have to run it in a terminal using ctrl-alt-f1 (ctrl-alt-f7 to get back), and do "sudo /etc/init.d/gdm stop; sudo rmmod nvidia" beforehand.

2) Boot into windows, go to the hp (or whoever your manufacturer is) website, grab the most recent bios update. Install it.

3) Boot into linux and be happy.

Worked for me.

sorry Brian, but it doesn't :(
I'm on HP DV2000 (dv2125nr), running Slackware 11.
Have hardtime with this issue a long time already.
Have my BIOS updated,
using NVIDIA latest drivers NVIDIA-Linux-x86-1.0-9631
And I'm trying every single ndiswrapper released since v 1.20
It does work for me perfectly with using kernels 2.4.XX only!
But I have the same ndiswrapper and nvidia IRQ conflict issue when I'm on any of 2.6.XX kernel. Unfortunately. But I wish to use exactly 2.6.XX kernel for my laptop, due to many reasons, as one of them is better ACPI.
So, I have not given up yet and still keep my fingers crossed to one Guru, I guess, that will be able to find solution.
All the best to all of you!
Cheers!
P.S. Merry Christmas and Happy New Year!

I'm taking all my words back. Sorry Brian :)
It WORKS!
I have tried everything, for me it works on following configuration only:
Kernel 2.6.20-rc3
ndiswrapper 1.34rc2
NVIDIA-Linux-x86-1.0-9631
Cheers!

papay,
That nvidia driver is pretty old...1.0-9746 doesnt work, with kernel 2.6.20-rc3 and ndiswrapper 1.34rc2?

yup, I missed that :)
However the newrst one does work for me either.
Cheers Marcos!

I am having the same troubles as you guys. I am running Archlinux and I have the bcm4312 (dell 1490).
running

Dist: Archlinux
kernel: 2.6.19.2
kernel parameters: acpi=noirq irqpoll noapic mem=1024M
nvidia drivers : 9131
ndiswrapper: 1.34

About nvidia 9746 driver: this is no different of 9131 except that it supports the newest 8800 series cards, so it should make no difference between 9131 and 9746.

Iooking at my dmesg I see that this is more likely bug to be related to smp (I have dual core cpu). I havent tried it with kernel 2.6.20 but hopefully it will have been fixed as it did for papay and others.

Do you guys have smp also? what cpu type do you have?

cheers
Jinn

DMESG:
wlan2: no IPv6 routers present
spurious 8259A interrupt: IRQ7.
BUG: soft lockup detected on CPU#0! <------------------------------ NOTICE THIS!
 [<c014d30b>] softlockup_tick+0xab/0x110
 [<c0131131>] update_process_times+0x31/0x80
 [<c01178a0>] smp_apic_timer_interrupt+0x90/0xb0
 [<c0103d43>] apic_timer_interrupt+0x1f/0x24
 [<c014d681>] handle_IRQ_event+0x11/0x60
 [<c014ef46>] handle_level_irq+0x96/0x120
 [<c0105da9>] do_IRQ+0x69/0xd0
 [<c0103cb2>] common_interrupt+0x1a/0x20
 [<c012c1a2>] __do_softirq+0x62/0xf0
 [<c012c26b>] do_softirq+0x3b/0x50
 [<c012c4c5>] irq_exit+0x45/0x50
 [<c01178a5>] smp_apic_timer_interrupt+0x95/0xb0
 [<c0103d43>] apic_timer_interrupt+0x1f/0x24
 [<c026a993>] acpi_processor_idle+0x1fe/0x3bb
 [<c0405250>] unknown_bootoption+0x0/0x270
 [<c0101c06>] cpu_idle+0x76/0xe0
 [<c0405826>] start_kernel+0x366/0x430
 [<c0405250>] unknown_bootoption+0x0/0x270
 =======================
irq 4: nobody cared (try booting with the "irqpoll" option)
 [<c014e154>] __report_bad_irq+0x24/0x90
 [<c014e3f8>] note_interrupt+0x238/0x270
 [<f8b4d232>] ndis_isr+0x52/0xc0 [ndiswrapper]
 [<c014ef9e>] handle_level_irq+0xee/0x120
 [<c0105da9>] do_IRQ+0x69/0xd0
 [<c0103cb2>] common_interrupt+0x1a/0x20
 [<c014d681>] handle_IRQ_event+0x11/0x60
 [<c014ef46>] handle_level_irq+0x96/0x120
 [<c0105da9>] do_IRQ+0x69/0xd0
 [<c0103cb2>] common_interrupt+0x1a/0x20
 [<c014d681>] handle_IRQ_event+0x11/0x60
 [<c014ef46>] handle_level_irq+0x96/0x120
 [<c0105da9>] do_IRQ+0x69/0xd0
 [<c0103cb2>] common_interrupt+0x1a/0x20
 [<c012c1a2>] __do_softirq+0x62/0xf0
 [<c012c26b>] do_softirq+0x3b/0x50
 [<c012c4c5>] irq_exit+0x45/0x50
 [<c01178a5>] smp_apic_timer_interrupt+0x95/0xb0
 [<c0103d43>] apic_timer_interrupt+0x1f/0x24
 [<c026a993>] acpi_processor_idle+0x1fe/0x3bb
 [<c0405250>] unknown_bootoption+0x0/0x270
 [<c0101c06>] cpu_idle+0x76/0xe0
 [<c0405826>] start_kernel+0x366/0x430
 [<c0405250>] unknown_bootoption+0x0/0x270
 =======================
handlers:
[<f8b4d1e0>] (ndis_isr+0x0/0xc0 [ndiswrapper])
Disabling IRQ #4
jinn@Estergon:0 >

Adding "nosmp" to kernel parameters seems to have done it for me. Works just fine and I am using ndiswrapper right now.

I would really like to know now, to ensure that this is smp related (and not only in my case) that if you have smp systems as well, and if nosmp helps you also.

thx
Jinn

I have tested both with smp and without smp support in my kernel and I
experience the problem both ways. I've tried driver arguments in my
xorg.conf, various kernel args at boot time, several different
versions of the nvidia driver, and the only think that worked (kind
of) was not running both the nvidia driver and and the broadcom
(ndiswrapper) driver at the same time. I can use one of either but
not both at the same time. (nv instead of nvidia driver is my
preferred way to go because the last time I tested it, the broadcom
driver in the kernel only allowed 802.11b for my particular 802.11g
card ( "Dell D820/Broadcom Corporation BCM4310 UART (rev 01)" ).

Chris Bozic

On 1/14/07, jinn <email address hidden> wrote:
> DMESG:
> wlan2: no IPv6 routers present
> spurious 8259A interrupt: IRQ7.
> BUG: soft lockup detected on CPU#0! <------------------------------ NOTICE THIS!
> [<c014d30b>] softlockup_tick+0xab/0x110
> [<c0131131>] update_process_times+0x31/0x80
> [<c01178a0>] smp_apic_timer_interrupt+0x90/0xb0
> [<c0103d43>] apic_timer_interrupt+0x1f/0x24
> [<c014d681>] handle_IRQ_event+0x11/0x60
> [<c014ef46>] handle_level_irq+0x96/0x120
> [<c0105da9>] do_IRQ+0x69/0xd0
> [<c0103cb2>] common_interrupt+0x1a/0x20
> [<c012c1a2>] __do_softirq+0x62/0xf0
> [<c012c26b>] do_softirq+0x3b/0x50
> [<c012c4c5>] irq_exit+0x45/0x50
> [<c01178a5>] smp_apic_timer_interrupt+0x95/0xb0
> [<c0103d43>] apic_timer_interrupt+0x1f/0x24
> [<c026a993>] acpi_processor_idle+0x1fe/0x3bb
> [<c0405250>] unknown_bootoption+0x0/0x270
> [<c0101c06>] cpu_idle+0x76/0xe0
> [<c0405826>] start_kernel+0x366/0x430
> [<c0405250>] unknown_bootoption+0x0/0x270
> =======================
> irq 4: nobody cared (try booting with the "irqpoll" option)
> [<c014e154>] __report_bad_irq+0x24/0x90
> [<c014e3f8>] note_interrupt+0x238/0x270
> [<f8b4d232>] ndis_isr+0x52/0xc0 [ndiswrapper]
> [<c014ef9e>] handle_level_irq+0xee/0x120
> [<c0105da9>] do_IRQ+0x69/0xd0
> [<c0103cb2>] common_interrupt+0x1a/0x20
> [<c014d681>] handle_IRQ_event+0x11/0x60
> [<c014ef46>] handle_level_irq+0x96/0x120
> [<c0105da9>] do_IRQ+0x69/0xd0
> [<c0103cb2>] common_interrupt+0x1a/0x20
> [<c014d681>] handle_IRQ_event+0x11/0x60
> [<c014ef46>] handle_level_irq+0x96/0x120
> [<c0105da9>] do_IRQ+0x69/0xd0
> [<c0103cb2>] common_interrupt+0x1a/0x20
> [<c012c1a2>] __do_softirq+0x62/0xf0
> [<c012c26b>] do_softirq+0x3b/0x50
> [<c012c4c5>] irq_exit+0x45/0x50
> [<c01178a5>] smp_apic_timer_interrupt+0x95/0xb0
> [<c0103d43>] apic_timer_interrupt+0x1f/0x24
> [<c026a993>] acpi_processor_idle+0x1fe/0x3bb
> [<c0405250>] unknown_bootoption+0x0/0x270
> [<c0101c06>] cpu_idle+0x76/0xe0
> [<c0405826>] start_kernel+0x366/0x430
> [<c0405250>] unknown_bootoption+0x0/0x270
> =======================
> handlers:
> [<f8b4d1e0>] (ndis_isr+0x0/0xc0 [ndiswrapper])
> Disabling IRQ #4
> jinn@Estergon:0 >
>
>
> Adding "nosmp" to kernel parameters seems to have done it for me. Works just fine and I am using ndiswrapper right now.
>
> I would really like to know now, to ensure that this is smp related (and
> not only in my case) that if you have smp systems as well, and if nosmp
> helps you also.
>
> ...

for crying out loud, just upgrade to a 2.6.20-rc3 or higher kernel as stated and you wont have problems. stop giving yourselves headaches

Even works with 2.6.19.1. nosmp dramatically reduces the performance

On 1/15/07, Marcos Pinto <email address hidden> wrote:
>
> for crying out loud, just upgrade to a 2.6.20-rc3 or higher kernel as
> stated and you wont have problems. stop giving yourselves headaches
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

upgraded to 2.6.20-rc5 hoping it would make a difference.. And it did. But now I doesn't connect to the damned AP unless i am very very close... Is this the same for you?

I gotta say that I plainly hate Broadcom for being so damn anti Linux and uncooperative.. Hopefully the reverse engineered driver for Linux gets bugsfree (it works right now but has power issues = speed is below 60kb/s)

bcm43xx.berlios.de

I agree (sort of). I upgraded to feisty herd2 (which uses the 2.6.20
kernel) the other day and I no longer get the IRQ problems. The
wireless works as it did for me with the old kernel when I used the nv
driver instead of nvidia. I don't notice any difference regarding my
proximity to the access point.

Chris Bozic

On 1/18/07, jinn <email address hidden> wrote:
> upgraded to 2.6.20-rc5 hoping it would make a difference.. And it did.
> But now I doesn't connect to the damned AP unless i am very very
> close... Is this the same for you?
>
> I gotta say that I plainly hate Broadcom for being so damn anti Linux
> and uncooperative.. Hopefully the reverse engineered driver for Linux
> gets bugsfree (it works right now but has power issues = speed is below
> 60kb/s)
>
> bcm43xx.berlios.de
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

that my issue for this problem :
For me it's ok with option nosmp to the kernel .
I'have, hp dv9000ea laptop , with broadcom 4310 , my graphic card is nvidia 7600 .
I use nvidia proprietary drivers and ndiswrapper for the wifi .
But the performance are reduced.....

I am currently running kernel 2.6.20 which was released yesterday. With this
kernel you dont need the nosmp, and it works just fine.

hope it works for you.
Jinn

2007/2/6, francksau <email address hidden>:
>
> that my issue for this problem :
> For me it's ok with option nosmp to the kernel .
> I'have, hp dv9000ea laptop , with broadcom 4310 , my graphic card is
> nvidia 7600 .
> I use nvidia proprietary drivers and ndiswrapper for the wifi .
> But the performance are reduced.....
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

upgrade your kernel dude
On 2/6/07, francksau <email address hidden> wrote:
>
> that my issue for this problem :
> For me it's ok with option nosmp to the kernel .
> I'have, hp dv9000ea laptop , with broadcom 4310 , my graphic card is
> nvidia 7600 .
> I use nvidia proprietary drivers and ndiswrapper for the wifi .
> But the performance are reduced.....
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

The txpower bug is fixed! full speed with bcm4311 and bcm4312! No need for ndiswrapper anymore.

This is what you need.
kernel 2.6.20 and 2 patches:

ftp://lwfinger.dynalias.org/patches/2.6.20_combined
this one fixes :
> 1. The fix for DMA with > 1 GB RAM.
> 2. The code change to operate the radio LED for systems with a switch.
> 3. The changes to get proper scaling for rates and frequencies in iwlist.
> 4. The fix for the oops that occurs in hwrng_unregister on resuming. By the definitions used to
> control the changes in a stable release, this is a BUG and should be fixed in 2.6.20.1.

And this patch for the txpower issue:

Index: linux-2.6/drivers/net/wireless/bcm43xx/bcm43xx_phy.c
===================================================================
--- linux-2.6.orig/drivers/net/wireless/bcm43xx/bcm43xx_phy.c
+++ linux-2.6/drivers/net/wireless/bcm43xx/bcm43xx_phy.c
@@ -1225,7 +1225,7 @@ static void bcm43xx_phy_initg(struct bcm
       }
       if (phy->rev < 3 && phy->connected)
               bcm43xx_phy_write(bcm, 0x047E, 0x0078);
- if (phy->rev >= 6 && phy->rev <= 8) {
+ if (phy->rev >= 6 && phy->rev < 8) {
               bcm43xx_phy_write(bcm, 0x0801, bcm43xx_phy_read(bcm, 0x0801) | 0x0080);
               bcm43xx_phy_write(bcm, 0x043E, bcm43xx_phy_read(bcm, 0x043E) | 0x0004);
       }
_______________________________________________
Bcm43xx-dev mailing list
<email address hidden>
https://lists.berlios.de/mailman/listinfo/bcm43xx-dev

--------------------------------------------------------------------

patch with -l option on patch 2 since you will need to ignore the whitespace when you copy and paste.

with kernel 2.6.20 , i'have a black screen .........
so i will keep last kernel which work ....

Sorry, I read the whole thread and am wasn't sure if there was a fix found for this problem?

I am on a Dell E1405 with an nvidia card and the 1390 wireless card. I am using the nvidia beta drivers, and ndiswrapper for the wireless, and experiencing the same bug. Any help on what to do to fix this problem would be greatly appreciated.

So far the fix is to use a more current version of the kernel. I
started using the development builds of Feisty and my ndiswrapper and
nvidia are working just fine.

On 3/6/07, Stefan <email address hidden> wrote:
> Sorry, I read the whole thread and am wasn't sure if there was a fix
> found for this problem?
>
> I am on a Dell E1405 with an nvidia card and the 1390 wireless card. I
> am using the nvidia beta drivers, and ndiswrapper for the wireless, and
> experiencing the same bug. Any help on what to do to fix this problem
> would be greatly appreciated.
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

Thats right, I think this could be marked as fixed upstream, however, remember, the kernel isn't official yet, and even though were getting pretty close to a release, don't use it if unless your aware of the fun that comes with untested/development kernels.

On a similar note, bcm43xx support is significantly better for the 4311 series cards in Feisty, and you probably will find yourself content with that as opposed to ndiswrapper...

OK So I installed the latest kernel 2.6.20.1 with ndiswrapper 1.38 yet still am having the same problems. I get the following in my dmesg when I try to do any type of scanning on the wireless card:

 ADDRCONF(NETDEV_UP): eth1: link is not ready

Is there something I am missing?

I'm not totally sure but that wasn't the error I was getting related
to this bug. My card would work initially but then cause IRQ
conflicts when I used the nvidia driver. I wish I could help more but
I think your problem may not be related.

On 3/6/07, Stefan <email address hidden> wrote:
> OK So I installed the latest kernel 2.6.20.1 with ndiswrapper 1.38 yet
> still am having the same problems. I get the following in my dmesg when
> I try to do any type of scanning on the wireless card:
>
> ADDRCONF(NETDEV_UP): eth1: link is not ready
>
> Is there something I am missing?
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

ifconfig eth1 up before using iwconfig

On 3/6/07, Stefan <email address hidden> wrote:
> OK So I installed the latest kernel 2.6.20.1 with ndiswrapper 1.38 yet
> still am having the same problems. I get the following in my dmesg when
> I try to do any type of scanning on the wireless card:
>
> ADDRCONF(NETDEV_UP): eth1: link is not ready
>
> Is there something I am missing?
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

I am understanding that you have the bcm4306, if so you dont need
ndiswrapper and can use the bcm43xx module from the 2.6.20.1 kernel.

2007/3/7, Chris Bozic <email address hidden>:
>
> I'm not totally sure but that wasn't the error I was getting related
> to this bug. My card would work initially but then cause IRQ
> conflicts when I used the nvidia driver. I wish I could help more but
> I think your problem may not be related.
>
> On 3/6/07, Stefan <email address hidden> wrote:
> > OK So I installed the latest kernel 2.6.20.1 with ndiswrapper 1.38 yet
> > still am having the same problems. I get the following in my dmesg when
> > I try to do any type of scanning on the wireless card:
> >
> > ADDRCONF(NETDEV_UP): eth1: link is not ready
> >
> > Is there something I am missing?
> >
> > --
> > BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> > https://launchpad.net/bugs/57355
> >
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

2.6.20 is offical dude

On 3/7/07, Kevin Kubasik <email address hidden> wrote:
>
> Thats right, I think this could be marked as fixed upstream, however,
> remember, the kernel isn't official yet, and even though were getting
> pretty close to a release, don't use it if unless your aware of the fun
> that comes with untested/development kernels.
>
> On a similar note, bcm43xx support is significantly better for the 4311
> series cards in Feisty, and you probably will find yourself content with
> that as opposed to ndiswrapper...
>
> --
> BCM1390M (broadcom 4311) with ndiswrapper and nvidia IRQ conflicts
> https://launchpad.net/bugs/57355
>

Hello hello,

So this did work with the ndiswrapper. Thanks everyone for your help. I feel like a big idiot because my problem was that my wireless adapter was shutoff (function F2 must have been hit). Anyway works like a charm, thanks all.

Stefan

I'm not sure if this is related or not, but would someone be able to look at the following crash I get about 50% of the time when I boot my Dell D620 with Ubuntu Feisty 64-BIT, Ndiswrapper and the Dell WLAN 1490. I notice that during boot, if the WiFI LED on the laptop doesn't light up, this error occurs. If I reboot, the LED will light up and everything is fine. Not sure what is causing it. Only thing I can think of is maybe the Dell driver I'm using with NDISWRAPPER doesn't like 64-bit OSes

Here's the guts:

May 19 08:44:10 abray-laptop kernel: [ 24.738007] ndiswrapper version 1.43 loaded (smp=yes)
May 19 08:44:10 abray-laptop kernel: [ 24.816669] ndiswrapper (link_pe_images:577): fixing KI_USER_SHARED_DATA address in the driver
May 19 08:44:10 abray-laptop kernel: [ 24.818264] ndiswrapper: driver bcmwl5 (Broadcom,10/12/2006, 4.100.15.5) loaded
May 19 08:44:10 abray-laptop kernel: [ 24.818537] ACPI: PCI Interrupt 0000:0c:00.0[A] -> GSI 17 (level, low) -> IRQ 17
May 19 08:44:10 abray-laptop kernel: [ 24.826218] PGD 7c832067 PUD 7c833067 PMD 7c29c067 PTE 0
May 19 08:44:10 abray-laptop kernel: [ 24.826224] CPU 0
May 19 08:44:10 abray-laptop kernel: [ 24.826226] Modules linked in: ndiswrapper parport_pc lp parport fuse joydev snd_hda_intel snd_hda_codec snd_pcm_oss snd_mixer_oss snd_pc
m snd_seq_dummy snd_seq_oss snd_seq_midi snd_rawmidi pcmcia snd_seq_midi_event nvidia(P) af_packet snd_seq hci_usb i2c_core snd_timer snd_seq_device bluetooth serio_raw yenta_so
cket rsrc_nonstatic pcmcia_core snd soundcore snd_page_alloc iTCO_wdt iTCO_vendor_support shpchp psmouse pci_hotplug intel_agp evdev tsdev ext3 jbd mbcache sg sr_mod cdrom sd_mo
d generic ata_generic tg3 ata_piix libata scsi_mod ehci_hcd uhci_hcd usbcore thermal processor fan fbcon tileblit font bitblit softcursor vesafb cfbcopyarea cfbimgblt cfbfillrec
t capability commoncap
May 19 08:44:10 abray-laptop kernel: [ 24.826262] Pid: 4302, comm: modprobe Tainted: P 2.6.20-15-generic #2
May 19 08:44:10 abray-laptop kernel: [ 24.826265] RIP: 0010:[phys_startup_64+754183/2147483392] [phys_startup_64+754183/2147483392]
May 19 08:44:10 abray-laptop kernel: [ 24.826268] RSP: 0018:ffff8100762df6e0 EFLAGS: 00010286
May 19 08:44:10 abray-laptop kernel: [ 24.826271] RAX: ffffc20000323814 RBX: ffffc20000057000 RCX: ffffc2000004d000
May 19 08:44:10 abray-laptop kernel: [ 24.826273] RDX: 00000000fffffffb RSI: 0000000000070000 RDI: ffffc2000004d000
May 19 08:44:10 abray-laptop kernel: [ 24.826276] RBP: ffff81007d13fa40 R08: 0000000000000000 R09: ffffc20000048000
May 19 08:44:10 abray-laptop kernel: [ 24.826278] R10: ffffffff88a4da00 R11: 00000000000000a5 R12: 0000000000000000
May 19 08:44:10 abray-laptop kernel: [ 24.826280] R13: ffffc2000004d000 R14: ffffc20000323434 R15: 0000000000000001
May 19 08:44:10 abray-laptop kernel: [ 24.826283] FS: 00002af3078856f0(0000) GS:ffffffff8054e000(0000) knlGS:0000000000000000
May 19 08:44:10 abray-laptop kernel: [ 24.826286] CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
May 19 08:44:10 abray-laptop kernel: [ 24.826288] CR2: ffffc20000323814 CR3: 0000000074984000 CR4: 00000000000006e0
May 19 08:44:10 abray-laptop kernel: [ 24.826291] Process modprobe (pid: 4302,...

FYI,

Just discovered NDISWRAPPER v1.44 was recently released and resolves an issue with 64-bit systems with a gig or more of RAM and Broadcom adapters. That matches perfectly! :P

Just updated and so far it worked - only booted once.

Just to close the loop, it looks like Ndiswrapper Version 1.44 corrected my issue.