Broadcom NetXtreme II (BCM5708) not detected by installer [karmic]

[Ronald McCollam]

Binary package hint: debian-installer

The karmic server install is not detecting/configuring BCM5708 network cards. This is occurring on multiple machines with cards with this chip on both amd64 and i386.

Performing 'modprobe bnx2' returns no errors but does not cause the device to be configured.

Output of lspci:

00:00.0 Host bridge: Intel Corporation 5000X Chipset Memory Controller Hub (rev)
00:02.0 PCI bridge: Intel Corporation 5000 Series Chipset PCI Express x4 Port 2)
00:03.0 PCI bridge: Intel Corporation 5000 Series Chipset PCI Express x4 Port 3)
00:04.0 PCI bridge: Intel Corporation 5000 Series Chipset PCI Express x8 Port 4)
00:05.0 PCI bridge: Intel Corporation 5000 Series Chipset PCI Express x4 Port 5)
00:06.0 PCI bridge: Intel Corporation 5000 Series Chipset PCI Express x8 Port 6)
00:07.0 PCI bridge: Intel Corporation 5000 Series Chipset PCI Express x4 Port 7)
00:10.0 Host bridge: Intel Corporation 5000 Series Chipset FSB Registers (rev 1)
00:10.1 Host bridge: Intel Corporation 5000 Series Chipset FSB Registers (rev 1)
00:10.2 Host bridge: Intel Corporation 5000 Series Chipset FSB Registers (rev 1)
00:11.0 Host bridge: Intel Corporation 5000 Series Chipset Reserved Registers ()
00:13.0 Host bridge: Intel Corporation 5000 Series Chipset Reserved Registers ()
00:15.0 Host bridge: Intel Corporation 5000 Series Chipset FBD Registers (rev 1)
00:16.0 Host bridge: Intel Corporation 5000 Series Chipset FBD Registers (rev 1)
00:1c.0 PCI bridge: Intel Corporation 631xESB/632xESB/3100 Chipset PCI Express )
00:1d.0 USB Controller: Intel Corporation 631xESB/632xESB/3100 Chipset UHCI USB)
00:1d.1 USB Controller: Intel Corporation 631xESB/632xESB/3100 Chipset UHCI USB)
00:1d.2 USB Controller: Intel Corporation 631xESB/632xESB/3100 Chipset UHCI USB)
00:1d.7 USB Controller: Intel Corporation 631xESB/632xESB/3100 Chipset EHCI USB)
00:1e.0 PCI bridge: Intel Corporation 82801 PCI Bridge (rev d9)
00:1f.0 ISA bridge: Intel Corporation 631xESB/632xESB/3100 Chipset LPC Interfac)
00:1f.1 IDE interface: Intel Corporation 631xESB/632xESB IDE Controller (rev 09)
01:00.0 PCI bridge: Intel Corporation 80333 Segment-A PCI Express-to-PCI Exprese
01:00.2 PCI bridge: Intel Corporation 80333 Segment-B PCI Express-to-PCI Exprese
02:0e.0 RAID bus controller: Dell PowerEdge Expandable RAID controller 5
04:00.0 PCI bridge: Broadcom EPB PCI-Express to PCI-X Bridge (rev c3)
05:00.0 Ethernet controller: Broadcom Corporation NetXtreme II BCM5708 Gigabit )
06:00.0 PCI bridge: Intel Corporation 6311ESB/6321ESB PCI Express Upstream Port)
06:00.3 PCI bridge: Intel Corporation 6311ESB/6321ESB PCI Express to PCI-X Brid)
07:00.0 PCI bridge: Intel Corporation 6311ESB/6321ESB PCI Express Downstream Po)
07:01.0 PCI bridge: Intel Corporation 6311ESB/6321ESB PCI Express Downstream Po)
08:00.0 PCI bridge: Broadcom EPB PCI-Express to PCI-X Bridge (rev c3)
09:00.0 Ethernet controller: Broadcom Corporation NetXtreme II BCM5708 Gigabit )
10:0d.0 VGA compatible controller: ATI Technologies Inc ES1000 (rev 02)

[Colin Watson (cjwatson)]

The kernel packaging needs to deliver the bnx2 firmware to the installer as well as the module.

[Colin Watson (cjwatson)]

Patch series which should make this work:

  https://lists.ubuntu.com/archives/kernel-team/2009-June/005963.html
  https://lists.ubuntu.com/archives/kernel-team/2009-June/005961.html
  https://lists.ubuntu.com/archives/kernel-team/2009-June/005964.html
  https://lists.ubuntu.com/archives/kernel-team/2009-June/005962.html

[Pablo Marchant (pamarca)]

After an updrage from Jaunty to karmic, my broadcom BCM4311 is not seen anymore. Is this due to the same issue??

Im using ndiswrapper right now, and I was wondering if this fix would allow me to use the regular drivers for my wireless. Is it already on the repositories?

[Colin Watson (cjwatson)]

On Wed, Jun 17, 2009 at 02:49:34PM -0000, Pablo Marchant wrote:
> After an updrage from Jaunty to karmic, my broadcom BCM4311 is not seen
> anymore. Is this due to the same issue??

Firstly, this bug is specifically and only about the installer. Any
issue with an installed system is a different bug.

Secondly, BCM4311 is a wireless device, while BCM5708 (the subject of
this report) is a gigabit Ethernet (wired) device.

In short, your problem is entirely unrelated to this one - I suggest you
file it separately.

[Launchpad Janitor (janitor)]

This bug was fixed in the package linux - 2.6.30-10.11

---------------
linux (2.6.30-10.11) karmic; urgency=low

  [ Amit Kucheria ]

  * [Config] Comment splitconfig.pl and misc cleanup
  * [Config] Rename all configs to the new naming scheme
  * [Config] Splitconfig rework
  * [Config] Rename scripts/misc/oldconfig to kernelconfig
  * [Config] Fix build system for new config split
  * [Config] Run updateconfigs after the splitconfig rework

  [ Andy Whitcroft ]

  * Revert "SAUCE: Default to i915.modeset=0 if CONFIG_DRM_I915_KMS=y"
  * [Config] standardise CONFIG_STAGING=y
  * [Config] standardise CONFIG_RD_LZMA=y
  * [Config] CONFIG_PCI_IOV=y
  * [Config] CONFIG_PCI_STUB=m
  * [Config] merge kernel configs more agressively

  [ Colin Watson ]

  * [Config] Run kernel-wedge in $(builddir) rather than at the top level
  * [Config] Add support for including firmware in udebs
  * [Config] Ship bnx2 firmware in nic-modules udeb
    - LP: #384861

  [ Luke Yelavich ]

  * [Config] ports - Import of ports architectures into kernel packaging
    infrastructure
  * [Config] ports - Do not update ports kernel configurations by default
  * [Config] ports - Disable ABI checking for ports architectures
  * [Config] ports - Build drivers in ubuntu sub-directory on powerpc
  * [Config] ports - Add control.d/vars.* files for ports architectures
  * [Config] ports - Add ports architectures for linux-libc-dev
  * [Config] ports - Create powerpc specific message-modules and
    block-modules udebs
  * [Config] ports - Add configuration files for ports architectures

  [ Manoj Iyer ]

  * [Config] Enable CONFIG_BLK_DEV_AEC62XX=m for amd64 and i386
    - LP: #329864

  [ Michael Casadevall ]

  * [Config] ports - Fix compression of kernels

  [ Stefan Bader ]

  * [Upstream] mmc: prevent dangling block device from accessing stale
    queues
    - LP: #383668

  [ Tim Gardner ]

  * [Config] Recommend grub-pc in linux-image
    - LP: #385741
  * [Config] Implement i386 generic and generic-pae flavours
  * [Config] ports - Add control info after integrating ports arches
  * [Config] Removed auto-generated files from git
  * [Config] Added netxen_nic to nic-modules
    - LP: #389603

  [ Matt Zimmerman ]

  * Rename linux-doc-PKGVER to linux-doc and clean up its description
    - LP: #382115

 -- Tim Gardner <email address hidden> Mon, 15 Jun 2009 14:38:26 -0600

[Ronald McCollam (fader)]

Still seeing this behavior. Colin Watson asked to see the syslog from a machine that is exhibiting this bug:

Jul 1 11:46:08 syslogd started: BusyBox v1.10.2
Jul 1 11:46:08 kernel: klogd started: BusyBox v1.10.2 (Ubuntu 1:1.10.2-2ubuntu7)
Jul 1 11:46:08 kernel: [ 0.000000] Initializing cgroup subsys cpuset
Jul 1 11:46:08 kernel: [ 0.000000] Initializing cgroup subsys cpu
Jul 1 11:46:08 kernel: [ 0.000000] Linux version 2.6.30-9-generic (buildd@crested) (gcc version 4.4.0 (Ubuntu 4.4.0-5ubuntu1) ) #10-Ubuntu SMP Fri Jun 12 13:08:18 UTC 2009 (Ubuntu 2.6.30-9.10-generic)
Jul 1 11:46:08 kernel: [ 0.000000] Command line: vga=normal initrd=initrd.gz DEBCONF_DEBUG=5 console=ttyS0,9600 debconf/priority=critical url=http://10.189.90.1/cgi-bin/preseed.cgi locale=en_US console-setup/ask_detect=false console-setup/layoutco
Jul 1 11:46:08 kernel: [ 0.000000] KERNEL supported cpus:
Jul 1 11:46:08 kernel: [ 0.000000] Intel GenuineIntel
Jul 1 11:46:08 kernel: [ 0.000000] AMD AuthenticAMD
Jul 1 11:46:08 kernel: [ 0.000000] Centaur CentaurHauls
Jul 1 11:46:08 kernel: [ 0.000000] BIOS-provided physical RAM map:
Jul 1 11:46:08 kernel: [ 0.000000] BIOS-e820: 0000000000000000 - 00000000000a0000 (usable)
Jul 1 11:46:08 kernel: [ 0.000000] BIOS-e820: 0000000000100000 - 00000000dfed2000 (usable)
Jul 1 11:46:08 kernel: [ 0.000000] BIOS-e820: 00000000dfed2000 - 00000000dfee8000 (reserved)
Jul 1 11:46:08 kernel: [ 0.000000] BIOS-e820: 00000000dfee8000 - 00000000dff07c00 (ACPI data)
Jul 1 11:46:08 kernel: [ 0.000000] BIOS-e820: 00000000dff07c00 - 00000000e0000000 (reserved)
Jul 1 11:46:08 kernel: [ 0.000000] BIOS-e820: 00000000f0000000 - 00000000f8000000 (reserved)
Jul 1 11:46:08 kernel: [ 0.000000] BIOS-e820: 00000000fe000000 - 0000000100000000 (reserved)
Jul 1 11:46:08 kernel: [ 0.000000] BIOS-e820: 0000000100000000 - 0000000120000000 (usable)
Jul 1 11:46:08 kernel: [ 0.000000] DMI 2.4 present.
Jul 1 11:46:08 kernel: [ 0.000000] last_pfn = 0x120000 max_arch_pfn = 0x100000000
Jul 1 11:46:08 kernel: [ 0.000000] MTRR default type: uncachable
Jul 1 11:46:08 kernel: [ 0.000000] MTRR fixed ranges enabled:
Jul 1 11:46:08 kernel: [ 0.000000] 00000-9FFFF write-back
Jul 1 11:46:08 kernel: [ 0.000000] A0000-BFFFF uncachable
Jul 1 11:46:08 kernel: [ 0.000000] C0000-D3FFF write-protect
Jul 1 11:46:08 kernel: [ 0.000000] D4000-EBFFF uncachable
Jul 1 11:46:08 kernel: [ 0.000000] EC000-EFFFF write-back
Jul 1 11:46:08 kernel: [ 0.000000] F0000-FFFFF write-protect
Jul 1 11:46:08 kernel: [ 0.000000] MTRR variable ranges enabled:
Jul 1 11:46:08 kernel: [ 0.000000] 0 base 0000000000 mask FF80000000 write-back
Jul 1 11:46:08 kernel: [ 0.000000] 1 base 0080000000 mask FFC0000000 write-back
Jul 1 11:46:08 kernel: [ 0.000000] 2 base 00C0000000 mask FFE0000000 write-back
Jul 1 11:46:08 kernel: [ 0.000000] 3 base 0100000000 mask FFE0000000 write-back
Jul 1 11:46:08 kernel: [ 0.000000] 4 disabled
Jul 1 11:46:08 kernel: [ 0.000000] 5 disabled
Jul 1 11:46:08 kernel: [ 0.000000] 6 disabled
Jul 1 11:46:08 kernel: [ 0...

[Colin Watson (cjwatson)]

This appears to be because the d-i netboot environment in use on the certification server is out of date. Ronald is investigating.

[Colin Watson (cjwatson)]

On IRC, Ronald confirmed this to be fixed with a newer image.