please enable rt2800usb and disable rt3070sta in the imx51 kernel

Bug #423767 reported by Oliver Grawert
18
This bug affects 1 person
Affects Status Importance Assigned to Milestone
linux-firmware (Ubuntu)
Won't Fix
Medium
Bryan Wu
Karmic
Won't Fix
Medium
Bryan Wu
linux-fsl-imx51 (Ubuntu)
Fix Released
Low
Bryan Wu
Karmic
Fix Released
Low
Bryan Wu

Bug Description

there are a few wireless rt3070sta cards that work with babbage and lange boards, please enable the driver from staging so we can test it.

Tags: armel
Oliver Grawert (ogra)
Changed in linux-fsl-imx51 (Ubuntu):
assignee: nobody → Amit Kucheria (amitk)
importance: Undecided → Medium
milestone: none → karmic-alpha-6
tags: added: armel
Revision history for this message
Loïc Minier (lool) wrote :

This is CONFIG_RT3070 and in the fsl-imx51 branch of ubuntu-karmic.git, I see:
debian.fsl-imx51/config/config.common.ubuntu:CONFIG_STAGING=y
debian.master/config/config.common.ubuntu:CONFIG_RT3070=m
debian.master/config/config.common.ubuntu:CONFIG_STAGING=y
(Note this driver isn't listed at all in the imx51 config)

Changed in linux-fsl-imx51 (Ubuntu Karmic):
status: New → Triaged
Revision history for this message
Amit Kucheria (amitk) wrote :

From the Kconfig:
Symbol: RT3070 [=n]
 Prompt: Ralink 3070 wireless support
 Defined at drivers/staging/rt3070/Kconfig:1
 Depends on: STAGING && !STAGING_EXCLUDE_BUILD && USB && X86 && WLAN_80211

The "Depends on" lines makes this driver X86-only for now. I can try removing the x86 and compiling the driver.

Changed in linux-fsl-imx51 (Ubuntu Karmic):
importance: Medium → Low
Revision history for this message
Matt Sealey (mwsealey) wrote :

At Genesi we can confirm it works already on Babbage and Lange, it pops up as rt2800usb and connects to all wireless networks we've tested (b, g, n, mixed, unencrypted, WEP and WPA2).

The X86 dependency doesn't seem to cause a problem here...

Revision history for this message
Loïc Minier (lool) wrote :

Matt, which kernel are you using? This bug is about the Ubuntu kernels, I don't think the driver is enabled in imx51 kernels in karmic yet, nor do I think it was in jaunty. When enabled it triggers loads of dmesg output and doesn't work right in the 2.6.31 kernels we're using.

Revision history for this message
Matt Sealey (mwsealey) wrote :

Hi Loic,

2.6.31-100.7-imx51 with a changed config.. actually we bumped our ABI in-house just to be safe (to the wonderfully random yet oddly apt "631") to ensure that we do not clash with the generic Ubuntu kernel, bumped it to 2.6.31 mainline, added a lot of our own patches, but it is mostly 2.6.31-100.7. Adding the wireless card was a matter of hitting the space bar..

Revision history for this message
Oliver Grawert (ogra) wrote :

rt2800usb is a completely different driver from rt3070sta ... the former is the in-kernel driver maintained by kernel.org upstream, the latter is from the staging branch ...

as loic mentioned, the rt3070sta definately doesnt work and only spams dmesg atm (it also required a hack to even enable it since the config only allows it to enable it for i386 by default atm), but we had an explicit request to enable it for testing purposes (which is why we have this bug)

Revision history for this message
Matt Sealey (mwsealey) wrote :

rt2800usb works fine on the RT3070 though. I thought the whole point of rt2800usb was to get away from the godawful mess that is the Ralink mainline sta driver? In 2.6.31 with the support for RT3070 rolled into rt2800usb, rt3870sta as a driver is practically obselete, and given that rt2800usb works out of the box here (it took ZERO effort to click NetworkManager and connect) and rt3070sta does not (it basically doesn't work properly on ANY platform unless you work some arcane magic with touching oddly named config files in /etc/Wireless/RT2870/*, and doesn't work well with wpa_supplicant or with it's own encryption on my network)....

... the question is, why would anyone want to use the rt3070sta driver, then?

Revision history for this message
Loïc Minier (lool) wrote :

Ohhh ok thanks for clarification; didn't know rt2800usb has support for this chip.

Matt, the support for this driver seems to be turned on as a module already, debian.fsl-imx51/config/config.common.ubuntu has CONFIG_RT2800USB=m, which other config did you have to flip to get this driver working?

Amit, please disable CONFIG_RT3070 again; it seems there's a better driver available for the chip we care about.

summary: - please enable rt3070sta in the imx51 kernel
+ please enable rt2800usb in the imx51 kernel
Revision history for this message
Matt Sealey (mwsealey) wrote : Re: please enable rt2800usb in the imx51 kernel

Loic,

Indeed:

http://git.kernel.org/?p=linux/kernel/git/torvalds/linux-2.6.git;a=commitdiff;h=d53d9e67b55f6a9fc3f836c5c392eb41ce5676f4

Note the problems, but they do not seem to affect me here (hooray for wpa_supplicant!) - to get it working all you need is the firmware from the rt3070sta driver (at ralinktech.com.tw) in /lib/firmware.. which, on looking, has no package ready yet in Ubuntu. I would think this needs to be done for release. The license is identical to the rt73, rt2x00 firmwares already packaged.

Revision history for this message
Oliver Grawert (ogra) wrote :

it should be in the linux-firmware package (which the linux-imx51 metapackage depends on) if it is not yet, a new bug should be opened to add it.

Loïc Minier (lool)
affects: linux-fsl-imx51 (Ubuntu Karmic) → linux-firmware (Ubuntu Karmic)
Oliver Grawert (ogra)
Changed in linux-firmware (Ubuntu Karmic):
milestone: karmic-alpha-6 → ubuntu-9.10-beta
Amit Kucheria (amitk)
summary: - please enable rt2800usb in the imx51 kernel
+ please enable rt2800usb and disable rt3070sta in the imx51 kernel
Changed in linux-firmware (Ubuntu Karmic):
status: Triaged → Fix Committed
affects: linux-firmware (Ubuntu Karmic) → linux-fsl-imx51 (Ubuntu Karmic)
Revision history for this message
Loïc Minier (lool) wrote :

Ok to move the task back to linux-fsl-imx51 as we need to disable the old staging driver

Changed in linux-firmware (Ubuntu Karmic):
assignee: nobody → Amit Kucheria (amitk)
status: New → Triaged
Amit Kucheria (amitk)
Changed in linux-firmware (Ubuntu Karmic):
importance: Undecided → Medium
Revision history for this message
Loïc Minier (lool) wrote :
Revision history for this message
Amit Kucheria (amitk) wrote :

The firmware shipped in the driver conflicts with the one we already carry in the distro. Initiating contact with Ralink to find out if we can simply upgrade to the new firmware w/o regressions.

Steve Langasek (vorlon)
Changed in linux-fsl-imx51 (Ubuntu Karmic):
milestone: ubuntu-9.10-beta → ubuntu-9.10
Revision history for this message
Nicholas Roberts (nicholasdavidroberts) wrote :

Apologies for this interruption, but I have been trying to get a 'convert's' laptop working on Ubuntu for 3 weeks now. Everything is fine apart from WiFi. He had two cards (one USB and one PCMCIA), I won't bore you with the detail but they do not work with Linux.

To keep him interested (and myself sane) I bought a USB stick with a 148f:2070 chipset. rt2800usb loads but neither Network Manager no WiCd can connect it. I have tried it with Jaunty and Karmic... Jaunty ignores it, Karmic loads rt2800usb and neither work.

Seeing as WiFi is so important to connectivity, it is sad that Linux/Ubuntu is soooooo very complicated/impossible on that issue. Needless to say that Windoze$$$ XP worked with all three within 3 minutes of plugging the card/dongle in.

If ANYBODY can tell me how to get a 148f:2070 chipset connected with ANYTHING other than Window$ I will donate my body to them on my death... creepy eh?

Revision history for this message
Nicholas Roberts (nicholasdavidroberts) wrote :

By the way, if anyone wonders what my chipset is, it is listed under...

http://cateee.net/lkddb/web-lkddb/RT2800USB.html

Should work, but no... grrrrrrr!

Revision history for this message
Paul Larson (pwlars) wrote :

@Nicholas, this bug pertains only to the armel port of the kernel. If you are having this problem on x86 with the latest daily karmic image, please open a bug against linux (ubuntu).

Revision history for this message
Launchpad Janitor (janitor) wrote :
Download full text (10.7 KiB)

This bug was fixed in the package linux-fsl-imx51 - 2.6.31-103.11

---------------
linux-fsl-imx51 (2.6.31-103.11) karmic; urgency=low

  [ Amit Kucheria ]

  * [Config] Disable LMPCM_USB
  * [Config] Disable rt3070sta in staging
    - LP: #423767
  * [Config] Add asix to d-i modules
    - LP: #431283
  * [Config] Enable MXC_SAHARA
  * ARM: IMX51: Change default_bpp to 32
    - LP: #440104

  [ Andy Whitcroft ]

  * SAUCE: ubuntu-insert-changes -- follow abstracted debian
  * rebase to Ubuntu-2.6.31-12.39
  * rt3070sta has been disabled

  [ Tim Gardner ]

  * [Config] Added cpio as a build dependency.

  [ Upstream Kernel Changes ]

  * ABI

  [ Ubuntu: 2.6.31-12.39 ]

  * install the full changelog with the binary package
  * changelog -- explicitly note rebases and clean history
  * Revert "SAUCE: (drop after 2.6.31) Disable iwl-1000 series wireless
    adapters"
  * Revert "SAUCE: (drop after 2.6.31) eCryptfs: Prevent lower dentry from
    going negative during unlink"
  * Revert "SAUCE: (drop after 2.6.31) em28xx: ir-kbd-i2c init data needs a
    persistent object"
  * Revert "SAUCE: (drop after 2.6.31) saa7134: ir-kbd-i2c init data needs
    a persistent object"
  * Revert "drm/i915: Check that the relocation points to within the
    target"
  * Revert "drm/i915: fix tiling on IGDNG"
  * Revert "Driver core: add new device to bus's list before probing"
  * Revert "HID: completely remove apple mightymouse from blacklist"
  * Revert "KVM: x86: check for cr3 validity in ioctl_set_sregs"
    All reverts superseded by corresponding v2.6.31.2 patches
  * [Config] Add d101m_ucode.bin to d-i/firmware/nic-modules
    - LP: #439456
  * [Config] Enabled CONFIG_SND_HDA_HWDEP
    - LP: #443117
  * [Config] Add cpio as a build dependency.
  * [Config] Set default I/O scheduler back to CFQ for desktop flavours
    - LP: #381300
  * KVM: VMX: Fix EPT with WP bit change during paging
  * pata_amd: do not filter out valid modes in nv_mode_filter
  * p54usb: add Zcomax XG-705A usbid
  * x86: Increase MIN_GAP to include randomized stack
  * serial: bfin_5xx: fix building as module when early printk is enabled
  * USB: option.c Add support for ZTE AC2726 EVDO modem
  * USB: option: TELIT UC864G support
  * video: s3c_fb.c: fix build with CONFIG_HOTPLUG=n
  * kbuild: fix cc1 options check to ensure we do not use -fPIC when
    compiling
  * drivers/mfd/ab3100-core.c: fix powerpc build error
  * thinkpad-acpi: don't ask about brightness_mode for fw. 1V and 1R
  * ACPI: pci_slot.ko wants a 64-bit _SUN
  * fbcon: only unbind from console if successfully registered
  * kallsyms: fix segfault in prefix_underscores_count()
  * sisfb: change SiS_DDC_Port type to SISIOADDRESS
  * mmc_spi: fail gracefully if host or card do not support the switch
    command
  * alpha: AGP update (fixes compile failure)
  * fs: make sure data stored into inode is properly seen before unlocking
    new inode
  * eCryptfs: Handle unrecognized tag 3 cipher codes
  * eCryptfs: Check for O_RDONLY lower inodes when opening lower files
  * eCryptfs: Filename encryption only supports password auth tokens
  * eCryptfs: Validate global auth tok keys
  * eCryptfs: Prevent lower dentry from goin...

Changed in linux-fsl-imx51 (Ubuntu Karmic):
status: Fix Committed → Fix Released
Revision history for this message
Loïc Minier (lool) wrote :

Amit, did you get an answer from Ralink?

Revision history for this message
Amit Kucheria (amitk) wrote : Re: [Bug 423767] Re: please enable rt2800usb and disable rt3070sta in the imx51 kernel

On 09 Oct 09, Loïc Minier wrote:
> Amit, did you get an answer from Ralink?
>

Not really :-/

I am afraid there is not much I can do wrt to firmware since there is a
potential for regressing other ralink drivers on x86 by replacing this one
firmware.

Revision history for this message
Loïc Minier (lool) wrote :

@Amit: perhaps we can patch the firmware name in the ARM driver and include the firmware under this new name?

Revision history for this message
lesnoland (lesnoland) wrote :

@Nicholas

I think your device has the same chipset as mine. I managed to get it to work like this:

http://ubuntuforums.org/showthread.php?p=8071728

Revision history for this message
Loïc Minier (lool) wrote :

Any progress on this?

Revision history for this message
Loïc Minier (lool) wrote :

I'm dropping the release target, but am happy to see this as a SRU.

Changed in linux-firmware (Ubuntu Karmic):
status: Triaged → Won't Fix
Revision history for this message
Chase Douglas (chasedouglas) wrote :

I just took a look at this bug. Unfortunately, we're rather late in the Lucid cycle, so we likely won't update the firmware due to fear of regressions without good justification. If anyone feels it's justified to put this into Lucid still, please leave a comment. Otherwise, this will get deferred to the next release.

Thanks

Revision history for this message
Amit Kucheria (amitk) wrote :

Re-assigning to bryan

Changed in linux-firmware (Ubuntu):
assignee: Amit Kucheria (amitk) → Bryan Wu (cooloney)
Changed in linux-fsl-imx51 (Ubuntu Karmic):
assignee: Amit Kucheria (amitk) → Bryan Wu (cooloney)
milestone: ubuntu-9.10 → none
Changed in linux-fsl-imx51 (Ubuntu):
assignee: Amit Kucheria (amitk) → Bryan Wu (cooloney)
milestone: ubuntu-9.10 → none
Changed in linux-firmware (Ubuntu Karmic):
assignee: Amit Kucheria (amitk) → Bryan Wu (cooloney)
Revision history for this message
Chase Douglas (chasedouglas) wrote :

Bryan, Amit,

Is the firmware in Maverick's package up to date (i.e. can we close the development task)?

Thanks

Changed in linux-firmware (Ubuntu):
status: Triaged → Incomplete
Changed in linux-firmware (Ubuntu):
status: Incomplete → Won't Fix
To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.