4331 support

4331 support

[james at jamesanslow.net]

Can anyone provide me with an update as to the current status of BCM  
4331 support? Is support going to be coming eventually? Is anyone  
working on it?

Many thanks

4331 support

[Larry Finger]

On 08/25/2011 05:38 PM, james at jamesanslow.net wrote:
> Can anyone provide me with an update as to the current status of BCM 4331
> support? Is support going to be coming eventually? Is anyone working on it?

If you pull the wireless-testing git tree, it is working, perhaps with minor 
changes. See a recent mail message:

========================================
Awesome guys,
thanks for the pointers, it worked! (i m sending this email using the b43 driver!)

I used the wireless-next git, and then additionally applied the "b43: HT-PHY: 
allow writing longer tables with a single call" patch.
(and the ones in Rafal's last mail)

Thanks for the great support and fast answers!
Congratulations on hacking a driver like this together, I am impressed!

Best regards,
Maxime

2011/8/22 Rafa? Mi?ecki <zajec5@gmail.com>

     Hi Maxime,

     2011/8/22 Maxime Vincent <maxime.vince@gmail.com>:
     > I'm sorry if this is a noobish question, or asking this kind of questions is
     > not done.
     > I'm not really familiar with dev mailing lists.
     > I'm new to kernel development, not new to linux.
     >
     > I was wondering what the status of support for the 4331 wifi chipset was?
     > As I can see there's already quite some code for the 4331 chipset, but I
     > can't really make up from that if it's supposed to work or not, or maybe
     > partially?
     >
     > I have compiled latest linux-next kernel including the b43 driver, extracted
     > firmware, but I can't seem to see an interface registering.
     > Only thing I can see in dmesg:
     > Broadcom 43xx driver loaded [ Features: PMNLS, Firmware-ID:  FW13 ]

     I'm going to clean HT-PHY (BCM4331) support in about a week, as David
     wrote, it should be quite stable.

     Most of the patches are already in wireless-next, I just need to test
     that tree, see if everything was sent&applied correctly and drop
     BROKEN from B43_PHY_HT.

     If you're "brave" enough, you can:
     1) Edit Kconfig and drop BROKEN from B43_PHY_HT
     2) Edit dma.h and define B43_DMA_RINGMEMSIZE to "2 * PAGE_SIZE"
     (or just wait a week).
===============================================


Larry

4331 support

[james at jamesanslow.net]

Hi Larry,

Thanks for getting back to me on this. I'm not overly experienced when  
it comes to driver development and so I may be best suited to waiting  
until this is made live. That is, unless compiling and patching  
experimental drivers is an easy thing to do (but from what I  
understand it's not?)

Kind regards,
James

Quoting Larry Finger <Larry.Finger@lwfinger.net>:

> On 08/25/2011 05:38 PM, james at jamesanslow.net wrote:
>> Can anyone provide me with an update as to the current status of BCM 4331
>> support? Is support going to be coming eventually? Is anyone working on it?
>
> If you pull the wireless-testing git tree, it is working, perhaps  
> with minor changes. See a recent mail message:
>
> ========================================
> Awesome guys,
> thanks for the pointers, it worked! (i m sending this email using  
> the b43 driver!)
>
> I used the wireless-next git, and then additionally applied the  
> "b43: HT-PHY: allow writing longer tables with a single call" patch.
> (and the ones in Rafal's last mail)
>
> Thanks for the great support and fast answers!
> Congratulations on hacking a driver like this together, I am impressed!
>
> Best regards,
> Maxime
>
> 2011/8/22 Rafa? Mi?ecki <zajec5@gmail.com>
>
>     Hi Maxime,
>
>     2011/8/22 Maxime Vincent <maxime.vince@gmail.com>:
>     > I'm sorry if this is a noobish question, or asking this kind  
> of questions is
>     > not done.
>     > I'm not really familiar with dev mailing lists.
>     > I'm new to kernel development, not new to linux.
>     >
>     > I was wondering what the status of support for the 4331 wifi  
> chipset was?
>     > As I can see there's already quite some code for the 4331  
> chipset, but I
>     > can't really make up from that if it's supposed to work or  
> not, or maybe
>     > partially?
>     >
>     > I have compiled latest linux-next kernel including the b43  
> driver, extracted
>     > firmware, but I can't seem to see an interface registering.
>     > Only thing I can see in dmesg:
>     > Broadcom 43xx driver loaded [ Features: PMNLS, Firmware-ID:  FW13 ]
>
>     I'm going to clean HT-PHY (BCM4331) support in about a week, as David
>     wrote, it should be quite stable.
>
>     Most of the patches are already in wireless-next, I just need to test
>     that tree, see if everything was sent&applied correctly and drop
>     BROKEN from B43_PHY_HT.
>
>     If you're "brave" enough, you can:
>     1) Edit Kconfig and drop BROKEN from B43_PHY_HT
>     2) Edit dma.h and define B43_DMA_RINGMEMSIZE to "2 * PAGE_SIZE"
>     (or just wait a week).
> ===============================================
>
>
> Larry
>

4331 support

[Larry Finger]

On 08/25/2011 06:27 PM, james at jamesanslow.net wrote:
> Hi Larry,
>
> Thanks for getting back to me on this. I'm not overly experienced when it comes
> to driver development and so I may be best suited to waiting until this is made
> live. That is, unless compiling and patching experimental drivers is an easy
> thing to do (but from what I understand it's not?)

If you want to wait, then it will be kernel 3.2 before this stuff is built in. 
FYI, that will ~4-5 months.

The material will be available in the compat-wireless package that may be 
available for your distro in a few weeks.

The difficulty in building the kernel depends on your distro. For openSUSE, it 
is pretty easy. For Debian-based distros, it is a bit more difficult; however, 
once you have all the tools installed, the commands are easy. It just takes a while.

Please do a "reply-to-all". I do not do private consulting unless you are paying.

Larry

4331 support

[james at jamesanslow.net]

Hi Larry,

I did reply-all. Please note the CC to the b43-dev mailing list in  
this and my previous mail.

Thanks for the guidance. I'll see how I go.

Kind regards

Quoting Larry Finger <Larry.Finger@lwfinger.net>:

> On 08/25/2011 06:27 PM, james at jamesanslow.net wrote:
>> Hi Larry,
>>
>> Thanks for getting back to me on this. I'm not overly experienced  
>> when it comes
>> to driver development and so I may be best suited to waiting until  
>> this is made
>> live. That is, unless compiling and patching experimental drivers is an easy
>> thing to do (but from what I understand it's not?)
>
> If you want to wait, then it will be kernel 3.2 before this stuff is  
> built in. FYI, that will ~4-5 months.
>
> The material will be available in the compat-wireless package that  
> may be available for your distro in a few weeks.
>
> The difficulty in building the kernel depends on your distro. For  
> openSUSE, it is pretty easy. For Debian-based distros, it is a bit  
> more difficult; however, once you have all the tools installed, the  
> commands are easy. It just takes a while.
>
> Please do a "reply-to-all". I do not do private consulting unless  
> you are paying.
>
> Larry
>
>

4331 support

[Stuart Longland]

On 08/26/11 09:27, james at jamesanslow.net wrote:
> Hi Larry,
> 
> Thanks for getting back to me on this. I'm not overly experienced when
> it comes to driver development and so I may be best suited to waiting
> until this is made live. That is, unless compiling and patching
> experimental drivers is an easy thing to do (but from what I understand
> it's not?)
> 
> Kind regards,
> James

Well, no more difficult than downloading a kernel source tree via git
and compiling it.  The wireless-testing tree is basically a branch of
the mainline Linux kernel with experimental versions of device drivers.

Failing that, maybe try switching to the latest stable 3.0.x release of
the kernel if you're not running that already.
-- 
Stuart Longland (aka Redhatter, VK4MSL)      .'''.
Gentoo Linux/MIPS Cobalt and Docs Developer  '.'` :
. . . . . . . . . . . . . . . . . . . . . .   .'.'
http://dev.gentoo.org/~redhatter             :.'

I haven't lost my mind...
  ...it's backed up on a tape somewhere.

4331 support

[Larry Finger]

On 08/25/2011 07:33 PM, Stuart Longland wrote:
>
> Well, no more difficult than downloading a kernel source tree via git
> and compiling it.  The wireless-testing tree is basically a branch of
> the mainline Linux kernel with experimental versions of device drivers.
>
> Failing that, maybe try switching to the latest stable 3.0.x release of
> the kernel if you're not running that already.

For the 4331, that will not help at all. The code that makes it function is 
very, very new.

Larry

4331 support

[James Anslow]

Hello all,

I followed the instructions as forwarded in this mail re Kconfig and
dma.h. However, when I try to modprobe my device is detected but I get:

  69.767145] b43-phy0: Broadcom 4331 WLAN found (core revision 29)
[   69.767602] b43-phy0 ERROR: FOUND UNSUPPORTED PHY (Analog 9, Type 7,
Revision 1)
[   69.767611] b43: probe of bcma0:0 failed with error -95


Can anyone offer assistance?

On 25/08/11 23:52, Larry Finger wrote:
> On 08/25/2011 05:38 PM, james at jamesanslow.net wrote:
>> Can anyone provide me with an update as to the current status of BCM
>> 4331
>> support? Is support going to be coming eventually? Is anyone working
>> on it?
>
> If you pull the wireless-testing git tree, it is working, perhaps with
> minor changes. See a recent mail message:
>
> ========================================
> Awesome guys,
> thanks for the pointers, it worked! (i m sending this email using the
> b43 driver!)
>
> I used the wireless-next git, and then additionally applied the "b43:
> HT-PHY: allow writing longer tables with a single call" patch.
> (and the ones in Rafal's last mail)
>
> Thanks for the great support and fast answers!
> Congratulations on hacking a driver like this together, I am impressed!
>
> Best regards,
> Maxime
>
> 2011/8/22 Rafa? Mi?ecki <zajec5@gmail.com>
>
>     Hi Maxime,
>
>     2011/8/22 Maxime Vincent <maxime.vince@gmail.com>:
>     > I'm sorry if this is a noobish question, or asking this kind of
> questions is
>     > not done.
>     > I'm not really familiar with dev mailing lists.
>     > I'm new to kernel development, not new to linux.
>     >
>     > I was wondering what the status of support for the 4331 wifi
> chipset was?
>     > As I can see there's already quite some code for the 4331
> chipset, but I
>     > can't really make up from that if it's supposed to work or not,
> or maybe
>     > partially?
>     >
>     > I have compiled latest linux-next kernel including the b43
> driver, extracted
>     > firmware, but I can't seem to see an interface registering.
>     > Only thing I can see in dmesg:
>     > Broadcom 43xx driver loaded [ Features: PMNLS, Firmware-ID:  FW13 ]
>
>     I'm going to clean HT-PHY (BCM4331) support in about a week, as David
>     wrote, it should be quite stable.
>
>     Most of the patches are already in wireless-next, I just need to test
>     that tree, see if everything was sent&applied correctly and drop
>     BROKEN from B43_PHY_HT.
>
>     If you're "brave" enough, you can:
>     1) Edit Kconfig and drop BROKEN from B43_PHY_HT
>     2) Edit dma.h and define B43_DMA_RINGMEMSIZE to "2 * PAGE_SIZE"
>     (or just wait a week).
> ===============================================
>
>
> Larry

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 900 bytes
Desc: OpenPGP digital signature
URL: <http://lists.infradead.org/pipermail/b43-dev/attachments/20110826/9c750a45/attachment.sig>

4331 support

[Larry Finger]

On 08/26/2011 10:29 AM, James Anslow wrote:
> Hello all,
>
> I followed the instructions as forwarded in this mail re Kconfig and
> dma.h. However, when I try to modprobe my device is detected but I get:
>
>    69.767145] b43-phy0: Broadcom 4331 WLAN found (core revision 29)
> [   69.767602] b43-phy0 ERROR: FOUND UNSUPPORTED PHY (Analog 9, Type 7,
> Revision 1)
> [   69.767611] b43: probe of bcma0:0 failed with error -95
>
>
> Can anyone offer assistance?

What kernel source did you use?

Larry

4331 support

[James Anslow]

I am using 2.6.38-11 if you mean linux kernel source.

On 26/08/11 16:35, Larry Finger wrote:
> On 08/26/2011 10:29 AM, James Anslow wrote:
>> Hello all,
>>
>> I followed the instructions as forwarded in this mail re Kconfig and
>> dma.h. However, when I try to modprobe my device is detected but I get:
>>
>>    69.767145] b43-phy0: Broadcom 4331 WLAN found (core revision 29)
>> [   69.767602] b43-phy0 ERROR: FOUND UNSUPPORTED PHY (Analog 9, Type 7,
>> Revision 1)
>> [   69.767611] b43: probe of bcma0:0 failed with error -95
>>
>>
>> Can anyone offer assistance?
>
> What kernel source did you use?
>
> Larry

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 900 bytes
Desc: OpenPGP digital signature
URL: <http://lists.infradead.org/pipermail/b43-dev/attachments/20110826/e21e7710/attachment-0001.sig>

4331 support

[Larry Finger]

On 08/26/2011 10:38 AM, James Anslow wrote:
> I am using 2.6.38-11 if you mean linux kernel source.

Yes, I mean the kernel source.

If you want to use the 4331, you will need the very latest kernel source from 
the wireless-testing git tree.


A. Because it messes up the way people read things.
Q. Why is top posting bad?

Larry

4331 support

[james at jamesanslow.net]

Thanks for the heads-up larry. Can you direct me towards where that's hosted?


------Original Message------
From: Larry Finger
Sender: Larry Finger
To: James Anslow
Cc: b43-dev at lists.infradead.org
Subject: Re: 4331 support
Sent: 26 Aug 2011 16:57

On 08/26/2011 10:38 AM, James Anslow wrote:
> I am using 2.6.38-11 if you mean linux kernel source.

Yes, I mean the kernel source.

If you want to use the 4331, you will need the very latest kernel source from 
the wireless-testing git tree.


A. Because it messes up the way people read things.
Q. Why is top posting bad?

Larry




Sent from my BlackBerry? wireless device

4331 support

[Larry Finger]

On 08/26/2011 11:13 AM, james at jamesanslow.net wrote:
> Thanks for the heads-up larry. Can you direct me towards where that's hosted?
>
>
> ------Original Message------
> From: Larry Finger
> Sender: Larry Finger
> To: James Anslow
> Cc: b43-dev at lists.infradead.org
> Subject: Re: 4331 support
> Sent: 26 Aug 2011 16:57
>
> On 08/26/2011 10:38 AM, James Anslow wrote:
>> I am using 2.6.38-11 if you mean linux kernel source.
>
> Yes, I mean the kernel source.
>
> If you want to use the 4331, you will need the very latest kernel source from
> the wireless-testing git tree.
>
>
> A. Because it messes up the way people read things.
> Q. Why is top posting bad?

You need to install git and do a

git clone 
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless-testing.git

The source will be in wireless-testing directory. To get future updates, cd to 
that directory and 'git pull'.

Obviously, my subtle A./Q. quote was not sufficient. Top posting in most mailing 
lists is discouraged. Look at where your question is in the flow of this 
posting. How is one to follow the reasoning?

Larry