I've been poking around in b43 and brcmsmac trying to wrap my head
around device detection. Obviously, b43 and brcmsmac both support
some devices, and since 4f3d09d (b43: add option to avoid duplicating
device support with brcmsmac, 2012-01-11) overlapping devices have
been masked behind CONFIG_B43_BCMA_EXTRA. At the time of that commit,
brcmsmac didn't support the r17 cores, but support has since been
added in 6f80f01 (brcmsmac: add support for cores with revision 17,
2012-12-07). Does that mean that b43 should move the r17 core entry
into the masked-by-CONFIG_B43_BCMA_EXTRA block?
Cheers,
Trevor
--
This email may be signed or encrypted with GnuPG (http://www.gnupg.org).
For more information, see http://en.wikipedia.org/wiki/Pretty_Good_Privacy
2013/2/21 W. Trevor King <[email protected]>:
> I've been poking around in b43 and brcmsmac trying to wrap my head
> around device detection. Obviously, b43 and brcmsmac both support
> some devices, and since 4f3d09d (b43: add option to avoid duplicating
> device support with brcmsmac, 2012-01-11) overlapping devices have
> been masked behind CONFIG_B43_BCMA_EXTRA. At the time of that commit,
> brcmsmac didn't support the r17 cores, but support has since been
> added in 6f80f01 (brcmsmac: add support for cores with revision 17,
> 2012-12-07). Does that mean that b43 should move the r17 core entry
> into the masked-by-CONFIG_B43_BCMA_EXTRA block?
I dislike CONFIG_B43_BCMA_EXTRA from the beginning, it should be
selectable as kernel-option according to me. Forcing user to recompile
the kernel, just to test a driver, sound like a really poor
user-friendliness to me.
--
Rafał
On Thu, Feb 21, 2013 at 02:07:23PM +0100, Hauke Mehrtens wrote:
> On 02/21/2013 01:17 PM, W. Trevor King wrote:
> > I've been poking around in b43 and brcmsmac trying to wrap my head
> > around device detection. Obviously, b43 and brcmsmac both support
> > some devices, and since 4f3d09d (b43: add option to avoid
> > duplicating device support with brcmsmac, 2012-01-11) overlapping
> > devices have been masked behind CONFIG_B43_BCMA_EXTRA. At the time
> > of that commit, brcmsmac didn't support the r17 cores, but support
> > has since been added in 6f80f01 (brcmsmac: add support for cores
> > with revision 17, 2012-12-07). Does that mean that b43 should move
> > the r17 core entry into the masked-by-CONFIG_B43_BCMA_EXTRA block?
>
> Yes you could move that core id into the CONFIG_B43_BCMA_EXTRA block,
> but this core was just found on Router SoCs and there you mostly want
> to use AP mode, which brcmsmac does not support.
So maybe a CONFIG_BRCM_BCMA_EXTRA to mask brcmsmac devices when b43 is
the recommended module?
--
This email may be signed or encrypted with GnuPG (http://www.gnupg.org).
For more information, see http://en.wikipedia.org/wiki/Pretty_Good_Privacy
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 02/21/2013 02:46 PM, W. Trevor King wrote:
> On Thu, Feb 21, 2013 at 02:07:23PM +0100, Hauke Mehrtens wrote:
>> On 02/21/2013 01:17 PM, W. Trevor King wrote:
>>> I've been poking around in b43 and brcmsmac trying to wrap my
>>> head around device detection. Obviously, b43 and brcmsmac both
>>> support some devices, and since 4f3d09d (b43: add option to
>>> avoid duplicating device support with brcmsmac, 2012-01-11)
>>> overlapping devices have been masked behind
>>> CONFIG_B43_BCMA_EXTRA. At the time of that commit, brcmsmac
>>> didn't support the r17 cores, but support has since been added
>>> in 6f80f01 (brcmsmac: add support for cores with revision 17,
>>> 2012-12-07). Does that mean that b43 should move the r17 core
>>> entry into the masked-by-CONFIG_B43_BCMA_EXTRA block?
>>
>> Yes you could move that core id into the CONFIG_B43_BCMA_EXTRA
>> block, but this core was just found on Router SoCs and there you
>> mostly want to use AP mode, which brcmsmac does not support.
>
> So maybe a CONFIG_BRCM_BCMA_EXTRA to mask brcmsmac devices when b43
> is the recommended module?
>
I would like to stay with the current situation. This
CONFIG_B43_BCMA_EXTRA is useful for distributions like Ubuntu where
you have one image used on many different devices and you are shipping
mostly every kernel model. No one used distributions like Ubuntu on
the BCM4716 where this core rev 17 wifi core is found. There you just
build an image with the modules you need so you end up with b43 or
brcmsmac in your image.
Hauke
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with undefined - http://www.enigmail.net/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=bdvI
-----END PGP SIGNATURE-----
-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
On 02/21/2013 01:17 PM, W. Trevor King wrote:
> I've been poking around in b43 and brcmsmac trying to wrap my head
> around device detection. Obviously, b43 and brcmsmac both support
> some devices, and since 4f3d09d (b43: add option to avoid
> duplicating device support with brcmsmac, 2012-01-11) overlapping
> devices have been masked behind CONFIG_B43_BCMA_EXTRA. At the time
> of that commit, brcmsmac didn't support the r17 cores, but support
> has since been added in 6f80f01 (brcmsmac: add support for cores
> with revision 17, 2012-12-07). Does that mean that b43 should move
> the r17 core entry into the masked-by-CONFIG_B43_BCMA_EXTRA block?
Yes you could move that core id into the CONFIG_B43_BCMA_EXTRA block,
but this core was just found on Router SoCs and there you mostly want
to use AP mode, which brcmsmac does not support.
Hauke
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.11 (GNU/Linux)
Comment: Using GnuPG with undefined - http://www.enigmail.net/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=EQlO
-----END PGP SIGNATURE-----