freebsd stable - May 2006 - CFT: new ath hal (take 2)

On Saturday 06 May 2006 21:17, Sam Leffler wrote:
> A new test snapshot is available at:
>
> http://people.freebsd.org/~sam/ath_hal-20060506.tgz
>
> This is 0.9.17.0 and is known to work on sparc (tested) and may well
> work on alpha (untested but likely the same issue that broke sparc).
>
> This code should be a drop-in replacement for what's in cvs in HEAD but
> be sure to rebuild both the driver and rate control module as there are
> ABI changes.  Note that to use this hal with RELENG_6 or other branches
> other than HEAD you must have rev 1.143 of if_ath.c.  To build a sparc
> module you must have rev 1.5 of modules/ath_hal/Makefile and rev 1.539
> of conf/options.
>
I attached a patch for if_ath.c on releng_6 
with it the hal should compile on i386 and amd64 fine

but at the end for my card, the DWL-AG530, nothing changed


ath_hal: 0.9.17.0 (AR5210, AR5211, AR5212, RF5111, RF5112, RF2413, RF5413)
ath0: <Atheros 5212> mem 0xfae00000-0xfae0ffff irq 16 at device 11.0 on
pci0
ath0: Ethernet address: 00:11:95:c2:2a:80
ath0: mac 5.9 phy 4.3 radio 3.6


setting to regdomain 16

Channel   1 : 2412  Mhz 11g          Channel  48 : 5240  Mhz 11a
Channel   2 : 2417  Mhz 11g          Channel  50 : 5250  Mhz 11a Turbo
Channel   3 : 2422  Mhz 11g          Channel  52 : 5260  Mhz 11a
Channel   4 : 2427  Mhz 11g          Channel  56 : 5280  Mhz 11a
Channel   5 : 2432  Mhz 11g          Channel  58 : 5290  Mhz 11a Turbo
Channel   6 : 2437  Mhz 11g          Channel  60 : 5300  Mhz 11a
Channel   7 : 2442  Mhz 11g          Channel  64 : 5320  Mhz 11a
Channel   8 : 2447  Mhz 11g          Channel 149 : 5745  Mhz 11a
Channel   9 : 2452  Mhz 11g          Channel 152 : 5760  Mhz 11a Turbo
Channel  10 : 2457  Mhz 11g          Channel 153 : 5765  Mhz 11a
Channel  11 : 2462  Mhz 11g          Channel 157 : 5785  Mhz 11a
Channel  36 : 5180  Mhz 11a          Channel 160 : 5800  Mhz 11a Turbo
Channel  40 : 5200  Mhz 11a          Channel 161 : 5805  Mhz 11a
Channel  42 : 5210  Mhz 11a Turbo    Channel 165 : 5825  Mhz 11a
Channel  44 : 5220  Mhz 11a


setting regdomain 18

Channel   1 : 2412  Mhz 11g          Channel 136 : 4942* Mhz 11a
Channel   2 : 2417  Mhz 11g          Channel 136 : 4945* Mhz 11a
Channel   3 : 2422  Mhz 11g          Channel 136 : 4950* Mhz 11a
Channel   4 : 2427  Mhz 11g          Channel 137 : 4955* Mhz 11a
Channel   5 : 2432  Mhz 11g          Channel 137 : 4960* Mhz 11a
Channel   6 : 2437  Mhz 11g          Channel 137 : 4965* Mhz 11a
Channel   7 : 2442  Mhz 11g          Channel 137 : 4970* Mhz 11a
Channel   8 : 2447  Mhz 11g          Channel 138 : 4975* Mhz 11a
Channel   9 : 2452  Mhz 11g          Channel 138 : 4980* Mhz 11a
Channel  10 : 2457  Mhz 11g          Channel 138 : 4985* Mhz 11a
Channel  11 : 2462  Mhz 11g



Jo?o

-------------- next part --------------
A non-text attachment was scrubbed...
Name: if_ath.patch
Type: text/x-diff
Size: 6408 bytes
Desc: not available
Url :
http://lists.freebsd.org/pipermail/freebsd-stable/attachments/20060508/057ba8b1/if_ath.bin

On Sunday 07 May 2006 21:39, JoaoBR wrote:

the former patch was the wrong one, sorry

Jo?o
-------------- next part --------------
A non-text attachment was scrubbed...
Name: if_ath.patch
Type: text/x-diff
Size: 5742 bytes
Desc: not available
Url :
http://lists.freebsd.org/pipermail/freebsd-stable/attachments/20060508/5c288db1/if_ath.bin

On Sunday 07 May 2006 22:09, Sam Leffler wrote:
> I don't know why you are posting patches like this; they include
changes
> that are not MFC'd for specific reasons.  I've already told you
> privately that the new hal has nothing to do with the issue of
> supporting the 1/2 and 1/4-rate channels in the public safety band (as
> enabled with regdomain 18).
>
> If people want to try the new hal on a branch other than HEAD they
> should follow the directions I gave in my original mail (modulo my
> obtuseness).  I did not tell people to take the driver off head or
> anything similar.  You need ONE change; the change in rev 1.143 of
> if_ath.c that explicitly casts the bus tag and bus handle parameters in
> the call to ath_hal_attach.  The main reason for the new hal is to fix
> sparc64 support (and possibly alpha).
>

you are really a piece of work my friend

first you give vague info and incorrect instructions and then you bark around 
when I try to help others which might have had the same dificulties to follow 
your brilliant mind

since you like to be specific, here it is:
> is known to work on sparc (tested) and may well work on alpha
does not mean at all that it is the "main reason" like you claim above

and what you sell here:
> Note that to use this hal with RELENG_6 or other branches  other than HEAD 
you must have rev 1.143 of if_ath.c

is incorrect or at least very very bad explained

and this way (my patch) using part of if_ath.c v.1.143, the 4.9Ghz channel 
mismatch (overlapping with 11g channels) is very well addressed 
and even if it has nothing to do with the 9.17 HAL it resolves for now the 
issue


and to answer your assumptions, I never said this is a MFC patch or whatever 
and is ment only to help whom may be interested to build the 9.17 HAL on 
releng_6

I hope you do not mean it like it sounds but FYI put perhaps you could 
overview your way of talking. Beeing the boss does not require beeing harsh.

Jo?o







A mensagem foi scaneada pelo sistema de e-mail e pode ser considerada segura.
Service fornecido pelo Datacenter Matik  https://datacenter.matik.com.br