Current linus git bcm4xxx broken for me

Current linus git bcm4xxx broken for me

[Benjamin Herrenschmidt]

I get that with current upstream git :

[   34.127738] bcm43xx: Chip ID 0x4318, rev 0x2
[   34.136739] bcm43xx: Number of cores: 4
[   34.145660] bcm43xx: Core 0: ID 0x800, rev 0xd, vendor 0x4243, enabled
[   34.154658] bcm43xx: Core 1: ID 0x812, rev 0x9, vendor 0x4243, disabled
[   34.163560] bcm43xx: Core 2: ID 0x804, rev 0xc, vendor 0x4243, enabled
[   34.172422] bcm43xx: Core 3: ID 0x80d, rev 0x7, vendor 0x4243, enabled
[   34.184508] bcm43xx: PHY connected
[   34.193390] bcm43xx: Detected PHY: Version: 3, Type 2, Revision 7
[   34.202256] bcm43xx: Detected Radio: ID: 8205017f (Manuf: 17f Ver: 2050 Rev: 8)
[   34.211067] bcm43xx: Radio turned off
[   34.219814] bcm43xx: Radio turned off
[   35.158260] bcm43xx: PHY connected
[   35.315532] bcm43xx: Radio turned on
[   35.340646] bcm43xx: ASSERTION FAILED (radio_attenuation < 10) at: drivers/net/wireless/bcm43xx/bcm43xx_phy.c:1485:bcm43xx_find_lopair()
[   35.350611] bcm43xx: WARNING: Writing invalid LOpair (low: 58, high: 32, index: 130)
[   35.360642] Call Trace:
[   35.370672] [8F8AFBE0] [800080AC] show_stack+0x5c/0x1a0 (unreliable)
[   35.380875] [8F8AFC10] [C24D394C] bcm43xx_phy_lo_adjust+0xcc/0x390 [bcm43xx]
[   35.391054] [8F8AFC30] [C24CA19C] bcm43xx_radio_set_txpower_bg+0x18c/0x310 [bcm43xx]
[   35.401223] [8F8AFC50] [C24D2314] bcm43xx_phy_initb6+0x264/0x920 [bcm43xx]
[   35.411399] [8F8AFC70] [C24D4BD8] bcm43xx_phy_initg+0x428/0xff0 [bcm43xx]
[   35.421520] [8F8AFCE0] [C24D5864] bcm43xx_phy_init+0xc4/0x7b0 [bcm43xx]
[   35.431634] [8F8AFD00] [C24C4554] bcm43xx_wireless_core_init+0xaa4/0x12c0 [bcm43xx]
[   35.441842] [8F8AFDA0] [C24C5FA0] bcm43xx_init_board+0x300/0x690 [bcm43xx]
[   35.452092] [8F8AFDF0] [8027C3D8] dev_open+0x88/0xe0
[   35.462274] [8F8AFE10] [8027B124] dev_change_flags+0x144/0x180
[   35.472505] [8F8AFE30] [802C4D98] devinet_ioctl+0x628/0x760
[   35.482661] [8F8AFEA0] [802C5174] inet_ioctl+0xb4/0xe0
[   35.492743] [8F8AFEB0] [8026E28C] sock_ioctl+0xfc/0x270
[   35.502832] [8F8AFED0] [80094B7C] do_ioctl+0x3c/0x90
[   35.512837] [8F8AFEE0] [80094C5C] vfs_ioctl+0x8c/0x460
[   35.522718] [8F8AFF10] [80095070] sys_ioctl+0x40/0x80
[   35.532524] [8F8AFF40] [8000FBB4] ret_from_syscall+0x0/0x38

The error (with backtrace) is repeated every second or so. The card doesn't appear to associate.
It used to work with real old versions of the driver (though I never had it working above 11Mbits),
and I haven't tried again in the past few monthes mostly because I was too busy and couldn't be
bothered to try the new git based stuff.

Ben.

Re: Current linus git bcm4xxx broken for me

[Benjamin Herrenschmidt]

On Fri, 2006-04-14 at 10:02 +1000, Benjamin Herrenschmidt wrote:
> I get that with current upstream git :

  .../...

Correction: It actually works after tickling softmac a bit (it's an open
network with non advertised essid, the init scripts are supposedly
setting the essid but softmac is having some vapors, changing the essid
manually a couple of times and it ends up associating).

Still, the dmesg filling up with those messages is pretty nasty :)

Ben.

Re: Current linus git bcm4xxx broken for me

[Michael Buesch]

On Friday 14 April 2006 02:02, Benjamin Herrenschmidt wrote:
> I get that with current upstream git :
> 
> [   34.127738] bcm43xx: Chip ID 0x4318, rev 0x2
                                  ^^^^^^
                                  known broken
We work at it.

-- 
Greetings Michael.

Re: Current linus git bcm4xxx broken for me

[Johannes Berg]

[-- Attachment #1: Type: text/plain, Size: 780 bytes --]

On Fri, 2006-04-14 at 10:35 +1000, Benjamin Herrenschmidt wrote:
> On Fri, 2006-04-14 at 10:02 +1000, Benjamin Herrenschmidt wrote:
> > I get that with current upstream git :
> 
>   .../...
> 
> Correction: It actually works after tickling softmac a bit (it's an open
> network with non advertised essid, the init scripts are supposedly
> setting the essid but softmac is having some vapors, changing the essid
> manually a couple of times and it ends up associating).

Below ...

> Still, the dmesg filling up with those messages is pretty nasty :)

There are two issues here. One is the messages (related to it being a
0x4318) and secondly that softmac isn't behaving will :)
I'm trying to fix the latter but can't really pinpoint the problem yet.

johannes

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 793 bytes --]