Date: Tue, 02 Mar 2010 19:57:26 +0100 From: =?ISO-8859-1?Q?Gustau_P=E9rez?= <gperez@entel.upc.edu> To: John Baldwin <jhb@freebsd.org> Cc: freebsd-current@freebsd.org, weongyo.jeong@gmail.com Subject: Re: Call for Test and Review: bwn(4) - another Broadcom Wireless driver Message-ID: <4B8D5F96.3010700@entel.upc.edu> In-Reply-To: <201003020801.32955.jhb@freebsd.org> References: <20091223035331.GA1293@weongyo> <20100228074056.GA3536@weongyo> <4B8C220A.1060402@entel.upc.edu> <201003020801.32955.jhb@freebsd.org>
next in thread | previous in thread | raw e-mail | index | archive | help
En/na John Baldwin ha escrit:
> On Monday 01 March 2010 3:22:34 pm Gustau Pérez wrote:
>
>> En/na Weongyo Jeong ha escrit:
>>
>>> On Fri, Feb 26, 2010 at 04:21:06PM -0800, Xin LI wrote:
>>>
>>>
>>>> Hi, Weongyo,
>>>>
>>>> On 2010/02/25 16:51, Weongyo Jeong wrote:
>>>>
>>>>
>>>>> FYI bwn(4) driver is committed into FreeBSD tree. I think the driver
>>>>> supports your LP PHY device. After cvsup please try to rebuild siba_bwn
>>>>> and bwn modules.
>>>>>
>>>>> Could you please test with it? Please let me know and send me your
>>>>> full dmesg when you encounters the following problems:
>>>>>
>>>>> - if the driver doesn't work or is unstable.
>>>>> - if it prints debugging or verbose messages.
>>>>>
>>>>>
>>>> Great! Thanks for the work!
>>>>
>>>> Is it possible to MFC the work back to 8-STABLE at some point?
>>>>
>>>>
>>> Of course yes. AFAIK it could be compiled and works without problems on
>>> 8-STABLE; I checked it. :-)
>>>
>>>
>>>
>> I'm trying to run it in STABLE right now. When kldloading if_bwm
>> (following the given instructions, the other modules are kldloaded) it
>> complains with :
>>
>> link_elf_obj: symbol _mtx_assert undefined
>>
>> looks like it fails in if_bwnvar.h. Culprit is this define :
>>
>> #define BWN_ASSERT_LOCKED(sc)
>> mtx_assert(&(sc)->sc_mtx, MA_OWNED)
>>
>> Do I need witness enabled to run this ? Is there any way to run it in
>> STABLE or do I need to run CURRENT ?
>>
>
> It sounds like you have INVARIANTS defined when the module was built, but your
> kernel does not have INVARIANT_SUPPORT defined.
>
>
That did it. Thank you. After recompiling and rebooting the kernel, I
noticed via dmesg that when kldloading ssb the module complains with :
ssb0: <Broadcom BCM4312 802.11a/b/g Wireless> mem
0xf9ffc000-0xf9ffffff irq 17 at device 0.0 on pci12
ssb0: unsupportted coreid 0x817
When kldloading if_bwn (after bwn_v4_ucode) the machine freezes. I
guess my hard isn't supported yet.
Regards,
Gus
--
PGP KEY : http://www-entel.upc.edu/gus/gus.asc
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?4B8D5F96.3010700>