arm/179532: wireless networking on ARM

Adrian Chadd adrian at freebsd.org
Fri Jun 14 08:22:40 UTC 2013


Yup, it's being handed an unaligned radiotap header address from the run driver.

Grr.



Adrian


On 14 June 2013 01:20, Guillaume Bibaut <yom at iaelu.net> wrote:
> Here is the output :
>
> set_channel: c=0xc289b214, p=0xc2892fbb, rc=0xc2892fbb
>
>
> picture :
>
> http://iaelu.net/~yom/efikamx/arm-eabi/photo6.JPG
>
>
> --
> Guillaume
>
>
> Le 14 juin 2013 à 07:32, Adrian Chadd <adrian at freebsd.org> a écrit :
>
> So that line does this:
>
>        rc->freq = htole16(c->ic_freq);
>
> Please print out the value of p, c and rc before that is called.
>
> printf("%s: c=%p, p=%p, rc=%p\n", __func__, c, p, rc);
>
> I bet one of them is unaligned.
>
> And I bet that 'c' is aligned, but 'rc' and 'p' aren't aligned.
> I bet they're unaligned because ic->ic_rxchan and/or ic->ic_txchan is
> unaligned.
>
> Now, why _those_ are unaligned, I dunno. I bet the pointers passed
> into ieee80211_radiotap_attach() (th and rh) are unaligned, and that
> leads to this unalignment mess.
>
> The whole tx and rx tap structure allocation in if_runvar.h is a bit sick.
> Sigh.
>
>
>
> Adrian
>
>
> On 13 June 2013 16:49, Adrian Chadd <adrian at freebsd.org> wrote:
>
> Ugh,
>
> Hi,
>
> An alignment fault? ugh. It works fine on MIPS; I wonder if it's
> something odd to do with the specific driver in use.
>
> I'll look at the source and get back to everyone.
>
> Thanks!
>
>
>
> adrian
>
> _______________________________________________
>
> freebsd-arm at freebsd.org mailing list
> http://lists.freebsd.org/mailman/listinfo/freebsd-arm
> To unsubscribe, send any mail to "freebsd-arm-unsubscribe at freebsd.org"
>
>


More information about the freebsd-arm mailing list