Re: [rfc] removing the NDISulator

2013-10-22 Thread Thomas Mueller
> The NDISulator is a crutch from a time when there wasn't _any_ real > alternative. > There are plenty of alternatives now. What's lacking is desire and > person-power. But the datasheets are there, or the vendor code has been > released, or there's linux/otherbsd drivers. > Leaving it in there

Re: 9.2-STABLE r255918 with GENERIC and iwn - core dump

2013-10-22 Thread Adrian Chadd
I know what's causing this! It's because when the management frame completes, there's a callback mbuf tag (M_TXCB) that causes the driver to call the net80211 TX completion callback. Now, because some drivers call the net80211 tx completion callback from within their driver locks, it causes locki

Re: 9.2-STABLE r255918 with GENERIC and iwn - core dump

2013-10-22 Thread claudiu vasadi
For when the time comes, I'm all in for any tests (if needed). On Tue, Oct 22, 2013 at 6:07 PM, Adrian Chadd wrote: > I know what's causing this! > > It's because when the management frame completes, there's a callback mbuf > tag (M_TXCB) that causes the driver to call the net80211 TX completio

Re: 9.2-STABLE r255918 with GENERIC and iwn - core dump

2013-10-22 Thread Adrian Chadd
Grr, it's slightly more complicated than that. That whole timer mess is actually used for two things: * if the management transmit succeeds - it acts as a short-interval (a few seconds) timer to ensure that the probe request ends up providing some response that transitions to auth; otherwise it a