Re: Kernel-Crash when working with ubt0

On Friday, 30 August 2019 21:50:28 CEST Maksim Yevmenkin wrote:
> [trimming the list of recipients]
>
good like this, so many mails gonna unsubscribe the list - many problems and 
bugs as far as eye reaches... 

> On Fri, Aug 30, 2019 at 12:08 PM Miranda Maria Sophie Van den
> 
> Breukelingen <mms.vanbreukelingen_at_gmail.com> wrote:
> > works for me as far as pairing concernes (with patch -C > .diff.txt
> > 
> > option         LOCK_PROFILING
> > option         WITNESS
> > 
> > is on in kern.config. My problem presists in the following form:
> > 
> > l2control -a c0:7a:a5:00:c7:11 read_channel_list
> > l2control: Could not bind socket, bdaddr=c0:7a:a5:00:c7:11: Network is
> > down
> 
> can you please tell what 'c0:7a:a5:00:c7:11' is? is this a remove address?
this is a Q13 - headphone - button.... it's unique with BT; recompiled kernel 
with the patch but internal stack is just available when you do the service 
start at least 5-6 times and then there's no info about the stack. 

Now I know how to apply a patch, what work must it be to write one!

 
> looking at your previous emails, i *think*, 'c0:7a:a5:00:c7:11' is a
> remote address. if this is the case, then, l2control is not going to
> work. from the l2control(8) man page
> 
> ==
>      -a local
>              Connect to the local device with the specified BD_ADDR or name.
> Example: -a 00:01:02:03:04:05 or -a bt_device.
> ==
> 
>  more specifically, the read_channel_list command will list l2cap
> channels open on the local device.
sometimes I get the read_channel_list open with one "port"; sometimes i 
doesn't. Sir Dice knows why!


> i'm not exactly sure what are you trying to do here.
connect BT-audio


> > l2ping is ok;
> 
> ok. this tells me that hci and l2cap communication works
yes, differently with 0 bytes or 16 bytes depending on the device. 

> > So, what to do with the "network down" - problem?
> 
> you have asked l2control to bind to the *local* bd_addr
> (c0:7a:a5:00:c7:11). and since there is no *local* bluetooth device
> with this bd_addr, l2control returns "network down". l2control will
> only work for *local* bd_addr's
I'm week with the whole /usr/local - at Arch on the dual boot machine I did a 
rm -Rf of /usr/local - since then you have to do a pacman with the --force 
commannd when updating packages. The meaning of this directory is differntly 
used with linux and bsd. 

I really don't know, what's the difference between them; natrually a bluetooth-
box that I can hold in my hands is loacally - but remote??? like LAN and WAN? 
MAN, I guess, like the switch between 192.168.8.1 and 10.42.0.1 on 
192.168.8.100 here! This address has to be "converted" into "BT-latin" and 
been connected like this. but how? 'bluetooth-control scan' is a good script 
but it doesn't connect, just pairs - and with the MagicBox there are a few 
more steps between binding and connecting.  

If there's time - and it is, as it is weekend - I'm gonna have a look what 
google tells me about A2DP and freeBSD. 

In the meantime I have a corrupt and infected mach_kernel at my VM (kvm on 
freeBSD is under heavy development, I don't think bhyve can replace it; 
anyway, the Mac Maverick doesn't run with freeBSD and VirtualBox is just the 
last thing I'm thinking of; I disabled hyperv on kernel). 

gonna write again when there are results. 

> thanks
it's up to me to say "thank you"!
> max
miranda