[OpenIndiana-discuss] LiveUSB and oi_151a5

[Jim Klimov]

2012-08-29 0:36, Milan Jurik wrote:
> Hi,
>
> Jim Klimov píše v út 28. 08. 2012 v 22:09 +0400:
> [...]
>>>     How lacking is USB3 support? Would the ports not work at all,
>>> or can they drop back to compatibility modes like USB1 or USB2?
>>> Can they be passed through "as is" to a VM (VirtualBox, KVM) that
>>> would have drivers to support them better? ;)
>>
>> The USB3 ports did not work with the OS - not even for the
>> mouse. One port is powered, so the mouse laser shined there ;)
>>
>
> no surprise, USB3 HCI chips are not supported by Illumos at all.

I knew that beforehand, from the list - so no surprises here.
It only remains to hope that as USB3 gains momentum on desktops,
servers, JBODs and so on, someone would step up with a new driver
or a port of one. I hope it won't take much more than a year ;)

>> Also the WiFi driver is missing, at least from the LiveCD -
>> apparently the laptop uses a BCM 4313 WiFi abgn chip, IIRC.
>> Does anyone know off the headtop if that driver is in OI,
>> or where it can be found? I'll go google for that now ;)
>> I've already seen that Broadcom opensourced their drivers
>> for Linux 2 years ago, and the code drop included support
>> for this chip...
>>
>
> Opensourced under GPLv2 license? Or published documentation?

I gather, opensourced -
* 
http://www.linuxpromagazine.com/Online/News/Broadcom-Releases-Open-Source-Wireless-Drivers

* https://www.illumos.org/issues/182
> With the surprising release by Broadcom of source for their
 > 802.11n chips
 > 
http://git.kernel.org/?p=linux/kernel/git/gregkh/staging-next-2.6.git;a=commit;h=a9533e7ea3c410fed2f4cd8b3e1e213e48529b75
 > (under what looks like a BSD or MIT license, as well as an
 > apparent usable license for the firmware blobs), it would
 > be useful to port these to Illumos

Did this proceed anywhere? Do we at least know the licensing
for that code drop, and if illumos can use it (or an encumbered
SFE repo?)


> No, Broadcom WiFi chips are not supported by Illumos :-(


By now I've seen several issues on this in the bug tracker,
where you were active too, and a suggested workaround to use
either a different wifi device (for which a driver exists),
or hack up an NDIS-wrapper driver. I plan to play with the
latter when I get a hold of the computer for long.

Are NDIS drivers known to cause some problems (stability of
radio link or the OS, capability for WiFi encryption and other
standards, etc.)? Does anything prohibit successful binary
builds from being redistributed (as in an IPS repo), or is
this a semi-legal hack due to reverse engineering involved?

//Jim