New CT firmware release is uploaded.

New CT firmware release is uploaded.

[Ben Greear]

Release notes:

      Improve RAM usage:  Re-organized, trimmed, and otherwise made better use
      of RAM.  Allow compiling out swbmiss, beacon filtering, roaming code.
      Images with 'full' in their name do NOT have the previously mentioned features
      compiled out.

      Over-all, freed up about 80k of extra RAM, which can be used for more vdevs,
      peers, buffers, etc.

      Supports 64 vdevs (one should be reserved for monitor interface, so
      effectively 63 vdevs for current kernels.)

I have only lightly tested this on my 3.14 kernel so far.

A lot of these changes could potentially introduce run-time
crashes due to extensive use of a different type of RAM for
storing many structures.  Should be a quick fix should problems
arise.

As always, let me know if you find any problems.

Thanks,
Ben

-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com


_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k

Re: New CT firmware release is uploaded.

[Ben Greear]

I've put up a new version (010) that fixes a crash bug in the 009
release.  64 stations (vdevs) have been fairly well tested at
this point (using my 3.14.14+ kernel).

http://www.candelatech.com/ath10k.php

Thanks,
Ben


On 07/21/2014 04:02 PM, Ben Greear wrote:
> Release notes:
> 
>       Improve RAM usage:  Re-organized, trimmed, and otherwise made better use
>       of RAM.  Allow compiling out swbmiss, beacon filtering, roaming code.
>       Images with 'full' in their name do NOT have the previously mentioned features
>       compiled out.
> 
>       Over-all, freed up about 80k of extra RAM, which can be used for more vdevs,
>       peers, buffers, etc.
> 
>       Supports 64 vdevs (one should be reserved for monitor interface, so
>       effectively 63 vdevs for current kernels.)
> 
> I have only lightly tested this on my 3.14 kernel so far.
> 
> A lot of these changes could potentially introduce run-time
> crashes due to extensive use of a different type of RAM for
> storing many structures.  Should be a quick fix should problems
> arise.
> 
> As always, let me know if you find any problems.
> 
> Thanks,
> Ben
> 


-- 
Ben Greear <greearb@candelatech.com>
Candela Technologies Inc  http://www.candelatech.com


_______________________________________________
ath10k mailing list
ath10k@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/ath10k