On Thu, Aug 27, 2015 at 02:38:00PM -0400, fred veldini wrote:
> Update on the performance of ATH10k Mesh.
>
> Bob, I applied the patches 950XXXX to the lastest wireless-testing and did
> some tests.
Thanks!
> I looked at the mac.c file to see if I could figure out why the rawmode was
> being refused.
> So I tried something and gave it a shot.
> I Recompiled the kernel this time I removed the follow
> drivers/net/wireless/ath/ath10k/mac.c
>
> 63 - case NL80211_IFTYPE_MESH_POINT:
> 64 - if (!test_bit(ATH10K_FLAG_RAW_MODE, &ar->dev_flags)) {
> 65 - ret = -EINVAL;
> 66 - ath10k_warn(ar, "must load driver with rawmode=1 to add
> mesh interfaces\n");
> 67 - goto err;
> 68 - }
OK - hmm, it looks like you lost part where the rawmode parameter is defined
and processed (it sets the ATH10K_FLAG_RAW_MODE on the dev). Another
option might be to use the nohwcrypt modparam which I believe also
enables raw mode. If you still have my patch that enables raw mode
unconditionally, well, you obviously don't need the modparam anyway.
In any case - it might be easier for us to get on the same page once
the outstanding patches land upstream, and especially in OpenWRT.
> # iperf -c 192.168.35.250 -t 30 -i 2 -P 25 |grep SUM
> [SUM] 0.0- 2.0 sec 48.9 MBytes 205 Mbits/sec
> [SUM] 2.0- 4.0 sec 51.4 MBytes 215 Mbits/sec
> [SUM] 4.0- 6.0 sec 52.8 MBytes 221 Mbits/sec
> [SUM] 6.0- 8.0 sec 52.8 MBytes 221 Mbits/sec
Not too bad! Thanks for testing.
--
Bob Copeland %% http://bobcopeland.com/
_______________________________________________
Devel mailing list
[email protected]
http://lists.open80211s.org/cgi-bin/mailman/listinfo/devel
