On Mittwoch, 6. Dezember 2017 22:26:09 CEST Sven Eckelmann wrote: > On Mittwoch, 6. Dezember 2017 18:52:01 CET Daniel Hernandez wrote: > > So been running this mesh patch on 4 routers in my home for over a > > month and it works great ! > > Reporting my feedback to see if it can be in the next release of > > LEDE . > > Felix Fietkau worked hard to get something into LEDE. I think all(?) > relevant > changes [1] are now in LEDE master (added 21 days ago). So you could try the > current master and send him some feedback. ...[
We have now noticed that your related change [1] will create a VHT80 mesh
interface when you specified VHT40 in for the wifi-device. For example like
the this configuration:
config wifi-device 'radio1'
option type 'mac80211'
option channel '44'
option disabled '0'
option beacon_int '100'
option distance '300'
option path 'platform/soc/a800000.wifi'
option htmode 'VHT40'
option noscan '1'
option ldpc '0'
option hwmode '11na'
option txpower '26'
config wifi-iface 'wmesh3'
option device 'radio1'
option ifname 'mesh3'
option network 'mesh3'
option mode 'mesh'
option mesh_id 'mesh'
option disabled '0'
option mcast_rate '18000'
option mesh_ttl '1'
option mesh_fwding '0'
option region_disable '0'
option macaddr 'AE:86:74:AB:77:F5'
option encryption 'psk2'
option key 'fb719b4e5ecd1bca734c1af1b59d28c7'
The new interface then looks like:
phy#1
Interface mesh3
ifindex 8
wdev 0x100000003
addr ae:86:74:ab:77:f5
type mesh point
channel 44 (5220 MHz), width: 80 MHz, center1: 5210 MHz
txpower 23.00 dBm
Problem is here that max_oper_chwidth is not parsed by
wpa_supplicant_join_mesh and instead overwritten by it. It will therefore just
use the highest oper mode it can use for the selected frequency. And because
vht is also overwritten, the same will happen with HT40.
Kind regards,
Sven
[1]
https://git.openwrt.org/?p=openwrt/openwrt.git;a=commit;h=d91494eedf06ac6b31c1aa9f7172871b16af96c8
signature.asc
Description: This is a digitally signed message part.
_______________________________________________ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
