Router Model: Netgear R8500
Firmware: DD-WRT v3.0-r36410 std (07/28/18)
Previous: DD-WRT v3.0-r35030 std (02/19/18)
Kernel: Linux 4.4.144 #3596 SMP Sat Jul 28 07:59:16 CEST 2018 armv7l
Status: Working. Both 5G are showing 650-866 mbps cabability, 2.4 only 72. Tried multiple settings, some would give higher speeds but not stable (devices would disconnect) current settings 72mbps stable (VHT40). Previous build had 144 stable.Changing some settings (preamble to short, shortslot overdrive:short, channel to 6-best channel for me and VHT40) would get to 144mbps but it would fluctuate very often down to 11mbps. Wouldn't be a big deal except I have 4 1st generation PS4s and due to their locations must be connected wirelessly.So far so good as not all my kids get on at once. Will see what happens when they all do. Note Amazon Devices could not see Wl2 5G-2 as it is set to channel 64. Wl0 is on channel 161. No issues with either radio (both set to VHT80).
Reset: No
Errors: None thus far, will report any I may find.
Fyi, I pointed out this ticket to BS, he replied:
"but its correct. its no typo. its a fix. this special algorithm has to be identical with the algorithm used in the wireless driver. since newer chipsets are using the dhd firmware blobs
the driver is fixed in these firmware. so i had to adjust the algorithm to be identical on all devices and drivers"
He didn't address the multiple wireless interfaces configure comment.
So then shouldn't vendor firmware have the same issue? Or is the "dhd blogs" he refers to something only open source firmware used, and vendors use different drivers for their closed source firmware??
So then shouldn't vendor firmware have the same issue? Or is the "dhd blogs" he refers to something only open source firmware used, and vendors use different drivers for their closed source firmware??
The vendor provides the driver; as it is closed-source (binary, a 'blob') BS integrated it and updates his code to accommodate. He's saying the reason it changed from using 'i' (as an iterator) to '1' (as an offset) is due to that changing in the (provided) driver. _________________ #NAT/SFE/CTF: limited speed w/ DD#Repeater issues#DD-WRT info: FAQ, Builds, Types, Modes, Changes, Demo#
OPNsense x64 5050e ITX|DD: DIR-810L, 2*EA6900@1GHz, R6300v1, RT-N66U@663, WNDR4000@533, E1500@353,
WRT54G{Lv1.1,Sv6}@250|FreshTomato: F7D8302@532|OpenWRT: F9K1119v1, RT-ACRH13, R6220, WNDR3700v4
Fyi, I pointed out this ticket to BS, he replied:
"but its correct. its no typo. its a fix. this special algorithm has to be identical with the algorithm used in the wireless driver. since newer chipsets are using the dhd firmware blobs
the driver is fixed in these firmware. so i had to adjust the algorithm to be identical on all devices and drivers"
He didn't address the multiple wireless interfaces configure comment.
The serial boot logs since my fun of de-bricking just make me cringe. That being said, I'm trying to decide exactly how far back to roll to in search of both radios working as APs simultaneously. I mean, I *could* go with using the 2.4 band as a wireless client and test that, perhaps.
EDIT: I'll have to re-check simultaneous radio access again, but some of the strange behavior that was going on seems to have stabilized. I switched over from the FreshTomato 5GHz AP to the DD-WRT 5GHz AP a few hours ago and it hasn't broken yet. It wouldn't surprise me if it was related to all of the multiple instances of configuring interfaces, invalid arguments, and otherwise in bringing up the network.
Joined: 08 May 2018 Posts: 14246 Location: Texas, USA
Posted: Thu Aug 09, 2018 15:48 Post subject:
quarkysg wrote:
Can someone running this build telnet/ssh into the router and see if the ‘lsmod’ command is working properly? My custom build from the same SVN repo is showing only 1 line output, which is wrong. If I use an older build busybox binary on my later build, it’s working normally.
Check your inbox. I think I may have stumbled upon something as to what broke.
Router/Version: Linksys EA6500 v1
Firmware: DD-WRT v3.0-r36410 giga
Kernel: (Sorry, don't remember, whatever Kernel this came with out of the box)
Previous: DD-WRT v3.0-r36079 giga
Setup: Factory Setup as Router & AP (with root user renamed to "admin" and new password)
Reset: Yes, both soft reset upon flashing as well as via router reset button after successful flash
Issues/Errors:Wireless Virtual Interface does not work at all !! Had to revert to previous build, where it at least works intermittendly.
Note: After factory reset and reboot, I added a virtual interface on the 2.4 GHz network with default options (bridged), applied settings, rebooted router, via web interface and then for good measure power-cycled router. I then double-checked that the virtual interface was mapped to br0.
When trying to connect to virtual WLAN interface, it never succeeded. Sometimes the attempt was stuck at connecting, other times it connected, but it did not receive and IP address from the DHCP server of the router. In the latter case, the router's monitoring page would show that it had issued an IP address, but the client did not receive it properly. (Tested with 2 Android 8.1 clients, as well as a Windows 10 client).
Router Model: Linksys E900
Firmware: dd-wrt.v24-36410_NEWD-2_K3.x_mega-e900.bin
Status: Basic setup working. WPA2 access point. DHCP server.
Reset: Yes
Errors: So far no obvious problems. Did encounter a not so obvious WiFi issue. Communication via WiFi stops working occasionally, but the connection doesn't drop completely. The WiFi client (MacBook) stays connected to the access point and has 192.168.1.109 assigned as ip address (according to network preferences in MacOS). But the MacBook can't ping anything. Not even the router ip address at 192.168.1.1. This problem lasts for some minutes, or I can fix it by manually clicking the "Renew DHCP lease" button on my MacBook. Then I can ping the router again and also connect to the internet. After pressing the button the ip address remains 192.168.1.109.
Happend about 5 times (that I noticed) in two days. Didn't happen on stock firmware.
Router Model:Linksys WRT610N V2
Firmware:DD-WRT.v24-36410_NEWD-2_K3.x_mega.bin
Previous:DD-WRT.v24-35531_NEWD-2_K3.x_mega.bin
Kernel:Linux 3.10.108-d4 #20244 Sat Jul 28 12:45:04 CEST 2018 mips
*
**Status:Gateway 2 local ap's
wl0 and wl1 using same SSID'S**
1 DHCP Server
*Netgear crap router with no krack patch for iot devices/plugged lan to wan
Alternate DHCP and SSID
net isolation
*1 WRT300N V1.1 Repeater
Alternate DHCP Server
Broadcasting same main SSID
Reset:No
Errors:None
*Overclocked to 533 MHz* (Very reliable)
*DNSSEC VIA CLOUDFLARE IPV6+IPV4 SERVERS*
*SFE Enabled*
Status: working, repeater like an WISP
Reset: yes
Errors: In some times, after the modification of some parameter, the wirelles does not work, i can connect though lan and see a thing different: Current Time Not available
and the date command shows a date:
Wed Dec 31 22:00:38 BRST 1969
I cant set the date throug comand line, and it needs restore router settings to default to work again
Joined: 08 May 2018 Posts: 14246 Location: Texas, USA
Posted: Wed Nov 07, 2018 4:29 Post subject:
@quarkysg, jwh7, slice1900:
The wlconf binary is built from source in DD-WRT, it is not a blob. I am not entirely sure if the actualy wl driver component is being compiled from the latest source code in DD-WRT or not, it does NOT appear as such. The ticket is referring to the wlconf binary, and after re-seeing your posts in this thread, seems my comment on the ticket may be completely wrong, but the VIF/VAP issue is a mess - and I am not entirely sure if it's the wlconf binary or DD-WRT's init or a combination of both.
