wildcat_paris

@Zador sure! you are right. but I forgot to config my favorite git in configuration.sh #KERNEL_DEV='git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable.git' KERNEL_DEV='git://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git' it is slow but limited to the latest tags for DEV

@AnyFox I guess you already know about uboot 16.01-rc4 and lamobo-r1 https://github.com/db260179/openwrt-bpi-r1/releases/tag/v1.1-stable-bpi-r1-openwrt @Zador I wanted to test but I am stuck @ [ o.k. ] ... updating [ linux-vanilla ] remote: Counting objects: 4421354, done. remote: Compressing objects: 100% (733075/733075), done. Receiving objects: 3% (154000/4421354), 65.41 MiB | 10.00 KiB/s

@tkaiser yep, I wrote "B53125 has only on GMAC" but I thought "A20 has only on GMAC", my bad. EDIT: Thomas, I have fixed my previous post "A20 SoC has only one GMAC link to the B53125"

@AnyFox is debootstrap-ng ready for 2 partitions yet? several tests, removed all the files

@Igor @Zador todo? it would not cost much to activate the kernel next/dev the "Minstrel 802.11ac support" (I tested it once with my dummy Alfa "AC")

@Zador thanks for the tip, I will build a new image for testing on odroid

@Igor I have seen the creation of debootstrap-ng with git pull before I create a new image with trusty... I also removed trusty cache (now I almost always remove...) But I leave Sherlock Holmes doing the investigation... or Dr Watson! Maybe Dr. Henry Jekyll as well but certainly not the evil Edward Hyde EDIT: I will be trying odroidxu4-next (again) as odroid-default has faulty ax88179_178a driver (Hardkernel image with kernel 3.10.92 works but BW is 1/3 of GMAC) I need ax88179_178a as part of my Plan B to move the "router" part of my Lamobo-r1 on OdroidXU4

@thom_arm Google is your friend, resize the larger partition but first do a backup of the sdcard on a harddrive (dd / dc3dd) try this https://www.howtoforge.com/linux_resizing_ext3_partitions or test gparted

@Igor (fresh water from France, Slovenia or Zadorland is great for ) FYI you have not published a trusty version for odroid, so you may know the issue it boots/reboots as expected UART => unable to use root as a login SSH => root login, then password replacement, then ssh drops me I will dig the sdcard! EDIT2: after modifying the sdcard /etc/passwd /etc/shadow /etc/group (adding a user) I can login!!!

@Tido I am not blaming the B53125, it is just nice, complete and afaik efficient 1/ Lamobo implemented 1 GMAC PHY with the A20 SoC for 2 (or more) VLAN. 2/ Lamobo may have added a separate PHY for the WAN on the B53125 (but only little better for BW) 3/ A20 SoC with 2 GMAC (one for WAN, one for LAN) with B53125 in the middle 4/ A20 SoC with 2 GMAC (one for B53125, one for WAN PHY) Your diagram (google docs), you are pretty right -- the only issue is => the WAN port number is arbitrary and the remaining "LAN" ports (0,1,2,4) for eth0.102 => all of the ports are linked to the A20-GMAC port 8 anyway More simply, when you setup eth0.101 and eth0.102 VLAN both are sharing the same A20-GMAC port 8 so linux routing/iptables (=CPU) handle every moves from eth0.101 to eth0.102 and vice-versa (VLAN are theorically separate LANs). in doing so, the BW is limited to (as a mean empiric value) A20-GMAC BW divided by 2 (not counting the terrible CPU/IRQ load) As A20-GMAC has hardware/software/etc issues RX/TX are not optimal, so you can divide the BW by 2 again (empiric value, worst case). So I am not that far stating the BW is 1/4 of GMAC BW when transfering data with Lamobo-r1 as a LAN-to-LAN or LAN-to-WAN "router" when eth0.101 and eth0.102 are in charge. EDIT: hand written schemas http://tinyurl.com/gw2m6z5 note : a completely separate WAN PHY GMAC would be far more secure! EDIT2: The only great thing with Lamobo-r1 you can spy data secretly on a LAN (flood the network to make switches working as hubs / or spy you ISP between their BOX and Internet, how do they handle SIP-like telephony)

@Zador This is what I supposed @Igor [ o.k. ] Install kernel [ linux-image-odroidxu4_4.83_armhf.deb ] dpkg-deb: error: failed to read archive `/home/gr/armbian/output/debs/linux-image-odroidxu4_4.83_armhf.deb': No such file or directory root@server1404:~/armbian# tree /home/gr/armbian/output/debs/ A week-end with Igor: Attention please EDIT: I was doing "ln -s" as a workaround... but... ok Igor has just fixed... my heroe EDIT2: still need to remove CLEAN="debs" and redo

FYI Just updated everything and I am using the lastest non modified compile.sh I wanted to build a OdroidXU4 + default kernel + CLI only + trusty (same issue with wheezy/jessie) => *** Configuration file ".config" not found! so "cp linux-odroidxu-default.config ../../sources/linux-odroidxu/odroidxu3-3.10.y/.config" then I get [ o.k. ] Using kernel config file [ lib/config/linux-odroidxu4-default.config ] cp: cannot stat '/home/gr/armbian/lib/config/linux-odroidxu4-default.config': No such file or directory (and some kernel config are not complete => CEPH related) root@server1404:~/armbian# ll /home/gr/armbian/lib/config/*odroid* -rw-r--r-- 1 root root 8528 Dec 25 23:51 /home/gr/armbian/lib/config/boot-odroid.ini -rw-r--r-- 1 root root 111026 Jan 8 18:03 /home/gr/armbian/lib/config/linux-odroidxu-default.config -rw-r--r-- 1 root root 103487 Jan 8 18:03 /home/gr/armbian/lib/config/linux-odroidxu-next.config

Armbian provides an example with bridge interface for lan/wlan like auto br0 iface br0 inet static bridge_ports eth0 wlan0 address 192.168.0.254 netmask 255.255.0.0 network 192.168.0.0 I guess you can do auto br0 iface br0 inet dhcp bridge_ports eth0 wlan0 br0 takes the MAC adress of the first PHY declared in the bridge_ports (in this example eth0 MAC) for DHCP I hope it may help

(moved from another thread about LUKS dmcrypt) As with my test on lamobo-r1, you can enable NEON and anything for Cortex A7 crypto # At least one emulation must be selected # CONFIG_VFP=y CONFIG_VFPv3=y CONFIG_NEON=y CONFIG_KERNEL_MODE_NEON=y as well as the crypto just in case my whole config for 4.4-rc8 (needs the PRNG patch for A20) EDIT: NEON related

even an interesting prototype if they have learned from their mistakes with lamobo-r1 But my guess is they have many Lamobo-r1 in stock, they just cannot sell Lamobo-r1s before they sell every piece of lamobo-r1 in stock!!! tkaiser has very interesting ideas about options to pick a "home router" but IMHO there are not as multipurpose as the lamobo-r1 would be for price/size/etc.

I have applied the patches / adapted the fix provided by openwrt for lamobo-r1 when zador suggested I use iperf3 to test local/internet (playing with the TCP window as well) I realized: - iperf3 results from lamobo-r1 to other (well know network linux) local machine is giving 850 Mbits/s for RX 450 Mbits/s for TX (with 10MB tcp window!) => it matches the statement from openwrt lamobo-r1 (so the fix is working, the bandwidth is CPU limited by the "too many interrupts" as well as TX issues with GMAC) - iperf3 results from lamobo-r1 to the internet (Max 500Mbit/s down 200 MBits/s up) are RX/450 Mbits/s TX/220 Mbits/s (which is "OK" for my Fiber Internet contract) - any local machine (my linux PC [former "router"], my Windows7, my odroidXU4), going through the lamobo-r1 is limited to RX/230Mbits TX/210Mbit/s I tought of tkaiser saying the B53125 has only on GMAC A20 SoC has only one GMAC link to the B53125 (it would need 2 GMAC one "in" one "out" = at least 2 Gbit/s) to handle traffic in/out so when the lamobo-r1 transfer data from one host to another (I limited my test to local network/Internet), it is doing RX from local/TX to internet (vice versa, handling IRQ for RX and IRQ for TX on the same cpu/thread even when changing this config having 2 cpu/thread available) => the limiting factor is TX to the internet / CPU handling the too many IRQ ok, my conclusions are flowed, I would need to do transfer local/local through lamobo-r1 to have a more reliable statement, as I have not studied the impact of IPtables/NAT/etc. so doing A20/GMAC in/out at the same time is limitation, my empiric idea is it cut the A20/GMAC bandwidth par 2 at best (ok it is flowed) NOTE: I am still hoping for a tech fix to reduce the number of IRQ

EDIT: Igor replied before I post my message @disconnect for the crypto part, it is already with Armbian vanilla version (kernel 4.3.3) but the Hardware PRNG patch is not applied (all the hardware prng won't be included in the linux kernel if the vendor do not provide a proper datasheet/spec, as there is a potential security threat) for the audio part, Igor worked on it, there was a git problem, now Zador seems to be working on with kernel 4.4.x so possibly available when the vanilla version moves to kernel 4.4

@zendus I have moved the topic, hoping a user owning the board can help. I have just read http://www.armbian.com/banana-pi-m2/ please edit your topic: are you using Debian Jessie or Ubuntu Trusty? I read "Jessie fails to boot on some boards."

@Shawn I read only "Debian Jessie" but you wrote "Armbian Debian Jessie" ok, Armbian has its separate 8192CU driver with AP Make sure you have rtl8192cu (Armbian) or rtl8xxxu (Kernel) => lsmod (please) But as you can read on the forum (search for tkaiser comments) the Lamobo-R1 has no stable wifi AP and it has only one GMAC connected to the B53125 switch (I own one Lamobo-r1) so connecting ethernet "one LAN to another LAN" mean at the very best (with some patches), you will get only the half (or 1/4) of a gigabit ethernet bandwidth => Lamobo-r1 would have needed 2 GMAC connected to the ethernet switch (one "IN", one "OUT"). and (alas), it is only USB2, so USB/GMAC ethernet converter with not help (I have tried), it is worse than A20 GMAC+switch.

for crypto, it will work with kernel 4.3.3 and later (so next and dev Armbian) for the prng, you will need the patch from Corentin Labbe http://sunxi.montjoie.ovh/ => http://sunxi.montjoie.ovh/patchs_current/0001-crypto-sun4i-ss-support-the-Security-System-PRNG.patch Corentin has made a nice tool to test https://github.com/montjoie/cryptotest There is some errors yet, he wrote me there are some general kernel pending issues. But working. FYI I have done my A20 prng testing posted some results here (very old "diehard" tool) http://www.cacert.at/cgi-bin/rngresults(search 2 proper results with "sunxi") and there (old "dieharder" csprng tool) http://forum.armbian.com/index.php/topic/404-stable-vs-mainline-for-next/?hl=diehard#entry2558

@bl4ckc00k1e as topic owner you can edit your topic title... but ok I have updated it as you wanted 1/ the brcmfmac of AP6181 says cfg80211/AP/IBSS 2/ "iw list" says "IBSS/managed/AP", no monitoring! so? I am not a wifi/kernel expert but I guess it is not registered in "iw list" don't expect monitoring to work.

@Shawn Are you using a recent kernel like 4.3.x / 4.4.x? the recent kernel includes a driver for 8192CU but only as a client I guess "Could not set interface to mode (3)" is setting in AP mode" which doesn't work with the kernel module just try gr@bpi:~$ iw list you will get this Supported interface modes: * managed * monitor instead of (AP compatible) Supported interface modes: * IBSS * managed * AP * AP/VLAN * WDS * monitor * mesh point btw gr@bpi:~$ lsmod Module Size Used by rtl8xxxu 51066 0 mac80211 353027 4 rtl8xxxu gr@bpi:~$ cat /proc/version Linux version 4.4.0-rc8-sunxi (root@server1404) (gcc version 4.8.2 (Ubuntu/Linaro 4.8.2-16ubuntu4) ) #1 SMP Tue Jan 5 02:34:34 CET 2016 as Rui says, just get a wifi dongle with proper Linux driver in AP mode, don't waste your time.

@zador yes thanks for the idea to test with "iperf3" (with different TCP windows size values) gives the value from OpenWRT patches I have also tested the public servers But it only works: Internet <=> Lamobo-r1 using the L-r1 as a "router" for others machines on the LAN limits the BW to 236/205 MBits/s (as with my previous tests) = more or less 400 MBits / 2 (+ IPtables in the middle) I will be testing soon the XU4 with an extra USB3/GMAC to act as a simple router. XU4 was my plan B for routing when I have ordered it. (ok my AMD Phenom2 965 PC with 2 GMAC is working very fine as a router but... power consumption is terrible as a 24/7 machine) private joke : with your spoiler, you make my "coffee pouring out of ..." my nose while laughing aloud

@zador I already patched the clocking @1GHz before using 2 threads for I/O you are mentioning stmmac patchs, are they included in Armbian next/dev? or are they the changes from Openwrt? please, thx.

I hope this may help http://forum.armbian.com/index.php/topic/211-kernel-update-procedure-has-been-changed/ and please, make first a backup of your sdcard / system.