I’m getting a bit lost here - where did you get a file named like this? There’s no such file produced when compiling OpenWRT from the fork this thread is about.
Thanks again frank I will try this (I don’t have it at home right now)
This may be my problem, I used gparted to create a partition table and then one fat16 partition for boot, one ext4 for the openwrt image… but those are p1 and p2, not sure how i can manage to create a “boot0” partition… What would be the command?
I don’t have any external hardware to test it, if this is what you refer to.
You are right, it is 18.06.4 from another thread… I tried to compile both other images and your code, using the official ubuntu on the R2 itself (on sd card) but both failed, so I figured I might as well try compiled images… If you have one for your fork I would love to try it!
Boot0 can be created with uboot (debug-uart needed to see and enter uboot-prompt) or mmc-utils (from running linux like your ubuntu).
https://www.fw-web.de/dokuwiki/doku.php?id=en/bpi-r2/storage#mmc-utils
You’ll find compiled mmc-utils in 4.14 branch on my github-repo (utils/mmc/mmc)
Debug-uart is recommended…without it you cannot see if board hang anywhere
https://www.fw-web.de/dokuwiki/doku.php?id=en:bpi-r2:debug-uart
Yes you are correct. I had a PC connected to one of the LAN ports (in bridge mode) when I did the speed tests and the network stall occured.
Can you find a way to reproduce it (config,hw-wiring,etc)?
I will try tomorrow to redo the steps and see if I can reproduce it again.
I just check the commit sha for your OpenWRT branch and I see these results:
$ git describe v18.06.2-bpir2-rc1 $ git rev-parse HEAD 2561dd6c4c3c4a14f0d3befd7b9240554af4f111
I’m not sure if these are the expected results as I see that they are different than what you provided. Or are those only some examples and not the real ones ?
I also looked into the git repository folder target/linux/mediatek/patches-4.14 and I did not see the file:
0027-net-next-mediatek-fix-DQL-support.patch
I just ran recently some speedtests and iperf and so far everything works fine (trace error did not appear).
1 Like
Mine were examples - I just executed command on whatever local git working copy I’ve had at that moment. Yours are an exact match with the expected values.
As you’re reporting that you are not able to reproduce the problem lately I’m wondering are you still on the same image you were when been affected by the problem? If that’s the case then it seems that this problem might be triggered by something else in addition to the 0027-net-next-mediatek-fix-DQL-support.patch. For me I had never experienced it since the moment I dropped out this patch for good. But it does not mean that the kernel is bug-free - it might be that the problem is rare with 0027-net-next-mediatek-fix-DQL-support.patch dropped out and I’m being lucky not to hit it (yet).
Hey, Thanks for such a lovely work and support for the R2.
I think you should update the checkout to the stable one as this is depreciated from your branch.
git checkout v18.06.2-bpir2-rc1, I used git checkout openwrt-18.06-bpir2
and in make menuconfig > I didnt change anything from this, except checked if its ext4 as mentioned.
Change this for normal users.
make -j16 to make -j(processor cores)
I had some missing dependencies and it was just shown as warning and it started to compile. Waiting for it to finish compiling then will update how successful it is.
make -j4
WARNING: Makefile 'package/utils/busybox/Makefile' has a dependency on 'libpam', which does not exist
WARNING: Makefile 'package/utils/busybox/Makefile' has a build dependency on 'libpam', which does not exist
WARNING: Makefile 'package/network/utils/curl/Makefile' has a dependency on 'libgnutls', which does not exist
WARNING: Makefile 'package/network/utils/curl/Makefile' has a dependency on 'libopenldap', which does not exist
WARNING: Makefile 'package/network/utils/curl/Makefile' has a dependency on 'libidn2', which does not exist
WARNING: Makefile 'package/network/utils/curl/Makefile' has a dependency on 'libssh2', which does not exist
WARNING: Makefile 'package/boot/kexec-tools/Makefile' has a dependency on 'liblzma', which does not exist
WARNING: Makefile 'package/network/services/lldpd/Makefile' has a dependency on 'libnetsnmp', which does not exist
time: target/linux/prereq#0.09#0.01#0.11
Should I install all the dependencies and then start again or let it continue? and how long does it normally take this to compile?
Error
`make -r world: build failed. Please re-run make with -j1 V=s or V=sc for a higher v
erbosity level to see what's going on
make: *** [/home/furkan/Downloads/Projects/BPI/R2/openwrt/openwrt/include/toplevel.m
k:218: world] Error 1
Trying V=s now.
It’s not about you missing some deps on the build box, it is about mishaps/errors/misconfigs inside official upstream package feed. Some packages optionally may use some other packages (like curl may be compiled with libssh2 support, e.t.c.) but these other packages are not enabled by default and thus are not found. For the most part warnings like these are safe to ignore.
i guess you mean libbssl 
@spikerguy you can do this to fix the warnings: BPI-R2 new image: OpenWrt 18.06.2 source code fork
That’s “on my list” and had been partially done in private testing repo but I hadn’t been able to test if things are flying fine when updated to latest 18.06 upstream release/branch so it is yet to be tested and then pushed to public repo.
These are different. v18.06.2-bpir2-rc1 is a tag and image you get when building it is expected to be 1-to-1 compatible with the upstream release v18.06.2. I.e. any package available out there in the Internet that was built for upstream openwrt 18.06.2 will be compatible with the image built from v18.06.2-bpir2-rc1 tag (with a possible exception being binary kernel-related deps), openwrt-18.06-bpir2 is a branch that is tracking (well, not yet, but is expected to be) upstream branch openwrt-18.06. This one is where upstream maintenance of 18.06.x releases is going on. So it you’re up to getting freshest-and-greatest developments on stable branch - using openwrt-18.06-bpir2 is a good choice. For a more safe (and binary compatible with third party packaged) side building from tag v18.06.2-bpir2-rc1 is a better choice. I don’t track master openwrt branch for now - it is an unstable territory and I don’t have enough spare time to do the required maintenance when tracking that one.
Nope, libssh2, there’s such a thing out there in the wild :-).
I tried this but it could’nt find this tag.
Depends on three things: (1) speed of your internet connection - a lot of source tarballs are downloaded in process and are placed into dl folder; (2) speed of your disk subsystem, good idea is to use fast SSD as a playing ground for the build; (3) CPU speed and amount of available RAM - these are obvious. make -j4 means that buildroot will try to build packages 4 at once so if you’ve got more than 4 core CPU good idea would be to use higher number in make command. On the other hand there are interdependencies between packages and parts of toolchain so buildroot won’t always be able to utilize all cores. For my rig (Ryzen 2700X, 64Gb of RAM, Samsung 970 NVMe as a storage for build area, 30MBit internel connection) typical time to build OpenWRT from the ground up is like 20-30 minutes.
Well, tag is obviously in there: https://github.com/lexa2/openwrt/commits/v18.06.2-bpir2-rc1 .
What does git tag | grep v18 show when executed inside your clone?
Ah, and just to be sure: had you executed lines below before proceeding with make menuconfig?
$ { echo CONFIG_TARGET_mediatek=y; echo CONFIG_TARGET_mediatek_mt7623=y; echo CONFIG_TARGET_mediatek_mt7623_DEVICE_7623n-bananapi-bpi-r2=y; } > .config
$ { echo CONFIG_DEVEL=y; echo CONFIG_CCACHE=y; } >> .config
$ make defconfig
Ofcource. I did with V=s and it compiled successfully.
I will try it to boot and get back to you.
HAHAHA Thanks for the a detailed information. Much Appreciated.
Booted fine. But I cant test the network as I don’t have any network cables at home. Wifi is not available on this build, Also I cannot connect my phone on USB tethering mode.
root@OpenWrt:/# uname -a
Linux OpenWrt 4.14.123 #0 SMP PREEMPT Fri Jun 7 09:50:53 2019 armv7l GNU/Linux
So I will have to wait till Saturday to test it in detail and install lucy. Thanks for all the effort you have put in to get this up for R2. Keep up the good work.
Here is the OpenWRT image with Luci installed and some additional packages (OpenVPN, Adblock etc) based on @lexa2 fork
@lexa2 is it possible to port your commits to the newer 19.07 branch or trunk ? I tried that but some of the patches do no apply and require manual changes
1 Like
Hello I tried installing luci on my build but I think the branch I used from lexa2’s repo was too old for luci and then I followed some instruction on openwrt forum and it leads me to believe that I need to upgrade to something higher where luci-base and luci-mode is available.
I will try @cioby23 build now. I tried you build and I cant get the land to work with my Laptop. While Wan is working from the router. when I connect LAN to my laptop I am getting these logs on UART.
root@OpenWrt:/# [ 370.874490] br-lan: port 4(lan3) entered blocking state
[ 370.879686] br-lan: port 4(lan3) entered forwarding state
[ 370.885340] mt7530 3.switch lan3: Link is Up - 1Gbps/Full - flow control rx/tx
[ 372.954523] br-lan: port 4(lan3) entered disabled state
[ 372.959923] mt7530 3.switch lan3: Link is Down
[ 391.674482] br-lan: port 4(lan3) entered blocking state
[ 391.679676] br-lan: port 4(lan3) entered forwarding state
[ 391.685424] mt7530 3.switch lan3: Link is Up - 1Gbps/Full - flow control rx/tx
[ 393.754275] br-lan: port 4(lan3) entered disabled state
[ 393.759712] mt7530 3.switch lan3: Link is Down
I haven’t done anything. As I see the bridge is assigned 192.168.1.1 I assume uhttp should work on 192.168.1.1 and I tried to assign my laptop Ip in the same subnet but still It is not connecting.
Any Advice?
Is ping working? Which kernel-version?