Thanks for that, handy trick! Here's the output. The only thing I've changed is updated the kernel, which I've done 100 times on the EA4500 and the Dockstar without problems. Might it be an issue with 4.11.3? I can try an earlier version if you'd like? Using the non-DSA dtb. ## Booting image at 00800000 ... Image Name: Linux-4.11.3-kirkwood-tld-2 Created:by megal0maniac - Debian
gets router out Give me what you've got :) OpenWRT seems to handle the switch quite well. I have yet to get DSA working on any of bodhi's builds, although I'm aware it's a work-in-progress. Pity that the wireless radios seem to perform so poorly (even with stock firmware) but that's an issue for another thread.by megal0maniac - Debian
Hey bodhi! Thanks as always for the updates. I'm playing with the EA4500 again, wanted to see if DSA is perhaps working in a newer kernel, but boot gets stuck here (with both .dtb files, with DSA and without): ## Booting image at 00800000 ... Image Name: Linux-4.11.3-kirkwood-tld-2 Created: 2017-06-21 18:47:17 UTC Image Type: ARM Linux Kernel Image (uncompressed)by megal0maniac - Debian
Me still! VLANs work on OpenWRT, just to reiterate. I don't know what black magic they're using, but running swconfig on Debian doesn't even see the switch chip.by megal0maniac - Debian
Cool, that makes sense. I am not able to make a .dtb out of the file you provided but comparing it to the one I'm currently booting with, I don't see any reason why this one would work if the other doesn't. My gut feeling is that it's an issue of not addressing external ports correctly. There seems to be incredibly limited documentation DSA, or utilities to debug it though,by megal0maniac - Debian
Hello Just an update, booting the 4.4.0 image but with the kirkwood-ea4500-dsa dtb file from the latest kernel (4.6) is successful. I get warnings in the boot log, looks quite familiar: [ 2.037378] Distributed Switch Architecture driver version 0.1 [ 2.043736] mv643xx_eth_port mv643xx_eth_port.0 eth0: [0]: could not detect attached switch [ 2.052128] mv643xx_eth_port mv643xx_eth_by megal0maniac - Debian
Thanks for the credit, bodhi :) Will continue to testby megal0maniac - Debian
Hi bodhi Have you managed to make any progress on this? I haven't played with it much, but I started poking again the other day.by megal0maniac - Debian
So it turns out that there are 2 zImages openwrt-15.05.1-kirkwood-dockstar-zImage This one is specific for the Dockstar. It is the generic zImage with the dockstar dts appended to the end It doesn't care about bootargs u-boot env var It doesn't care about the .dtb you load with bootz It will ONLY use the arguments which it was compiled with. Changing the .dts in the OpenWRT sourby megal0maniac - Off-Topic
bobafetthotmail Wrote: ------------------------------------------------------- > did you try looking at the (crappy, partial) guide > I put down on openwrt forums? I had a look at that, thanks. It shed a bit of light but thankfully the Dockstar is already a "supported" target. One question, though. In step 5, why do you keep 1 eraseblock between u-boot environment and root?by megal0maniac - Off-Topic
bodhi Wrote: ------------------------------------------------------- > Your bootargs looks wrong. Does it have an "=" > sign after the bootargs like below? Sorry, it is right but it came from fw_printenv. These are my current (relevant) env vars: mtdparts=mtdparts=orion_nand:0x100000@0x0(u-boot),-@0x100000(rootfs) console=ttyS0,115200 bootargs_root=ubi.mtd=1 root=ubi0:by megal0maniac - Off-Topic
Hi all. I'm trying desperately to get OpenWRT 15.05.1 running on my Dockstar's NAND for a university project. No matter what I do, OpenWRT breaks the partitioning on boot and consequently cannot find my rootfs. I'm following the guide (loosely): http://blog.qnology.com/2015/04/openwrt-on-pogoplug-v3oxnas-proclassic.html My partitioning is as follows: mtdparts=mtdparts=orionby megal0maniac - Off-Topic
Sorry that this silly switch is being such a pain to work with. I feel I should just point out that the wireless radios do not work either with the Dockstar u-boot. I'm not being "that guy" and saying you need to fix that too, just mentioning it since there may be a connection. ;) Likely not, as the switch is connected over 2 RGMII ports and the radios over 2 PCIe ports...by megal0maniac - Debian
Just a shot in the dark, I contacted Claudio Leite (one of the authors of the driver) and he had this to say: QuoteI remember this being an issue when I tested some work Luka Perkov (OpenWrt dev) had done for that model. I ended up not working on it, since u-boot changed architecture shortly thereafter (and his patches were never accepted...) The data sheet requires an NDA with Marvell. I wouldby megal0maniac - Debian
I pulled the rootfs from a firmware image, but /etc/passwd and /etc/shadow both live in /tmp and I'm not sure how they're created. So unfortunately no luck there. Amusingly, there are a number of comments in shell scripts like "TODO: Figure out what this does and why" It's like they didn't even finish the product before it shipped :)by megal0maniac - Debian
bodhi, I found that there were issues with your .dtb I got "couldn't create dsa switch instance (error -12)" when using your .dtb with stock u-boot. With the OpenWRT .dtb I don't get this and the switch works. I'm not sure why we're using a different .dtb when the existing one works?by megal0maniac - Debian
Ah, sorry about that. I must have accidentally used the Dockstar one the 2nd time (when I pasted the logs) but the result is similar. This is the output using the correct u-boot: U-Boot 2015.10-tld-2.1 (Apr 02 2016 - 17:01:55 -0700) Linksys EA4500 SoC: Kirkwood 88F6282_A1 DRAM: 128 MiB (ECC not enabled) WARNING: Caches not enabled NAND: 128 MiB *** Warning - bad CRC, using defaultby megal0maniac - Debian
EDIT: Just tried booting this again with stock u-boot and it still kernel panics with the new .dtb :/ [ 101.051609] mv643xx_eth_port mv643xx_eth_port.1 eth1: link up, 1000 Mb/s, full duplex, flow control disabled [ 111.669041] mv643xx_eth_port mv643xx_eth_port.0 eth0: link up, 1000 Mb/s, full duplex, flow control disabled [ 111.704340] Unable to handle kernel NULL pointer dereference at vby megal0maniac - Debian
The latest log was using the dtb from the latest tarball (the one including zimage and 4.5 kernel)by megal0maniac - Debian
Hi I installed the new kernel, built uImage using zImage and the .dtb and booted with the Dockstar u-boot. Unfortunately it kernel-panics but it looks like a wrong address thing. Starting kernel ... Uncompressing Linux... done, booting the kernel. [ 0.000000] Booting Linux on physical CPU 0x0 [ 0.000000] Linux version 4.5.0-kirkwood-tld-1.1 (root@tldDebian) (gcc version 4.9.2by megal0maniac - Debian
bodhi Wrote: ------------------------------------------------------- > megal0maniac, > > > Here's a 12mp shot of the CPU section of the > > board: http://imgur.com/oLqPpuY > > Were this picture taken after you've taken the > largest shield off? was it difficult to do that? Yes, that was with the shield off. It just clips on and it's quite easby megal0maniac - Debian
Still fails unfortunately [ 1.979108] mv643xx_eth_port mv643xx_eth_port.0 eth0: [0]: detected a Broadcom Starfighter 2 switch [ 1.988275] unable to find register: core [ 1.992313] mv643xx_eth_port mv643xx_eth_port.0 eth0: [0]: couldn't create dsa switch instance (error -12)by megal0maniac - Debian
I couldn't build uImage with the .dtb integrated as there's no zImage for the 4.5 kernel you provided, but the Dockstar u-boot works with the .dtb as the 3rd argument of bootm so I did that. With Dockstar u-boot: MAC addresses still show as generic/null in the boot log even though they're specified in the env Earlier on, this is shown: : /ocp@f1000000/ethernet-controller@72by megal0maniac - Debian
Anything else I can do to assist at this point?by megal0maniac - Debian
pengu Wrote: ------------------------------------------------------- > When trying to reboot, it doesn't reboot, and runs > / stays in an unknown state. That is strange, I don't have that issue. How did you install OpenWRT and which version are you currently running > The other thing is, that the box runs a little bit > hot (54+ degrees when idle, testet with kernelby megal0maniac - Debian
You know what? It's 2016. Why settle for low-res images? :P Here's a 12mp shot of the CPU section of the board: http://imgur.com/oLqPpuY It looks as though the CPU is connected directly to the switch chip using MII channels, and the switch chip directly to the physical ports. There is no other network IC. The radios are connected via PCIe. The only IC on the underside of the boardby megal0maniac - Debian
bodhi Wrote: ------------------------------------------------------- > So this must have a different network chip. It > might be the 88e1318s like other new plugs such as > NSA325 or NSA310S/320S. I think you can find out > if you take a look inside. The picture you've > posted before was too blury to tell. I think the primary difference between the current supportedby megal0maniac - Debian
bodhi Wrote: ------------------------------------------------------- > OK. Then it probably means that more to be done in > u-boot. > > However, I think eth0 works if you wait long > enough. Because it did come up OK in dmesg. Keep in mind that eth0 being up just means that there's a link between the CPU and port 5 on the switch. We're plugging into external poby megal0maniac - Debian
It still doesn't work with the dockstar u-boot. I'm convinced it's because of the switch - the LEDs on the port are not lighting up at all. These should come on if there's a link, irrespective of whether the interface is configured or has an address. And of course the wireless interfaces don't appear at all. root@debian:~# ifconfig eth0 Link encap:Ethernetby megal0maniac - Debian
Interesting... Not sure what else to do about that though Hit any key to stop autoboot: 0 DockStar> setenv ethaddr 20:AA:4B:E5:D2:CF DockStar> setenv eth1addr 20:AA:4B:E5:D2:D0 DockStar> printenv (snip) eth1addr=20:AA:4B:E5:D2:D0 ethaddr=20:AA:4B:E5:D2:CF (snip) Environment size: 1459/131068 bytes DockStar> boot starting USB... USB0: USB EHCI 1.00 scanning bus 0by megal0maniac - Debian