<hitech95>
Can someone give me any hint on how to troubleshoot why a FIT image is not correctly flashed into the ubi volumes? The process doesn't throw any error on the syslog. but the update is not applied.
KGB-1 has quit [Quit: KGB-1]
<robimarko>
You sure that its getting flashed in the volumes your bootloader expects?
<raenye>
rmilecki: thanks for the bc53xx patches you sent upstream, but they do not include ea9200. I'll test both the status=disabled and the fixed link alternatives on my EA9200.
<rmilecki>
raenye: in issue i linked to my branch
<rmilecki>
with ea9200 change included
<raenye>
rmilecki: I saw these, thanks
<raenye>
compiling now
<raenye>
but maybe the r8000 solution from your [2/2] upstream patch would work too
<rmilecki>
raenye: except for MAC address
<rmilecki>
raenye: using port 5 means using eth0 which doesn't have MAC assigned in NVRAM
<rmilecki>
(tested on my R8000)
<raenye>
BTW, why it makes more sense to disable unused ports rather that only enable used ports?
<rmilecki>
because those ports are there
<raenye>
rmilecki: What should I look for in my EA9200 nvram? etXmacaddr for X=0,1,2?
mkrle has joined #openwrt-devel
<rmilecki>
we should describe hardware mainly
<rmilecki>
raenye: "nvram show | grep ports"
<rmilecki>
there is vlanXports variable for LAN
<rmilecki>
(well for WAN too I think)
<rmilecki>
check what ports are there
<raenye>
rmilecki: but when a SoC supports, say, USB, I often see usb nodes set as disabled
<raenye>
vlan2ports=4 8
<raenye>
vlan1ports=0 1 2 3 5 7 8*
<rmilecki>
raenye: so by vendor design port 8 should be used as CPU
<rmilecki>
in this case ports 5 and 7 are use for offloadig with FA engine which we don't support
<raenye>
rmilecki: also, et0macaddr and et1macaddr are zeroed
<rmilecki>
most likely yes
<raenye>
FA accelerates L3 routing, right? like CTF?
<raenye>
rmilecki: there's also this -
<raenye>
# nvram show | grep devs
<raenye>
landevs=vlan1 wl0 wl1 wl2
<raenye>
fwd_wlandevs=eth1 eth2 eth3
<raenye>
fwddevs=fwd0 fwd1
<raenye>
wandevs=et2
<rmilecki>
yeah
<raenye>
rmilecki: BTW, did you figure out the single device per profile issue? (I still see one binary built with the patch and 20+ without it)
<rmilecki>
raenye: i've pushed the fix
<raenye>
I did see you committed it, but you were saying it did nothing on your side
<rmilecki>
raenye: just extended its description
<rmilecki>
raenye: it was about CONFIG_TARGET_MULTI_PROFILE and I put that info in commit body
<raenye>
rmilecki: I see. And may I ask why bcm53xx relies on upstream DTS?
<rmilecki>
because I'm too lazy to maintain downstream
<raenye>
:)
<raenye>
But that requires a lot of quiltery
opwrt-ddg_ has joined #openwrt-devel
<rmilecki>
also you can use buildroot to build "bcm53xx"
<rmilecki>
northstar actually
<raenye>
what is buildroot?
<raenye>
I build my own, with all my patches
<rmilecki>
try google ;)
opwrt-ddg_ has quit [Quit: Page closed]
<raenye>
rmilecki: I couldn't understand how is that different from my current build setup (git cloned from openwrt github)
opwrt-orsella has joined #openwrt-devel
<rmilecki>
raenye: buildroot?
<rmilecki>
it's a totally separated project
<rmilecki>
OpenWrt contains bits of some ancient buildroot though
<opwrt-orsella>
Hello All, I am currently testing openwrt 23.05.0 on a Banana Pi R3. board. I noticed that the recovery partition on the emmc does not start. By flashing the version of u-boot present in the 23.05-rc2 release using tftp I am able to start the recovery partition.
<opwrt-orsella>
I also noticed that the recovery partition on the emmc is not changed when flashing the emmc from NAND. I first flashed NAND booting from SD and then flashed emmc.
<mkrle>
hi folks, I haven't contributed to the project in a while, so I was wondering if it's enough to file a PR on GitHub or there is some additional process? not asking for priority/attention, just to make sure I'm not missing anything,
<raenye>
opwrt-orsella: how did you enter recovery? press the reset button after poweron?
<opwrt-orsella>
Hello, I selected "Boot recovery from emmc" from u-boot using the serial console
valku has joined #openwrt-devel
hanetzer1 is now known as hanetzer
<opwrt-orsella>
MMC read: dev # 0, block # 24576, count 256 ... 256 blocks read: OK
<opwrt-orsella>
this is the last line that is printed when I select "boot recovery system from emmc"
<hitech95>
robimarko, no idea. I only have a single ubi partition in the nand, and the image is working if I flash it via uboot. But I have no idea how to investigate. I was thiking too that there was something wrong with ubi volumes.
<raenye>
mkrle: my first PR took a while to be accepted. You could post the PR link here and hopefully get some reviews
<raenye>
opwrt-orsella: does it work when booting from nand? from sd?
mentalow has quit [Quit: :]]
mentalow has joined #openwrt-devel
<opwrt-orsella>
:raenye I have just tried booting recovery from NAND and SD and, unfortunately, they both do not work
<mkrle>
thanks raenye, I'll be patient. there's ~400 open PRs and mine is the same as any other. just wanted to make sure I'm not missing anything
<robimarko>
hitech95: Then check the upgrade recipe in OpenWrt
<hitech95>
robimarko, the bootloader is expecting a volume called "fit" in the "ubi" partition of the nand. but I don't find any reference of it in the definition of the sysupgrade fit image: https://pastebin.com/68fk5m9w
<hitech95>
am I looking in the wrong place?
<robimarko>
Beats me, Daniel is probably the one to ask
<rmilecki>
raenye: oh, i didn't see that, thanks for that patch
<rmilecki>
raenye: please send it for upstream
<rmilecki>
raenye: can you confirm that switch ports labels in DT are correct?
<rmilecki>
raenye: that "lan1" interface is actually "LAN 1" on device?
<hitech95>
robimarko, Yea... I've sent him an email some time ago I'm hoping for a reply. I'm 100% sure it has something to do with the partitions/volumes but I'm not sure on how to proceed.
mkrle has quit []
<mrkiko>
opwrt-orsella: your device name? Your openwrt version?
<mrkiko>
opwrt-orsella: oh sorry, you told it
<opwrt-orsella>
:mrkiko it is a banana pi r3 and I am testing the last OpenWrt stable version 23.05.0
raenye has quit [Ping timeout: 480 seconds]
robimarko has quit [Remote host closed the connection]
robimarko has joined #openwrt-devel
<opwrt-orsella>
:mrkiko I have just tested rc4 version also. The issue is present also on that version.
xback has quit [Remote host closed the connection]
xback has joined #openwrt-devel
Tapper has quit [Read error: Connection reset by peer]
Borromini has joined #openwrt-devel
<Borromini>
hi guys
<Borromini>
is anyone looking after the wiki? Both tmomas and bobafetthotmail seem to be AWOL.
<opwrt-orsella>
so no one is having issues running the recovery partition with the OpenWrt 23.05.0 on the Banana Pi R3?