mangix has quit [Read error: Connection reset by peer]
mangix has joined #openwrt-devel
ptudor has quit [Ping timeout: 480 seconds]
ptudor has joined #openwrt-devel
<russell-->
mangix: how?
<mangix>
russell--: my device keeps rebooting. notice how I keep connecting/disconnecting. I think it's an mt76 issue. I've reverted the last commit for now
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
goliath has quit [Quit: SIGSEGV]
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
Dopff16 has joined #openwrt-devel
<Dopff16>
hi, I need to log the mac addresses of users who have connected to the router. how can I do this?
<Dopff16>
nvm
Dopff16 has left #openwrt-devel [#openwrt-devel]
minimal has quit [Quit: Leaving]
srslypascal has quit [Ping timeout: 480 seconds]
danitool has quit [Quit: Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos]
<neggles>
mangix: hmm. weird
<neggles>
same error?
<russell-->
i'm still seeing circular dependencies, fwiw, since weeks
philipp64 has joined #openwrt-devel
<Grommish>
I use HOST_BUILD_DEPENDS:=python3/host to make python3 available to the build system.. is there a way to also allow it access to python-sqlite3 so the staging_dir python has access to sqlite3?
<Grommish>
libsqlite3/etc looks to be device-install rather than host available
gladiac is now known as Guest1528
gladiac has joined #openwrt-devel
Guest1528 has quit [Ping timeout: 480 seconds]
floof58_ has joined #openwrt-devel
floof58_ has quit []
floof58_ has joined #openwrt-devel
floof58 has quit [Ping timeout: 480 seconds]
floof58_ has quit []
floof58 has joined #openwrt-devel
<mangix>
neggles: yes
<mangix>
there's a difference in how long it takes for it to fail but it still fails
mangix has quit [Remote host closed the connection]
mangix has joined #openwrt-devel
<russell-->
mangix: do you have a serial console to see why it might be rebooting?
<neggles>
mangix: hmm, have you tried with the relocated kernel and no compression?
<neggles>
russell--: it's failing at unzlma
<neggles>
s/unzlma/unlzma
<neggles>
a problem hurricos is quite familiar with *cough* AP330 *cough*
<russell-->
neggles: it didn't sound like that was his problem
<mangix>
two different problems, yes
<mangix>
neggles: no compression == size too big
floof58 has quit [Ping timeout: 480 seconds]
Atomicly| has joined #openwrt-devel
AtomiclyCursed has quit [Ping timeout: 480 seconds]
AtomiclyCursed has joined #openwrt-devel
Atomicly| has quit [Ping timeout: 480 seconds]
floof58 has joined #openwrt-devel
AtomiclyCursed2 has joined #openwrt-devel
AtomiclyCursed has quit [Ping timeout: 480 seconds]
AtomiclyCursed2 is now known as AtomiclyCursed
<neggles>
mangix: yes, no compression will make it too big for the build system unless you tell the build system to ignore that - goal is not to build an image that fits in the kernel partition but find out if the issue is decompression or not
<neggles>
Uncompressing Kernel Image ... lzma compressed: uncompress error 1 == "decompressed image size is too big to fit in CONFIG_SYS_BOOTM_LEN" (which is set to 32MiB in the GPL sources, or 8MiB by default - I can check if someone wants to give me a dump of the u-boot partition)
<neggles>
return code 1 indicates 'SZ_ERROR_DATA' in theory, rather than SZ_ERROR_OUTPUT_EOF (image too big), so it may be that it doesn't support an lzma option we're using or something along those lines - worth giving gzip a try
<neggles>
but that's assuming it's propagating the error right
<neggles>
....and these gpl sources are accurate
torv has quit [Remote host closed the connection]
torv has joined #openwrt-devel
srslypascal has joined #openwrt-devel
<mangix>
neggles: I tried gzip before. Different but similar error
<neggles>
I don't suppose you saved the specific error?
<neggles>
mangix: Error: inflate() returned <something> ?
<nick[m]1234>
ahhhh maybe I need qca,no-eeprom in my dts
<nick[m]1234>
oh my god finally it works!!! :D
robimarko_ has joined #openwrt-devel
robimarko has quit [Ping timeout: 480 seconds]
<nbd>
mirko: i think the easiest solution is probably to wrap the definition in #ifndef __cplusplus in libubox
Lynx- has joined #openwrt-devel
<Lynx->
Anyone that could help get hping3 back into OpenWrt?
<Lynx->
nping seems completely broken, and otherwise there is no good ping utility that supports ICMP type 13
<mirko>
nbd: sounds good to me
<mirko>
nbd: but wait, wouldn't that still clash? i mean, we /are/ in c++ context - aren't we?
<nbd>
right, that's why i wrote #ifndef instead of #ifdef
<nbd>
fallthrough should only be defined for C
<mirko>
sorry, right (re ifndef)
<mirko>
but if it should only be defined in C context - why does Qt do it then (which is all C++)?
<nbd>
it has its own definition
<nbd>
but it uses clang::fallthrough for C++, where that fallthrough part gets overwritten by the macro definition
<nbd>
and that causes it to fail
<nbd>
(if I read the code correctly)
<mirko>
i'm just wondering what happens with the next (c++-)lib doing also its down definition, which would again clash then.. but yeah, for the moment the ifndef __cplusplus within ubox would definitely help
ptudor_ has joined #openwrt-devel
ptudor has quit [Read error: Connection reset by peer]
<nbd>
mirko: well, Qt doesn't use a generic name, maybe other libs also won't use a generic name
<nbd>
i only picked 'fallthrough' for libubox to mirror the kernel's definition
<mirko>
i see
shibboleth has quit [Quit: shibboleth]
rua has quit [Quit: Leaving.]
rua has joined #openwrt-devel
Tapper has quit [Ping timeout: 480 seconds]
cbeznea has quit [Quit: Leaving.]
<nick[m]1234>
What does IMAGE_SIZE needs to be in a Device Makefile?
ekathva has joined #openwrt-devel
ekathva has quit [Remote host closed the connection]
<nick[m]1234>
neggles: thanks!!! :) because firmware is 0x790000 size? but why 7744k?
<nick[m]1234>
ahhh 0x790000/1024
<nick[m]1234>
okay thanks! :D
<neggles>
yep
<neggles>
in reality the actual max image size is less, since you need room for rootfs_data
rua has quit [Quit: Leaving.]
<neggles>
but it's really just a failsafe type deal - if the final image is over IMAGE_SIZE, the build system will fail/error the build, rather than generate an image you physically cannot flash
<nick[m]1234>
great. :)
<neggles>
it does get mildly annoying when you've replaced the flash chip in a device with a bigger one, but that's not exactly common :P
<nick[m]1234>
why should that break anything, if it just a failsafe? if your flash is bigger than the actual one it should work?
<neggles>
I have a tp-link which is a 4? 8?MiB device from factory but I've installed a 32MiB chip in it
<neggles>
I can't use the imagebuilder to make a firmware for it with packages pre-baked-in because it makes the image too big to fit in the original flash, so it errors
<neggles>
have to use full build system & edit the makefile. it's not a big deal
SamantazFox has joined #openwrt-devel
<neggles>
oh the minor inconvenience of it all! :P
valku has joined #openwrt-devel
<nick[m]1234>
:D
rua has joined #openwrt-devel
<nick[m]1234>
I would be soooo happy, if someone could merge xP I was sitting here all day long to figure out. And now everything works perfectly. :D
Misanthropos has joined #openwrt-devel
hexagonwin has quit [Quit: [ Quitting client ]]
<nick[m]1234>
argh, I destroyed the initram
<nick[m]1234>
initramfs needs to be uncompressed. now everything is working :)
pepe2k has quit [Remote host closed the connection]
danitool has quit [Quit: Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos]
minimal has joined #openwrt-devel
hexagonwin has joined #openwrt-devel
hexagonwin has quit [Ping timeout: 480 seconds]
cbeznea has joined #openwrt-devel
danitool has joined #openwrt-devel
Tapper has joined #openwrt-devel
rua has quit [Ping timeout: 480 seconds]
swegener has joined #openwrt-devel
rua has joined #openwrt-devel
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
pepe2k has joined #openwrt-devel
danitool has quit [Quit: Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos]
danitool has joined #openwrt-devel
arthur_melo has joined #openwrt-devel
pepe2k has quit [Remote host closed the connection]
arthur_melo has quit []
mrkiko has quit [Quit: leaving]
ptudor_ is now known as ptudor
rua has quit [Ping timeout: 480 seconds]
ecloud has quit [Ping timeout: 480 seconds]
ecloud has joined #openwrt-devel
sf has quit [Quit: Page closed]
minimal has quit [Remote host closed the connection]
minimal has joined #openwrt-devel
minimal has quit [Read error: Network is unreachable]
<hurricos>
nick[m]1234: You should actually skip kmod-owl-loader and use nvmem-cells implementation to load WiFi :^)
GNUmoon has joined #openwrt-devel
<hurricos>
Ultimately someone like Ansuel or Chunkeey is going to convert your implementation to use nvmem-cells instead, but the drawback is that you won't have tested it yourself.
<hurricos>
The problem then is, of course, figuring out how the f*** to write your device tree such that when loading the card, ath9k reads all the way down to your nvmem-cells in the right part of the device tree ;)
<hurricos>
kmod-owl-loader isn't that bad of a compromise but ideally it gets dropped in exchange for something being upstreamed :P
<hurricos>
congrats on the port tho
rua has joined #openwrt-devel
rua has quit [Ping timeout: 480 seconds]
<hurricos>
blogic: RE: realtek-poe: What was your development environment for PoE devices? If I suspected that the daemon isn't properly configuring the STM MCU, how could I send frames directly to the firmware?
<nick[m]1234>
hurricos: thanks :) Do you have any idea how I can rewrite it? Or give me a reference to a working file? Does the order in the dts change anything?
<hurricos>
I was thinking, "I should add a ubus call that lets me ask realtek-poe to send a frame directly."
<hurricos>
But you might have something more clever already somewhere.
<hurricos>
nick[m]1234: order matters surprisingly often in device trees.
<hurricos>
nick[m]1234: dig through the ath79 dts tree's git log, and go find the ath9k patch in the OpenWrt tree which enables nvmem-cells lookups to work for MAC and calibration data
<hurricos>
the nvmem-names should be "calibration" (ath10ks have "pre-calibration" as well)
<nick[m]1234>
thanks. :) so there is definitely a way to arrange the dts that it will work even without kmod-own-loader?
dansan has quit [Read error: Connection reset by peer]
dansan has joined #openwrt-devel
rua has joined #openwrt-devel
<nick[m]1234>
thanks I will test it, however the device is currently not reachable for me.