danitool has quit [Quit: Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos]
felix has quit []
felix has joined #openwrt-devel
ekathva has joined #openwrt-devel
srslypascal has quit [Ping timeout: 480 seconds]
goliath has joined #openwrt-devel
srslypascal has joined #openwrt-devel
Tapper has joined #openwrt-devel
cbeznea has joined #openwrt-devel
goliath has quit [Quit: SIGSEGV]
Tapper has quit [Read error: Connection reset by peer]
Tapper has joined #openwrt-devel
goliath has joined #openwrt-devel
goliath has quit []
borek has joined #openwrt-devel
ly_ has joined #openwrt-devel
<ly_>
hi
<ly_>
anyone ?
<mrkiko>
:D ??
<ly_>
I test if can use
<mrkiko>
ly_: guess youcan
<ly_>
How does make package compile the package? which make file to use
<jow>
hurricos: my 2c: introducing a dedicated read/write thread for the i2c io would be overkill. Also what use is it accepting ubus calls very fast if they end up writing to the same io constrained resource (i2c) anyway
<jow>
hurricos: so instead of waiting for the ubus request to get handled, the request will get handled immediately and ends up waiting for a pthread semaphore
<jow>
easier to do a single thread io multiplexing (using epoll through uloop) program architecture
ly_ has quit [Quit: Page closed]
SlimeyX has quit [Read error: Connection reset by peer]
rua has quit [Quit: Leaving.]
goliath has joined #openwrt-devel
danitool has joined #openwrt-devel
SlimeyX has joined #openwrt-devel
foxtrot has quit [Remote host closed the connection]
foxtrot has joined #openwrt-devel
foxtrot is now known as Guest6665
Guest6665 is now known as foxtrot
Tapper has quit [Ping timeout: 480 seconds]
borek1 has joined #openwrt-devel
borek has quit [Ping timeout: 480 seconds]
borek1 is now known as borek
cbeznea1 has joined #openwrt-devel
robimarko has joined #openwrt-devel
cbeznea has quit [Ping timeout: 480 seconds]
borek1 has joined #openwrt-devel
borek has quit [Ping timeout: 480 seconds]
borek1 is now known as borek
borek has quit [Quit: borek]
borek has joined #openwrt-devel
ekathva has quit [Ping timeout: 480 seconds]
<fpsusername[m]>
Harm_: Why did you decide to apply the patches/device support on the main branch and not 22.03 (latest stable)?
<fpsusername[m]>
I just can't build it every time lmao. The only time building worked was when I first build it with quite some old commits
<mrkiko>
any chance the John Kohl posting on openwrt Forum about the GL-B2200 is around here in IRC?
<mrkiko>
or - anyone having any experience / tool to operate in discourse from CLI?
rua has joined #openwrt-devel
cbeznea has joined #openwrt-devel
hitech95 has joined #openwrt-devel
cbeznea1 has quit [Ping timeout: 480 seconds]
ekathva has joined #openwrt-devel
ekathva has quit [Ping timeout: 480 seconds]
<\x>
robimarko: so i have built 4721 with the 256-qam AP patch, i installed it to two ipq4019 units but as you see, it does not negotiate as VHT, seems also wpa_supplicant is needed to be patched?
<\x>
i tried connecting two of them over wireless but i still only see 300 Mbps on luci web
<robimarko>
You need CONFIG_DRIVER_NL80211_QCA enabled for sure
<robimarko>
As 2.4G VHT is done by using vendor extensions
<\x>
i am using ath10k-ct so i did not use ath10k
<\x>
out of the four patchfiles i only used three, i did not need "package/kernel/mac80211/patches/ath10k/983-ath10k-allow-vht-on-2g.patch
<\x>
"
ekathva has joined #openwrt-devel
<\x>
I hope this thing gets more traction but I guess I cannot complain that its not as this is not a standard
<robimarko>
It wont
<robimarko>
Try enabling CONFIG_DRIVER_NL80211_QCA for wpa supplicant as that is disabled even in full config
<robimarko>
And hostapd as well
<\x>
hostapd hack already works well, but yeah i dont know how to do the supplicant one, i only know how to copy paste haha
<f00b4r0>
jow: about to lunch now, will test immediately after
ekathva has joined #openwrt-devel
ekathva has quit [Quit: Leaving]
<f00b4r0>
jow: it works 👍
Tapper has joined #openwrt-devel
hanetzer1 has joined #openwrt-devel
hanetzer has quit [Ping timeout: 480 seconds]
<aparcar[m]>
what is the proper way to install openwrt on a x86 device? We provide ISO images but they lack an "installer". Just booting the iso and then dd the content of the USB stick?
<f00b4r0>
the iso is the install
<f00b4r0>
copy the iso to your boot media, done.
<f00b4r0>
s/iso/image/
<aparcar[m]>
f00b4r0: sure, in that case we should include our other images in the ISO images
<aparcar[m]>
so you can dd it over
<aparcar[m]>
my point is, if you have Debian.iso it gives you an installer
<f00b4r0>
what iso image, there are no iso image
<aparcar[m]>
right, we don't offer it by default anymore
<aparcar[m]>
but there used to be one, point being, we should offer an install mechanism
<f00b4r0>
the img.gz files are ready-to-boot content, they're not meant to be written to CD/USB intermediary media
<f00b4r0>
what for? Are you suggesting that it's too hard to use dd?
<f00b4r0>
as explained in the wiki?
<f00b4r0>
an install mechanism in the debian sense would imply the ability to choose what to install on the fly, I don't see that happening nor being something to waste however little resources we have on, imho.
<aparcar[m]>
f00b4r0: not having that makes installing OpenWrt on VMs more complicated. My point is not a custom image but the current approach seems flaky: dd image onto usb stick, boot usb stick and have openwrt running live, dd usb stick to target device
<f00b4r0>
that's not what happens.
<f00b4r0>
dd to target device, boot.
<aparcar[m]>
s/VMs/VMs and x86 hardware
<f00b4r0>
for VMs, there are tools available that can convert the provided imgs to vm-native disk images
<aparcar[m]>
then I open up the server, dd openwrt on the hard drive and so on?
<f00b4r0>
I used one to test on virtualbox
<aparcar[m]>
okay let's turn this around, I need to install OpenWrt on multiple hardware servers, I can attach via serial and it has USB ports. I don't plan to open up every server and attach my laptop to the internal hard drive
<Habbie>
grab any decent live image, note the image url from downloads, wget, dd?
<f00b4r0>
aparcar[m]: I don't see openwrt running off of your typical server bare metal, (although it can certainly be done). In the most likely case of running off of e.g. APU-like devices, you copy the content onto your boot media (SD card, M2 SSD, etc) and you're done. It's way simpler and quicker than having to muck with an intermediary installer, especially when you build your own images (from buildroot or image builder)
<Habbie>
(that decent live image could in fact be openwrt)
<f00b4r0>
or do what Habbie just described.
<aparcar[m]>
I see OpenWrt running bare metal so I need a proper way to install it. Not having that option upstream seems a flaw which likely prevents more people from using it bare metal
<f00b4r0>
feel free to submit a patch :P
hanetzer2 has joined #openwrt-devel
<aparcar[m]>
that's what I'll do
<f00b4r0>
considering the limited hardware subset that the x86 image support, I think your approach does not make sense, but it's your call :)
<f00b4r0>
as long as you do not remove the existing image files that everybody else uses :)
rmilecki has quit [Quit: Konversation terminated!]
<jow>
aparcar[m]: you could boot the iso and dd an image from there
<jow>
or boot any other live linux offering a shell and dd
<jow>
systemrescuecd, debian installer, ...
<aparcar[m]>
currently I'm wondering if we can't add another grub entry called "install to first hard drive" which call dd instead of /sbin/init or something
<jow>
or maybe add a simple, x86 specific base-files "installtodisk" script
<aparcar[m]>
yea installtodisk would be my goto, that could also automatically extend the root disk etc
<jow>
yes
<aparcar[m]>
well using init=/bin/ash gives me an empty /proc and /dev for now...
<jow>
that is epxected
<aparcar[m]>
but I need /dev/...
<aparcar[m]>
could you give me a pointer here please
<jow>
umount -l /dev; mount -t tmpfs none /dev; udevtrigger
<hurricos>
jow: I agree but I want to check what epoll actually does on i2c read. For that I'm going to grab the hardware once it gets the OpenWrt port.
<jow>
umount -l /dev; mount -t sysfs none /sys; mount -t tmpfs none /dev; udevtrigger
<hurricos>
jow: it may turn out that epoll on i2c just magically works, which would disturb me but also make things very simple
<jow>
hurricos: why wouldn't it work?
<hurricos>
jow: my suspicion is that epoll will return immediately, because /dev/i2c is technically "ready for reading" -- but it's not a regular file at all
<hurricos>
it's a kernel-backed character device.
<hurricos>
It will not work in a defined way, just like select(2) behavior isn't defined for non-stream, non-tty, non-block, non-fifo, non-pipe devices
<hurricos>
and I don't care that /dev/i2c is 'ready for reading', I care that the i2c transaction has finished
<jow>
you're using /dev/i2c ?
<hurricos>
Correct
<svanheule>
hurricos: any preference on which i2c-poe HW you want to get your hands on?
<hurricos>
The one bkobl has :^)
<hurricos>
since rtl839x is bit-banged gpio
<hurricos>
i2c-gpio*, is what I meant
<hurricos>
I don't want nice hardware, is the point. I want the crappy stuff to see how bad crap flies
<svanheule>
:P
<hurricos>
netgear GS728 somesuch
<svanheule>
GS728TPv2
<svanheule>
be sure to get the v2
<hurricos>
I'd rather find out what happens than speculate, I am not a strong C developer so speculation has not gotten me far.
<hurricos>
will do
<hurricos>
and I'm waiting on the OpenWrt port for it as well
<hurricos>
svanheule: has anyone torn apart the fs728?
<svanheule>
hurricos: fast ethernet!? what decade is this?
<hurricos>
does it use broadcom switching hardware? At this price point I'd rather buy the wrong board and butcher it, I don't have $500 to spare
<hurricos>
as per mrnuke_'s horrible, sarcastic suggestion
<hurricos>
s/switching/poe
<svanheule>
I have modified my €50 (or so) GS1900-8 to be able to talk to an RTL8231 via bit-banged I2C. On that board it's normally wired up for MDIO.
<svanheule>
but, rtl838x sadly does not appear to support muxing the uart1 pins as GPIO :(
rua has joined #openwrt-devel
<svanheule>
and the PoE MCU also needs different strapping pin config, I imagine...
AtomiclyCursed has quit [Quit: ZNC 1.8.2 - https://znc.in]
AtomiclyCursed has joined #openwrt-devel
MaxSoniX has quit [Quit: Konversation terminated!]
Tapper1 has joined #openwrt-devel
minimalist has joined #openwrt-devel
slate has joined #openwrt-devel
<xdarklight>
hauke: could you please take a look at https://github.com/xdarklight/linux/commit/31e123ede758a6d6a006448a4708c1b9f737e93a and let me know if you think the fix is correct? it still seems like GSWIP doesn't want to use the FDB for "single port bridges", so it's not a fix for the problem I'm seeing. but I'm thinking it's still a fix
digitalcircuits has joined #openwrt-devel
guifipedro__ has joined #openwrt-devel
rsalvaterra_ has joined #openwrt-devel
wigyori_ has joined #openwrt-devel
fblaese_ has joined #openwrt-devel
zarzarzar_ has joined #openwrt-devel
tmn505_ has joined #openwrt-devel
aiyion has joined #openwrt-devel
svanheule_ has joined #openwrt-devel
enyc_ has joined #openwrt-devel
PtitGNU_ has joined #openwrt-devel
slh_ has joined #openwrt-devel
_embargo_ has joined #openwrt-devel
hauke1 has joined #openwrt-devel
aiyion_ has quit [charon.oftc.net coulomb.oftc.net]
Tapper has quit [charon.oftc.net coulomb.oftc.net]
rua has quit [charon.oftc.net coulomb.oftc.net]
minimal has quit [charon.oftc.net coulomb.oftc.net]
slh64 has quit [charon.oftc.net coulomb.oftc.net]
Misanthropos has quit [charon.oftc.net coulomb.oftc.net]
clandmeter has quit [charon.oftc.net coulomb.oftc.net]
gladiac has quit [charon.oftc.net coulomb.oftc.net]
indy has quit [charon.oftc.net coulomb.oftc.net]
stintel has quit [charon.oftc.net coulomb.oftc.net]
ryd has quit [charon.oftc.net coulomb.oftc.net]
svanheule has quit [charon.oftc.net coulomb.oftc.net]
rsalvaterra has quit [charon.oftc.net coulomb.oftc.net]
fblaese has quit [charon.oftc.net coulomb.oftc.net]
Habbie has quit [charon.oftc.net coulomb.oftc.net]
ephemer0l has quit [charon.oftc.net coulomb.oftc.net]
zarzarzar has quit [charon.oftc.net coulomb.oftc.net]
enyc has quit [charon.oftc.net coulomb.oftc.net]
guifipedro has quit [charon.oftc.net coulomb.oftc.net]
xes_ has quit [charon.oftc.net coulomb.oftc.net]
PtitGNU has quit [charon.oftc.net coulomb.oftc.net]
_slate_ has quit [charon.oftc.net coulomb.oftc.net]
embargo has quit [charon.oftc.net coulomb.oftc.net]
digitalcircuit has quit [charon.oftc.net coulomb.oftc.net]
wigyori has quit [charon.oftc.net coulomb.oftc.net]
decke[m] has quit [charon.oftc.net coulomb.oftc.net]
lipnitsk has quit [charon.oftc.net coulomb.oftc.net]
owrt-snap-builds has quit [charon.oftc.net coulomb.oftc.net]
fpsusername[m] has quit [charon.oftc.net coulomb.oftc.net]
evils[m]1 has quit [charon.oftc.net coulomb.oftc.net]
JuniorJPDJ has quit [charon.oftc.net coulomb.oftc.net]
lykos153 has quit [charon.oftc.net coulomb.oftc.net]
tohojo has quit [charon.oftc.net coulomb.oftc.net]
f12 has quit [charon.oftc.net coulomb.oftc.net]
jschwart has quit [charon.oftc.net coulomb.oftc.net]