<\x>
Ansuel: can you do regdb updates on ipq60xx now? a colleague of mine sent a patch to wireless-regdb to fix channel 36~48 for my country, would be nice to get that updated ;)
<CN_SZTL>
<nick[m]1234> "Tianling Shen" <- ah nice to hear that
<CN_SZTL>
and i saw they released 1.70.0
tlj has quit [Ping timeout: 480 seconds]
<\x>
seems on intel Z790 refresh some motherboards will ship with Intel BE200 (Gale Peak 2), and on AMD Advantage laptops there will be MT7927. so, WiFi7/be (pre-certified) adapters comes early huh
robimarko has joined #openwrt-devel
<\x>
so those intel be200 cards will be out by september ish
<robimarko>
Yeah, but I know they have newer FW as its in QSDK
<robimarko>
BTW, this new multi PD remoteproc is working like a treat
<robimarko>
On IPQ5018 cause I dont have IPQ9574
<robimarko>
At this rate QCA is going to actually upstream IPQ9574
<Ansuel>
so ipq5018 is closer to ipq9574 than ipq807x?
<robimarko>
Only in the remoteproc
<Ansuel>
they also use edma v2 ?
<Ansuel>
oh ok they just used the new coprocessor way
<robimarko>
Its just a FW design
<robimarko>
Where they offloaded all of the magic to TZ via SCM
<robimarko>
So you enable the radio via an SCM call
<robimarko>
Its still on EDMA v1
<robimarko>
Only IPQ95xx and IPQ53xx are EDMAv2 as they have no NSS offloading
<Ansuel>
but ipq95xx have nss cores afaik
<Ansuel>
think they are used only for wifi offload
<Znevna>
oh no where do you hope that mt7621 is going?
<Znevna>
:p
<robimarko>
To EoL as far as new products are aware
<Znevna>
:(
<Znevna>
but it's one of the most affordble and supported chip
<robimarko>
I said in regards to new products
<Znevna>
ah
<Znevna>
got it :p
<robimarko>
Cause they looks we have AX crap they pulled with it was ridicolous
<Znevna>
my ax53u units run fine tho
<Znevna>
but they are only APs so ..
<robimarko>
Ansuel: Well, there are claims of network accelerating, but nothing about special cores etc
<robimarko>
Hence why they beefed up the CPU
<Ansuel>
them putting more powerful CPU is just them finally finishing all the defective cpu from mobile repurposed for ipq and using more recent defective cpu LOL
<robimarko>
Well, at least its A73
<Ansuel>
with an arm x1 you wound't need to offload anything
<robimarko>
Other than your wallet to buy it
<Ansuel>
wifi 7 ipq is already overpriced and i don't think it comes from manufactoring cost but just marketing of stupid feature like putting an oled screen on a damn router
<robimarko>
Well, its been like that for every new generation
<Znevna>
is that a real thing?
<robimarko>
Yeah
<robimarko>
TP-Link makes them
<Znevna>
rofl
<robimarko>
Though its so crazy priced that I dont know who is buying
<Znevna>
I remember the CPU spikes cause by the lcd/touch on rb2011 and rb3011 :D
<Ansuel>
Znevna don't remember if it was gigabyte but they did a thing that moves the entire device to follow a wireless client with the antenna
<Znevna>
oh yes i've seen that
<Znevna>
hilarious
<robimarko>
And there is your beam steering
<Znevna>
"oy, my router has a broken actuator"
<Znevna>
I remember it was a little noisy too? but they said it will be resolved in production units ? :p
<Ansuel>
the fun thing is that considering the quality of the script if something broke (for real the actuator is blocked or other problem) the router will probably start to bootloop or be inaccessible
<Znevna>
I imagine cats having fun with those
<Znevna>
so a little question, ubi can be resized (up) but can it also be made smaller?
<Znevna>
or that requires a wipe and reinstall x.x
<Ansuel>
i think wipe
<Ansuel>
ubi grow is actually just putting empty nand
<Znevna>
there was a script in the mail list somewhere posted
<Znevna>
that ran before ubi mount
<Znevna>
if the dts changed the ubi partition size, it'll run ubi resize to increase it and avoid a bootloop
<Znevna>
but downgrading will surely bootloop
<Znevna>
.. as an alternative to having a device duplicate
<Ansuel>
ok we have a downstream driver that export the same symbol
<robimarko>
Its the old B53 swconfig driver
<Ansuel>
and conflicts with the dsa driver
<Ansuel>
guess in 5.15 the dsa driver wasn't compiled?
<robimarko>
WTF: cc1: error: /home/robimarko/Building/AX3600/ipq807x-5.15/staging_dir/target-mips_mips32_musl/usr/include: No such file or directory [-Werror=missing-include-dirs]
<robimarko>
It just compiled fine and with ALL_KMOD now this
<robimarko>
Well, this beats me
* f00b4r0
curses. Just wiped 1h of work
<robimarko>
Ansuel: Does CI even use ALL_KMOD?
<Znevna>
was it good work? :p
<f00b4r0>
it was documentation work
<robimarko>
Ansuel: Ok, so if kmod-switch-bcm53xx and kmod-switch-bcm53xx-mdio are included then there is a clash
<robimarko>
Also, I kind of have a feeling that its not the point update but 6.1 in general on bmips
<robimarko>
Il just rename the symbols to not clash
rua has joined #openwrt-devel
fakuivan has quit [Remote host closed the connection]
fakuivan has joined #openwrt-devel
fakuivan has quit [Remote host closed the connection]
<f00b4r0>
ynezz: missed your buildbot PR from yesterday, I've adjusted my repo watch setting hopefully I should get notified of everything now :)
tlj has quit [Remote host closed the connection]
tlj has joined #openwrt-devel
tlj has quit [Remote host closed the connection]
tlj has joined #openwrt-devel
<f00b4r0>
well I think I need to stop touching a keyboard for the day. ran rm -rf on the wrong folder this time -_-
<stintel>
btrfs snapshots saved me once or twice from that ;)
<f00b4r0>
stintel: but can btrfs save you from itself? ;)
* f00b4r0
hides
<stintel>
hahahaha
<stintel>
for that you have nightly backups ;)
<f00b4r0>
:-D
<Ansuel>
f00b4r0 OH GOD NO
<f00b4r0>
for everything important I have zfs
<f00b4r0>
+ multiple offsite replications
<f00b4r0>
still, I'm pissed. on the plus side I freed a whole bunch of space. still, it's a good clue I should probably stay away from my computer for a while :)
<f00b4r0>
guidosarducci: re bfptool-minimal: i guess it does drop some deps, but then quite a bit of the presented interface fails with "cannot load library foo" errors, which is probably not ideal. My use case is debugging a bpf issue, and having less than 140kB left on rootfs. Fortunately it does work just fine when extracted by hand in /tmp :)
tlj has quit [Remote host closed the connection]
tlj has joined #openwrt-devel
fakuivan has quit [Remote host closed the connection]
schwicht has joined #openwrt-devel
fakuivan has joined #openwrt-devel
schwicht has quit [Remote host closed the connection]
<hurricos>
Slimey: In the top left, it says "The Beating Heart of Technology". This, of course, is a lie. Crabs don't have hearts, they have open circulatory systems.
<Slimey>
true
tlj has quit [Ping timeout: 480 seconds]
<robimarko>
Luckily its "One Unified SDK" so it will never be upstream
Tapper has quit [Ping timeout: 480 seconds]
tlj has joined #openwrt-devel
rmb has joined #openwrt-devel
sigurasg has joined #openwrt-devel
schwicht has joined #openwrt-devel
<sigurasg>
I'm trying to mess around with the MT7621, and I'm snooping for docs on it. Found some on hardkr, but so far I've been unable to register with my Canadian mobile. Any hints, tips or tricks, or maybe you have programming manuals or data sheets you can share?
schwicht has quit []
tlj has quit [Remote host closed the connection]
tlj has joined #openwrt-devel
tlj has quit [Ping timeout: 480 seconds]
djfe is now known as Guest2025
djfe has joined #openwrt-devel
<robimarko>
Ansuel: Well, CI failed on bmips patches but kernel build succeeded so that b53 fix works
<robimarko>
Patches are refreshed as part of 6.1.31 bump