<f00b4r0>
hauke: how do I go about submitting this for 22.03? Should I send the series to m-l or would you grab it from github? For context this is the patch series that make wireless properly working on mikrotik ipq40xx devices
<robimarko>
LGTM, cant do runtime testing but it was tested on 5.10 before and it worked fine
<f00b4r0>
yeah, no reason why it would break
mzvd has quit [Read error: Connection reset by peer]
mzvd has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
<zorun>
f00b4r0: I can run-test and pull them
<f00b4r0>
zorun: awesome
<zorun>
it's only necessary for some devices, right? I'm running 22.03-rc4 on a hap-ac2 and wireless is working
mzvd has joined #openwrt-devel
<f00b4r0>
it's necessary for all mikrotik devices. Wireless will work but at reduced range/throughput unless you're the happy few with the same BDF onboard as the was that was shipped in the ipq-wifi package
<f00b4r0>
the one that was shipped*
<robimarko>
Pretty much only the older ac2-s will work fine as those use the currently packaged BDF
<robimarko>
But there is no way to know which ones use which BDF
<zorun>
ok, I'm probably one of the happy few :) the device is several years old already
<zorun>
good, I can check if there's no regression for these devices then
<f00b4r0>
exactly
goliath has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
<zorun>
ah, 2203dbdf is a branch name, not a commit
<zorun>
I was confused for a moment
<f00b4r0>
yeah :)
<f00b4r0>
6 commits in there, on top of current 22.03
mzvd has joined #openwrt-devel
borek has joined #openwrt-devel
<stintel>
nitroshift: o/
mzvd has quit [Read error: Connection reset by peer]
nitroshift has quit [Quit: Gone that way --->]
mzvd has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
nitroshift has joined #openwrt-devel
mzvd has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
mzvd has joined #openwrt-devel
nitroshift has quit [Quit: Gone that way --->]
mzvd has quit [Ping timeout: 480 seconds]
mzvd has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
mzvd has joined #openwrt-devel
<zorun>
f00b4r0: everything looks good, the ipq-wifi-mikrotik_hap-ac2 package is no longer there, but wireless is still working fine
<f00b4r0>
zorun: great. Feel free to merge then, AIUI rc5 will be tagged this weekend
<f00b4r0>
ah, stintel just pushed a few things, I'll rebase :)
<stintel>
had to make use of the last chance to break things in rc5 :D
<f00b4r0>
;D
<f00b4r0>
done
<stintel>
I'll cherry-pick firewall4 bump to 22.03 too
<f00b4r0>
sounds good. Btw I don't know if someone has write access to patchwork but several of my "New" patches there have been merged
<zorun>
pushed, thanks for the patches!
<stintel>
hmmm I should look into the m300 watchdog really
<f00b4r0>
(i know you ACK'd on IRC, but that may not be enough for whoever deals with this later)
Tapper has quit [Ping timeout: 480 seconds]
<zorun>
ah, near-concurrent updates again :-)
mzvd has quit [Read error: Connection reset by peer]
<f00b4r0>
22.03 looking good ;-)
borek has joined #openwrt-devel
Tapper has joined #openwrt-devel
robimarko_ has joined #openwrt-devel
mzvd has joined #openwrt-devel
robimarko has quit [Ping timeout: 480 seconds]
minimal has joined #openwrt-devel
fda has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
fda- has quit [Ping timeout: 480 seconds]
mzvd has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
mzvd has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
srslypascal is now known as Guest3928
srslypascal has joined #openwrt-devel
cbeznea has quit [Read error: Connection reset by peer]
cbeznea has joined #openwrt-devel
srslypascal has quit [Remote host closed the connection]
srslypascal has joined #openwrt-devel
mzvd has joined #openwrt-devel
Guest3928 has quit [Ping timeout: 480 seconds]
Tapper has quit [Ping timeout: 480 seconds]
mzvd has quit [Remote host closed the connection]
mzvd has joined #openwrt-devel
Tapper has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
mzvd has joined #openwrt-devel
<robimarko_>
f00b4r0: Thanks for dealing with backporting dynamic BDF stuff
<f00b4r0>
np, yw
mzvd has quit [Read error: Connection reset by peer]
mzvd has joined #openwrt-devel
<stintel>
hah, found a solution for the broken watchdog
<stintel>
very lazy one :)
<stintel>
+CONFIG_BOOKE_WDT=y
<f00b4r0>
heh
mzvd has quit [Read error: Connection reset by peer]
goliath has quit [Quit: SIGSEGV]
<stintel>
don't you just love simple solutions like that :)
<f00b4r0>
;)
mzvd has joined #openwrt-devel
<rsalvaterra>
I HATE BCM27XX
<rsalvaterra>
I mean, "hi".
<Habbie>
the keys are like right next to eachother
<rsalvaterra>
Habbie: :D
<rsalvaterra>
God, this 5.15.51 bump is getting quite painful.
goliath has joined #openwrt-devel
<stintel>
ah, you're working on that
<stintel>
good
<stintel>
I can go to the park then
<stintel>
34C good for my tan :D
<robimarko_>
Thats low, it just hit 37C here
<robimarko_>
Its been a little warm for the last week
<robimarko_>
Spent half of day in a pool instead of working
<stintel>
lovely
<stintel>
I might do Croatia for a bit at the end of the month
<robimarko_>
Hopefully seaside, cause its just crazy high anywhere else
<robimarko_>
Mountains are good also
<stintel>
close enough, Oprtalj in Istria
<robimarko_>
Thats nice
<rsalvaterra>
stintel: Why, oh, why doesn't bcm27xx keep the device trees in a files directory, like other sane architectures (ath79 comes to my mind)?
<stintel>
one of my clients has a few holiday houses over there, I've been there quite some times
<rsalvaterra>
Nooo, let's just carry a humougous patch and risk breaking every rebase.
<stintel>
rsalvaterra: maybe I should revive my idea of creating bcm283x target
<stintel>
which would be vanilla without all the rpi kernel backports
<robimarko_>
I think that bcm27xx is basically the Pi kernel backported to OpenWrt one
<stintel>
but as long as we're keeping bcm27xx that won't help the rebase problem of course
<stintel>
but I would feel better running a less hacked up kernel
<rsalvaterra>
So, here's what I tried to do: I applied all patches to 5.15.50 and copied the device tree that broke the rebase.
<rsalvaterra>
Then, I did the kernel bump, applied the relevant patch with quilt -f and tried replacing the (broken) device tree with the copied one.
<rsalvaterra>
Then I did the changes upstream did.
<rsalvaterra>
And now quilt refresh hates me.
<rsalvaterra>
diff: arch/arm64/boot/dts/overlays/null: No such file or directory
<rsalvaterra>
Diff failed on file 'arch/arm64/boot/dts/overlays', aborting
goliath has quit [Quit: SIGSEGV]
<rsalvaterra>
Ugh…
<stintel>
:P
<stintel>
good luck with that
cbeznea has quit [Quit: Leaving.]
<rsalvaterra>
stintel: You're not helping. :P
<rsalvaterra>
Oh, well…
mzvd has quit [Read error: Connection reset by peer]
Tapper has quit [Ping timeout: 480 seconds]
mzvd has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
fda- has joined #openwrt-devel
goliath has joined #openwrt-devel
fda has quit [Ping timeout: 480 seconds]
<f00b4r0>
stintel: if I'm completely honest I don't fully understand what you described in your commit message ;P
mzvd has quit [Read error: Connection reset by peer]
mzvd has joined #openwrt-devel
fda has joined #openwrt-devel
fda- has quit [Ping timeout: 480 seconds]
mzvd has quit [Read error: Connection reset by peer]
mzvd has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
mzvd has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
<Ansuel>
question question
<Ansuel>
i'm in openwrt/staging_dir/toolchain-aarch64_cortex-a53_gcc-12.1.0_musl/initial
<Ansuel>
i do ln -sf lib /home/ansuel/openwrt/staging_dir/toolchain-aarch64_cortex-a53_gcc-12.1.0_musl/initial/lib64
<Ansuel>
why a lib symbolic link is created in openwrt/staging_dir/toolchain-aarch64_cortex-a53_gcc-12.1.0_musl/initial/lib
<Ansuel>
???
<Ansuel>
(/home/ansuel/openwrt/staging_dir/toolchain-aarch64_cortex-a53_gcc-12.1.0_musl/initial/lib64 is a simbolic link to the directory lib in openwrt/staging_dir/toolchain-aarch64_cortex-a53_gcc-12.1.0_musl/initial)
<Ansuel>
ok this is fun
<Ansuel>
openwrt/staging_dir/toolchain-aarch64_cortex-a53_gcc-12.1.0_musl/initial/lib64 is a simbolic link to openwrt/staging_dir/toolchain-aarch64_cortex-a53_gcc-12.1.0_musl/initial/lib
<Ansuel>
ln dereference the link to openwrt/staging_dir/toolchain-aarch64_cortex-a53_gcc-12.1.0_musl/initial/lib
mzvd has quit [Read error: Connection reset by peer]
<Ansuel>
Mangix mh?
<Ansuel>
let me read the discussion
<Ansuel>
Mangix the leds stuff is still pending
gladiac has quit [Quit: k thx bye]
<stintel>
ah so matrix oftc bridge is _still_ bugged
<stintel>
wtf
<stintel>
f00b4r0: I tried ... want me to explain it in more detail ?
<stintel>
rsalvaterra: I wasn't trying to :P
<f00b4r0>
well for my education, sure :) I don't fully follow the connection between the Book-E watchdog and how it prevents the W83793's one to bug?
<stintel>
W83793 is modular. procd starts wdt before kmodloader, so if you don't manually start the W83793 WDT in e.g. /etc/rc.local, it's not running
<stintel>
but procd starts the WDT during handover to sysupgrade, so you run sysupgrade, the W83793 WDT starts
<stintel>
unfortunately procd default (30s timeout) does not work well with W83793 WDT - start it manually, and you'll see that your system resets
<stintel>
start it with 120s timeout, and it's fine
<stintel>
*but* even then, with 120s timeout, which is fine for normal operation, it does not like handover to sysupgrade, because it still resets the system
<stintel>
enable Book-E WDT in-kernel, it's there when procd starts WDT, so it's active from the start, then kmodloader loads W83793, you'll get /dev/watchdog1, but procd will be using /dev/watchdog which is the Book-E
<f00b4r0>
oh!
<f00b4r0>
i see
<f00b4r0>
so w83 is never used
<stintel>
exactly
<f00b4r0>
got it. Clever :)
<stintel>
magic :)
<stintel>
much better imho than handling the w8xxxx weirdness in procd
<stintel>
because once we go down that path ...
<stintel>
before we know it the procd watchdog implementation grows larger than systemd 😂
<f00b4r0>
lol
<stintel>
now with that the m300 should be rock solid, just in time for 22.03 :)
<f00b4r0>
👍
<stintel>
time to find a replacment 😂
<f00b4r0>
mind if I copy paste your explanation to the wiki, eventually, so it's not lost? Because TBH the commit message doesn't really convey all that info ;)
<stintel>
sure
<stintel>
it kind of contains that info but you need to know how procd handles wdt
<stintel>
should have asked you to review the commit first
<stintel>
sorry
<f00b4r0>
no problem :)
<f00b4r0>
the important part is, the problem is "solved" :)
<stintel>
without writing code!
<f00b4r0>
yay!
<stintel>
allowing me to have a lazy Friday
philipp64 has quit [Quit: philipp64]
<f00b4r0>
hehe
<stintel>
4 day work week ftw
<f00b4r0>
amen to that!
<stintel>
afterwards bought a new backpack that can take my inline skates, helmet, and stuff I usually take to the park, like a frisbee, ball, beach palletes, bluetooth speaker
<stintel>
now I'm totally ready for summer :P
<f00b4r0>
nice
<Habbie>
very good
<mangix>
Ansuel: fair enough. I wouldn't consider it so. Most ramips devices don't use GPIO controlled switch LEDs. I think it's mostly me that suggests it...
<Ansuel>
Mangix: lets see if he manage to do it
mzvd has joined #openwrt-devel
mzvd has quit [Read error: Connection reset by peer]
<mrkiko>
nbd: thanks a lot for the hint! But given Iam running master anyway, I'll take the chance totest mt76 HEAD
<mrkiko>
nbd: to see how things go. Thanks a lot for the help. I was really awaiting for this fix.
borek has joined #openwrt-devel
mzvd has joined #openwrt-devel
<mrnuke>
robimarko_, Ansuel: have you had to deal with the "header" binary that comes packaged in some ipq807x FW images?
<Ansuel>
mhh what do you mean ?
<mrnuke>
Found out that wax218 firmware is scrambled. I can scp it to the vendor FW, run "header -x firmware.bin", and it will unscramble it.
<mrnuke>
then scp it back, and gues what, it's flat_dt image :D
danitool has quit [Remote host closed the connection]
<mrnuke>
By "firmware" I mean the firmware update images presented by the vendor
danitool has joined #openwrt-devel
borek has quit [Ping timeout: 480 seconds]
Borromini has joined #openwrt-devel
Borromini has quit [Quit: Lost terminal]
borek has joined #openwrt-devel
<mrkiko>
nbd: cmonroe: running mt76 HEAD with pushed mt7622 firmware - so far all seems working fine. Can not see "impossible" messages. thanks for the help.
<cmonroe>
mrkiko: great! it's the same firmware build I did my testing on and saw the same behavior. in cases where the impossible messages would start I instead see it clearing the BMCQ.
<cmonroe>
mrkiko: i also left my phone connected (with auto-join enabled) for several hours with no issues.
borek has quit [Quit: borek]
<robimarko_>
mrnuke: Cant say that I saw that before
robimarko_ has quit [Quit: Leaving]
minimal has quit [Quit: Leaving]
Piraty has quit [Quit: No Ping reply in 180 seconds.]