Znevna has quit [Quit: ZNC 1.8.1+deb1~bpo10+1 - https://znc.in]
Znevna has joined #openwrt-devel
goliath is now known as Guest2749
goliath has joined #openwrt-devel
Guest2749 has quit [Ping timeout: 480 seconds]
<ynezz>
f00b4r0: AFAIK there is for example "EXTRA_DEPENDS:=ucode (>= 2022-03-22)" in firewall4
<f00b4r0>
oh, neat
<f00b4r0>
that's exactly what I need, thx
<f00b4r0>
i was looking at packages in the packages/net/ folder and couldn't find such examples (despite many ones where it seems they would be most needed)
<f00b4r0>
nbd: it seems something really wrong is happening at the test location. I noticed frequent spikes of "system" cpu time on collectd and now that I logged into the device dmesg is completely spammed with
<f00b4r0>
[92632.805643] ieee80211 phy0: WM: (2365.095842:65:MURU-E)[MuruMumGetGrpIdxByPfid]No Group searched, MumNum=1,Pfid0-3=4/00/0
<f00b4r0>
[92632.816752] ieee80211 phy0: WM: (2365.095873:66:MURU-E)[MuruMumGetGrpIdxByPfid]No Group searched, MumNum=1,Pfid0-3=4/2550/0
<f00b4r0>
happening every 10ms afaict
<f00b4r0>
judging by the current size of slab_unreclaimed and the telltale gap in measurements I think the bug occured again. Only debug is lost due to current MURU-E spam
<f00b4r0>
there are no client associated anymore as you'd expect
<f00b4r0>
anything I should collect to help debugging?
kapcom01 has joined #openwrt-devel
<kapcom01>
ynezz: thank you. i cant use any other version other than 19.07 because i want to emulate a device with this specific version. can you help me with padding the rootfs with dd? can you guide me or give a link with details on how to do that?
<f00b4r0>
nbd: in case that's relevant, napi/phy0-8 is eating 24% cpu sustained
<nbd>
thx. i think i might have an idea on what's going on