max__ has quit [Remote host closed the connection]
max__ has joined #openwrt-devel
goliath has joined #openwrt-devel
goliath has quit [Quit: SIGSEGV]
goliath has joined #openwrt-devel
Strykar has quit [Quit: /quit]
Strykar has joined #openwrt-devel
max__ has quit [Ping timeout: 480 seconds]
robimarko has joined #openwrt-devel
fakuivan has quit [Quit: No Ping reply in 180 seconds.]
fakuivan has joined #openwrt-devel
<christop1e>
I asked about an ubsan issue in the mt7621 clk driver a while ago? I got in touch with the developer Sergio Paracuellos and he submitted a patch to lkml. https://lkml.org/lkml/2023/2/6/165 yay! I found a similar issue in the mediatek clk driver and "fixed" it the same way. Should I submit those to openwrt?
<robimarko>
Ideally, send it upstream first and then backport it
<christop1e>
"upstream" means submitted to lkml or merged? also I'm not the right person to send the second patch to lkml.
<robimarko>
Ideally merged, but if stuff is really broken then I guess submited is enough
<robimarko>
Why are you not the rigth person?
<christop1e>
Because I don't known what I'm doing when i'm removing a BIT() macro in a clk driver. (;
<johnf>
at least as I see it, you kind of can't use the SDK following the instructions
rua has joined #openwrt-devel
rua1 has joined #openwrt-devel
rua has quit [Remote host closed the connection]
bluew has quit [Ping timeout: 480 seconds]
<tmn505>
johnf: did You install base-files from feeds? Aside from that, the automatical signing won't work because of a bug which this series tries to address: https://github.com/openwrt/openwrt/pull/10255
<johnf>
tmn505: hmm, no, I didn't install base-files
<johnf>
that other PR looks significant
<johnf>
oh, though it also looks stalled
<johnf>
:(
floof58 has quit [Read error: Connection reset by peer]
floof58 has joined #openwrt-devel
<johnf>
tmn505: ok, so I was able to install base-files, I'm doing another build but I suspect this will solve the problem
<johnf>
tmn505: mixed results, I generated the required key-build files
<johnf>
however the compile of base-files fails with a core dump
<karlp>
christop1e: if you found it, and you fixed it, that does actually make you qualified :)
<johnf>
also if you submit it, and there's something terribly wrong with it, you'll get the feedback needed to fix it
<rmilecki>
christop1e: for some reason it didn't reach linux-5.15.y stable branch (even though it uses proper Fixes pointing to the commit present in the linux-5.15.y)
<rmilecki>
so we can either try to make it reach linux-5.15.y or backport it on our own
floof58 has quit [Read error: Connection reset by peer]
<stintel>
it gets better. it's been eating CPU since the 9th, but it didn't impact network traffic until half an hour ago
<stintel>
so I guess a kernel or libbpf update happened and due to git-src link still being there I didn't get the update qosify with fixes for that new version
dgcampea has joined #openwrt-devel
goliath has quit [Quit: SIGSEGV]
cbeznea has joined #openwrt-devel
Borromini has joined #openwrt-devel
dgcampea has quit [Remote host closed the connection]
dgcampea has joined #openwrt-devel
goliath has joined #openwrt-devel
cbeznea has quit [Quit: Leaving.]
robimarko has quit [Quit: Leaving]
audun has joined #openwrt-devel
Borromini has quit [Quit: leaving]
tlj_ has quit [Remote host closed the connection]
f00b4r0 has joined #openwrt-devel
tlj has joined #openwrt-devel
T-Bone has quit [Ping timeout: 480 seconds]
rsalvaterra has quit []
rsalvaterra has joined #openwrt-devel
<djfe>
do 32bit openwrt installations support 64bit datatypes? curl 8 just released and requires them now.
goliath has quit [Quit: SIGSEGV]
Ansuel has joined #openwrt-devel
<Ansuel>
aparcar ping?
<Ansuel>
(I should really find a solution for this damn irc....
<aparcar[m]>
Ansuel: I'm wondering if this "detect changes action" works for us. It looks like it's running way to many kernel builds
<Ansuel>
anyway one way that should speedup testing is stop caching external toolchain
<Ansuel>
that should give more space to ccache and speedup compilation for everything
<aparcar[m]>
ok
<aparcar[m]>
sounds good
<aparcar[m]>
I saw some cases where i.e. only files in ath79 kernel were changed but still ALL kernel target were tested
<aparcar[m]>
is that intended?
<Ansuel>
do you have an example for that? yes that action sometimes detect more changes but i notice this with push to master (caused by the action using more commit)
<Ansuel>
aparcar if it's ok i would merge that pr and disable caching external toolchain
<Ansuel>
would result in more load on our cdn but i don't think it would be problematic?
<aparcar[m]>
ehh right now we don't really have the cdn in place
<aparcar[m]>
what is being downloaded/?
<Ansuel>
sdk for stable branch and external toolchain for master
<Ansuel>
this for each run... but consider that the load is still present (since the buildbot compile an external toolchain so the entire caching system is broken)
<aparcar[m]>
if we dockerize those things, doesn't it solve the bandwidth?
<Ansuel>
yep
<Ansuel>
and that is actually already there
bluew has joined #openwrt-devel
<Ansuel>
in another pr
<aparcar[m]>
where^^
<aparcar[m]>
well also I reworked some docker stuff, please check it out