goliath has quit [Quit: SIGSEGV]
raenye has quit [Ping timeout: 480 seconds]
<KGB-2> https://tests.reproducible-builds.org/openwrt/openwrt_bcm47xx.html has been updated. (100.0% images and 99.9% packages reproducible in our current test framework.)
rua has quit [Ping timeout: 480 seconds]
tSYS has quit [Quit: *squeak*]
tSYS has joined #openwrt-devel
KGB has quit [autokilled: Possible spambot. Mail support@oftc.net if you think this is in error. (2023-10-15 08:00:24)]
KGB has joined #openwrt-devel
danitool has joined #openwrt-devel
rmilecki has quit [Quit: Konversation terminated!]
rmilecki has joined #openwrt-devel
rmilecki has quit []
rmilecki has joined #openwrt-devel
robimarko has joined #openwrt-devel
rua has joined #openwrt-devel
Strykar has quit [singleton.oftc.net coherence.oftc.net]
mcbridematt has quit [singleton.oftc.net coherence.oftc.net]
madwoota has quit [singleton.oftc.net coherence.oftc.net]
cp- has quit [singleton.oftc.net coherence.oftc.net]
dannyAAM has quit [singleton.oftc.net coherence.oftc.net]
Danct12 has quit [singleton.oftc.net coherence.oftc.net]
madwoota has joined #openwrt-devel
cp- has joined #openwrt-devel
cp- has quit [Read error: Connection reset by peer]
Strykar has joined #openwrt-devel
dannyAAM has joined #openwrt-devel
Danct12 has joined #openwrt-devel
robimarko has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
cmonroe has quit [Remote host closed the connection]
cmonroe has joined #openwrt-devel
robodogspok has joined #openwrt-devel
<robodogspok> hello there. I was wondering with my friend why OpenWrt is using LTS kernel instead of using newest stable kernel?
guerby has joined #openwrt-devel
<dwfreed> robodogspok: because trying to keep up with newest would require a team of paid people working on it
<dwfreed> keeping up with LTS is hard enough
<robodogspok> dwfreed: What kind of work it involves?
<dwfreed> refreshing patches, fixing bugs, testing devices
<robodogspok> Is it always same risk with LTS kernels? More regressions when backporting things to LTS kernels?
<dwfreed> sure, but non-LTS kernels have a much shorter life
<dwfreed> once a release is done, the kernel does not see major updates
<dwfreed> so newest kernels would go EOL in the middle of OpenWrt support
DLange_ has joined #openwrt-devel
Shados has quit [Read error: Connection reset by peer]
<KanjiMonster> "$ find target/linux/ -iname '*.patch' | wc -l => 2689" (23.05) - and that's ignoring the drivers added via files directly
Obi-Wan has quit [Quit: ZNC by prozac - http://znc.sourceforge.net]
DLange has quit [Remote host closed the connection]
EqUaTe has quit [Remote host closed the connection]
xdarklight has quit [Read error: Connection reset by peer]
rotanid has quit [Remote host closed the connection]
xdarklight_ has joined #openwrt-devel
lynxis has quit [Read error: Connection reset by peer]
EqUaTe has joined #openwrt-devel
rotanid has joined #openwrt-devel
schmars[m] has quit [Ping timeout: 480 seconds]
fakuivan has quit [Ping timeout: 480 seconds]
<robodogspok> dwfreed: but isn't a stable kernel a stable environment?
<dwfreed> yes, but it's only supported until the next stable releases
<dwfreed> if there's a bug found that would affect a stable kernel that's EOL, the fix will not be ported to that kernel by upstream
goliath has joined #openwrt-devel
<robodogspok> would be possible to just update newer stable kernel between .point releases? Or there just too much work involved with testing?
<dwfreed> too much work
<robodogspok> is there regularly breaking things in between stable mainland kernels?
Danct12 has quit [singleton.oftc.net coherence.oftc.net]
<KanjiMonster> APIs get constantly updated/changed, so any driver (changes) in OpenWrt will potentially need to be updated/adapted to the changed APIs
cp- has joined #openwrt-devel
<KanjiMonster> and these updates tend to be non-trivial
<robodogspok> In stable kernels? Did Torvalds say don't break APIs and userland?
<KanjiMonster> only userland
<robodogspok> okay
Danct12 has joined #openwrt-devel
<KanjiMonster> for kernel land it's "we don't care about out-of-tree"
<KanjiMonster> and that can include dropping APIs or features that are only used by out of tree drivers
rua has quit [Quit: Leaving.]
DLange_ is now known as DLange
rua has joined #openwrt-devel
xdarklight_ is now known as xdarklight
minimal has joined #openwrt-devel
schmars[m] has joined #openwrt-devel
Spr0cket has joined #openwrt-devel
Spr0cket has quit [Remote host closed the connection]
Spr0cket has joined #openwrt-devel
Spr0cket has quit []
Spr0cket has joined #openwrt-devel
robimarko has joined #openwrt-devel
Spr0cket has quit []
Spr0cket has joined #openwrt-devel
Spr0cket has quit [Read error: Connection reset by peer]
Spr0cket has joined #openwrt-devel
Spr0cket has quit [Read error: Connection reset by peer]
Spr0cket has joined #openwrt-devel
<dermoth> slh if you're interested here's my download script (with current params for my hw)... it would be quite easy I think to generate params automatically from the TOH. https://0x0.st/H4nC.sh
<dermoth> if you think of using/redistributing it i'll be happy to add better error reporting, check/dryrun modes, etc... All that's required is bash and rsync, md5sum for checking (I guess there's newer checksum slgos now, I probably haven't touched that part in years)
<dermoth> oh I guess I have been using that script for much longer actually, even my lolder version doesn'T have md5sums, and I've been lazy, not checking if i'm actually checking anythiong!
<dermoth> well oldest version of the script I have traces of gets backfire/10.03.1/brcm-2.4 :)
<dermoth> oh and find for checking, but could rewrite this in pure bash.... Even if I can script it with find tests only find doesn't allow me to return non-zero based on those tests :(
<dermoth> slh updated one - https://0x0.st/H45p.sh - removed packages check (there's been no checksum file for ages apparently but I could write functions to validate from ipkg metadata) and added check for sha1sums and sha256sums (only the later is present in the oldest mirrors I kept)
<dermoth> that's fine for myself but if you think it could be worth redistributing it I'm happy to work further on it
robimarko has quit [Remote host closed the connection]
robimarko has joined #openwrt-devel
dermoth has quit [Ping timeout: 480 seconds]
<hitech95> Ansuel, updated the PR now targetting master as requested.
dermoth has joined #openwrt-devel
Borromini has joined #openwrt-devel
<Mangix> robimarko: ping
cmonroe has quit [Ping timeout: 480 seconds]
Borromini has quit [Ping timeout: 480 seconds]
Borromini has joined #openwrt-devel
Borromini has quit [Quit: Lost terminal]
<robimarko> Mangix: Sorry, was watching footbal
<Mangix> know where QCA's kernel lives now?
<Mangix> codeaurora went down
<robimarko> Its all on Codelinaro since last year
<Mangix> alright, time to find their ag71xx driver
<Mangix> found it
<Mangix> oh wait this is not their driver. this is the upstream one
<Mangix> agh
<robimarko> There is mention of inline checksum engine
<Mangix> great
<Ansuel> robimarko present there but not on edma
<Ansuel> logik
<Mangix> I doubt this will result in anything, but still good to have
<robimarko> Havent all of these MIPS datasheets leaked already?
<robimarko> Ansuel: There has got to be checksum offloading in HW on EDMAv1 as well but we have no clue how to enable it
<robimarko> Like, its been present in even basic HW for decades
<Mangix> the issue is someone to write the code.
<robimarko> Well, that is ultimately always the issue
<robimarko> BTW, it seems that QCA-s fix for small packet stall was to disable inline checksum engine for QCA956x
<Mangix> what was it in OpenWrt?
<Ansuel> small packet stall ?
<robimarko> Same thing is present in QCA-
<robimarko> *QCA-s kernel
<Mangix> except it's using properly named macros
<Mangix> Might as well port them to OpenWrt
<Ansuel> robimarko very disappointing
<Ansuel> datasheet?
<swalker> updated openwrt/upstream, https://sdwalker.github.io/uscan/index.html
<Ansuel> reading the MAC conf 2 of qca9563
<Ansuel> there are 2 way of crc
<Ansuel> one enable simple crc the other enable crc + padding
<Ansuel> maybe the padding it the one fkedup
<Mangix> grepping around the tree shows they don't implement GRO
cmonroe has joined #openwrt-devel
robimarko has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
iocampomx has joined #openwrt-devel
<iocampomx> Hi there! I want to thank you all for your support on the last few weeks of feedback/suggestions/ideas on how to build an OpenWrt image for an un-supported device. I've decided to post an article about the reasoning behind it, and how OpenWrt helped me to bring new life to old hardware. https://blog.nafiux.com/posts/cnpilot_r190w_openwrt_bufferbloat_fqcodel_cake/
<Mangix> looks like I should hold off on upgrading
rmilecki has quit [Quit: Konversation terminated!]
rmilecki has joined #openwrt-devel
rmilecki has quit []
rmilecki has joined #openwrt-devel
raenye has joined #openwrt-devel
<djfe> when was openwrt 23 removed again from the firmware selector?
<dwfreed> i've poked, doubt it is intentional
<djfe> oh well, just wanted to mention it. downloads are still available of course :)
minimal has quit [Quit: Leaving]