philipp64 is now known as Guest567
philipp64 has joined #openwrt-devel
Guest567 has quit [Ping timeout: 480 seconds]
madwoota- has joined #openwrt-devel
madwoota has quit [Ping timeout: 480 seconds]
madwoota- is now known as madwoota
cyrozap has joined #openwrt-devel
<Tapper> Hi philipp64 I am like you and use the ./scripts/env for my builds
<Tapper> I had a x86 build set up and did the kernel_menuconfig to refresh
<Tapper> not it will not build. My other builds will tho. Just a heads up
<Tapper> Now it will not build.
<Tapper> *
<philipp64> Why are we getting rid of $(AUTORELEASE) again? It saved a lot of trouble...
<stintel> philipp64: it also created trouble
<stintel> I'm not able to point you to specifics atm
<stintel> maybe we need a faq bot
rua has quit [Ping timeout: 480 seconds]
<philipp64> I liked that if PR's were approved out of order (or merged out of order) then no one got sent back to redo their PR...
<Tapper> Any one have any pointers on how I can fix this bug. I set up a x86 build using make menuconfig. It builds and all is good. Then if you open make kernel_menuconfig and close it again the build will fail.
<Tapper> I never changed any settings in Kernel_menuconfig
<Tapper> Even after doing a make dirclean it still will not build
<stintel> Tapper: you need to make BUILD_LOG=1 and check logs/ directory
<Tapper> stintel building now I will paste the log.
<Tapper> Paste bin.
<Tapper> WTF log says ERROR: target/linux failed to build.
<Tapper> that's it
<Tapper> No I have more. Hang on.
<Tapper> I did not make any changes to make menuconfig or Kernelmenuconfig
<stintel> Tapper: so something in your .config or a kernel bump exposed CPU_UNRET_ENTRY
<Tapper> stintel can I change that back?
<stintel> sure, if you find what did
<stintel> I'm searching
<Tapper> k
<stintel> but I'm close to shitfaced
<Tapper> hahaha
<Tapper> it's all good mate.
<Tapper> If it gets to mutch then leve it till the morning.
<stintel> I would like to introduce a CI change that catches this breakage
<stintel> but I've not thought of how exactly to pull that off
<Tapper> Yeah does not sound simple
<stintel> maybe if your native language is GNU Make
<Tapper> I am no good with my native language for spelling never mind make.
<Tapper> I can talk the back end of a dog tho. lol
<stintel> I'm not sure what that means :D
<stintel> I make a good cow impression ..... mooooooooooooo
<stintel> f43b9876e857c (Peter Zijlstra 2022-06-27 22:21:17 +0000 2577) config CPU_UNRET_ENTRY
<stintel> let's see if this floated in a stable bump
<stintel> Tapper: 5.10 or 5.15 ?
<Tapper> 5.15 but I can quick test with k5.10 if you like
<stintel> wait
<Tapper> No x86 is on k5.15 now
<Tapper> So I cant test with 5.10
<stintel> yeah looks like f43b9876e857c was added in a stable series
<stintel> in 5.15.57
rua has joined #openwrt-devel
<stintel> I should not git push while drunk but ...
<stintel> Tapper: this is x86?
<Tapper> stintel I will test for you mate
<Tapper> yes x86
<stintel> hmmm
<Tapper> If you never tutch Kernel menuconfig it builds just fine
<stintel> CPU_UNRET_ENTRY is already in target/linux/x86/config-5.15
<Tapper> I opened Kernel_menuconfig and all went to shit.
<Tapper> I did not make any changes just closed it again.
<stintel> ah it's not in x86/64
<stintel> I'm not sure if target/linux/x86/config-5.15 is supposed to leak into target/linux/x86/64/config-5.15
<stintel> Tapper: try adding CONFIG_CPU_UNRET_ENTRY=y to target/linux/x86/64/config-5.15
<stintel> I'm on too many Belgian beers to pull that off without accidentally pushing to upstream
<Tapper> OK
<Tapper> Belgian beers sound good.
<stintel> taste good too
<stintel> and hit you hard
<Tapper> stintel I am from the UK so we have some nice ones here to.
<Tapper> I like this one. https://www.oldspeckledhen.co.uk/
<Tapper> it's a ale tho not a larger
<minimal> the larger the better ;-)
dangole has quit [Remote host closed the connection]
<Tapper> stintel I added that line to the file, but no go. Still does not build.
danitool has quit [Quit: Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos]
minimal has quit [Quit: Leaving]
<stintel> Tapper: sorry mate, then I'm out of ideas for today
<Tapper> It's all good mate. I will ask around in the morning.
<stintel> can you check the same log if it's still CPU_UNRET_ENTRY that says NEW?
torv is now known as Guest579
torv has joined #openwrt-devel
Guest579 has quit [Ping timeout: 480 seconds]
tSYS has quit [Quit: *squeak*]
tSYS has joined #openwrt-devel
<Tapper> I posted my bug to github. https://github.com/openwrt/openwrt/issues/11625
<stintel> Tapper: could you add diffconfig ?
<Tapper> stintel can I get difconfig to a txt file?
<stintel> ./scripts/diffconfig.sh > /tmp/foo.txt
<stintel> iirc
<stintel> I think it's suggested in the bugreport template but I'll cut you some slack because iirc you're one of our blind users
<stintel> since we're on the subject, if you think the template could use improving there, please do tell
<Tapper> stintel template is good.
<Tapper> Cant find that txt file tho. I am on WSL
<stintel> which one? command I gave you should just create it
<stintel> unless WSL doesn't have /tmp?
<Tapper> tmp is not the same as /tapper/home/openwrt/tmp
<Tapper> and says I cant write the file there.
<stintel> so choose a path that is writeable
<stintel> I don't have any windows at hand so I can't easily figure that out for you
<Tapper> k
<mrkiko> stintel: hi!! Happy to see the bad times are over, at least for now
<stintel> mrkiko: hey no worries. I'm an IRC regular for ~20 years. it was actually weird realizing we hadn't had such a troll in almost 5y
<dwfreed> heh
<stintel> dwfreed: I guess for you this is at least weekly
<mrkiko> stintel: you're an OP now so maybe next time it will be little bit easier
<dwfreed> stintel: something like that
<mrkiko> dwfreed: on what channel ?
<dwfreed> mrkiko: the network :)
<dwfreed> I am OFTC staff
<mrkiko> dwfreed: ah, nice!! :D
<Tapper> stintel: is that OK?
<stintel> Tapper: it's not OK, but these things happen
<stintel> if someone has no power, they vent at random places (e.g. today on #openwrt-devel), if someone does have power they might invade a country
<stintel> it sucks, it should not happen, but it happens
<Tapper> stintel: Sorry I was talking about the edit I made to my bug report.
<stintel> ahh sorry dude
<Tapper> It's OK.
<mrkiko> stintel: so, if I understand right, you say it's better to have no power? Sorry, my question is naive. I'm not even native english speaker
<Tapper> mrkiko It's what you do with that power that maters.
<stintel> mrkiko: I think I'm saying it doesn't matter
<stintel> mrkiko: anyone can have a bad day
<mrkiko> stintel: Tapper: ok, thanks
<Tapper> This is getting a bit deep for 3.41 AM lol
<stintel> some take it out on random people on IRC which is mostly negligble
<stintel> others invade countries and get away with it
<mrkiko> mybe it's even too late for preparing pasta, but that's what I'm doing :D :D
<stintel> mrkiko: enjoy ;)
<mrkiko> stintel: yeah, now the concept is clearer. Thanks.
<stintel> fsck
<stintel> now I'm hungry
<mrkiko> stintel: sorry :D :D
<stintel> don't be
<mrkiko> stintel: would happily hand some pasta to you but guess we're little bit far away from each other - italy here
<dwfreed> I should eat too
<stintel> the slightest mention of food usually gets me hungry
<stintel> or at least craving
<Tapper> I am off to bed people have a good day nite or what ever it is for you.
<stintel> Tapper: have a good night
<stintel> mrkiko: not _that_ far
* Tapper waves
<stintel> but I flew to Belgium
<mrkiko> Tapper: good night Tapper and thanks for all! And your recommendations. :)
<mrkiko> stintel: :D
<stintel> with my car it could be a matter or 8-24 hours depending on where you at
<stintel> but I don't have my car, and my gf is flying in on Frida
<stintel> the lack of words and characters here and there suggests I should really go to bed
<mrkiko> stintel: :D :D well, if the occasion arrives, I'll be there !!
<stintel> maybe 2023
<mrkiko> have a good night stintel
<stintel> thanks mrkiko. does this mean it's good morning for you?
<mrkiko> stintel: yes, in a sense yes
<mrkiko> stintel: but I think i'll sleep again today :D
ledeFail has joined #openwrt-devel
<ledeFail> lede was a mistake for openwrt
<ledeFail> it's why buildroot crushes it
ledeFail is now known as slavintel
slavintel is now known as ledeDisaster
ledeDisaster is now known as convolutedMakefileHell
convolutedMakefileHell is now known as ledeGarbage
<ledeGarbage> let's meet up in bulgaria: https://a.uguu.se/eWrgMTJO.jpg
ledeGarbage is now known as brcmdidnothingWrong
brcmdidnothingWrong is now known as ledeShitshow
<ledeShitshow> _ _ __ _ _
<ledeShitshow> | | ___ __| | ___ __ ____ _ ___ __ _ / _| __ _(_) |_ _ _ __ ___
<ledeShitshow> | |/ _ \/ _` |/ _ \ \ \ /\ / / _` / __| / _` | | |_ / _` | | | | | | '__/ _ \
<ledeShitshow> | | __/ (_| | __/ \ V V / (_| \__ \ | (_| | | _| (_| | | | |_| | | | __/
<ledeShitshow> |_|\___|\__,_|\___| \_/\_/ \__,_|___/ \__,_| |_| \__,_|_|_|\__,_|_| \___|
<ledeShitshow>
<ledeShitshow> slavic oper can't even figure out services
ledeShitshow is now known as opnfukingwrt
opnfukingwrt is now known as ledeDisaster
ledeDisaster is now known as FraudWRT
FraudWRT is now known as LEDEdilapidated
LEDEdilapidated is now known as YWNBAM
YWNBAM is now known as LEDE
LEDE is now known as LEDEclown
<LEDEclown> /bin/sh: 1: Syntax error: "(" unexpected
<LEDEclown> aw
<LEDEclown>  ____ __ _______ _______ ____ _ _ 
<LEDEclown>  / __ \ \ \ / / __ \__ __/ / _|
<LEDEclown> | | (_) 
<LEDEclown> | | | |_ __ ___ _ _\ \ /\ / /| |__)
<LEDEclown> | | | | | |_ _ _ ___| | ___ _ __ 
<LEDEclown> | | | | '_ \ / _ \ '_ \ \/ \/ /
<LEDEclown> | _ / | | | | _| | | |/ __| |/ / |
<LEDEclown> '_ \ 
<LEDEclown> | |__| | |_) | __/ | | \ /\ / | | \
<LEDEclown> \ | | | | | | |_| | (__| <| | | | |
<LEDEclown>  \____/| .__/ \___|_| |_|\/ \/ |_| \_\
<LEDEclown> |_| | |_| \__,_|\___|_|\_\_|_| |_|
<LEDEclown>  | | \_\ 
<LEDEclown>  |_| 
<LEDEclown>  ____ _______ _______ 
<LEDEclown>  \ \ \ / / __ \__ __| 
<LEDEclown>  __ _ | \ \ /\ / /| |__) | | | 
<LEDEclown>  / _` || |\ \/ \/ / | _ / | | 
LEDEclown has quit [autokilled: This host violated network policy. Mail support@oftc.net if you think this is in error. (2022-12-28 06:24:13)]
openwrtDev has joined #openwrt-devel
openwrtDev has quit [autokilled: This host violated network policy. Mail support@oftc.net if you think this is in error. (2022-12-28 06:29:17)]
mrkiko_ has joined #openwrt-devel
mrkiko_ has quit [autokilled: This host violated network policy. Mail support@oftc.net if you think this is in error. (2022-12-28 06:36:11)]
bcmfan has joined #openwrt-devel
srslypascal is now known as Guest600
srslypascal has joined #openwrt-devel
Guest600 has quit [Ping timeout: 480 seconds]
philipp64 is now known as Guest603
philipp64 has joined #openwrt-devel
Guest603 has quit [Ping timeout: 480 seconds]
chuck48 has joined #openwrt-devel
bcmfan has quit [Ping timeout: 480 seconds]
gladiac has joined #openwrt-devel
Luke-Jr has quit [Ping timeout: 480 seconds]
danitool has joined #openwrt-devel
gladiac has quit [Quit: k thx bye]
Luke-Jr has joined #openwrt-devel
<mrkiko> Anyone knows if TP-Link TL-XDR608{6,8} are available only in some markets or some shops? Can't find them easily right now
<dwfreed> mrkiko: probably limited to the chinese market
<dwfreed> since all my search results only show chinese
<mrkiko> dwfreed: thanks a lot! Ok...
<mrkiko> dwfreed: thanks!!
Luke-Jr has quit [Ping timeout: 480 seconds]
robimarko has joined #openwrt-devel
Luke-Jr has joined #openwrt-devel
noltari has quit [Ping timeout: 480 seconds]
<aiyion> karlp: any suggestions on whether or how to reduce/hash the 16 byte Chip-ID to a 6 byte mac?
bbezak2 has quit []
bbezak has joined #openwrt-devel
<aiyion> karlp: I think this is what u-boot does:
<aiyion> the first three words supposedly share quite a few bits across different models.
<aiyion> As u-boot already does the work, is there a sane way to just take the mac-adress uboot generates for h2+/h3/h5 devices?
<aiyion> I think it would be clean to calculate the same address as uboot does.
<aiyion> Or use uboots if possible.
noltari has joined #openwrt-devel
<aiyion> are we doing a similar thing in other targets already?
<aiyion> getting a mac address from an uboot env?
<aiyion> If I understand this correctly, robimarko answered the question for IPQ40xx in april:
<robimarko> There is a U-boot NVMEM driver
<robimarko> So you can get the MAC as a cell that it provides
<robimarko> Or add an alias so bootloader patches it as thats been the default U-boot behaviour for ages now
<aiyion> which would be preferred? And do you have a link where we do it already?
<aiyion> My current patch extracts the raw 16 bytes from /sys/bus/nvmem/devices/sunxi-sid0/nvmem .
<aiyion> Your ways sound way cleaner.
<robimarko> Does the bootloader already save the MAC as ethaddr?
<robimarko> If it does then you can use the U-boot env NVMEM driver to fetch it
<aiyion> I haven't found a section where it does yet, but it's my first or second visit to uboots codebase, so thats not worth much.
<robimarko> You dont need the source code
<robimarko> Just read the default u-boot env
<robimarko> If ethaddr is there then it does by default
<aiyion> will do once I switched trains, thanks
<robimarko> It could also calculate on each boot and newer save it
<robimarko> Then using an alias should work best
<aiyion> I expect the latter to be the case but will report back in a short while.
xutaxkamay has joined #openwrt-devel
hanetzer has quit [Ping timeout: 480 seconds]
hanetzer has joined #openwrt-devel
gladiac has joined #openwrt-devel
guidosarducci has quit [Remote host closed the connection]
guidosarducci has joined #openwrt-devel
gladiac has quit [Quit: k thx bye]
robimarko has quit [Remote host closed the connection]
robimarko has joined #openwrt-devel
blogic has joined #openwrt-devel
<Tapper> robimarko I posted the info you asked for: https://github.com/openwrt/openwrt/issues/11625
<Tapper> I am not shure I did it write tho. Because of my screen reader I had to send the output to a txt file then post it.
<robimarko> Tapper: As far as I can tell
<robimarko> It really changes the x86 config
<Tapper> robimarko do you know how I can change it back?
<Tapper> I did a make dirclean but still broken
<Tapper> I don't want to get rid of my env settings for my other routers.
<PaulFertser> Tapper: if git diff is showing it you can revert to what's in repo with "git reset --hard" (this resets everything you changed in all files)
<PaulFertser> dirclean doesn't affect it of course
<Tapper> PaulFertser will this mess up my env settings?
<PaulFertser> Tapper: do not do that in the env subdirectory, do it where git diff shows the kernel config difference.
<PaulFertser> Tapper: check "git status", only files mentioned there are affected by reset --hard.
<Tapper> PaulFertser OK will look at it. I don't know why the config for x86 gets so messed up tho.
<PaulFertser> Neither do I.
<Tapper> I get modified: target/linux/ramips/mt7621/config-5.10
<Tapper> modified: target/linux/x86/64/config-5.15
<Tapper> blank
<Tapper> no changes added to commit (use "git add" and/or "git commit -a")
<Tapper> modified: target/linux/x86/config-5.15
<PaulFertser> Tapper: so reset --hard would reset also your changes to mt7621 kernel_menuconfig.
<Tapper> I did not make any changes to mt7621 kernel_menuconfig.
<Tapper> I opened Kernel_menuconfig and shut it again. that's all
<Tapper> never changed anything.
<PaulFertser> Tapper: then feel free to reset.
<Tapper> PaulFertser works again thanks.
<Tapper> Kernel config is messed up still.
<Tapper> If I open and then close it it messes my config
<PaulFertser> Tapper: yep, to be expected.
chuck48 has quit [Remote host closed the connection]
gch981213 has joined #openwrt-devel
<robimarko> git restore would probably work
<robimarko> Its expected that it would remove symbols that are in generic
<robimarko> Even if you dont save it
chuck48 has joined #openwrt-devel
<PaulFertser> But it's not expected that it breaks the build altogether :)
<robimarko> Thats a function of generic and target config diverging too much
<robimarko> Is this maybe on 22.03?
guidosarducci_ has joined #openwrt-devel
guidosarducci has quit []
DRX has joined #openwrt-devel
cyrozap has quit [Quit: ZNC 1.7.4 - https://znc.in]
<DRX> What package contains the act_police kernel module?
<blogic> DRX: sched-act-police
chuck48 has quit [Remote host closed the connection]
<DRX> Thanks for the help; I found act_police in the package kmod-sched. Hopefully I can get the police filter working on qdisc ingress now with tc.
cyrozap has joined #openwrt-devel
hanetzer has quit [Ping timeout: 480 seconds]
DR_X has joined #openwrt-devel
DRX has quit [Read error: Connection reset by peer]
DR_X has left #openwrt-devel [Leaving]
philipp64 has quit [Ping timeout: 480 seconds]
philipp64 has joined #openwrt-devel
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
<aiyion> robimarko: back at it; at least until my better half is back. This is the output of printenv from uboot: https://bpa.st/T6OJ2 other than 'serial' which is on byte to long I don't see anything,that resembles a mac adress.
<robimarko> You are gonna have to do it from the userspace then
<aiyion> My current patch does not work at all, as /sys/bus/nvmem/devices/sunxi-sid0 does not contain `nvmem` on the NanoPi R1.
<aiyion> It should though, I think; as it's an Allwinner H3
<aiyion> Now that's a name I recognize. https://github.com/torvalds/linux/blob/master/drivers/nvmem/sunxi_sid.c oliv3r[m]
<robimarko> World is a small place
<aiyion> Might I just be missing a kmod?
<robimarko> NVMEM stuff is usually compiled in
<robimarko> And its enabled in the sunxi config: CONFIG_NVMEM_SUNXI_SID=y
<aiyion> huh.
<robimarko> And the compatible is there in the DTS
<robimarko> You wouldnt have /sys/bus/nvmem/devices/sunxi-sid0 if the driver wasnt probed
<robimarko> And there should be nvmem sysfs entry to provide you the binary
<aiyion> Well, the device is running a pathced master from like six days ago, `/sys/bus/nvmem/devices/sunxi-sid0/` does exist, but does only contain power, uevent and two symlinks.
* aiyion is afk for 30m
danitool has quit [Ping timeout: 480 seconds]
bluew has quit [Quit: Leaving]
bluew has joined #openwrt-devel
<lucenera> What could such a thing come from? https://clbin.com/UjhU4
<lucenera> What does unknown option mean in that context?
dangole has joined #openwrt-devel
<Znevna> after so much spam from that stintel impersonator I've read @zone[1].stintel...
<robimarko> aiyion: Beats me why the binary interface is missing
<lucenera> '-_-
danitool has joined #openwrt-devel
<philipp64> who is the maintainer for target/linux/x86?
<robimarko> There are no target maintainers
<Tapper> philipp64wotch out for this bug. https://github.com/openwrt/openwrt/issues/11625
<robimarko> Tapper: Can you try saving the config, I have a feeling that not saving only removes the symbols and does not add new defaults
<Tapper> robimarko sorry what do you meen?
<robimarko> Tapper, once you are asked to save or not save the config
<robimarko> You said that you selected to not save it
<robimarko> How about trying to save it
<Tapper> I was not asked.
<Tapper> If you don't make changes it will not ask you on exit
<robimarko> Thats the thing, doesnt matter if you make any changes
<robimarko> If there is a diff between generic and target config it will ask you to save
<Tapper> It did not ask mate.
<Tapper> I will test again if you would like me to, but let me eat this food.
<Tapper> lol
<Tapper> brb
<robimarko> No worries, this is just new to me
<aiyion> robimarko: I'll try it with a recent release or nightly.
<freistil90> Hi, I have a problem compiling a package with the SDK
<freistil90> I'm trying to build a custom package with the SDK and I get issues with the index step
<freistil90> Something along the line "key-build was not found"
<freistil90> Currently running `make V=sc` to see what the exact problem is, can't paste the actual error here.
<freistil90> (until that has finished)
<freistil90> O
<freistil90> I'm a bit puzzled about what to do.
<freistil90> I did: 1) download the SDK and unpack it
<freistil90> 2) Create some local folder and put the makefile in it
<freistil90> 3) added `src-link local /path/to/Packages` to my feeds.conf.default
<freistil90> 4) Run `./scripts/feeds update -a`
<freistil90> 5) Now link build process `./scripts/feeds install <package name>`
<freistil90> and then 6) `make -j $(nproc) download world`
<freistil90> The problem goes into the direction of https://forum.openwrt.org/t/cant-compile-ipk-with-sdk-key-build-signing-key-is-missing/2266 but I'm not sure what to do with it (as also the links are not working anymore)
<freistil90> For completeness, the Makefile I'm trying to build is https://github.com/stangri/source.openwrt.melmac.net/blob/master/taskserver/Makefile
<freistil90> A bit of advice would be appreciated :)
<freistil90> ... okay, it worked. This was missing: `./staging_dir/host/bin/usign -G -s ./key-build -p ./key-build.pub -c "Local build key"`. This is missing from the guides maybe...
<aiyion> robimarko: oh fuck -.-' 22.03.2 does still have nvmem
<aiyion> I'll go and rebuild that locally to make sure its there as well.
rua has quit [Remote host closed the connection]
<xdarklight> aiyion: there's a CONFIG_NVMEM_SYSFS kernel config option. is it enabled in your kernel?
rua has joined #openwrt-devel
<robimarko> Looks like it isnt enabled for Sunxi
<aiyion> xdarklight: I stumbled across that as well. I'm not sure yet, is on my list.
<aiyion> looks like a regression then?
<robimarko> Well, it isnt set in the sunxi 5.10 config as well
<robimarko> Its only set in the subtarget config
<robimarko> A53 has it set
<robimarko> But A7 doesnt
<aiyion> that file has nvmem sysfs access.
<robimarko> No idea how
<robimarko> As even in 22.03 its not enabled in the target config
<robimarko> Nor the A7 subtarget config
<aiyion> Is something introducing it as dependency or sth?
<aiyion> That flag is not new, is it?
<robimarko> Nope
<robimarko> 5.10 has it as well
<aiyion> huh.
<robimarko> Nanopi R1 is A7 isnt it?
<aiyion> yes
<aiyion> Allwinner H3, Quad-core Cortex-A7
<robimarko> Then no idea how its getting enabled in 22.03
<robimarko> As its clearly not enabled in kernel config
<robimarko> It should just be enabled in the target config for all SoC-s instead of just in A53
<robimarko> BTW, I just learned one crazy thing
<robimarko> FW for WCN6855 is getting stripped during building
<philipp64> Tapper: I'm wondering why the x86 doesn't have pciutils and usbutils packaged by default, and then just remove them for special targets...
<robimarko> So, does anybody know how to disable stripping per package
<robimarko> As its clearly wrong to strip linux-firmware
<philipp64> Not that we have any special targets right now... they've all been done away with.
<hauke> robimarko: set STRIP:=:
<hauke> see package/firmware/layerscape/ppfe-firmware/Makefile
<robimarko> hauke: Thanks
<hauke> which FW gets stripped?
<robimarko> WCN6855 one
<robimarko> Both amss.bin and m3.bin
<robimarko> I am sure there will be more
<robimarko> As any ELF binary will get stripped
<hauke> yes we should deactivate it for all FW files
<hauke> often FW files are not in elf format, but if they are it causes problems
<robimarko> Yeah, this is the first time I have hit this
<robimarko> Usually its just binary data, but it looks like FW for ath11k PCI cards are just ELF
Borromini has joined #openwrt-devel
Borromini has quit [Quit: Lost terminal]
robimarko has quit [Quit: Leaving]
<aiyion> a shame he's gone. I think I found the difference.
<aiyion> CONFIG_ALL_KMODS=y
<aiyion> not sure though whether a kmod can introduce a the sysfs setting
<hauke> aiyion: a kmod can add arbitary kernel settings
<aiyion> My build is still ongoing, but that is currently the only lead I have.
<aiyion> Question is, is it even worthwhile to find how the flag is enabled in releases; and furthermore, is it harmful to enable it for sunxi in general?
<hauke> aiyion: I think we can activate it sunxi in general
<hauke> all devices have biug flash chips
<aiyion> Will submit a patch once I've found the respective file to flip the flag.
minimal has joined #openwrt-devel