Slimey has quit [Quit: AdiIRC is updating to v4.1]
Ansuel_ has quit [Quit: Probably my PC crashed or time to sleep.]
floof58_ has joined #openwrt-devel
Slimey has joined #openwrt-devel
floof58 has quit [Ping timeout: 480 seconds]
dansan has joined #openwrt-devel
hitech95 has quit [Remote host closed the connection]
minimal has quit []
zx2c4 has quit [Ping timeout: 480 seconds]
hubvu has quit [Ping timeout: 480 seconds]
zx2c4 has joined #openwrt-devel
danitool has quit [Ping timeout: 480 seconds]
zx2c4 has quit [Ping timeout: 480 seconds]
zx2c4 has joined #openwrt-devel
victhor has quit [Ping timeout: 480 seconds]
philipp64|work has joined #openwrt-devel
hubvu has joined #openwrt-devel
<JiiPee> Anyone to comment this https://github.com/openwrt/openwrt/pull/4671
<JiiPee> should I just roll back led and network changes and separate them?
<JiiPee> even that they are clone devices
<JiiPee> rest of the combined changes should be fine because they are specific for jboot devices
dansan has quit [Quit: The C preprocessor is a pathway to many abilities some consider to be unnatural.]
<genuser1> slh its late and my package finished compiling for the second time. I am curious, when I switch to the master branch I can see the wac510 option. When I switch to any of the 21.02.0 branch wac510 is not an option. Where is the "switch" or parameter that determines if the wac510 (or any other device) is available for a branch? Thanks.
<JiiPee> genuser1: if it wasn't supported when 21.02 was released, then it's not available when you use 21.02 branch unless you backport it yourself to your fork
felix_ has quit [Ping timeout: 480 seconds]
felix has joined #openwrt-devel
felix_ has joined #openwrt-devel
felix has quit [Ping timeout: 480 seconds]
rua has quit [Quit: Leaving.]
felix has joined #openwrt-devel
felix_ has quit [Ping timeout: 480 seconds]
felix_ has joined #openwrt-devel
felix has quit [Ping timeout: 480 seconds]
felix has joined #openwrt-devel
felix_ has quit [Ping timeout: 480 seconds]
goliath has joined #openwrt-devel
goliath has quit [Quit: SIGSEGV]
rua has joined #openwrt-devel
pmelange has joined #openwrt-devel
goliath has joined #openwrt-devel
dedeckeh has joined #openwrt-devel
pmelange has left #openwrt-devel [#openwrt-devel]
dedeckeh has quit [Remote host closed the connection]
lmore377 has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
pmelange has joined #openwrt-devel
lmore377 has joined #openwrt-devel
Tapper has joined #openwrt-devel
goliath has quit [Quit: SIGSEGV]
rua has quit [Quit: Leaving.]
dangole has joined #openwrt-devel
_lore_ has quit [Ping timeout: 480 seconds]
rmilecki has joined #openwrt-devel
robimarko has joined #openwrt-devel
<robimarko> Hi, is anybody working on 5.15 support for OpenWrt?
<robimarko> I am currently playing whack a mole with ath11k as it depends on a lot of stuff that is newer than 5.10
<hauke> robimarko: I wanted to look into kernel 5.15 to integrate the Allwinner D1, there are only patches available based on 5.14 for the D1 SoC
<hauke> but it will probably take some time till I start
<hauke> robimarko: in which component are all these changes needed for ath11k?
<robimarko> hauke: Well, MHI and remoteproc are the biggest one
<robimarko> But relayfs changed in 5.11, that was easy to backport though
<robimarko> There is other stuff I probably forgot but had to backport
victhor has joined #openwrt-devel
goliath has joined #openwrt-devel
<hauke> robimarko: hmm, so it is spread over multiple subsystems, if it would only be one used only for QCA stuff we could just backport the complete subsystem
Acinonyx_ has joined #openwrt-devel
<hauke> robimarko: does ath11k for PCIe cards have the same problem?
Acinonyx has quit [Ping timeout: 480 seconds]
<robimarko> hauke: Yeah, PCI is especially problematic as it uses MHI for controlling it
<robimarko> And that is really bad in 5.10
<robimarko> Issue is that you cant just backport ath11k like other drivers
<robimarko> But you also need to backport the changes to other subsystems it uses
<robimarko> So remoteproc and MHI are number one
<robimarko> So for PCI variant you need to backport MHI, QRTR is also kind of broken in 5.10
<hauke> robimarko: for the braodcom b43 driver we backproted the complete ssb and bcma subsystem from a recent kernel to our kernel, but that was only used for broadcom stuff and notthing generic
<hauke> with ath11k it looks like it is different
<robimarko> Yeah, PCI cards are becoming more spread and they can be used on any target
hanetzer has quit [Quit: WeeChat 3.3]
Tapper has quit [Ping timeout: 480 seconds]
hanetzer has joined #openwrt-devel
_lore_ has joined #openwrt-devel
_lore_ has quit [Read error: No route to host]
_lore_ has joined #openwrt-devel
DLange has quit [Quit: \o/ CU soon]
DLange has joined #openwrt-devel
pmelange1 has joined #openwrt-devel
pmelange has quit [Read error: Connection reset by peer]
pmelange has joined #openwrt-devel
pmelange1 has quit [Ping timeout: 480 seconds]
spacewrench has joined #openwrt-devel
<robimarko> hauke: I started on 5.15
<robimarko> Unfortunatelly most of the hacks that we are carrying for years wont apply
<robimarko> Got patches aplying until 613 in pending
dangole has quit [Ping timeout: 480 seconds]
Tapper has joined #openwrt-devel
danitool has joined #openwrt-devel
Ansuel has joined #openwrt-devel
<Ansuel> mangix: e9hack reported some problem with the updated pr. did you notice anything strange on your side?
rmilecki has quit [Ping timeout: 480 seconds]
<mangix> Nope.
<mangix> I found something interesting though
<mangix> Out of curiosity, I added led-open-drain and saw Ethernet performance suffer. No idea why.
kabel has quit [Remote host closed the connection]
<Ansuel> o.O
<Ansuel> mangix: then i can't understand the problem that e9hack is having
<mangix> He has ignore-power-on-cel missing too.
rua has joined #openwrt-devel
shibboleth has joined #openwrt-devel
<Ansuel> it seems his device is very sensible to rx / tx delay
<Ansuel> i just discovered we can actually configure lags on qca8k O_O
<Ansuel> i mean there is only one reference in the entire documentation but i can see how that can be configured
<robimarko> Ansuel: Seriously it has LAG support?
<Ansuel> robimarko: there is some reference in the ACL table wonder if it does support a very basic type?
<robimarko> Oh, they actually say "Trunking support" in the features section of the 8337 datasheet
<Ansuel> i'm checking if i can implement the multicast table for dsa and it looks like it all works putting entry in the ACL table
<robimarko> But they dont have any other mentions other than that
<Ansuel> Indicate which ports are associated with this MAC address when they are
<Ansuel> set to 1. E.g. bit[48] is assigned to port0, bit[49] is assigned to port1, bit[50]
<Ansuel> to port2, etc. If all bits are set to 0 and the entry is static, the packet should
<Ansuel> be dropped. For multicast address and unicast for link aggregation, more
<Ansuel> than one bit is set to 1.
<Ansuel> this is in the arl table
<robimarko> Ok, I see it now
<Ansuel> i was starting to write some code to implement multicast offload
<Ansuel> and i just notice that
<Ansuel> the ARL table seems OP
<Ansuel> also should not be that difficult to implement (the multicast part) i just need to exapnd the current fdb part...
<robimarko> Doesnt DSA have MDB functions?
<robimarko> Just for this?
<Ansuel> yes and i want to implement them
<Ansuel> there is mdb join and leave
<robimarko> Ok, that is what I was thinking
<robimarko> Its essentially offloading IGMP snooping to the HW
<Ansuel> yes and it's all handled by arl just declaring how the switch should forward the packet
<Ansuel> but with that reference to link aggregation i think arl can also be used for lags
<robimarko> Awesome, the N variant also can do some primitive HNAT for IP/GRE and handfull of protocols
<robimarko> And it uses the ARL as well for that
<Ansuel> yes all the hnat stuff is present in the qsdk
<Ansuel> but i'm scared to work on that...
<robimarko> But its limited to I think 1024 entries or so
<robimarko> Honestly, its probably slower than the CPU doing it
<Ansuel> the ssdk have all the code for that but it's really an hack as it's a bit hardcoded
<Ansuel> this switch can even offload pppoe lol...
<robimarko> Yeah, I mean its in the datasheet so its not all black magic
<robimarko> Yeah
<robimarko> DSA should technically be able to support hat
<Ansuel> do you think the switch will be slower ? lol
<robimarko> There is an offload function, but no idea how to use it
<Ansuel> mhh can't see any api for that in dsa
<robimarko> Well, it only has 1024 entries
<Ansuel> ow really?
<Ansuel> do you have any link to this offload function ? i'm curious
<robimarko> Its basically hooking up TC flower
<robimarko> Which the the current standardised way of offloading filters etc to HW
<robimarko> I think that Hellcreek and Ocelot support some of it
<robimarko> They are the only ones I know
<Ansuel> oww the tc api... still a mistery to me :/
<robimarko> Trust me, I find it to be black magic
<Ansuel> anyway tell me if i'm wrong... the lag in theory should have a mac address so to join the lag we should just create and acl rule with the port set
<robimarko> I honestly dont know
<robimarko> Newer had to implement it
dangole has joined #openwrt-devel
<Ansuel> they probably have this crazy idea of implementing multi-cpu support creating a lag
<Ansuel> across the 2 cpu port
shibboleth has quit [Remote host closed the connection]
shibboleth has joined #openwrt-devel
wvdakker1 has joined #openwrt-devel
wvdakker has quit [Ping timeout: 480 seconds]
rmilecki has joined #openwrt-devel
<genuser1> Is there documention/links somewhere that explains all the options in menuconfig and kernel_menuconfig? Trying to just build a working image for the netgear wac510 with luci baked in and I am getting overwhelmed with all the options. I only plan to use my wac510 as a dumb ap and managed it with luci.
<Slimey> in make menuconfig just turn on luci
<Slimey> or download the 21.02.0 release it should have luci baked in
<Habbie> mangix, everything works, with just the two PRs, no patch
<genuser1> 21.02.0 doesn't "support" the netgear wac510 yet
<genuser1> I have to build against the master
<Slimey> ah
<Slimey> its under luci > collections > luci then
<Habbie> mangix, well, i only tried one LAN port, will test the others soon
<genuser1> I just checked mine, seems the option M is wrong...should be a *
<Slimey> i need to submit my pr also
<Slimey> a few things to work out first
f00b4r0 has quit [Read error: Connection reset by peer]
f00b4r0 has joined #openwrt-devel
bluew has joined #openwrt-devel
hanetzer has quit [Ping timeout: 480 seconds]
hanetzer has joined #openwrt-devel
hgl has quit [Remote host closed the connection]
hgl has joined #openwrt-devel
hgl has quit [Read error: Connection reset by peer]
hgl has joined #openwrt-devel
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
<slh> genuser1: you're apparently confusing "make menuconfig" with "make kernel_menuconfig", you shouldn't need the later, which gives you very fine grained access to kernel options, not target or package specifics
rua has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
<genuser1> slh yes and no. I was sure the difference between the two. I just wasnt sure what all the options/features were and if I really should be changing much. No worries either way :). I have compiled the master branch a couple of times now but I am still unable to upgrade the base-files package. it is still complaining about a read only system. Any pointers?
<russell--> genuser1: df -h show an overlay?
robimarko has quit [Quit: Page closed]
<genuser1> russell hopefully this pastebin works https://pastebin.com/LAnzGAvX
<genuser1> thats df -h
<russell--> that looks reasonable
<russell--> what does "still complaining" look like?
<genuser1> russell hopefully this image goes through https://ibb.co/ydkZrnz
<russell--> hmm. i never (very rarely) update packages except by building new firmwares
<russell--> especially not sure about base-files
<russell--> the expeditious thing would be to rebuild a whole firmware
<hauke> genuser1: netgear wac510 is not supported in OpenWrt 21.02 it is only supported in master
<genuser1> hauke I understand it is only supported in master. Does that also mean that any packages showing updates cannot be upgraded during operation?
<genuser1> This is my first time trying to compile any kind of openwrt so I am learning all this as I go
<Habbie> there's a master/snapshot package repo
<Habbie> which works fine in my experience except during big transitions
<Habbie> like last month when the libc (musl) got upgraded
<Habbie> but in that case building a new image is wise anyway
<genuser1> So that's a question...if I build the image, then load it and it already shows those packages have updates...how do I rebuild with those updates?
<Habbie> ah, then the feeds in your build environment might be outdated
<Habbie> run 'scripts/feeds' and read the help text
shibboleth has quit [Quit: shibboleth]
pmelange has left #openwrt-devel [#openwrt-devel]
<russell--> genuser1: i use something like this: git checkout master ; git pull ; scripts/feeds update -a ; scripts/feeds install -a
rmilecki has quit [Ping timeout: 480 seconds]
<genuser1> thanks all. I guess I figured if I did the feeds update and install once it was good enough. Didnt realize I should do it before every build.
<Tapper> genuser1: for a build that is for a AP then you should do make menuconfig set the make and model of your router. then remove dnsmasq, firewall, odhcp and odhcp6
<Tapper> cool
<genuser1> tapper thats funny you say that because I had to disable those services anyways for my other owrt dumb aps. facepalm on myself for not thinking that
<Tapper> under luci I pick luci-base, luci-mod-admin-full, luci-mod-network, luci-mod-status, luci-mod-system and luci-theme-openwrt-2020. If you pick luci-ssl you will not beable to remove luci-app-firewall and things.
<Tapper> If you pick luci packages with out using one of the collections then don't forget to add uhttpd
<genuser1> tapper awesome. Gonna save those notes and rebuild :)
Tapper has quit [Ping timeout: 480 seconds]
goliath has quit [Quit: SIGSEGV]
dangole has quit [Remote host closed the connection]