beholders-eye has quit [Ping timeout: 480 seconds]
<slh>
digitalcircuit: I'll try to run your PR on my nbg6817 this weekend, no promises yet - I need to juggle quite a few pending/ unmerged PRs to retain a 'working' image (qca8k-DSA for ipq806x, ipq40xx, ath79; ipq807x, etc.) - but it's great that you found something
victhor has quit [Remote host closed the connection]
rsa9000 has quit [Remote host closed the connection]
Tapper has quit [Read error: Connection reset by peer]
Tapper has joined #openwrt-devel
jow_ has joined #openwrt-devel
jow has quit [Read error: Connection reset by peer]
<rsa9000>
He only notified me that the new option is awailable
<Grommish>
to which you answered that's good news, but you didn't fix it?
<rsa9000>
There are actually nothing to fix. Switch perfectly handles LED
<rsa9000>
Do we need to make 'everything software'?
<Grommish>
that's my best guess, you could always ask him *shrug*
<rsa9000>
Besides that, I check other ramips DTS and no one utilize the new switch GPIO interface.
<stintel>
that's no reason to use it i.e. if the new switch GPIO interface was only recently added ...
<russell-->
you have to options available: 1) be patient; 2) poke people who might apply it
rua has quit [Ping timeout: 480 seconds]
<Grommish>
Ah, so you'd rather have someone else have to fix it later? :D
<rsa9000>
Nope
<stintel>
as for the patch you used as example that was accepted in a few hours, it's probably also familiarity with the target
<stintel>
you can git log --author=chunkeey -- target/linux/ath79 vs git log --author=chunkeey -- target/linux/ramips and you'll see what I mean
<rsa9000>
Srommish: nope, I think that the switch themself perfectly handle port indication
<Grommish>
If there was a commit for it, then whoever submitted it doesn't seem to think so.
<rsa9000>
Grommish: also he do not bother to convert some boards to the new interface
<Grommish>
But, I digress. in any case, "Rosen" who commented on your PR is mangix in this chat. I'm sure he'll be able to tell you why it's yet to be processed, be it because it was holidays or whatever
<rsa9000>
or introduce a new board that utilizes it
<rsa9000>
so I assume that this interface will be used latter to something other than a LED control
<Grommish>
But for me, if someone goes out of there way to link to a commit, saying "this issue was fixed", and it related to whatever I'm working on, I'm probably going to put it in
<russell-->
luckily, git makes it relative easy to carry your support-adding patch while you are waiting for it to be applied
<Grommish>
russell--: or when they never will be applied :D
<rsa9000>
:)
<russell-->
distributed version control++
<stintel>
I would also suggest to test with v5.10 kernel and mention that - we're in the process of moving all targets to 5.10
<Grommish>
My device is triple isolated boot, OpenWrt only allows 1 of the 3 slots, I maintain the other 2 locally
<stintel>
if you add a device which is not tested with 5.10 this might be enough reason to ignore the patch
<Grommish>
it sits with gold-ld linker and some zram crapz from rsalvaterra that will probably never make it into the tree
* russell--
has one of those never-will-be-applied (for a wdr3600 with a bigger flash part soldered on)
<stintel>
:P
<rsa9000>
stintel: is absense of 5.10 compatibility statement is the only reason to ignore a patch?
rua has joined #openwrt-devel
<russell-->
there are many reasons
<Grommish>
this sysupgrade file is 77mb hahaha ugh
<russell-->
the most common reason is probably that people are preoccupied with other things
<rsa9000>
russell--: that is why I asked for expected time of applying
<rsa9000>
i do not want to bother dev with everyday pings
<stintel>
a ping once a week seems fine by me
<rsa9000>
on the other hand waiting for a couple of weeks without any response looks like the work will never be applied
<stintel>
I would review it myself, but I'm too busy with other things (infra, firewall4, qoriq, testing mt7613 stability0
<Grommish>
If they can't tell you what's wrong with it, then they should tell you why it can't be accepted
<stintel>
also I send DTS stuff for review myself, because I'm not too experienced with it
<Grommish>
Hot damn it works!
Borromini has joined #openwrt-devel
<rsa9000>
stintel: review is best what was invented by a human race :)
<russell-->
someone will eventually get bored and/or embarrassed after multiple months, don't give up hope
<stintel>
hanetzer: when 's your eap615 due to arrive ?
<hanetzer>
uh, lemme check.
<hanetzer>
monday
<russell-->
rsa9000: finding someone else with the hardware to test your patch can also help
<rsa9000>
help with applying?
<SwedeMike>
rsa9000: regarding yesterdays DTS discussion, I am against describing hardware in a file regardless, I want discovery to work.
<rsa9000>
SwedeMike: there are no autodiscovery option for this board
<rsa9000>
either i create a full dts or i have a partially working router
<SwedeMike>
rsa9000: I know, it was a general complaint about MIPS/ARM etc
<rsa9000>
choise is simple in this particular case
<rsa9000>
SwedeMike: vendors will never invest their resources in the ability to boot a 3rd party firmware on their device
<rsa9000>
SwedeMike: otherwise their become a noname hardware manufacturers
<SwedeMike>
rsa9000: I know vendors who do intend to let people do that, who don't have it either. Is there even such a thing in ARM-land?
danitool has joined #openwrt-devel
<rsa9000>
SwedeMike: almost no
<rsa9000>
SwedeMike: this is not a MIPS-land or ARM-land issue
<rsa9000>
this is more a device class issue
<rsa9000>
Universal PC survive due to its ability to run any software
<rsa9000>
Highly specialized devices survive due to its ability to do their work very well
<rsa9000>
How many people will ask for an ability to run a custom firmware on thier microwave own?
<rsa9000>
*owen
* hanetzer
gives a side-eye to that reflow oven people seem to like
<rsa9000>
summing up the discussion, can I say that there are no estimate for patch applying time and the only option for submitter is the weekly ping?
<jow>
yes
<rsa9000>
ok, will keep in mind
<rsa9000>
:D
Borromini has quit [Quit: Lost terminal]
<hanetzer>
while you're pinging, don't forget your pongs :P
clayface has joined #openwrt-devel
<hanetzer>
stintel: eap610 appears to be interesting from the perspective of something to replace ubiquiti-style poe APs
<\x>
hi, I wonder now that on DSA devices we dont see port status on luci anymore, is there a replacement to that switch panel atleast just for port status?
clayface_ has quit [Ping timeout: 480 seconds]
victhor has joined #openwrt-devel
rua has quit [Quit: Leaving.]
minimal has joined #openwrt-devel
rua has joined #openwrt-devel
dedeckeh has joined #openwrt-devel
goliath has joined #openwrt-devel
rua has quit [Quit: Leaving.]
rsalvaterra has quit [Ping timeout: 480 seconds]
rua has joined #openwrt-devel
rsalvaterra has joined #openwrt-devel
pmelange has joined #openwrt-devel
pmelange has left #openwrt-devel [#openwrt-devel]
rsa9000 has left #openwrt-devel [#openwrt-devel]
<hanetzer>
stintel: huh, just got another look at the wall variant. didn't realize the 'internet in' port was on the back of it :P
<stintel>
I just hung mine off a bookshelf :D
<stintel>
the 235 model that is
<hanetzer>
noted.
beholders-eye has joined #openwrt-devel
nmrh has joined #openwrt-devel
victhor has quit [Remote host closed the connection]
<hanetzer>
I suppose there's nothing preventing you from reconfiguring the internet in port to out, eh?
<stintel>
PoE
<stintel>
other than that, you can configure it how you want
<stintel>
but PoE is hardwired
<hanetzer>
yeah, there *is* that. but the poe on the internet in port is also poe in, not out, so wouldn't fry anything eh?
<stintel>
I'm not following ?
<hanetzer>
I may be assuming. Does the wall you have, have a power-in port that's not the poe? as in, normal barrell jack?
<stintel>
no
<stintel>
but same with U6-Lite or U6-LR afair
<hanetzer>
huh. guess I went crosseyed and saw the reset button as a power port :P
<stintel>
:D
<stintel>
I hope you have a poe switch :D
<stintel>
well or injector
<hanetzer>
three of them.
<stintel>
ok no harm done then :)
<hanetzer>
may make dev a bit annoying tho. since the cable that feeds my pc is from one of those POE switches, will have to cable shuffle a bit.
wb9688 has joined #openwrt-devel
<wb9688>
What would be the recommended way to package a Node.js application for OpenWrt?
nickaser has joined #openwrt-devel
<nickaser>
Does WRT1900AC v1 allows card replacement? It loooks like it has mini pcie on it, can I just replace the card with an AX wifi enabled one?
Tapper has quit [Ping timeout: 480 seconds]
<will[m]>
wb9688: do you know how to create a basic OPKG package?
<hanetzer>
nickaser: good question. one would think it could, but you may have to install a different kmod.
<nickaser>
I see that ax200 is supported, can I use it to make it an AX capable router?
<stintel>
intel firmware is very limited in AP mode, if it is supported at all
<stintel>
I don't recommend it
<wb9688>
will[m]: Yes, I know that
<nickaser>
Thanks for the information.
nickaser has quit [Quit: Page closed]
<wb9688>
But with Node.js applications you of course have hundreds of libraries it depends, so I was wondering how OpenWrt would generally handle that, and if OpenWrt even has Node.js
<will[m]>
Ah that's the part I don't know. Start installing OPKG stuff and see if you can make it run I guess
pmelange has joined #openwrt-devel
pmelange has left #openwrt-devel [#openwrt-devel]
Tapper has joined #openwrt-devel
nmrh has quit [Ping timeout: 480 seconds]
<jow>
wb9688: node.js is one of the worst imaginable language runtimes for typical OpenWrt targets
<jow>
the hard requirement for floating point math, nan boxing tricks, the overall bloatyness of the entire ecosystem, vendoring issue due to impossible to pacakge npm dependency hell etc.
<wb9688>
I definitely understand that, though I'm running OpenWrt on a Raspberry Pi
<jow>
aside from that, there are nodejs packages for the few targets that support tham
<jow>
*them
<wb9688>
But yes, the dependency hell sucks
<jow>
my packaging approach would likely be running npm install etc. on the host, then somehow package the resulting node_modules/ as-is
<jow>
and use a version-locked packages.json to make the build reproducible
<jow>
(and pray that the pulled in stuff still is available and free of severe security issues three weeks later)
<PaulFertser>
btw, this ucode, was it just an experiment or do you have some plan about it?
<jow>
PaulFertser: mid to long term I plan to reimplement the remaining Lua parts of LuCI in ucode
<jow>
currently working on a ucode scripting host for rpcd as well to be able to replace the various libexec plugins used by luc
<jow>
*LuCI
<jow>
and firewall4 uses it for templating the nft ruleset
<PaulFertser>
jow: I see. It's quite an unorthodox decision to implement a new script language but since it works for you I guess it's that rare case where there's a clear win by being inventive.
<jow>
yes, unorthodox indeed
<PaulFertser>
The interpreter looks surprisingly minimal for the level of syntax it has to offer.
<jow>
I was looking for something with C / ECMAScript like syntax, non-async code flow, a Perl-esque stdlib (just a few dozen basic functions like split(), length() etc.), PHP style templating and the "smallness" of Lua without the associated syntactic verbosity, builtin bitops and very simple way to read and write json
<PaulFertser>
For those wondering what we're talking about, example: https://git.openwrt.org/?p=project/firewall4.git;a=blob;f=root/usr/share/firewall4/templates/ruleset.uc;h=e67fd7eca861f820e19272351a6546ea6c17eccc;hb=HEAD
<jow>
and for fw4 in particular I didn't want another C implementation since rendering rulesets from uci is essentially just doing lots of array and dictionary juggling, something that is very tedious to do in C
<jow>
and I wanted the "builtin" ruleset template to be editable
<jow>
hence the need for a more template-oriented approach
<jow>
wanted to write down all those considerations as preamble to the documentation but didn't get around to it yezt
<jow>
*yet
beholder has joined #openwrt-devel
beholders-eye has quit [Ping timeout: 480 seconds]
Tapper has quit [Ping timeout: 480 seconds]
Strykar has quit [Quit: /quit]
Strykar has joined #openwrt-devel
dedeckeh has quit [Remote host closed the connection]
<PaulFertser>
I'd be also curious to learn how you ended up writing your own lexer and parser; flex+bison seems to be an established solution and they do not seem to impose any runtime dependencies and probably no significant overhead.
<jow>
ecmascript syntax has various syntactic corner cases which are hard to disambiguate using context-free LR(1) grammar (like { ... } as block scope vs. { ... } as object literal)
<jow>
and the parser needs to drive the lexer to pass context information (e.g. is the next "/" a division operator or the beginning of a regexp literal)
<jow>
or is the next reserved word an identifier or a statement keyword
<jow>
this can be done with bison and lex but then you end up e.g. duplicating grammar rules to disambiguate and in general you quickly start to bend around the tools
<PaulFertser>
Understood, thank you for the clarifications!
<jow>
(I guess a (un)healthy portion of nih syndrome is involved too...)
<PaulFertser>
That's a well-justified effort and looks like you managed to hit the sweet spot and had fun doing that so that's a win either way.
beholders-eye has quit [Read error: Connection reset by peer]
beholders-eye has joined #openwrt-devel
lmore377 has quit [Quit: No Ping reply in 180 seconds.]
beholders-eye has quit [Remote host closed the connection]
beholder1 has quit [Ping timeout: 480 seconds]
lmore377 has joined #openwrt-devel
<Borromini>
enyc: serious bugfixes and security holes?
beholder has quit [Ping timeout: 480 seconds]
beholder has joined #openwrt-devel
beholder1 has joined #openwrt-devel
beholder has quit [Ping timeout: 480 seconds]
beholder1 has quit [Ping timeout: 480 seconds]
nmrh has joined #openwrt-devel
beholder1 has joined #openwrt-devel
Lynx- has quit [Quit: Going offline, see ya! (www.adiirc.com)]
beholder1 has quit [Read error: Connection reset by peer]
beholder1 has joined #openwrt-devel
beholder1 has quit [Remote host closed the connection]
beholder1 has joined #openwrt-devel
beholder1 has quit [Read error: Connection reset by peer]
Borromini has quit [Quit: Lost terminal]
beholder1 has joined #openwrt-devel
beholder1 has quit [Read error: Connection reset by peer]
beholder1 has joined #openwrt-devel
beholder1 has quit [Read error: Connection reset by peer]
beholder1 has joined #openwrt-devel
beholders-eye has joined #openwrt-devel
beholder1 has quit [Read error: Connection reset by peer]
beholder has joined #openwrt-devel
beholder1 has joined #openwrt-devel
beholder has quit [Remote host closed the connection]
beholder has joined #openwrt-devel
beholder1 has quit [Read error: Connection reset by peer]
beholder has quit [Read error: Connection reset by peer]
beholder has joined #openwrt-devel
beholders-eye has quit [Ping timeout: 480 seconds]
beholder has quit [Remote host closed the connection]
beholder has joined #openwrt-devel
beholder1 has joined #openwrt-devel
beholder1 has quit [Read error: Connection reset by peer]
beholder has quit [Read error: Connection reset by peer]
beholder1 has joined #openwrt-devel
beholders-eye has joined #openwrt-devel
beholder1 has quit [Read error: Connection reset by peer]
beholder has joined #openwrt-devel
beholder1 has joined #openwrt-devel
beholders-eye has quit [Ping timeout: 480 seconds]
beholder1 has quit [Remote host closed the connection]
beholder has quit [Ping timeout: 480 seconds]
beholder1 has joined #openwrt-devel
beholders-eye has joined #openwrt-devel
beholder1 has quit [Read error: Connection reset by peer]
beholder has joined #openwrt-devel
beholder1 has joined #openwrt-devel
beholder has quit [Read error: Connection reset by peer]
beholder has joined #openwrt-devel
beholder2 has joined #openwrt-devel
beholder has quit [Read error: Connection reset by peer]
beholders-eye has quit [Ping timeout: 480 seconds]
beholder2 has quit [Remote host closed the connection]
beholder1 has quit [Ping timeout: 480 seconds]
beholder2 has joined #openwrt-devel
beholder2 has quit [Remote host closed the connection]
beholder2 has joined #openwrt-devel
beholder2 has quit [Read error: Connection reset by peer]
beholder2 has joined #openwrt-devel
Tapper has joined #openwrt-devel
beholder2 has quit [Remote host closed the connection]
beholder2 has joined #openwrt-devel
<hauke>
enyc: 21.02.2 is done either if there is a serious security problem or when too much time passed since the last release and someone thinks it should be done
beholder2 has quit [Read error: Connection reset by peer]
beholder2 has joined #openwrt-devel
<hauke>
beholder2: please fix your client
beholder2 has quit [Read error: Connection reset by peer]
beholder2 has joined #openwrt-devel
beholders-eye has joined #openwrt-devel
<enyc>
hrrm i see
beholder2 has quit [Ping timeout: 480 seconds]
<hanetzer>
hauke: they blinkin? lotta eyes to do that with ;P