guerby has quit [Remote host closed the connection]
Rentong has quit [Remote host closed the connection]
guerby has joined #openwrt-devel
Rentong has joined #openwrt-devel
Rentong has quit [Ping timeout: 480 seconds]
Rentong has joined #openwrt-devel
Rentong has quit [Ping timeout: 480 seconds]
goliath has quit [Quit: SIGSEGV]
<digitalcircuit>
Missed the "CONFIG_ARM_QCOM_CPUFREQ_KRAIT=y" line my first time; I've properly git cherry-pick'd the CPU governor changes now. First backup running (haven't yet figured out the exact stress-ng parameters, I got all sorts of exciting new crashes), and I'll update once I've got a notable result. Fingers crossed!
Rentong has joined #openwrt-devel
<digitalcircuit>
I'm realizing that with 21.02.0-rc3 already out the door, if backporting the ipq806x CPU governor works, it's unfortunately probably unlikely we can get this in 21.02.0. I'd be happy to try to do additional testing if there's any way I can help.
<russell-->
the ended up settling, and cambium paid an undisclosed sum, which is weird and unexplained
<russell-->
they*
* russell--
got a hint that some of the dispute was over airview, but just wild guessing
<zorun>
so, cambium settled/paid *and* stopped doing it?
<russell-->
i gather they EOL'd the product but let existing users continue using it or something
Rentong has joined #openwrt-devel
<aparcar[m]>
zorun: sorry for the worker mess then, I'll connect to the sunshine admin
<russell-->
apparently boiled down to 18 files, some of which were u-boot and ubiquiti's changes which they won't disclose or something.
<russell-->
the court ordered ubiquiti to disclose the files, but on a non-networked computer to only attorneys with windows and horrible sounding tools for inspection
<russell-->
*shudder*
<russell-->
i am embarrassed to say how many words i just read and how little they said
<djStolen>
PaulFertser: would djStolen be possible?
<djStolen>
Email is djstolen69@gmail.com
<djStolen>
PaulFertser: regarding the device. Actually, I think that pics there show the same device as mine (pics are really low quality though) but the flash chip and ram chip are GigaDevice and EtronTech not Spansion/Winbond or Nanya as shown. Also found specs for chips in my device.
<djStolen>
PaulFertser: I would give my best with pics as soon as I get some time
rejoicetreat has quit []
<PaulFertser>
djStolen: user, added, check your mail. I do not think specific vendor of flash and RAM chips is essential, there're plenty of interchangable chips.
<djStolen>
PaulFertser: thanks, will check asap. Jup, I agree, was just puzzled with all discrepancies, and the fact that breed is not open source, and of course the bricking of the device even though followed instructions
Rentong has quit [Ping timeout: 480 seconds]
<PaulFertser>
djStolen: do you already have tools for direct SPI NOR manipulation?
danitool has quit [Quit: Cubum autem in duos cubos, aut quadratoquadratum in duos quadratoquadratos]
<djStolen>
PaulFertser: some xPi is free, I still need to find my 8 pin clamp but is definitely somewhere around, this I am sure. I just hope the clamp method works and no desolder needed
<PaulFertser>
djStolen: could be tricky. Sometimes you need to find the reset pin of the SoC and pull it low. Sometimes you need to limit the current you feed into Vcc bus so that the SoC would keep itself in reset.
retoatwork has quit [Quit: leaving]
floof58 has quit [Remote host closed the connection]
floof58 has joined #openwrt-devel
JohnA has quit [Remote host closed the connection]
<djStolen>
PaulFertser: thanks for hints. Will try, hopefully it works. :-)
<djStolen>
Last resort is to desolder. :-)
<PaulFertser>
djStolen: pulling just the vcc leg usually works too
djStolen has quit [Remote host closed the connection]
retoatwork has joined #openwrt-devel
<retoatwork>
Is anyone here working on MediaTek MT76x8 support for DSA? Or aware of such work? Would be interested in it, and also willing to put in effort from my side.
wearethepads has quit [Remote host closed the connection]
JohnA_ has left #openwrt-devel [Leaving]
goliath has quit [Quit: SIGSEGV]
rejoicetreat has quit [Ping timeout: 480 seconds]
<retoatwork>
blocktrron: Thanks for the link. However, this is not DSA related.
<retoatwork>
Sorry, should have read a bit further down. Yes, it sort of is. Actaully this is work getting done for me/my employer :P
<retoatwork>
Problem is that we can not really upstream our hackish driver, because the hardware we are developing on (GARDENA smart gateway) has just one single Ethernet port and we'd need to support the "whole thing", including all ports, etc. when trying to upstream.
<blocktrron>
retoatwork: do you have a link to your driver?
Borromini has quit [Quit: leaving]
feriman has quit [Ping timeout: 480 seconds]
rmilecki has quit [Ping timeout: 480 seconds]
Rentong has quit [Ping timeout: 480 seconds]
<digitalcircuit>
I'm going to wait until I get two successful backups in a row before I make a new comment on https://bugs.openwrt.org/index.php?do=details&task_id=3099 (i.e. generating email notifications). Should find out by tomorrow evening. (I wasn't able to easily figure out the right stress-ng parameters to simulate as I kept getting exciting new panics/hangs from stress-ng instead.)
<digitalcircuit>
If there's any sort of other testing or whatnot I can do to help (e.g. regression testing), feel free to ping me.
<digitalcircuit>
Good news everyone (well, for a very small group of folks, maybe just me) - OpenWRT 21.02.0-rc3 + git cherry-pick'ng the ipq806x cpufreq driver from the OpenWRT master branch ( https://git.openwrt.org/?p=openwrt/openwrt.git;a=commitdiff;h=6e411b8416388a9c8be1b2291be9b5adeeb07784 ) appears to have resolved the crash-on-SFTP-upload-to-USB-drive issue.
<retoatwork>
very much WIP, but progressing slowly due to other work
<retoatwork>
what I am currently wondering: Does the MT7688 hardware support the the same tagging format as the upstream tag_mtk.c driver implements?
<zorun>
digitalcircuit: cool. this single commit was enough? no need to backport later ones or adapt the DTS for 5.4?
<zorun>
digitalcircuit: once tested, please send a backport patch (with git cherry-pick -x and "PATCH 21.02" in subject), add Ansuel in Cc
<digitalcircuit>
zorun: Tentatively, that seems true! However, I'm doing at least one more test (and will continue after that) before I say this with confidence. Rarely (10% of the time?) this test would succeed without the fix.
<digitalcircuit>
zorun: And thank you! Figuring out the backport patch submission process was on my agenda, and your information just now (plus the wiki, existing patch submissions, etc) is a major help :)