[LEDE-DEV] adding kernel module packages (kmod-rtc-ds1307) to opkg repos?
Hi, just started working on adding rtc to one project and saw that kernel module kmod-rtc-ds1307 can be compiled manually but it is not available as a package. Can this package be added to be build in the future so it is available as a package in Lede package repositories? What is correct procedure to make this happen? Thanks, Valent. ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
[LEDE-DEV] Serial data getting lost - bug or a feature?
Hi, I'm working with Lede on ar71xx platform (Carambola2 wifi module) to gather sensor data over serial, save it and transfer it to "could". Previous firmware was based on OpenWrt Chaos Calmer but latest one is based upon Lede 17.01.01. I'll test it also with 17.01.03 soon. What I have discovered is that now with using Lede sometimes there are missing characters while reading data from serial connection. Data is not constantly being read, but sensor device is logging data usually for 8 hours and then data is read via serial in one big "download". During these downloads I have never seen a character go missing while Chaos Calmer was used. What was even stranger is that sometimes I did get these "bad logs" (ie ones with few missing characters) but sometimes log download went perfectly! After some head scratching and troubleshooting I managed to reproduce this "bug" 100% and it was due to wifi! But only if wifi was enabled and not being used! So if radio was enabled, in sta mode but not connected to AP errors would happen! It I would disable wifi or if radio was associated with AP then there would be no errors during reading data from serial connection. Is there any explanation for this behaviour? Is this a bug or a feature? Is there any change between OpenWrt CC version and Lede that could explain this? Thanks in advance for any insight and answers, Valent. ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
Re: [LEDE-DEV] Serial data getting lost - bug or a feature?
Does ath9k driver when not connected do periodical scans to find ssid and to connect to? Does this cause some interrupts that could cause reading of serial data to loose one character? Does ath9k driver reset every time it doesn't find known ssid? ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
Re: [LEDE-DEV] Serial data getting lost - bug or a feature?
Hi Piotr, I'm using internal uart, but I have disabled cron from writing to serial console, and also disabled serial as login tty. Cheers, Valent. On Tue, Oct 17, 2017 at 1:08 PM, Piotr Dymacz wrote: > Hi Valent, > > > On 16.10.2017 22:04, Valent Turkovic wrote: >> >> Hi, >> I'm working with Lede on ar71xx platform (Carambola2 wifi module) to >> gather sensor data over serial, save it and transfer it to "could". >> >> Previous firmware was based on OpenWrt Chaos Calmer but latest one is >> based upon Lede 17.01.01. I'll test it also with 17.01.03 soon. >> >> What I have discovered is that now with using Lede sometimes there are >> missing characters while reading data from serial connection. >> >> Data is not constantly being read, but sensor device is logging data >> usually for 8 hours and then data is read via serial in one big >> "download". During these downloads I have never seen a character go >> missing while Chaos Calmer was used. >> >> What was even stranger is that sometimes I did get these "bad logs" >> (ie ones with few missing characters) but sometimes log download went >> perfectly! >> >> After some head scratching and troubleshooting I managed to reproduce >> this "bug" 100% and it was due to wifi! But only if wifi was enabled >> and not being used! >> >> So if radio was enabled, in sta mode but not connected to AP errors >> would happen! It I would disable wifi or if radio was associated with >> AP then there would be no errors during reading data from serial >> connection. >> >> Is there any explanation for this behaviour? Is this a bug or a >> feature? Is there any change between OpenWrt CC version and Lede that >> could explain this? > > > Are you using AR9331 built-in UART or some external, USB to UART adapter? > > -- > Cheers, > Piotr > >> >> Thanks in advance for any insight and answers, >> Valent. >> >> ___ >> Lede-dev mailing list >> Lede-dev@lists.infradead.org >> http://lists.infradead.org/mailman/listinfo/lede-dev >> > ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
Re: [LEDE-DEV] Serial data getting lost - bug or a feature?
I have just finished testing with Lede 17.01.3 but I couldn't use default image but I build new one via Imagebuilder because we have quite a few additional packages and we use wpad instead of wpad-mini. These are the packages I added to image: blkid block-mount kmod-fs-ext4 kmod-fs-vfat kmod-nls-base kmod-usb-storage e2fsprogs chat comgt kmod-usb-serial kmod-usb-serial-option kmod-usb-serial-wwan usb-modeswitch kmod-crypto-hash kmod-crypto-manager kmod-crypto-pcompress openvpn-openssl rsync coreutils coreutils-stat coreutils-timeout glib2 kmod-hid kmod-input-core kmod-lib-crc16 nano ntpclient zlib wireless-tools tcpdump tmux coreutils-nohup diffutils tmux -6relayd wpad coreutils-stty picocom python-light python-codecs python-pyserial cfdisk e2fsprogs kmod-nls-base kmod-nls-cp437 kmod-nls-iso8859-1 vnstat kmod-fs-f2fs f2fs-tools ca-certificates wget snmpd netcat And after that set log levels in /etc/config/system as follows: option conloglevel '1' option klogconloglevel '1' option cronloglevel '9' and then I comment out uart to be used as login terminal in /etc/inittab : #::askconsole:/usr/libexec/login.sh And after all of that I could reproduce issue 100%. So if wifi is active, in sta mode but not associated to AP then serial communication is having issues and is loosing characters when receiving them. Do you need some more info from my side that would help? ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
Re: [LEDE-DEV] Serial data getting lost - bug or a feature?
Thanks Jay for very detailed answer. You reminded me about Ymodem and Zmodem that I used when I was a studetn to tranfer apps to my HP48+ calculator, but which I still own! :) We do have our own protocol that does block transfers of 30 lines at a time, and it has crc check for each line so it is not an issue for us, but more a curiosity. But to now we were using OpenWrt CC and didn't have any issues in serial transfer, not we are starting to deploy devices with Lede we noticed this issue with wifi causing serial communication issues. So I'm really interested in getting to root cause of this. > -- Forwarded message -- > From: Jay Carlson > To: Valent Turkovic > Cc: Piotr Dymacz , LEDE Development List > > Bcc: > Date: Fri, 20 Oct 2017 11:09:27 -0400 > Subject: Re: [LEDE-DEV] Serial data getting lost - bug or a feature? > On 2017-10-18 Wed, at 06:32, Valent Turkovic wrote: > >> So if wifi is active, in sta mode but not associated to AP then serial >> communication is having issues and is loosing characters when >> receiving them. > > Regardless of this poor behavior, I think you're going to regret it > eventually if you don't have some kind of checksum and retransmit > logic--implying you're also doing framing. Async serial lines can be tricky, > and have non-zero bit error rates. And the framing mechanism needs to be > resilient against incomplete packets (meaning, bytes dropped between > transmitter and receiver). > > Newline-separated ASCII records with a hex CRC are not the worst idea, but > you need to deal with the situation where a newline gets lost, or you get > trash or fewer bytes than you needed. I've used YMODEM before for bulk data. > > Note that the Arduino Yún uses packets. > > Jay > > ___ > Lede-dev mailing list > Lede-dev@lists.infradead.org > http://lists.infradead.org/mailman/listinfo/lede-dev > ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
[LEDE-DEV] MeshPoint is one of the 12 finalists of Hackaday 2017!
I'm proud to announce that MeshPoint has become one of the 12 finalists of The Hackaday Prize 2017. We’re competing for first place in two categories: Best Product and as best project overall. Without awesome work from OpenWrt and Lede maintainers and contributors we couldn't have done this. So a really big thanks to all of you! You have our deepest gratitude. Thanks to the hard work of MeshPoint team members, the collaborative project has gotten another confirmation that we are working on technology that’s going to make a significant impact on the IT industry and the humanitarian innovation sector. In the past months, we have been working hard to perfect our device’s efficiency, performance and appliance. Out of the 12 chosen finalists, 5 are selected to win 5,000 up to 50,000 USD. Honestly, just being selected as one of the best by leading IT experts, engineers, entrepreneurs and professors from reputable American universities (such as MIT) is great motivation and a clear message to our team that we’re on the right track. The announcement will be made on the Hackaday Superconference 2017, the largest international conference dedicated to open source hardware projects and hardware development. The conference starts on Saturday, 11th of November. The announcements will be made on: Saturday, 11th of November at 18:30 California time (Sunday 03:30 CET) You can follow the event live on the Hackaday Youtube channel or on the Hackaday Facebook page. Thanks to all MeshPoint team members: Milijana Micunovic (Crisis Innovation Lab), Valent Turkovic (Crisis Innovation Lab), Robert Marko, David Blazevic, Ivan Vidakovic (HexaWorx), Luka Perkov (Sartura), Davor Popovic (Sartura), Luka Mustafa (Institute IRNAS), Luka Simic, Marin Stević, Robert Sorić and David Zovko (e-radionica). ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
Re: [LEDE-DEV] [PATCH] kernel: build RTC kmods when RTC_SUPPORT is activated
On Sat, Jul 8, 2017 at 11:16 AM, Felix Fietkau wrote: > On 2017-07-08 11:04, Hauke Mehrtens wrote: >> The build bots are not setting CONFIG_ALL any more, so activate this >> module by default why any of the ALL* options is selected. >> >> Signed-off-by: Hauke Mehrtens > I think it would be more useful for ALL to select ALL_NONSHARED and > ALL_NONSHARED to select ALL_KMODS. This might also make some further > cleanups possible... > > - Felix I have just picked up latest Lede release version and my i2c-rtc project, and checked it kmod-rtc-ds1307 would be maybe available now, but it still is not being compiled. Any plans to make these changes to selecfing kmods to be build in next release maybe? Valent. ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
Re: [LEDE-DEV] LEDE v17.01.0 final
Congrats! An impressive amount of work and progress has been shown! ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
Re: [LEDE-DEV] [OpenWrt-Devel] Introducing the LEDE project
I'm an outsider, have nothing to do with OpenWrt developement but still work on few projects which depend on OpenWrt as awesome project that enables us to do our projects (wifi mesh networking) but also do professional jobs for clients using OpenWrt as embedded os for lots of different applications. We have "suffered" both on volunteers side (wifi mesh) and on professional side by infrastructure failing... mirrors, wiki, forum... you name it, all of it was down quite a lot in last few years, and on IRC or mailing list you couldn't get concrete and timely answers. This is a HUGE red flag for any size project. Also even if we hd really good developers who know what they were doing (I'm not that one) we couldn't get patches into trunk without knowing somebody in the inside circle who has commit access, so patches would get ignored... and kept in our own git... this is also a really huge red flag. I have convinced companies that I work for to donate money towards OpenWrt so you have some budget for infrastructure, and I'm sure that there are lots of other people who could also get some funding, but answers I got on IRC were really dissapointing - that there is now was to give donations towards better infrastructure or any other kind of collecting funds... These are all signs of poorly managed open source project/community and I welcome any change in any direction that aims to fix any of these issues... so fork away and make things better. Go LEDE team! ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
[LEDE-DEV] Sync data from tmpfs to internal or external flash periodically !
While reading some articles on Arch wiki [1] I have discovered anything-sync-daemon [2] which is just awesome if you are embedded developer or enthusiast using OpenWrt. Function of the asd (anything-sync-daemon) is to periodically sync contents of tmpfs (ram drive) to some physical memory (hdd, flash drive, sd card, internal flash), this speeds up things considerably while also lowering the wear of physical memory storage device. Most OpenWrt devices have really small internal flash (4MB being most common, some models have up to 16MB of internal flash), and if you write to internal flash you can wear it out quite fast and render it unusable. I have worked on few projects which all collect and log some data, we usually use sd cards to save data to, but still they also also wear out after 6-12 months of use. I see asd (anything-sync-daemon) like an awesome tool that would collect data and sync every hour data from tmpfs to sc card so it would reduce wear and tear of sd card. You can also use asd for having permanent openwrt logs, and sync them to internal flash so that you have logs even after a reboot, of is you use rrdtool graphs you can also have your rrdtool database synced so you have data and rrd graphs also after reboot... There are lots of use cases I can think of... But currently there is no package for asd in lede or openwrt package repos. Has nobody discovered this package yet? Is there some other package that provides similar functionality already to Lede/OpenWrt? Thanks! [1] https://wiki.archlinux.org/index.php/Anything-sync-daemon [2] https://github.com/graysky2/anything-sync-daemon ___ Lede-dev mailing list Lede-dev@lists.infradead.org http://lists.infradead.org/mailman/listinfo/lede-dev
