Differences

This shows you the differences between two versions of the page.

--- eg-connectivity:meetings [2021/09/09 13:27]
waltminer
+++ eg-connectivity:meetings [2021/10/07 13:15]
wminer
@@ Line 12: / Line 12: @@
 -------
+==== October 7, 2021 ====
+Attendees: Walt, Scott, Chi-Ming, Sergio
+   * ACAIS V2C Demo - Chih-Ming
+     * 10/7
+       * No progress on getting Connection Manager working due to illness.
+       * Got GPS (VK-162) working with RPI4.
+   * CAN/ Vehicle Signaling
+     * 10/7
+       * No progress
+   * Ref HW
+     * 10/7
+       * No progress.
+   * 5G/LTE module
+     * 10/7
+New:
+==== September 23, 2021 ====
+Attendees: Walt, Jan-Simon, Scott, Chih-Ming
+   * Chih-Ming is using the RPI3 for their telematics demo. Discussed best way to go about it. May revisit removal of Pi3 support for Lamprey for Telematics profile only if Chih-Ming has a need.
+     * 7/15 : Chih-Ming compiled the koi branch and got the telematics demo working
+       * For master / lamprey 12.0.1 : lets look to get MACHINE=raspberrypi3-64 back.
+       * Revert https://gerrit.automotivelinux.org/gerrit/c/AGL/meta-agl/+/26429 and uprev to pi3-64bit
+       * Chih-Ming will check if they can use 64-bit
+   * Discussed Daniel's latest work
+     * 7/29
+       * Update: Platform can use pi4 underneath, so no pi3 required.
+       * Trying to integrate their application using the SDK
+   * 8/12
+     * Chih-Ming - switched to RPI4 with [[https://www.waveshare.com/sim8200ea-m2-5g-hat.htm | Waveshare 5g modem]]. Able to build RPI 64-bit image and run their software. Able to receive CAN data via CAN board.
+   * 8/26
+     * No update from Chih-Ming this week.
+   * 9/9
+     * Chih-Ming having problems with 4 of the 6 RPI 4 modules that ACAIS has do not boot due to not seeing the partitions.  Might be related to https://gerrit.automotivelinux.org/gerrit/c/AGL/meta-agl/+/26624 which fixes an issue seen with newer versions of the RPI 4 hardware.  Chi-Ming to check the versions of his boards. Scott and Jan-Simon will look at the best way to get this into the build (either directly from upstream or this change.)
+   * 9/23
+     * Chih-Ming's issues with all of the RPI4s were fixed with the updates suggested by Scott.  Working on connection to AWS.
+   * Vehicle Signaling
+       * 9/17 - Jira ticket open for the low-can binding generator not in sync with binding (SPEC-3551)
+       * 10/1 - Thomas back from holidays and crunch-time and is starting to look at the new version of the low-can generator.
+       * 11/12 - Looked at options for remote CAN support via [[https://github.com/mguentner/cannelloni | Cannelloni]]
+       * 11/26 - Cannelloni working. Thomas showed remote operation. Will send patches when ready.
+         * straight to meta-agl-demo (e.g. recipes-connectivity)
+       * 12/10 - Cannelloni added and it is working.
+         * Thomas reports that his  conversion from Canoe to JSON is progressing. Messages are being sent from the driving simulator to AGL and low CAN service is decoding the messages properly.
+       * 1/7 - No updates from Thomas.
+         * CAN test failures.  SPEC-3756, SPEC-3755 Scott noticed that we lost the configuration file that was in /etc as part of one of the merges last year.  This is needed for the green machine demo, but would not be noticed by most users.
+       * 1/21
+         * Scott's fixes went into master.
+       * 2/4
+         * Thomas did some basic testing of packet/message loss with Cannelloni. Test was about a minute long and both computers were on the same network.  He is going to do some more stressful testing over the next few weeks,
+       * 2/18
+         * Thomas did some more testing of Cannelloni usage. Values up to 255 are received and sent ok. Larger values are not received incorrectly. Scott suggests it is in the message JSON or maybe has todo with the CAN binding changes that IoT.bzh did in 2019. (SPEC-3810)
+         * Scott received CAN-FD Pi hats so he can test that out in the near future.
+       * 3/4
+         * No updates due to other priorities.
+       * 4/1
+         * SPEC-3810 - CAN decoding. Scott can reproduce the issue. Has not had a lot of time to debug.
+         * Discussed options for improving the CAN binding and signal composer. Would like to get some feedback about using Canoe format and whether pursuing Kayak as a tool for formatting messages would be something OEMs are interested in.
+         * Most likely would get rid of Signal Composer depending on the App FW rework proposal from Collabora.
+       * 4/15
+         * Scott had no time to debug the CAN decoding issue.
+       * 5/20
+         * No update
+       * 6/4
+         * Scott is now back full-time on AGL and will be able to debug CAN FD and decoding issues.
+         * Thomas looking at OpenOBD2
+           * https://github.com/commaai/opendbc
+           * https://github.com/alan707/openOBD2
+           * https://www.csselectronics.com/screen/page/simple-intro-obd2-explained/language/en
+       * 6/18
+         * Scott no-progress due to other topics
+         * Thomas has only about 3 more weeks at Ruetlingen University.  He is not sure who if anyone will take over the vehicle simulator.
+       * 7/1
+         * Thomas last day is Monday, Starting his presentation for AGL Tech Day. He shared a few student apps from Github.  Will include the final links in his Tech Day presentation.
+         * Scott hoping to get back to CAN next week.
+       * 7/15
+         * WIP on CAN.
+       * 8.12
+         * Scott has been on vacation, no update.
+       * 8/26
+         * No update
+       * 9/23
+         * Scott tested CAN FD Pi hats with Debian and they work. Will need a 5.10 kernel to get it to work on RPI 4.
+         * Tested CAN on the reference hardware. Works but the driver is throwing a lot of errors. Might be a problem with the way he has it wired.
+   * Ref HW
+     * 1/7
+       * Panasonic will provide the TI based Bluetooth/Wifi boards at the end of January.  Would like to know if Panasonic will provide drivers and firmware as well as what testing will be done prior to shipping it. Also will need an installation procedure for how to add the board to the existing hardware.
+     * 2/4
+       * Scott wired up the RF HW for CAN. Hardware is set up as CAN FD, but he does not have CAN FD capable hardware to test it. Ordered a RPI hat that is supposed to support CAN FD.  He theorizes that the reference hardware can be configured for non-FD, but needs to poke around the device tree/ driver to see if it can be reconfigured.
+     * 2/18
+       * TI Wifi/BT modules should be received in LF Yokohama office by the end of the month, Delivery plan for the software from Panasonic is TBD.
+     * 3/4
+       * Modules not received in Yokohama.
+     * 4/1
+       * Sending out from Yokohama
+     * 5/20
+       * Still need to ship.
+     * 7/29
+       * Board arrived and Scott is looking at the integration of the wifi/bt board.
+     * 8/12
+       * No updates.
+     * 9/9
+       * Scott working with Ito-san on the CAN power enablement patch that he submitted.
+       * Having issues with BT audio. A2DP stuttering so badly as to be unusable (same issue seen on KF). Scott suspects a PipeWire issues and will reach out to George.  HFP does not work at all. Most likely needs a special PipeWire or WirePlumber change to enable this.
+   * 5G/LTE module
+     * Richard received the 5G module and it works
+     * Sorting out module
+New:
 ==== September 9, 2021 ====

Automotive Linux Wiki

User Tools

Site Tools

Differences

Page Tools