*** bharmon has quit IRC | 00:14 | |
*** bbranch has quit IRC | 00:44 | |
*** bbranch has joined #automotive | 01:56 | |
*** bbranch has quit IRC | 02:08 | |
*** bbranch has joined #automotive | 02:37 | |
*** bbranch has left #automotive | 02:39 | |
*** weston has joined #automotive | 05:25 | |
weston | good morning all | 05:35 |
---|---|---|
CTtpollard | morning weston | 07:35 |
*** KlausUhl has joined #automotive | 08:05 | |
*** fredcadete has quit IRC | 08:08 | |
*** jonathanmaw has joined #automotive | 08:41 | |
*** jacobo has joined #automotive | 08:46 | |
CTtpollard | ok just did a quite upate to the readme for meta-gdp, both koelsh and porter point to newer commits | 08:49 |
CTtpollard | fixes and re-inclusion of driver scripts | 08:49 |
* CTtpollard replies to some mailing list questions | 09:01 | |
*** fredcadete has joined #automotive | 09:02 | |
*** fredcadete has left #automotive | 09:05 | |
*** fredcadete has joined #automotive | 09:05 | |
*** weston has quit IRC | 09:07 | |
*** wschaller has joined #automotive | 09:08 | |
*** fredcadete has quit IRC | 09:23 | |
*** fredcadete has joined #automotive | 09:39 | |
*** FaHfah has joined #automotive | 09:53 | |
*** FaH_auto has joined #automotive | 09:55 | |
*** FaH_auto has quit IRC | 09:57 | |
*** FaH_auto has joined #automotive | 09:59 | |
*** jacobo has quit IRC | 09:59 | |
*** FaH_auto has quit IRC | 10:01 | |
*** FaH_auto has joined #automotive | 10:07 | |
*** jacobo has joined #automotive | 10:08 | |
*** wschaller has quit IRC | 10:26 | |
*** fury has joined #automotive | 10:41 | |
*** FaHFah has joined #automotive | 11:34 | |
*** FaH_auto has quit IRC | 11:36 | |
*** weston has joined #automotive | 11:40 | |
*** wschaller has joined #automotive | 11:49 | |
*** FaH_auto has joined #automotive | 12:12 | |
*** FaHFah has quit IRC | 12:12 | |
weston | hi guys, any idea on how to do static analyis for C++ code written in QT | 12:24 |
KlausUhl | weston: is there anything that would make static analysis of Qt based C++ code any different from "plain" C++ code? | 12:29 |
KlausUhl | Otherwise, you can probably go through the following list and take what best suits your needs feature and, potentially, money wise: https://en.wikipedia.org/wiki/List_of_tools_for_static_code_analysis#C.2FC.2B.2B | 12:30 |
weston | The Qt creator can check it during coding itself in editor right? | 12:30 |
KlausUhl | Probably not. But that is only a limitation of Qt Creator. You will have to perform the static analysis outside of it. | 12:31 |
KlausUhl | I think I have been wrong. | 12:32 |
KlausUhl | There seems to be a Qt Creator plug-in for at least Cppcheck: http://sourceforge.net/projects/qtc-cppcheck/ | 12:32 |
rjek | I've had good success with clang's scan-build | 12:32 |
rjek | It's kinda like coverity, but somewhat less advanced. | 12:33 |
KlausUhl | You may want to research if there is a Qt Creator plug-in for that one. | 12:33 |
rjek | You launch scan-build with "scan-build make" | 12:33 |
rjek | Assuming you use make | 12:33 |
KlausUhl | Check this: http://doc.qt.io/qtcreator/creator-clang-static-analyzer.html | 12:34 |
rjek | scan-build is a wrapper that replaces calls to your compiler with clang's static analyser (while also generating code) | 12:34 |
rjek | So it's pretty transparent | 12:34 |
KlausUhl | clang scan-build seems to be integrated into Qt Creator by default | 12:34 |
weston | KlausUhl: great thanks | 12:36 |
KlausUhl | you are welcome | 12:36 |
rjek | The open source project I'm involved in uses all of scan-build, coverity, and cppcheck as part of CI | 12:37 |
*** CTtpollard has quit IRC | 12:37 | |
KlausUhl | That's the big gun :-) | 12:40 |
KlausUhl | But you are doing it right. In my experience, no single static analysis tool is able to catch everything. | 12:41 |
KlausUhl | The only drawback of using multiple static analysis tools it that you also get false positives from multiple sources. | 12:42 |
weston | KlausUhl: I was using QAC and QAC ++ | 12:44 |
weston | it is good | 12:45 |
weston | and Tessy for Unit testing | 12:45 |
*** emaj has joined #automotive | 12:46 | |
*** kratz00_work has quit IRC | 12:50 | |
*** FaHFah has joined #automotive | 12:50 | |
*** CTtpollard has joined #automotive | 12:52 | |
*** FaH_auto has quit IRC | 12:53 | |
fredcadete | clang scan-build integration in qtcreator seems to be only for the commercial version though | 13:14 |
rjek | You should still be able to use it in an unintegrated fashion though, yeah? | 13:16 |
*** alex_h has quit IRC | 13:26 | |
*** wschaller_ has joined #automotive | 13:34 | |
*** wschaller has quit IRC | 13:34 | |
CTtpollard | gah, doing a fresh checkout of everything has made me realise gdp still can't build with the specified meta-ivi commit | 13:34 |
*** alex_h has joined #automotive | 13:40 | |
*** alex_h has joined #automotive | 13:41 | |
* CTtpollard points meta-ivi to a newer commit as suggest by gunnar and rebuilds | 13:43 | |
*** kratz00_work has joined #automotive | 13:50 | |
*** FaHFah has quit IRC | 14:33 | |
*** bbranch has joined #automotive | 14:50 | |
*** bbranch has left #automotive | 14:58 | |
*** jlrmagnus has joined #automotive | 15:05 | |
jlrmagnus | MOrning | 15:06 |
CTtpollard | when building GDP has anyone ever had the error "No recipes available for meta-genivi-demo/recipes-multimedia/audiomanager/audiomanager_6.1.bbappend" | 15:06 |
CTtpollard | I've never had this error before, this is after switching to a newer commit of meta-ivi which fixes the common api c++ error | 15:07 |
FelixH | Hi jlrmagnus | 15:09 |
jlrmagnus | Hello Felix | 15:09 |
FelixH | I saw your can firewall yesterday, looks very interesting! | 15:09 |
jlrmagnus | Thanks. | 15:10 |
* rjek too | 15:10 | |
rjek | Sweet little PCB design too! | 15:11 |
jlrmagnus | Yeah. We did miss to add a CAN controller to the PCB, which is a bit embarrasing. | 15:16 |
jlrmagnus | *embarrassing | 15:17 |
jlrmagnus | That will be fixed. | 15:17 |
jlrmagnus | I didn't double check the design and thought the tranceiver had the necessary features to implement the protocol itself. | 15:17 |
* rjek finds that lots of small micros can only support one CANbus interface at a time :-/ | 15:21 | |
rjek | It also sounds like something that could almost be implemented in a CPLD or FPGA: list of masks to allow, otherwise store and forward | 15:22 |
jlrmagnus | Yes. | 15:25 |
jlrmagnus | It depends on how much frame transformation you want. | 15:26 |
*** bharmon has joined #automotive | 15:26 | |
*** KlausUhl has quit IRC | 15:30 | |
jlrmagnus | Down the road you may want to implement a very simplistic scripting language (lua?) that you transfer OTA to implement frame filtering and transformation. | 15:35 |
jlrmagnus | That allows you to keep state and create outbound frames with a payload composed of the input from several frames. | 15:36 |
* rjek rubs thighs at mention of Lua | 15:36 | |
rjek | Although it's beginning to sound too complex for my taste | 15:36 |
rjek | I will always advocate Lua, though. But for complete disclosure, I host lua.org (and the mailing lists) and lua-users.org on my personal infrastructure. So perhaps I'm bias. | 15:38 |
jlrmagnus | If we are to keep this safe, we need to delete recursion and non-finite loops from the equation. | 15:50 |
jlrmagnus | Else we risk locking the firewall through shitty scripts. | 15:51 |
*** jonathanmaw has quit IRC | 16:07 | |
*** bbranch has joined #automotive | 16:07 | |
*** wschaller_ has quit IRC | 16:34 | |
*** persia has quit IRC | 16:42 | |
myself | rjek: most small micros speaking CAN are using SPI to hang an arbitrary number of MCP2515's off | 16:54 |
myself | I haven't looked to see if there's a dropin CanFD-capable replacement | 16:55 |
jlrmagnus | We are taking the MCP2515 route. | 16:55 |
jlrmagnus | We blew the original design (that is checked into github) and just put the tranceiver in there, forgetting the CAN controller. | 16:56 |
jlrmagnus | In my review I thought the board had a combined CAN controller / tranceiver. We will fix that. | 16:57 |
myself | https://github.com/CANBus-Triple/CANBus-Triple-Hardware/blob/master/CANBusTriple_f.pdf | 16:57 |
myself | good example of multiple 2515's off a single 8-bit micro | 16:57 |
*** ulf` has quit IRC | 16:59 | |
*** ulf` has joined #automotive | 17:00 | |
*** jlrmagnus has quit IRC | 17:02 | |
*** joone has quit IRC | 17:08 | |
*** ulf` has quit IRC | 17:10 | |
*** bbranch has quit IRC | 17:12 | |
*** joone has joined #automotive | 17:12 | |
*** persia has joined #automotive | 17:24 | |
*** persia has joined #automotive | 17:24 | |
*** bbranch has joined #automotive | 17:26 | |
*** gmacario has quit IRC | 17:28 | |
*** jimmy1 has joined #automotive | 17:31 | |
*** jacobo has quit IRC | 17:44 | |
*** ulf` has joined #automotive | 18:08 | |
*** jlrmagnus has joined #automotive | 18:27 | |
jlrmagnus | myself: Thank you. I've sent this off to our hardware designer. | 18:28 |
*** jlrmagnus has quit IRC | 18:34 | |
*** jlrmagnus has joined #automotive | 18:50 | |
*** jlrmagnus has quit IRC | 19:13 | |
*** jlrmagnus has joined #automotive | 19:57 | |
*** bharmon has quit IRC | 20:15 | |
*** bharmon has joined #automotive | 20:22 | |
*** emaj has quit IRC | 20:29 | |
*** bbranch1 has joined #automotive | 21:03 | |
*** bbranch has quit IRC | 21:05 | |
*** bbranch1 has quit IRC | 21:32 | |
*** bbranch has joined #automotive | 21:48 | |
*** jlrmagnus has quit IRC | 22:04 | |
*** jlrmagnus has joined #automotive | 22:32 | |
*** bharmon has quit IRC | 23:02 | |
*** jlrmagnus has quit IRC | 23:31 | |
*** bharmon has joined #automotive | 23:36 |
Generated by irclog2html.py 2.14.0 by Marius Gedminas - find it at mg.pov.lt!