Hi, I recently watched ‘Hardware Design for Linux Engineers’, 52 minutes, https://www.youtube.com/watch?v=ziHhcBoRjQk. It was okay, more interesting towards the end where the programmer talked about the issues he'd had with his hardware design, like getting oscillations from his transparent level shifters when taking I²C signals off the board.
The main point of the video was KiCAD is a lot better now than a few years ago, https://www.kicad-pcb.org. CERN have been putting money into it, it's recently joined the Linux Foundation, etc. Version 6 later this year is meant to be a big improvement again. Eagle's two-layer limitation in its free-beer version has seen more people move to KiCAD as it has improved and that feedback has helped further. Features include routing of differential pairs, a push-and-shove router that will move traces out of the way as you draw, and interactive trace-length tuning. Anyway, the point of this email is one of the Q&A about fifty minutes referred to what sounded like a ‘DT layer’ and I don't know what that is. ‘What's the state of code generation for tools like KiCAD? Like are we at the point where we can turn out DT files based on the schematic yet.’ ‘No.’ ‘Is anyone trying that?’ ‘I don't think so.’ ‘KiCAD works at the level of schematic and it doesn't have a whole lot in terms of what the logical flow of the device is so what it would need is some way to describe the logical connections between the devices and then from that be able to generate a DT file. That would be incredibly useful. It would be possible I think to create an ad-hoc tool using a convention on the signal names so going through level shifters would be able to use a signal-name convention to associate components with buses and then from there generate a DT file.’ Anyone here know? -- Cheers, Ralph. -- Next meeting: BEC, Bournemouth, Tuesday, 2020-02-04 20:00 Check to whom you are replying Meetings, mailing list, IRC, ... http://dorset.lug.org.uk/ New thread, don't hijack: mailto:dorset@mailman.lug.org.uk
