17:30 <mtrbot-ml> [mattermost] <astro> @sb10q I prepared the PRs like the existing support for zedboard: peripheral pin definitions in migen, axi pins in migen-axi. just ignore me asking about that at all.

21:17 <annirak> daveshah: I mentioned I was having flashing problems yesterday and you asked me if the bitstream was built for the correct device. I've tracked down the command line passed to both nextpnr and to ecppack.

21:17 <annirak> nextpnr-ecp5 --json top.json --lpf top.lpf --textcfg top.config --25k --package CABGA256 --speed 6 --timing-allow-fail

21:18 <annirak> ecppack top.config --svf top.svf --bit top.bit

21:18 <daveshah> annirak: that looks correct to me

21:18 <annirak> I'm wondering if it's the SPI config to ecppack that could be the problem.

21:19 <daveshah> If you are using JTAG then you should pass no SPI config

21:19 <daveshah> previous versions of ecppack would produce a broken bitstream if you gave it SPI settings but this has been fixed for a few months

21:19 <annirak> I'm using JTAG so that should be fine.

21:21 <annirak> I have tried dropping the clock frequency and changing to falling-edge sampling for TDI. I consistently have the same, identical failure.

21:22 <daveshah> Can you post the full openocd output?

21:22 <annirak> The error messages are opaque to me. I'm not even sure whose docs to read. I'm considering breaking out the DSLogic and recording the whole transaction.

21:22 <annirak> Sure

21:24 <annirak> https://pastebin.com/6Yb2tEAq

21:26 <daveshah> I believe that status (BSE error code, bits 25:23 = 0b011) means a CRC error has occurred

21:26 <daveshah> Although it sounds like you've tried the obvious corruption causes already, a pulldown on TCK might be worth trying too

21:27 <daveshah> In a past life I had some big hassles with a MachXO2 board that turned out to need a TCK pulldown

21:27 <annirak> Interesting. That's an easy test.

21:29 <annirak> Thanks for your help, I really appreciate it! I'll let you know how it goes.

21:42 <annirak> Well, the error has changed. I'm now reading 0x0600000

21:42 <annirak> Well, the error has changed. I'm now reading 0x0600000

21:44 <daveshah> Hmm, that's not a code that makes sense

21:44 <daveshah> what is the JTAG clock frequency?

21:44 <annirak> 10MHz. I'll drop it.

21:48 <annirak> 500kHz with rising or falling-edge TDO-sampling produces the same result. Power cycling the board and running it again, same result. Power cycling the FT2232 and running again, same result.

21:50 <annirak> I'm going to go through everything again, because this doesn't make sense.

21:53 <mtrbot-ml> [mattermost] <hartytp> @sb10q did TS send you a thermostat or do you want me to post one. If the latter, which address to?

21:54 <mtrbot-ml> [mattermost] <hartytp> @astro when you get to looking at firmware for the new thermostat let me know. There are a few changes to the hw and it’s probably easiest if we talk through what needs doing

22:59 <annirak> *eyeroll* I changed to the other port of the FT2232HL and now it works.