Debugging the BeagleBone PRU in CCS

RE: Debugging the BeagleBone PRU in CCS

Fred27 — Wed, 18 Aug 2021 12:11:51 GMT

I added bb_pru_startup.js to the body of this post so it win't get lost again. (Why is pasting in code always to difficult!)

RE: Debugging the BeagleBone PRU in CCS

joshua1609 — Fri, 13 Aug 2021 21:45:06 GMT

Could you share a copy of the startup file bbb_pru_startup.js. Ive tried both links but in 2021 they are both dead.

RE: Debugging the BeagleBone PRU in CCS

Jan Cumps — Wed, 21 Aug 2019 16:43:29 GMT

Prevent Linux from booting

You need to make sure the Linux isn't currently running on the BeagleBone. Unfortunately I don't think you can debug the PRU whilst the Sitara's A8 core is busy running Linux. Obviously this means you can't debug any interaction between your Linux code and the PRU. Maybe it's possible, but I've not heard of it.

I'm going to try the without-linux approach first. It'll be a little messy because I'm currently booting and running from the eMMC.

Then I'll try to find out why it should be without Linux - because I want to debug in a situation where data is shared from OS to PRU.

RE: Debugging the BeagleBone PRU in CCS

Jan Cumps — Mon, 08 Jul 2019 15:50:28 GMT

I just bought the same debug probe as you - it arrived today. I'm going to try this.

I know that the probe is OK - it works on a TI CC1310, TMS570LS04 (both tested on LaunchPads) and a CC2650 (the SensorTag). I'll now test the setup you describe above.

(edit: you named it XDS100 in the blog post. The hyperlink correctly points to XDS110)

RE: Debugging the BeagleBone PRU in CCS

Jan Cumps — Sun, 26 May 2019 10:09:14 GMT

What worries me a bit is that TI is decommisioning the wiki that contains the .js debug init script.

Although it’s a bit unorganised, the wiki is full with info you can’t find anywhere else. For the BB, but also for the Hercules controller family...

RE: Debugging the BeagleBone PRU in CCS

shabaz — Sat, 18 May 2019 23:07:22 GMT

Another awesome post, this is very helpful! I think I have that connector so I will be soldering it onto my BBB soon too.

As I understand, there is some stuff possible from Linux, in terms of stopping and single-stepping, but whether anyone has written a recent debugger, I've no idea. There was something old but I couldn't get it to run a few years ago.

But I think with the method you describe, the workflow could be to code and test the PRU, and then run it with Linux later, with the code added to send/receive into a Linux process any necessary data. This way seems to be the best of both worlds, with a decent graphical CCS debugger, and rather than try to debug both PRU side and Linux application side at the same time.