10:42 <JamesMunns[m]> (Starter packs are lists of users you can follow, they're a major discoverability tool there).

10:42 <JamesMunns[m]> Btw, someone made an embedded rust starter pack on bluesky: https://go.bsky.app/G3wxTcN

10:42 <JamesMunns[m]> It's not an official wg list, but you can also ping the owner if you'd like to be added so folks can find you

16:46 <chrenderle[m]> diondokter: I have a question regarding device-driver. I'm writing a driver for ST25DV. This device distinguishes between user memory and system memory. They are addressed over different I2c addresses. And both memory regions start at address 0 (not I2C address). So there are 2 registers which have the address 0 and are distinguished with the I2C address. What is the best way to handle this with your crate? My best guess would

16:46 <chrenderle[m]> be to create two devices with seperate registers.

16:47 <diondokter[m]> chrenderle[m]: Yeah I'm thinking the same thing. Only thing that would be annoying is that you'd need some I2C sharing...

16:48 <diondokter[m]> Or make two blocks maybe?

16:49 <diondokter[m]> One block can use the normal addresses and the other block has an adress offset of say 100. Then in your interface you can detect which register set it is, subtract the 100 and use the other I2C address

16:50 <diondokter[m]> diondokter[m]: I think I like this one

16:51 <chrenderle[m]> diondokter[m]: Yeah that was also an idea

16:53 <chrenderle[m]> Another idea I had would be to enable touples as addresses. So I could use (u8, u16). Then i could distinguish in the u8 for which memory region I want

16:53 <chrenderle[m]> But as far as I understand this is currently not possible in 1.0.0-rc.0

16:55 <diondokter[m]> chrenderle[m]: Ah interesting. But also that'd be very impractical in the backend sadly

16:56 <diondokter[m]> chrenderle[m]: No indeed, it isn't

16:57 <diondokter[m]> But if you choose the first exponent of 2 larger than the highest address, you just get a bit that is on or off

16:57 <diondokter[m]> Very cheap

16:57 <diondokter[m]> And if you put it as an address offset on the block, then your register definitions don't have to be affected by it

16:59 <chrenderle[m]> diondokter[m]: Ah yes. That's a nice solution which doesn't require much implicit knowledge in the code. Thank you. Awesome crate you created there

17:00 <diondokter[m]> chrenderle[m]: Thanks! If you've got a version out, feel free to let me know. I'll add it to the book so people can get inspiration sources. Especially an example like this would be nice 🙂

18:47 <diondokter[m]> Ok, I might be able to make this work for multi-register support:... (full message at <https://catircservices.org/_irc/v1/media/download/AcLOeFBk5jtPn2TqNa2oJznJfFZIZZC_bJaU7xElhnfnbOSw5OAZglYOMjn2A9jDiIYePGc88Qu7RjdODO4s3EW_8AAAAAAAAGNhdGlyY3NlcnZpY2VzLm9yZy9nZVFXa0RpTnZGcWF4c21WZWZpQXNraHI>)