Dario Nieuwenhuis
Dario Nieuwenhuis
I think an inherent (not trait) `read_until_idle` impl makes sense, yes! :+1:
> if you started an async operation and then also did a blocking operation while it is outstanding, then you'd screw up the hardware state for the async operation. The...
> It still disallows blocking IO on an async I2c - I'm happy to reverse that if there's a reason, but I don't personally see one. Yes, it should be...
code size comparison with default build settings: ``` text data bss dec hex filename 14656 48 1552 16256 3f80 old, uart0 14720 48 1568 16336 3fd0 new, uart0: +64 bytes...
> Should the DMA be paused Oof this is cursed. I guess yes, or this will be a problem when DMAing from flash. > If so, how to properly do...
> What about a combination that allows both uses: a GenericFlash using const generics, and a type alias Flash based on cargo features? Or would that be too confusing? It...
Sounds good to me! :+1: I think having to write RAM-linked "flash driver" blobs is inevitable given the ROM limitations... > adding a "custom" peripheral for flash, such that you...
Ah! then all our peripherals are already "custom". They're all invented by the HAL, we don't use the PAC ownership model. (rp and stm32 PACs don't have owned singletons at...
the CI has a Pico W, not sure if it matters..?