On Thu, Feb 22, 2018, at 2:57 AM, markus wrote: > A few weeks ago I started porting mynewt to the stm32f3 family. I've > been using it for quite a while and it seems to work so I thought about > opening a PR (see > https://github.com/mlampert/mynewt-core/tree/feature/stm32f3) > > There seems to be a refactoring in progress to eliminate some of the > duplication between the different stm32 ports - which I think is a good > thing! > > So I was wondering if I should hold off until that is done?
Open your PR, please! > > Also, comments on the actual port are most welcome. It diverts from the > other stm32 family ports mainly in two areas: > > * accounting of external interrupts is on a per ISR basis, not on a per > HW interrupt basis, see > https://github.com/apache/mynewt-core/compare/master...mlampert:feature/stm32f3?expand=1#diff-12c253fbcf44cd336f08b3e2ed4eca8e > > I found this more useful during debugging than the external HW interrupt > counts - YMMV. > > * The stm32f303k8 is rather resource constrained with 64k flash and 12k > sram. I didn't see much value in trying to squeeze a bootloader and two > app images in there - so the nucleo-f303k8 bsp doesn't support a > bootloader and is set up to load a single app image directly. Agreed. I have a f3-discovery here, which has 256kb, so I can add/test a BSP for this one as well. Fabio Utzig
