Okay, more like an hour, but most of that was writing docs. Just for you:
github.com/jamesmunns/pel
Conversation
I don't expose the allocator yet (64KiB), so you'll have to do your own memory management in the 128K of space you have. But I do plan on supporting that soon, or at least automatically mapping the unused space out of 128K to be a user heap with a syscall interface.
1
1
Replying to
Not an issue for me at this point. Thanks for the patch. I'll send word when I have something going. Hope to get to it tonight. Time for emergency nap first :-)
1
1
Replying to
Lemme know if it works! I didn't really test it yet. Though the biggest part is making sure the linker script looks right. If it freaks out at link time, send me the error. If it freaks out at run time, send me the output of `arm-none-eabi-nm -nSC path/to/app.elf`.
1
1
Replying to
It's a perfect setup for me to be forced to look at this :-) I've been postponing this for so long, dreading that Rust/C border... I have an app for work I'd like to gradually rewrite but don't know where to start. This would be a good start.
1
1
Replying to
Honestly, working at the border is one of the worst ways to start: You have to be an expert at Rust AND C (or whatever C-emulating language you are using for FFI), which is a tricky thing to do.
That being said, hopefully I don't steer you wrong :)
1
1
Replying to
Realistically, I'm never going to be able to promote Rust if I don't have a gradual migration path with C/Rust interop. For my own solo projects I do what I want but changing the world is not going to fly for anything else.
1
1
Replying to
I mean, C+Rust totally work together well! But any time you are stitching two things together, you should understand both well so you don't get caught somewhere tricky.
1
1
Replying to
This is mostly what I'm after. To understand this border well. Currently it's a bit opaque to me...
1
1
I like this probe-run thingy. Painless.