Adam Greig

> (Additive-only is a misfeature, I'm increasingly convinced.) I agree; I already use it for mutually exclusive options in stm32f4 etc. I don't think it's disastrous to create a negative...

> (The take() calls in general assume that all users are compiled together. With separately compiled tasks linked together, each one has its own copy of the global being set...

> This issue is probably not the right place; shall I file a new one, or is there already a discussion in progress? Please go ahead with a new issue,...

The `disable_jtag()` function on the F1xx HAL writes 0b010 to SWJ_CFG to disable JTAG-DP and leave SW-DP enabled, which according to the reference manual _should_ leave PB3 available for SWO...

Unless I'm missing something, this was intentional - it was cut from the 0.7.x backports branch, while master is for 0.8.0 onwards and contains breaking changes (such as bumping bare-metal...

Cargo uses a [slightly variant definition of semantic versioning](https://doc.rust-lang.org/cargo/reference/specifying-dependencies.html#specifying-dependencies-from-cratesio): > This compatibility convention is different from SemVer in the way it treats versions before 1.0.0. While SemVer says there is...

Are you setting the linker file using `build.rs`, `.cargo/config.toml`, or both? Setting RUSTFLAGS should only take effect once, I thought, so that if you have two competing `.cargo/config.toml` only one...

I'm not totally happy allowing anything in the non-lto blobs (though I guess I'd probably re-run xtask before a release anyway); maybe we could have a check similar to previous...

> If we want to compare the disassembly, then we'd have to check that in as well. The old CI script didn't have them checked in: it dumped the checked-in...

For the -lto files we can continue to compare directly anyway. They don't really contain thumb assembly anyway, right? Do the non-lto file dumps compare exactly the same between Linux...