push last meeting minutes

Author: japaric

minutes/2018-07-02.md

@@ -0,0 +1,165 @@
+# Agenda
+
+- rust-std for `thumbv6m-none-eabi` to be temporarily unavailable after upgrading to LLVM 7. cf.
+  https://github.com/rust-lang/rust/pull/51966
+
+- Rust 2018: the home stretch. cf. https://internals.rust-lang.org/t/rust-2018-the-home-stretch/7810
+
+- Maintaining the cortex-m ecosystem. cf. #46,
+  https://github.com/rust-lang-nursery/embedded-wg/issues/109#issuecomment-398783889
+
+- Triaging issues on the [2018 edition milestone].
+
+- Commercial testimonials. cf. #108
+
+[2018 edition milestone]: https://github.com/rust-lang-nursery/embedded-wg/milestone/1
+
+# Minutes
+
+## `thumbv6m-none-eabi`
+
+We spent quite a bit of time discussing what to do regarding this upcoming problem. This target is
+part of what we want to have working on stable for the edition release, but we don't want to block
+upgrading LLVM which benefits the WASM ecosystem and improves LLD, which the no_std / embedded
+ecosystem also uses.
+
+We decided to not block the PR but to prioritize fixing the LLVM bug to re-enable the `thumbv6m`
+target before the next cycle. This would involve making `rust-std` unavailable for `thumbv6m` in
+nightly for as long it takes to fix the LLVM bug.
+
+Between the time the meeting was held and these minutes were posted [@pftbest] promptly [fixed] the
+LLVM bug so the problem was completely adverted. Thanks @pftbest! :heart:
+
+[@pftbest]: https://github.com/pftbest
+[fixed]: https://reviews.llvm.org/D48846
+
+## Rust 2018: home stretch
+
+We worked through [the issues in the 2018 edition milestone][edition] to assess how the work that
+still needs to be done can be accommodated in the roughly 3 cycles that we have before the edition
+release.
+
+### embedded Rust on stable
+
+[rust-lang-nursery/embedded-wg#42](https://github.com/rust-lang-nursery/embedded-wg/issues/42)
+
+This is just waiting for `#[panic_implementation]` to be stabilized. There's no work left for us to
+do. We just have to wait for [the proposed FCP process] to finish.
+
+[edition]: https://github.com/rust-lang-nursery/embedded-wg/milestone/1
+[the proposed FCP process]: https://github.com/rust-lang/rust/issues/44489#issuecomment-398965881
+
+### Stable assembly operations
+
+[rust-lang-nursery/embedded-wg#63](https://github.com/rust-lang-nursery/embedded-wg/issues/63)
+
+This is more of a nice to have as it's not required for embedded Rust on stable (we have a
+workaround for `asm!` / `global_asm!` on stable that involves external assembly files). If we hope
+to get this done in time by the edition release then these would be the final deadlines.
+
+- 1.28 (2018-08-02). implement [rust-lang-nursery/stdsimd#437]
+
+- 1.29 (2018-09-13). subset of most commonly used intrinsics stabilized in beta.
+
+[rust-lang-nursery/stdsimd#437]: https://github.com/rust-lang-nursery/stdsimd/issues/437
+
+### LLVM tools
+
+[rust-lang-nursery/embedded-wg#51](https://github.com/rust-lang-nursery/embedded-wg/issues/51)
+
+This is done as in the LLVM tools can be installed via rustup and the Cargo subcommands exist. We
+will continue to polish the user interface. To that end we'll dogfood the tools in the embedded Rust
+book.
+
+### The embedded Rust book
+
+[rust-lang-nursery/embedded-wg#56](https://github.com/rust-lang-nursery/embedded-wg/issues/56)
+
+We have to decide on the final outline, prioritization of topics and establish a process for others
+to help us write / review the book. The next meeting will focus on this topic.
+
+### Integration with C codebases
+
+[rust-lang-nursery/embedded-wg#48](https://github.com/rust-lang-nursery/embedded-wg/issues/48)
+
+This topic has a section in the embedded Rust book. This issue may be closed once we have a clear
+idea of how we are going to track progress on the book.
+
+### Test embedded targets in rust-lang/rust CI
+
+The extending Rust reach participants mentored by [@jamesmunns] are working on this. Among their
+goals they have:
+
+[@jamesmunns]: https://github.com/jamesmunns
+
+- Adding "it compiles" tests. libcore, or some other crate, compiles for $TARGET
+- Adding "it links" tests. A minimal embedded program links with different compiler options
+  (opt-level, LTO, incremental, etc.) and the presence of some symbols is checked (e.g. `main`).
+- Adding "it runs" tests. A embedded program is executed in QEMU and completes successfully.
+- Adding binary size regression tests. The binary size of a program, compiled with opt-level=s, is
+  tracked over time (per PR)   and the CI fails if the binary size of the program regresses.
+
+In their next meeting they'll discuss which of these goals are must have for edition and which ones
+are stretch goals, and a potential timeline.
+
+### MSP430 support
+
+[rust-lang-nursery/embedded-wg#48](https://github.com/rust-lang-nursery/embedded-wg/issues/48)
+
+The goal is to have a rust-std component working on stable plus up to date svd2rust support by the
+edition release. The deadlines for those are:
+
+- 1.29 (2018-09-13). rust-std component landed and enabled in rust-lang/rust
+
+- 1.30 (2018-10-25). svd2rust support up to date. This requires updating the msp430-rt crate.
+
+Compiler support for interrupts will be left unstable for this architecture. The MSP430 community
+can explore interrupt support using FFI (at a cost), but this is not a goal for the edition release.
+
+### AVR support
+
+[@japaric] will reach out to the owners of the [avr-rust] project to see if we unfork the project in
+time for the edition release. The goal would be able to develop AVR programs using Xargo with an
+official rustc compiler and an unmodified rust-src component.
+
+[@japaric]: https://github.com/japaric
+[avr-rust]: https://github.com/avr-rust
+
+### RISCV support
+
+With the upcoming LLVM 7 upgrade enabling this backend needs to be re-assessed: is it relatively bug
+free for wider use?
+
+### Material for the website
+
+As discussed during the Rust All Hands the website will include:
+
+- Videos of embedded projects
+- "Attributes" that answer the question "why Rust for embedded development?". e.g. Portability,
+  interop, etc.
+- Links to documentation: embedded Rust book, embedonomicon, etc.
+- Commercial testimonials
+
+We have issue #108 for collecting commercial testimonials. Videos will come from commercial users
+and from popular hobby projects like the [monotron]. Work on books is tracked elsewhere. The
+attributes section is yet to be written.
+
+[monotron]: https://github.com/thejpster/monotron
+
+## Commercial testimonials
+
+[rust-lang-nursery/embedded-wg#108](https://github.com/rust-lang-nursery/embedded-wg/issues/108)
+
+We quickly revisited the question of whether to include embedded Linux in the WG webpage. For now we
+agreed to just collect all sort of testimonials.
+
+Members of the WG will reach out to their contacts to see who of them are interested in providing
+testimonials.
+
+## Maintaining the Cortex-M / embedded ecosystem
+
+[rust-lang-nursery/embedded-wg#46](https://github.com/rust-lang-nursery/embedded-wg/issues/46)
+
+We didn't have time to discuss this but @japaric will revisit #46 to try to carve a more fleshed out
+proposal for assembling embedded-rust and / or rust-$ARCH GitHub orgs that will maintain the
+embedded ecosystem.