Nicholas Nethercote
Nicholas Nethercote
The tiny icount regressions are matched or exceeded by the tiny icount improvements. This seems fine. @rustbot label: +perf-regression-triaged
@jyn514: I have updated with a new commit that uses `spawn_scoped`. Thanks for the fast review!
Ugh, we have a test failure, but only on dist-x86_64-msvc-alt. It panics with `'WorkerLocal can only be used on the thread pool it was created on', C:\Users\runneradmin\.cargo\registry\src\github.com-1ecc6299db9ec823\rustc-rayon-core-0.4.1\src\worker_local.rs:49:17`, which is [here](https://github.com/rust-lang/rustc-rayon/blob/9fcf0ffe28db6ffe396fb13dbf01cbb925674187/rayon-core/src/worker_local.rs#L49)....
@jyn514: I ended up using `interface::run_compiler`, because (a) `run_in_thread_pool_with_globals` isn't visible outside its crate, and (b) probably best to enter the compiler via the front door anyway, as we do...
r? @jyn514
@bors try @rust-timer queue
The instruction count results aren't a win, but there are hints of goodness in the results for cycles, wall-time, max-rss, and especially binary size. The current version only merges the...
@bors try @rust-timer queue
Disappointing results here. The code is working as intended, and is merging lots of basic blocks. Here are some measurements for three metrics: - `wc`: size of LLVM IR as...
To summarize: - MIR uses one definition of BBs, and LLVM IR uses another. Most notably, function calls end a MIR BB but don't end an LLVM IR BB. -...