Zoxc
Zoxc
If https://github.com/rust-lang/rust/pull/122362 works out we could probably just do this change with a `#[global_allocator]`.
I'd like to see the memory use of a mimalloc + jemalloc mix.
> why can't this be done in `rustc-main`? The global allocator must be statically linked with `std`. > Will this PR require all custom drivers (including our tools) to use...
This could probably use a perf run as there could be some ThinLTO interactions.
So the `rustdoc` binary itself doesn't have the `rustc_private` feature, should we just add the feature as needed?
@lqd It all seems to be in the `native_library` query, which is a bit unexpected. We could give it another perf run?
I've made `-Z incremental_verify_ich` or `cfg!(debug_assertions)` enable these checks.
> One thing we could try to reduce the perf impact Not sure if it's worth optimizing given that it's opt-in.
> In a typical compilation, when happens the first use of this reciprocal index? > I'm tempted to say very early, as the first query to be invoked needs it....
Probably should do a perf run here as the result in https://github.com/rust-lang/rust/pull/122070 was more mixed than expected.