Ben Kimock

Hm, I'm wary of this heuristic because I tried something like it to solve the inverse problem and didn't get good results. Do you have any demonstrations of workloads where...

I'm not a GC expert, so if you haven't already, it might be good to at least draw inspiration from what techniques the really good mark-and-sweep GCs use to get...

Sure, here's all the logs: [logs.gz](https://github.com/awslabs/aws-sdk-rust/files/12787153/logs.gz) I agree that multipart is clearly the correct way to upload this, but I'm pretty sure that a 500 of all things indicates a...

I do not think it makes sense to hold this one part of the library to a standard that the rest of it, let alone the compiler's output, does not...

I think if people are going to claim that this feature has a cost to all users of Rust, they should be able to demonstrate that cost measured by our...

The maximum binary size overhead measured in our benchmark suite is 0.04%, and most benchmarks reported the size change is below their significance threshold for binary size changes. https://github.com/rust-lang/rust/pull/132081#issuecomment-2434978157 The...

There is already a perf run for that in the PR linked above, which I can link again here: https://github.com/rust-lang/rust/pull/117431

> I'd expect file paths (without nul terminator) to be substrings of formatted strings such as panic messages. Really? Are people using `file!()` in panic messages or something like that?...

> I usually enable debuginfo when profiling the compiler, but that of course doesn't help when profiling the distributed optimized artifacts :) I always set `debuginfo-level = 1` because it...

@bors r+ roll-up=iffy (may perturb perf)