Rollup merge of #113872 - nnethercote:tweak-cgu-sorting, r=pnkfelix

Tweak CGU sorting in a couple of places.

In `base.rs`, tweak how the CGU size interleaving works. Since #113777, it's much more common to have multiple CGUs with identical sizes. With the existing code these same-sized items ended up in the opposite-to-desired order due to the stable sorting. The code now starts with a reverse sort (like is done in `partitioning.rs`) which gives the behaviour we want. This doesn't matter much for perf, but makes profiles in `samply` look more like what we expect.

In `partitioning.rs`, we can use `sort_by_key` instead of `sort_by_cached_key` because `CGU::size_estimate()` is cheap. (There is an identical CGU sort earlier in that function that already uses `sort_by_key`.)

r? `@pnkfelix`

1 files changed, 3 insertions, 2 deletions

diff --git a/compiler/rustc_codegen_ssa/src/base.rs b/compiler/rustc_codegen_ssa/src/base.rs
index dc862803274..79c284ecfbf 100644
--- a/compiler/rustc_codegen_ssa/src/base.rs
+++ b/compiler/rustc_codegen_ssa/src/base.rs
@@ -38,6 +38,7 @@ use rustc_span::symbol::sym;
 use rustc_span::Symbol;
 use rustc_target::abi::{Align, FIRST_VARIANT};
 
+use std::cmp;
 use std::collections::BTreeSet;
 use std::time::{Duration, Instant};
 
@@ -682,10 +683,10 @@ pub fn codegen_crate<B: ExtraBackendMethods>(
     // are large size variations, this can reduce memory usage significantly.
     let codegen_units: Vec<_> = {
         let mut sorted_cgus = codegen_units.iter().collect::<Vec<_>>();
-        sorted_cgus.sort_by_cached_key(|cgu| cgu.size_estimate());
+        sorted_cgus.sort_by_key(|cgu| cmp::Reverse(cgu.size_estimate()));
 
         let (first_half, second_half) = sorted_cgus.split_at(sorted_cgus.len() / 2);
-        second_half.iter().rev().interleave(first_half).copied().collect()
+        first_half.iter().interleave(second_half.iter().rev()).copied().collect()
     };
 
     // Calculate the CGU reuse