Make ShardedHashMap::with_capacity split capacity between shards

[Zoxc]

This makes ShardedHashMap::with_capacity split capacity between shards. The callers in CtxtInterners::new did not divide these by the shard count, so it reserved 32 times greater capacity, significantly contributing to startup costs.

Spotted by Claude Opus while investigating startup stack usage.

[rustbot]

r? @nnethercote

rustbot has assigned @nnethercote.
They will have a look at your PR within the next two weeks and either review your PR or reassign to another reviewer.

Use r? to explicitly pick a reviewer

Why was this reviewer chosen?

The reviewer was selected based on:

• Owners of files modified in this PR: compiler, incremental, query-system
• compiler, incremental, query-system expanded to 75 candidates
• Random selection from 19 candidates

[Zoxc]

With -Z threads=sync:

Benchmark	Before	After		Before	After		Before	After
	Time	Time	%	Physical Memory	Physical Memory	%	Committed Memory	Committed Memory	%
🟣 helloworld:check	0.0612s	0.0530s	💚 -13.44%	55.64 MiB	41.71 MiB	💚 -25.04%	165.45 MiB	100.23 MiB	💚 -39.42%
Total	0.0612s	0.0530s	💚 -13.44%	55.64 MiB	41.71 MiB	💚 -25.04%	165.45 MiB	100.23 MiB	💚 -39.42%
Summary	1.0000s	0.8656s	💚 -13.44%	1 byte	0.75 bytes	💚 -25.04%	1 byte	0.61 bytes	💚 -39.42%

[rust-log-analyzer]

The job x86_64-gnu-miri failed! Check out the build log: (web) (plain enhanced) (plain)

Click to see the possible cause of the failure (guessed by this bot)

tests/pass-dep/libc/libc-socketpair.rs (revision `run`) ... ok
tests/pass-dep/libc/libc-time.rs (revision `run`) ... ok

FAILED TEST: tests/pass-dep/libc/libc-epoll-blocking.rs (revision `edge_triggered.run`)
command: MIRI_BE_RUSTC="host" MIRI_ENV_VAR_TEST="0" MIRI_TEMP="/tmp/miri-uitest-ePNOAC" RUST_BACKTRACE="1" "/checkout/obj/build/x86_64-unknown-linux-gnu/stage2-tools/x86_64-unknown-linux-gnu/tmp/miri_ui/0/tests/pass-dep/libc/libc-epoll-blockingedge_triggered"

error: test got exit status: 101, but expected 0
##[error]   --> tests/pass-dep/libc/libc-epoll-blocking.rs:316:58
    |
316 |                 let data = read_exact_array::<4>(fds[0]).unwrap();
    |                                                          ^^^^^^^^^ called `Result::unwrap()` on an `Err` value: Os { code: 9, kind: Uncategorized, message: "Bad file descriptor" }
    |

full stderr:

---
stack backtrace:
   0: __rustc::rust_begin_unwind
   1: core::panicking::panic_fmt
   2: core::result::unwrap_failed
   3: libc_epoll_blocking::waiting_threads_unblocked_after_socketpair_close::{closure#0}
   4: std::thread::scoped::scope::<libc_epoll_blocking::waiting_threads_unblocked_after_socketpair_close::{closure#0}, ()>::{closure#0}
   5: <core::panic::unwind_safe::AssertUnwindSafe<std::thread::scoped::scope<libc_epoll_blocking::waiting_threads_unblocked_after_socketpair_close::{closure#0}, ()>::{closure#0}> as core::ops::function::FnOnce<()>>::call_once
   6: std::panicking::catch_unwind::do_call::<core::panic::unwind_safe::AssertUnwindSafe<std::thread::scoped::scope<libc_epoll_blocking::waiting_threads_unblocked_after_socketpair_close::{closure#0}, ()>::{closure#0}>, ()>
   7: __rust_try
   8: std::panicking::catch_unwind::<(), core::panic::unwind_safe::AssertUnwindSafe<std::thread::scoped::scope<libc_epoll_blocking::waiting_threads_unblocked_after_socketpair_close::{closure#0}, ()>::{closure#0}>>
   9: std::thread::scoped::scope::<libc_epoll_blocking::waiting_threads_unblocked_after_socketpair_close::{closure#0}, ()>
  10: libc_epoll_blocking::waiting_threads_unblocked_after_socketpair_close
  11: libc_epoll_blocking::main
  12: <fn() as core::ops::function::FnOnce<()>>::call_once
note: Some details are omitted, run with `RUST_BACKTRACE=full` for a verbose backtrace.

full stdout:
---

Location:
   /cargo/registry/src/index.crates.io-1949cf8c6b5b557f/ui_test-0.30.7/src/lib.rs:365

Backtrace omitted. Run with RUST_BACKTRACE=1 environment variable to display it.
Run with RUST_BACKTRACE=full to include source snippets.
error: test failed, to rerun pass `--test ui`

Caused by:
  process didn't exit successfully: `/checkout/obj/build/x86_64-unknown-linux-gnu/stage2-tools/x86_64-unknown-linux-gnu/release/build/miri/855274d73082e308/out/ui-855274d73082e308 tests/pass tests/panic` (exit status: 1)
Command `/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/cargo test --target x86_64-unknown-linux-gnu -Zbinary-dep-depinfo -j 4 -Zroot-dir=/checkout --locked --color=always --profile=release --manifest-path /checkout/src/tools/miri/Cargo.toml -- tests/pass tests/panic [workdir=/checkout]` failed with exit code 1
Created at: src/bootstrap/src/core/build_steps/tool.rs:196:21
Executed at: src/bootstrap/src/core/build_steps/test.rs:771:19

Command has failed. Rerun with -v to see more details.
Bootstrap failed while executing `test --stage 2 miri -- tests/pass tests/panic`
Currently active steps:
test::Miri { target: x86_64-unknown-linux-gnu } at src/bootstrap/src/core/build_steps/test.rs:697
Build completed unsuccessfully in 0:01:55
  local time: Sat Aug 15 09:41:50 UTC 2026

[nnethercote]

Oh dear. Let's see if this affects any of the parallel benchmarks on CI.

@bors try @rust-timer queue

[rust-timer]

Awaiting bors try build completion.

@rustbot label: +S-waiting-on-perf

[rust-bors]

⌛ Trying commit 76c314f with merge 346a635…

To cancel the try build, run the command @bors try cancel.

Workflow: https://github.com/rust-lang/rust/actions/runs/31881292297