BTreeMap testing: introduce symbolic constants and use height consistently#70506
BTreeMap testing: introduce symbolic constants and use height consistently#70506bors merged 1 commit intorust-lang:masterfrom ssomers:btreemap_testing_consts
Conversation
rust-highfive
added
the
S-waiting-on-review
|
Thanks a lot, this helps to make these tests much less magic. :) |
|
📌 Commit e92d740 has been approved by |
bors
added
S-waiting-on-bors
Rollup of 3 pull requests Successful merges: - rust-lang#68692 (impl From<[T; N]> for Vec<T>) - rust-lang#70101 (Add copy bound to atomic & numeric intrinsics) - rust-lang#70506 (BTreeMap testing: introduce symbolic constants and use height consistently) Failed merges: r? @ghost
| let size = 1000; | ||
| #[cfg(miri)] | ||
| let size = 144; // to obtain height 3 tree (having edges to both kinds of nodes) | ||
| let size = MIN_INSERTS_HEIGHT_2; |
There was a problem hiding this comment.
Ah, this unfortunately broke the build in Miri as the type is now usize instead of u32... see #70559
There was a problem hiding this comment.
I also forgot to check Linux builds and debug assertions lately. So I'll just push here, I guess.
There was a problem hiding this comment.
WDYM "push here"? This PR landed already...
There was a problem hiding this comment.
miri-test-libstd also enables debug assertions, so that part is covered (also I believe some CI runners have debug assertions enabled).
There was a problem hiding this comment.
Oops, I misinterpreted. Thanks for fixing.
I'm pretty sure the ordinary PR checks a few months ago, when there were 4 of them, did not include debug assertions, so I did those myself. And --no-doc to test benchmarks.
There was a problem hiding this comment.
Yeah the PR runner probably has debug assertions disabled, but some others at least used to have them enabled.
…mulacrum fix BTreeMap test compilation with Miri This got broken by rust-lang#70506
4 participants
Doesn't change what or how much is tested, except for some exact integer types, just for convenience and because
node::CAPACITYis a usize.r? @RalfJung