Tracking Issue for `const_collections_with_hasher` #102575

aDotInTheVoid · 2022-10-02T12:06:03Z

Feature gate: #![feature(const_collections_with_hasher)]

This is a tracking issue for making HashMap::with_hasher and HashSet::with_hasher const.

Public API

// std::collections
impl<K, V, S> HashMap<K, V, S> {
    pub const fn with_hasher(hash_builder: S) -> HashMap<K, V, S> { ... }
}

impl<T, S> HashSet<T, S> {
    pub const fn with_hasher(hasher: S) -> HashSet<T, S> { ... }
}

Steps / History

Implementation: Make Hash{Set,Map}::with_hasher unstably const #102574
Final comment period (FCP)¹
Stabilization PR ( const-stabilize HashMap/HashSet::with_hasher (+ required compiller changes) #118427)

Unresolved Questions

None yet.

https://std-dev-guide.rust-lang.org/feature-lifecycle/stabilization.html ↩

The text was updated successfully, but these errors were encountered:

…th_hasher, r=oli-obk,fee1-dead Make Hash{Set,Map}::with_hasher unstably const Makes [`HashMap::with_hasher`](https://doc.rust-lang.org/stable/std/collections/hash_map/struct.HashMap.html#method.with_hasher) and [`HashSet::with_hasher`](https://doc.rust-lang.org/stable/std/collections/hash_set/struct.HashSet.html#method.with_hasher) `const`. This allows ```rust static GlobalState: Mutex<HashMap<i32, i32, SomeHasher>> = Mutex::new(HashMap::with_hasher(SomeHasher::new())) ``` Tracking issue: rust-lang#102575

GoldsteinE · 2023-01-03T11:17:52Z

What blocks this from stabilization? Can I do something to move it forward?

aDotInTheVoid · 2023-01-05T17:08:02Z

@GoldsteinE: see https://std-dev-guide.rust-lang.org/feature-lifecycle/stabilization.html

Main questions for stabilization are do we want to add the constraint that an empty hashmap can be const constructed (ie doesn't alloc and more). This is true for the current hashbrown design, but might not be true for other alternatives. I'm not framiliar enough with hashmap design to say.

Amanieu · 2023-08-03T01:02:51Z

This is fairly straightforward, we are essentially guaranteeing that a newly created HashMap does not allocate any memory.

@rfcbot fcp merge

rfcbot · 2023-08-03T01:02:52Z

Team member @Amanieu has proposed to merge this. The next step is review by the rest of the tagged team members:

No concerns currently listed.

Once a majority of reviewers approve (and at most 2 approvals are outstanding), this will enter its final comment period. If you spot a major issue that hasn't been raised at any point in this process, please speak up!

See this document for info about what commands tagged team members can give me.

rfcbot · 2023-08-03T08:45:35Z

🔔 This is now entering its final comment period, as per the review above. 🔔

rfcbot · 2023-08-13T08:48:07Z

The final comment period, with a disposition to merge, as per the review above, is now complete.

As the automated representative of the governance process, I would like to thank the author for their work and everyone else who contributed.

This will be merged soon.

aDotInTheVoid · 2023-11-07T17:57:14Z

Trying to stabilize this, I run into this error:

error: `hashbrown::HashMap::<K, V, S>::with_hasher` is not yet stable as a const fn
   --> library/std/src/collections/hash/map.rs:292:25
    |
292 |         HashMap { base: base::HashMap::with_hasher(hash_builder) }
    |                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
    |

This code here

rust/compiler/rustc_const_eval/src/transform/check_consts/check.rs

Lines 984 to 995 in 9bd71af

    
           // FIXME(ecstaticmorse); For compatibility, we consider `unstable` callees that 
        
           // have no `rustc_const_stable` attributes to be const-unstable as well. This 
        
           // should be fixed later. 
        
           let callee_is_unstable_unmarked = tcx.lookup_const_stability(callee).is_none() 
        
               && tcx.lookup_stability(callee).is_some_and(|s| s.is_unstable()); 
        
           if callee_is_unstable_unmarked { 
        
               trace!("callee_is_unstable_unmarked"); 
        
               // We do not use `const` modifiers for intrinsic "functions", as intrinsics are 
        
               // `extern` functions, and these have no way to get marked `const`. So instead we 
        
               // use `rustc_const_(un)stable` attributes to mean that the intrinsic is `const` 
        
               if self.ccx.is_const_stable_const_fn() || tcx.is_intrinsic(callee) { 
        
                   self.check_op(ops::FnCallUnstable(callee, None));

seems to think it's unstable, despite hashbrown having no stability annotations. CC @ecstatic-morse

aDotInTheVoid · 2023-11-28T20:23:12Z

Solved in #118423 by changing the compiler to respect rustc_allow_const_fn_unstable here. This requires new semantics as there's no feature gate on the hashbrown side, so I said any will work here. Not sure if this is OK.

This also means a load of ugly things are needed to be cfg_attr(bootstrap, ...)ed

@rustbot claim

Because std's dependencies are build with `-Zforce-unstable-if-unmarked` normally, std is unable to call them in const-stable function. However, becuase they had no explicit feature gates, there was no way to use `rustc_allow_const_fn_unstable` to allow this. Therefor we add `#[rustc_allow_const_fn_unstable(any)]` to allow using these functions. The reson to do this is rust-lang#102575, which relies on calling hashbrow function in const-stable std functions.

zk4x · 2024-10-24T06:30:29Z

Can we get an ETA how many years will this approximately take to stabilize? How can I speed this process? Or should I just write my own hashmap or use alloc::collections::BTreeMap? Thanks ahead for any info.

aDotInTheVoid · 2024-10-24T08:29:43Z

Sorry, I should have given an update to this. The FCP’s been approved, so the only thing remaining it to actually do the stabilisation. Unfortunately that’s actually tricky in this case, due to HashMap being implemented in an external crate (hashbrown), it has really painful interactions with the const-stability checker.

Those checks are currently being rewritten in #131349, so after that lands it should be possible to try again if you (or someone else) wants to work on it.

RalfJung · 2024-11-02T10:32:29Z

I am planning to make this possible as part of finishing my work on #129815. The external hashbrown crate will also need some changes.

However, you won't be able to construct a HashMap<K, V> in const since that one uses randomized keys. You will have to set a non-default hasher builder. See the test being added in #132503.

better test for const HashMap; remove const_hash leftovers The existing `const_with_hasher` test is kind of silly since the HashMap it constructs can never contain any elements. So this adjusts the test to construct a usable HashMap, which is a bit non-trivial since the default hash builder cannot be built in `const`. `BuildHasherDefault::new()` helps but is unstable (rust-lang#123197), so we also have a test that does not involve that type. The second commit removes the last remnants of rust-lang#104061, since they aren't actually useful -- without const traits, you can't do any hashing in `const`. Cc `@rust-lang/libs-api` `@rust-lang/wg-const-eval` Closes rust-lang#104061 Related to rust-lang#102575

better test for const HashMap; remove const_hash leftovers The existing `const_with_hasher` test is kind of silly since the HashMap it constructs can never contain any elements. So this adjusts the test to construct a usable HashMap, which is a bit non-trivial since the default hash builder cannot be built in `const`. `BuildHasherDefault::new()` helps but is unstable (rust-lang#123197), so we also have a test that does not involve that type. The second commit removes the last remnants of rust-lang#104061, since they aren't actually useful -- without const traits, you can't do any hashing in `const`. Cc ``@rust-lang/libs-api`` ``@rust-lang/wg-const-eval`` Closes rust-lang#104061 Related to rust-lang#102575

better test for const HashMap; remove const_hash leftovers The existing `const_with_hasher` test is kind of silly since the HashMap it constructs can never contain any elements. So this adjusts the test to construct a usable HashMap, which is a bit non-trivial since the default hash builder cannot be built in `const`. `BuildHasherDefault::new()` helps but is unstable (rust-lang#123197), so we also have a test that does not involve that type. The second commit removes the last remnants of rust-lang#104061, since they aren't actually useful -- without const traits, you can't do any hashing in `const`. Cc ```@rust-lang/libs-api``` ```@rust-lang/wg-const-eval``` Closes rust-lang#104061 Related to rust-lang#102575

better test for const HashMap; remove const_hash leftovers The existing `const_with_hasher` test is kind of silly since the HashMap it constructs can never contain any elements. So this adjusts the test to construct a usable HashMap, which is a bit non-trivial since the default hash builder cannot be built in `const`. `BuildHasherDefault::new()` helps but is unstable (rust-lang#123197), so we also have a test that does not involve that type. The second commit removes the last remnants of rust-lang#104061, since they aren't actually useful -- without const traits, you can't do any hashing in `const`. Cc `@rust-lang/libs-api` `@rust-lang/wg-const-eval` Closes rust-lang#104061 Related to rust-lang#102575

better test for const HashMap; remove const_hash leftovers The existing `const_with_hasher` test is kind of silly since the HashMap it constructs can never contain any elements. So this adjusts the test to construct a usable HashMap, which is a bit non-trivial since the default hash builder cannot be built in `const`. `BuildHasherDefault::new()` helps but is unstable (rust-lang#123197), so we also have a test that does not involve that type. The second commit removes the last remnants of rust-lang#104061, since they aren't actually useful -- without const traits, you can't do any hashing in `const`. Cc ``@rust-lang/libs-api`` ``@rust-lang/wg-const-eval`` Closes rust-lang#104061 Related to rust-lang#102575

Rollup merge of rust-lang#132503 - RalfJung:const-hash-map, r=Amanieu better test for const HashMap; remove const_hash leftovers The existing `const_with_hasher` test is kind of silly since the HashMap it constructs can never contain any elements. So this adjusts the test to construct a usable HashMap, which is a bit non-trivial since the default hash builder cannot be built in `const`. `BuildHasherDefault::new()` helps but is unstable (rust-lang#123197), so we also have a test that does not involve that type. The second commit removes the last remnants of rust-lang#104061, since they aren't actually useful -- without const traits, you can't do any hashing in `const`. Cc ``@rust-lang/libs-api`` ``@rust-lang/wg-const-eval`` Closes rust-lang#104061 Related to rust-lang#102575

RalfJung · 2024-11-03T19:03:22Z

The PR that should enable us to safely expose const fn from hashbrown is up now: #132541.

(That's just the infrastructure for this, once it lands, a bit of wiring-up needs to happen in hashbrown.)

RalfJung · 2024-11-12T06:57:31Z

Cc @rust-lang/wg-const-eval

bump hashbrown version This pulls in rust-lang/hashbrown#586, in preparation for rust-lang#102575. Cc `@Amanieu`

bump hashbrown version This pulls in rust-lang/hashbrown#586, in preparation for rust-lang#102575. Cc ``@Amanieu``

Rollup merge of rust-lang#133670 - RalfJung:hashbrown, r=Amanieu bump hashbrown version This pulls in rust-lang/hashbrown#586, in preparation for rust-lang#102575. Cc ``@Amanieu``

aDotInTheVoid added C-tracking-issue Category: An issue tracking the progress of sth. like the implementation of an RFC T-libs-api Relevant to the library API team, which will review and decide on the PR/issue. labels Oct 2, 2022

aDotInTheVoid mentioned this issue Oct 2, 2022

Make Hash{Set,Map}::with_hasher unstably const #102574

Merged

apiraino removed the to-announce Announce this issue on triage meeting label Aug 18, 2023

aDotInTheVoid self-assigned this Aug 20, 2023

aDotInTheVoid removed their assignment Sep 2, 2023

rustbot assigned aDotInTheVoid Nov 28, 2023

aDotInTheVoid mentioned this issue Nov 29, 2023

const-stabilize HashMap/HashSet::with_hasher (+ required compiller changes) #118427

Closed

aDotInTheVoid removed their assignment Oct 24, 2024

RalfJung mentioned this issue Nov 2, 2024

better test for const HashMap; remove const_hash leftovers #132503

Merged

RalfJung mentioned this issue Nov 12, 2024

Tracking Issue for fn const BuildHasherDefault::new() #123197

Open

3 tasks

RalfJung mentioned this issue Nov 13, 2024

mark const fn constructors as rustc_const_stable_indirect rust-lang/hashbrown#586

Merged

RalfJung mentioned this issue Nov 30, 2024

bump hashbrown version #133670

Merged

RalfJung added a commit to RalfJung/rust that referenced this issue Nov 30, 2024

Rollup merge of rust-lang#133670 - RalfJung:hashbrown, r=Amanieu

61adb5b

bump hashbrown version This pulls in rust-lang/hashbrown#586, in preparation for rust-lang#102575. Cc `@Amanieu`

RalfJung added a commit to RalfJung/rust that referenced this issue Nov 30, 2024

Rollup merge of rust-lang#133670 - RalfJung:hashbrown, r=Amanieu

ec000a7

bump hashbrown version This pulls in rust-lang/hashbrown#586, in preparation for rust-lang#102575. Cc ``@Amanieu``

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Tracking Issue for `const_collections_with_hasher` #102575

Tracking Issue for `const_collections_with_hasher` #102575

aDotInTheVoid commented Oct 2, 2022 •

edited

Loading

GoldsteinE commented Jan 3, 2023

aDotInTheVoid commented Jan 5, 2023

Amanieu commented Aug 3, 2023

rfcbot commented Aug 3, 2023 •

edited by m-ou-se

Loading

rfcbot commented Aug 3, 2023

rfcbot commented Aug 13, 2023

aDotInTheVoid commented Nov 7, 2023

aDotInTheVoid commented Nov 28, 2023 •

edited

Loading

zk4x commented Oct 24, 2024

aDotInTheVoid commented Oct 24, 2024

RalfJung commented Nov 2, 2024 •

edited

Loading

RalfJung commented Nov 3, 2024

RalfJung commented Nov 12, 2024

Tracking Issue for const_collections_with_hasher #102575

Tracking Issue for const_collections_with_hasher #102575

Comments

aDotInTheVoid commented Oct 2, 2022 • edited Loading

Public API

Steps / History

Unresolved Questions

Footnotes

GoldsteinE commented Jan 3, 2023

aDotInTheVoid commented Jan 5, 2023

Amanieu commented Aug 3, 2023

rfcbot commented Aug 3, 2023 • edited by m-ou-se Loading

rfcbot commented Aug 3, 2023

rfcbot commented Aug 13, 2023

aDotInTheVoid commented Nov 7, 2023

aDotInTheVoid commented Nov 28, 2023 • edited Loading

zk4x commented Oct 24, 2024

aDotInTheVoid commented Oct 24, 2024

RalfJung commented Nov 2, 2024 • edited Loading

RalfJung commented Nov 3, 2024

RalfJung commented Nov 12, 2024

Tracking Issue for `const_collections_with_hasher` #102575

Tracking Issue for `const_collections_with_hasher` #102575

aDotInTheVoid commented Oct 2, 2022 •

edited

Loading

rfcbot commented Aug 3, 2023 •

edited by m-ou-se

Loading

aDotInTheVoid commented Nov 28, 2023 •

edited

Loading

RalfJung commented Nov 2, 2024 •

edited

Loading