abby: always store nextgen region constraints in canonical form by BoxyUwU · Pull Request #161306 · rust-lang/rust

BoxyUwU · 2026-08-18T14:30:27Z

title. introduce an And/Or/LeafConstraint/ types to reason about the structure of our region constraints. Never produce arbitrarily nested or/ands and always have constraints in an evaluated form.

I kinda mucked up this PR and accidentally did two things at the same time. Not only do we immediately put everything into canonical form, we also change what it means for a region constraint to be in canonical form. Whoops :>

Rough overview of what a RegionConstraint is:

RegionConstraint contains two things: an AND of LEAFs and an OR of AND of LEAFs. Another way of thinking about it would be to say its an AND consisting of arbitrarily many LEAFs and a single OR of AND of LEAFs
There are never any region constraints shared between all ANDs of the OR, instead they're moved into the top level AND
If the OR constraint is false then we wipe the top level AND as it doesn't matter what they are, the constraint is always going to be false
ORs never have two equivalent ANDs within them. Similarly, ANDs never have two equivalent LEAFs within them

this simplifies a lot of things conceptually as we now no longer need to worry about what state our region constraints are in. and our algorithms also don't need to handle arbitrary nesting of ors/ands :) and its a lot easier to read the debug logs 😅

I also wound up needing to do this while trying to compile std/core with -Zassumptions-on-binders as we would otherwise OOM from having both:

Lots of duplicate region constraints (e.g. And('a: 'b, 'a: 'b))
Lots of region constraints shared across all elements of an OR (e.g. Or(And('a: 'b, 'b: 'c), And('a: 'b, 'b: 'd)))

Some future work:

Remove RegionConstraint::splatted_and_constraints it's kind of weird to even need it and probably encourages bad-for-perf patterns
we Probably want some kind of fast path for pushing new leaf constraints to the region constraint storage. slash have a way to register a leaf constraint directly rather than having to make a RegionConstraint. Perf stuff :3

In theory this PR should mostly not have functional changes. In practice it might affect some things due to changing the exact repr of things affecting query responses. There's probably also some behaviour differences here due to us falling on our face more or less in WIP parts of abby now that we have different region constraints. I don't think any of this should be meaningful though. This PR is intended to not fundamentally change the abby algorithm :3

This PR should be reviewed commit-by-commit. There are a bunch of commits restructuring existing logic to assume their input is in canonical form as it will be by the end of the PR.

Then there's the core change in always canonical form which actually replaces RegionConstraint with all the new types and updates all the locations using them.

Finally there's propagate ambiguity not evaluate which deals with the leftover evaluate_solver_constraint which was mostly unnecessary now due to moving its main logic into construction of RegionConstraint and friends. I didn't want to make actual bug fixes in this PR so I just left some FIXMEs about some of the issues that propagate_ambiguity has instead of fixing them here.

Fixes rust-lang/project-assumptions-on-binders#14

BoxyUwU · 2026-08-21T16:24:22Z

r? @lcnr @khyperia

rustbot · 2026-08-27T19:32:15Z

This PR was rebased onto a different main commit. Here's a range-diff highlighting what actually changed.

Rebasing is a normal part of keeping PRs up to date, so no action is needed—this note is just to help reviewers.

BoxyUwU · 2026-08-27T20:13:12Z

did it take me too long to rebase and I need to rebase again lolsob

khyperia

looks good to me! didn't rereview region_constraint.rs suuuper closely, mostly just assuming you haven't committed terrible crimes in the rebase ✨

View changes since this review

khyperia · 2026-08-28T05:13:19Z

+                .into_iter()
+                .map(|item| self.lower_test_binder_constraint(item))
+                .reduce(SolverRegionConstraint::new_or)
+                .unwrap_or(SolverRegionConstraint::new_true()),


this took me a moment, aaah, the SolverRegionConstraint::new_{and,or} auto-canonicalize as they're constructing. neat! (if I did this myself I might have rejected noncanonical syntax as invalid, but canonicalizing it makes sense too~)

an empty OR should be false, should it not?

It is not if you implement it wrong like I did

BoxyUwU · 2026-08-28T09:18:18Z

Should move the trait bounds on the span generic parameter into a new trait with a blanket impl to avoid the boilerplate everywhere actually

lcnr · 2026-08-28T09:43:43Z

-            }
-
-            Or(candidates.into_boxed_slice())
+            rewrite_placeholder_ty_outlives_constraints_in_universe_for_eager_placeholder_handling(


consider moving these functions into an eager_placeholder_handling module and removing the for_X from their name :> horribly long name

View changes since the review

lcnr · 2026-08-28T09:47:47Z

-    /// Converts the region constraint into an ORs of ANDs of "leaf" constraints. Where
-    /// a leaf constraint is a non-or/and constraint.
-    #[instrument(level = "debug", ret)]
-    pub fn canonical_form(self) -> Self {


removing canonical_form in this commit feels wrong :>

View changes since the review

hmm should just rename the commit. this was mostly intended as a "delete a bunch of junk to make the diff nicer"

lcnr · 2026-08-28T10:07:56Z

-            // The alias is either rigid or ambiguous in which case we'll return with ambiguity.
            Alias(_, alias) => self.destructure_alias_outlives(*alias, r),
-            UnresolvedInferenceVariable(_) => RegionConstraint::Ambiguity(()),
+            UnresolvedInferenceVariable(_) => Or::new_ambig(()),


vibe for future PR, have a NoSpans type instead of using () as () here is slightly confusing.

View changes since the review

rust-lang/project-assumptions-on-binders#34

lcnr

none of this is blocking this PR itself

View changes since this review

lcnr · 2026-08-28T10:16:51Z

+    }
+
+    pub fn new_and(a: Or<I, S>, b: Or<I, S>) -> Self {
+        // I think this returns false if either a or b is false?


horrible comment 🤣

yes, this returns Empty if one of the two is empty. We're building the cartesian product here.

This function also doesn't return something in canonical form, does it? if you have

[[A, B], [A]] and [[C], [B, C]] you end up with [A, B, C], [A, B, C] (deduped by And::new 🤔), [A, C], and another [A, B, C]

might be good to have a new_raw with a debug_assert that we're in canonical form?

Or::new does all the actual canonicalizing which is I guess the equivalent to your new_raw 🤔 having a debug assert would be nice though

BoxyUwU · 2026-08-28T13:57:05Z

going to make canonical form be span insensitive

rust-log-analyzer · 2026-08-28T14:37:04Z

The job pr-check-2 failed! Check out the build log: (web) (plain enhanced) (plain)

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

 Documenting rustc_middle v0.0.0 (/checkout/compiler/rustc_middle)
[RUSTC-TIMING] rustc_next_trait_solver test:false 1.740
 Documenting rustc_llvm v0.0.0 (/checkout/compiler/rustc_llvm)
 Documenting rustc_ast_passes v0.0.0 (/checkout/compiler/rustc_ast_passes)
warning: unresolved link to `RegionConstraint::AliasTyOutlivesViaEnv`
   --> compiler/rustc_next_trait_solver/src/solve/eval_ctxt/solver_region_constraints.rs:191:52
    |
191 |     ///     when exiting the current binder. See [`RegionConstraint::AliasTyOutlivesViaEnv`].
    |                                                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ no item named `RegionConstraint` in scope
    |
    = note: `#[warn(rustdoc::broken_intra_doc_links)]` on by default

 Documenting rustc_windows_rc v0.0.0 (/checkout/compiler/rustc_windows_rc)
error: `rustc_next_trait_solver` (lib doc) generated 1 warning
error: warnings are denied by `build.warnings` configuration
warning: build failed, waiting for other jobs to finish...
[RUSTC-TIMING] rustc_middle test:false 13.522
Bootstrap failed while executing `doc compiler --stage 1`
Currently active steps:
doc::Rustc { build_compiler: Compiler { stage: 0, host: x86_64-unknown-linux-gnu, forced_compiler: false }, target: x86_64-unknown-linux-gnu, crates: ["rustc-main", "rustc_abi", "rustc_arena", "rustc_ast", "rustc_ast_ir", "rustc_ast_lowering", "rustc_ast_passes", "rustc_ast_pretty", "rustc_attr_ir", "rustc_attr_parsing", "rustc_baked_icu_data", "rustc_borrowck", "rustc_builtin_macros", "rustc_codegen_llvm", "rustc_codegen_ssa", "rustc_const_eval", "rustc_crate_store", "rustc_data_structures", "rustc_driver", "rustc_driver_impl", "rustc_error_codes", "rustc_error_messages", "rustc_errors", "rustc_expand", "rustc_feature", "rustc_fs_util", "rustc_graphviz", "rustc_hashes", "rustc_hir", "rustc_hir_analysis", "rustc_hir_id", "rustc_hir_pretty", "rustc_hir_typeck", "rustc_incremental", "rustc_index", "rustc_index_macros", "rustc_infer", "rustc_interface", "rustc_lexer", "rustc_lint", "rustc_lint_defs", "rustc_llvm", "rustc_log", "rustc_macros", "rustc_metadata", "rustc_middle", "rustc_mir_build", "rustc_mir_dataflow", "rustc_mir_transform", "rustc_monomorphize", "rustc_next_trait_solver", "rustc_parse", "rustc_parse_format", "rustc_passes", "rustc_pattern_analysis", "rustc_privacy", "rustc_proc_macro", "rustc_public", "rustc_public_bridge", "rustc_query_impl", "rustc_resolve", "rustc_sanitizers", "rustc_serialize", "rustc_session", "rustc_span", "rustc_structures", "rustc_symbol_mangling", "rustc_target", "rustc_thread_pool", "rustc_trait_selection", "rustc_traits", "rustc_transmute", "rustc_ty_utils", "rustc_ty_walk", "rustc_type_ir", "rustc_type_ir_macros", "rustc_windows_rc"] } at src/bootstrap/src/core/build_steps/doc.rs:916
Command `/checkout/obj/build/x86_64-unknown-linux-gnu/stage0/bin/cargo doc --target x86_64-unknown-linux-gnu -Zbinary-dep-depinfo -j 4 -Zroot-dir=/checkout -Zembed-metadata=no --locked --color=always --profile=release --features 'llvm rustc' --manifest-path /checkout/compiler/rustc/Cargo.toml -Zskip-rustdoc-fingerprint --no-deps -Zrustdoc-map -p rustc-main -p rustc_abi -p rustc_arena -p rustc_ast -p rustc_ast_ir -p rustc_ast_lowering -p rustc_ast_passes -p rustc_ast_pretty -p rustc_attr_ir -p rustc_attr_parsing -p rustc_baked_icu_data -p rustc_borrowck -p rustc_builtin_macros -p rustc_codegen_llvm -p rustc_codegen_ssa -p rustc_const_eval -p rustc_crate_store -p rustc_data_structures -p rustc_driver -p rustc_driver_impl -p rustc_error_codes -p rustc_error_messages -p rustc_errors -p rustc_expand -p rustc_feature -p rustc_fs_util -p rustc_graphviz -p rustc_hashes -p rustc_hir -p rustc_hir_analysis -p rustc_hir_id -p rustc_hir_pretty -p rustc_hir_typeck -p rustc_incremental -p rustc_index -p rustc_index_macros -p rustc_infer -p rustc_interface -p rustc_lexer -p rustc_lint -p rustc_lint_defs -p rustc_llvm -p rustc_log -p rustc_macros -p rustc_metadata -p rustc_middle -p rustc_mir_build -p rustc_mir_dataflow -p rustc_mir_transform -p rustc_monomorphize -p rustc_next_trait_solver -p rustc_parse -p rustc_parse_format -p rustc_passes -p rustc_pattern_analysis -p rustc_privacy -p rustc_proc_macro -p rustc_public -p rustc_public_bridge -p rustc_query_impl -p rustc_resolve -p rustc_sanitizers -p rustc_serialize -p rustc_session -p rustc_span -p rustc_structures -p rustc_symbol_mangling -p rustc_target -p rustc_thread_pool -p rustc_trait_selection -p rustc_traits -p rustc_transmute -p rustc_ty_utils -p rustc_ty_walk -p rustc_type_ir -p rustc_type_ir_macros -p rustc_windows_rc [workdir=/checkout]` failed with exit code 101
Created at: src/bootstrap/src/core/build_steps/doc.rs:946:25
Executed at: src/bootstrap/src/core/build_steps/doc.rs:1007:26

Command has failed. Rerun with -v to see more details.
Build completed unsuccessfully in 0:00:37
  local time: Fri Aug 28 14:36:53 UTC 2026
  network time: Fri, 28 Aug 2026 14:36:53 GMT
##[error]Process completed with exit code 1.
##[group]Run echo "disk usage:"