Commit Graph

839 Commits

Author SHA1 Message Date
Michael Goulet
b096f0e0f0 Make permit_uninit/zero_init fallible 2023-02-14 22:37:30 +00:00
Nicholas Nethercote
9a53cee868 Replace an unnecessary mk_ty call with mk_array. 2023-02-15 09:04:21 +11:00
Oli Scherer
936bf29d4c s/eval_usize/eval_target_usize/ for clarity 2023-02-14 08:51:19 +00:00
Camille GILLOT
09797a463c Typo. 2023-02-13 17:01:03 +00:00
Ben Kimock
640ede7b0a Enable CopyProp by default, tune the impl a bit 2023-02-12 13:23:53 -05:00
bors
585f3eef26 Auto merge of #107851 - cjgillot:sroa-const, r=oli-obk
Put deaggregated statements after original constant.

Fixes https://github.com/rust-lang/rust/issues/107818
2023-02-11 20:59:18 +00:00
bors
5a8dfd933a Auto merge of #85158 - JulianKnodt:array_const_val, r=cjgillot
Mir-Opt for copying enums with large discrepancies

I have been meaning to make this for quite a while, based off of this [hackmd](https://hackmd.io/`@ft4bxUsFT5CEUBmRKYHr7w/rJM8BBPzD).`

I'm not sure where to put this opt now that I've made it, so I'd appreciate suggestions on that!
It's also one long chain of statements, not sure if there's a more friendly format to make it.

r? `@tmiasko`
I would `r` oli but he's on leave so he suggested I `r` tmiasko or wesleywiser.
2023-02-10 21:49:31 +00:00
Camille GILLOT
221ea3080d Put deaggregated statements after original constant. 2023-02-09 17:27:31 +00:00
Matthias Krüger
05748c66a0
Rollup merge of #107271 - Zeegomo:drop-rmw, r=oli-obk
Treat Drop as a rmw operation

Previously, a Drop terminator was considered a move in MIR. This commit changes the behavior to only treat Drop as a mutable access to the dropped place.

In order for this change to be correct, we need to guarantee that

1.  A dropped value won't be used again
   2.  Places that appear in a drop won't be used again before a
     subsequent initialization.

We can ensure this to be correct at MIR construction because Drop will only be emitted when a variable goes out of scope, thus having:
*   (1) as there is no way of reaching the old value. drop-elaboration
     will also remove any uninitialized drop.
 * (2) as the place can't be named following the end of the scope.

However, the initialization status, previously tracked by moves, should also be tied to the execution of a Drop, hence the additional logic in the dataflow analyses.

From discussion in [this thread](https://rust-lang.zulipchat.com/#narrow/stream/233931-t-compiler.2Fmajor-changes/topic/.60DROP.60.20to.20.60DROP_IF.60.20compiler-team.23558), originating from https://github.com/rust-lang/compiler-team/issues/558.
See also https://github.com/rust-lang/rust/pull/104488#discussion_r1085556010
2023-02-08 18:32:41 +01:00
kadmin
15d4728cda Add de-init to destination place 2023-02-08 02:04:07 +00:00
Matthias Krüger
306dbaf574
Rollup merge of #107662 - cjgillot:copy-projection, r=oli-obk
Turn projections into copies in CopyProp.

The current implementation can leave behind projections that are moved out several times.

This PR widens the check to turn such moves into copies: a move out of a projection of a copy is equivalent to a copy of the original projection.
2023-02-07 17:57:15 +01:00
kadmin
15f4eec7a9 Leave FIXME for wasm layout difference.
There is a distinction between running this on wasm and i686, even though they should be
identical. This technically is not _incorrect_, it's just an unexpected difference, which is
worth investigating, but not for correctness.
2023-02-07 09:37:55 +00:00
kadmin
5d9f5145ac Rm allocation in candidate
Instead of storing an extra array for discriminant values, create an allocation there and store
those in an allocation immediately.
2023-02-07 09:37:55 +00:00
kadmin
33b4d20338 Clean up MIR transform 2023-02-07 09:37:55 +00:00
Camille GILLOT
df889c9821 Rename assign_idx methods. 2023-02-06 22:05:05 +00:00
Camille GILLOT
c48756cdbf Limit creation of tracked place directly. 2023-02-06 21:55:05 +00:00
Camille GILLOT
9a6c04f5d0 Handle discriminants in dataflow-const-prop. 2023-02-06 21:51:47 +00:00
Camille GILLOT
cd3649b2a5 Only exclude locals if the place is not indirect. 2023-02-06 21:51:45 +00:00
Camille Gillot
9c5add14e7
Comment move->copy transform. 2023-02-06 18:14:24 +01:00
Camille GILLOT
29856180a8 Simplify ReplacementMap. 2023-02-05 13:41:24 +00:00
Camille GILLOT
8e05ab04e5 Run SROA to fixpoint. 2023-02-05 12:08:42 +00:00
Camille GILLOT
42c9514629 Simplify construction of replacement map. 2023-02-05 11:44:18 +00:00
Camille GILLOT
e465d647b1 Introduce helper. 2023-02-05 11:42:12 +00:00
Camille GILLOT
dc4fe8e295 Make SROA expand assignments. 2023-02-05 11:42:11 +00:00
Camille GILLOT
0843acbea6 Fix SROA without deaggregation. 2023-02-05 08:37:03 +00:00
Camille GILLOT
5c1cb5bbc6 Turn projections into copies in CopyProp. 2023-02-04 23:33:33 +00:00
Camille GILLOT
134d819072 Stop deaggegating MIR. 2023-02-02 23:20:29 +00:00
Camille GILLOT
6a0b218161 Stop deaggregating enums in MIR. 2023-02-02 23:20:27 +00:00
Camille GILLOT
b62a9da0c8 Handle aggregates in DataflowConstProp. 2023-02-02 23:09:51 +00:00
Camille GILLOT
0241c29123 Put a DefId in AggregateKind. 2023-02-02 23:09:51 +00:00
Matthias Krüger
6917040cf0
Rollup merge of #107524 - cjgillot:both-storage, r=RalfJung
Remove both StorageLive and StorageDead in CopyProp.

Fixes https://github.com/rust-lang/rust/issues/107511

https://github.com/rust-lang/rust/pull/106908 removed StorageDead without the accompanying StorageLive. In loops, execution would see repeated StorageLive, without any StorageDead, which is UB.

So when removing storage statements, we have to remove both StorageLive and StorageDead.

~I also added a MIR validation pass for StorageLive. It may be a bit overzealous.~
2023-02-02 17:14:06 +01:00
bors
ad8e1dc286 Auto merge of #107536 - GuillaumeGomez:rollup-xv7dx2h, r=GuillaumeGomez
Rollup of 12 pull requests

Successful merges:

 - #106898 (Include both md and yaml ICE ticket templates)
 - #107331 (Clean up eslint annotations and remove unused JS function)
 - #107348 (small refactor to new projection code)
 - #107354 (rustdoc: update Source Serif 4 from 4.004 to 4.005)
 - #107412 (avoid needless checks)
 - #107467 (Improve enum checks)
 - #107486 (Track bound types like bound regions)
 - #107491 (rustdoc: remove unused CSS from `.setting-check`)
 - #107508 (`Edition` micro refactor)
 - #107525 (PointeeInfo is advisory only)
 - #107527 (rustdoc: stop making unstable items transparent)
 - #107535 (Replace unwrap with ? in TcpListener doc)

Failed merges:

r? `@ghost`
`@rustbot` modify labels: rollup
2023-02-01 01:15:02 +00:00
Guillaume Gomez
53bb6322db
Rollup merge of #107467 - WaffleLapkin:uneq, r=oli-obk
Improve enum checks

Some light refactoring.
2023-01-31 23:38:52 +01:00
Ralf Jung
dfc4a7b2d0 make unaligned_reference a hard error 2023-01-31 20:28:11 +01:00
Camille GILLOT
3c10cf088a Remove both StorageLive and StorageDead in CopyProp. 2023-01-31 17:50:04 +00:00
bors
dc3e59cb3f Auto merge of #107443 - cjgillot:generator-less-query, r=compiler-errors
Test drop_tracking_mir before querying generator.

r? `@ghost`
2023-01-31 02:46:11 +00:00
Matthias Krüger
db9774951d
Rollup merge of #107172 - cjgillot:no-nal, r=nagisa
Reimplement NormalizeArrayLen based on SsaLocals

Based on https://github.com/rust-lang/rust/pull/106908
Fixes https://github.com/rust-lang/rust/issues/105929

Only the last commit "Reimplement NormalizeArrayLen" is relevant.
2023-01-30 17:50:09 +01:00
Maybe Waffle
f1d273cbfb Replace some _ == _ || _ == _s with matches!(_, _ | _)s 2023-01-30 12:26:26 +00:00
Maybe Waffle
4d75f61832 Use Mutability::{is_mut, is_not} 2023-01-30 12:26:26 +00:00
Maybe Waffle
fd649a3cc5 Replace enum ==s with matches where it makes sense 2023-01-30 12:26:26 +00:00
Nicholas Nethercote
2e93f2c92f Allow more deriving on packed structs.
Currently, deriving on packed structs has some non-trivial limitations,
related to the fact that taking references on unaligned fields is UB.

The current approach to field accesses in derived code:
- Normal case: `&self.0`
- In a packed struct that derives `Copy`: `&{self.0}`
- In a packed struct that doesn't derive `Copy`: `&self.0`

Plus, we disallow deriving any builtin traits other than `Default` for any
packed generic type, because it's possible that there might be
misaligned fields. This is a fairly broad restriction.

Plus, we disallow deriving any builtin traits other than `Default` for most
packed types that don't derive `Copy`. (The exceptions are those where the
alignments inherently satisfy the packing, e.g. in a type with
`repr(packed(N))` where all the fields have alignments of `N` or less
anyway. Such types are pretty strange, because the `packed` attribute is
not having any effect.)

This commit introduces a new, simpler approach to field accesses:
- Normal case: `&self.0`
- In a packed struct: `&{self.0}`

In the latter case, this requires that all fields impl `Copy`, which is
a new restriction. This means that the following example compiles under
the old approach and doesn't compile under the new approach.
```
 #[derive(Debug)]
 struct NonCopy(u8);

 #[derive(Debug)
 #[repr(packed)]
 struct MyType(NonCopy);
```
(Note that the old approach's support for cases like this was brittle.
Changing the `u8` to a `u16` would be enough to stop it working. So not
much capability is lost here.)

However, the other constraints from the old rules are removed. We can now
derive builtin traits for packed generic structs like this:
```
 trait Trait { type A; }

 #[derive(Hash)]
 #[repr(packed)]
 pub struct Foo<T: Trait>(T, T::A);
```
To allow this, we add a `T: Copy` bound in the derived impl and a `T::A:
Copy` bound in where clauses. So `T` and `T::A` must impl `Copy`.

We can now also derive builtin traits for packed structs that don't derive
`Copy`, so long as the fields impl `Copy`:
```
 #[derive(Hash)]
 #[repr(packed)]
 pub struct Foo(u32);
```
This includes types that hand-impl `Copy` rather than deriving it, such as the
following, that show up in winapi-0.2:
```
 #[derive(Clone)]
 #[repr(packed)]
 struct MyType(i32);

 impl Copy for MyType {}
```
The new approach is simpler to understand and implement, and it avoids
the need for the `unsafe_derive_on_repr_packed` check.

One exception is required for backwards-compatibility: we allow `[u8]`
fields for now. There is a new lint for this,
`byte_slice_in_packed_struct_with_derive`.
2023-01-30 12:00:42 +11:00
Giacomo Pasini
68c1e2fd48
Treat Drop as a rmw operation
Previously, a Drop terminator was considered a move in MIR.
This commit changes the behavior to only treat Drop as a mutable
access to the dropped place.

In order for this change to be correct, we need to guarantee that
  a) A dropped value won't be used again
  b) Places that appear in a drop won't be used again before a
     subsequent initialization.

We can ensure this to be correct at MIR construction because Drop
will only be emitted when a variable goes out of scope,
thus having:
  (a) as there is no way of reaching the old value. drop-elaboration
     will also remove any uninitialized drop.
  (b) as the place can't be named following the end of the scope.

However, the initialization status, previously tracked by moves,
should also be tied to the execution of a Drop, hence the
additional logic in the dataflow analyses.
2023-01-30 00:20:40 +01:00
Camille GILLOT
b456307cb1 Remove obsolete comment. 2023-01-29 22:09:51 +00:00
Camille GILLOT
a9aed861ac Reimplement NormalizeArrayLen. 2023-01-29 21:19:02 +00:00
Camille GILLOT
1d3f5b49d6 Test drop_tracking_mir before querying generator. 2023-01-29 13:50:07 +00:00
bors
2a4b00beaa Auto merge of #106908 - cjgillot:copyprop-ssa, r=oli-obk
Implement simple CopyPropagation based on SSA analysis

This PR extracts the "copy propagation" logic from https://github.com/rust-lang/rust/pull/106285.

MIR may produce chains of assignment between locals, like `_x = move? _y`.
This PR attempts to remove such chains by unifying locals.

The current implementation is a bit overzealous in turning moves into copies, and in removing storage statements.
2023-01-29 13:01:06 +00:00
bors
3cdd0197e7 Auto merge of #106227 - bryangarza:ctfe-limit, r=oli-obk
Use stable metric for const eval limit instead of current terminator-based logic

This patch adds a `MirPass` that inserts a new MIR instruction `ConstEvalCounter` to any loops and function calls in the CFG. This instruction is used during Const Eval to count against the `const_eval_limit`, and emit the `StepLimitReached` error, replacing the current logic which uses Terminators only.

The new method of counting loops and function calls should be more stable across compiler versions (i.e., not cause crates that compiled successfully before, to no longer compile when changes to the MIR generation/optimization are made).

Also see: #103877
2023-01-29 04:11:27 +00:00
bors
bcb064a7f4 Auto merge of #107406 - cjgillot:eliminate-witnesses, r=compiler-errors
Only compute mir_generator_witnesses query in drop_tracking_mir mode.

Attempt to fix the perf regression in https://github.com/rust-lang/rust/pull/101692

r? `@ghost`
2023-01-29 01:27:11 +00:00
Camille GILLOT
15d6325747 Remove HirId -> LocalDefId map from HIR. 2023-01-28 09:55:26 +00:00
Camille GILLOT
3175d03d3b Take a LocalDefId in hir::Visitor::visit_fn. 2023-01-28 09:51:50 +00:00
Camille GILLOT
4db4860503 Only compute mir_generator_witnesses query in drop_tracking_mir mode. 2023-01-28 08:41:22 +00:00
Camille GILLOT
65c3c90f3e Restrict amount of ignored locals. 2023-01-27 22:01:12 +00:00
Camille GILLOT
400cb9aa41 Separate witness type computation from the generator transform. 2023-01-27 19:00:26 +00:00
Camille GILLOT
e2387ad484 Remember where a type was kept in MIR. 2023-01-27 18:59:32 +00:00
Camille GILLOT
263da251af Use successor location for dominator check.
The assignment is complete only after the statement.
This marks self-assignments `x = x + 1` as non-sSA.
2023-01-27 18:22:45 +00:00
Camille GILLOT
d29dc057ba Do not merge locals that have their address taken. 2023-01-27 18:22:45 +00:00
Camille GILLOT
9096d31dcc Extract SsaLocals abstraction. 2023-01-27 18:22:45 +00:00
Camille GILLOT
bec73b09fd Pacify tidy. 2023-01-27 18:22:45 +00:00
Camille GILLOT
8f1dbe54ea Discard raw pointers from SSA locals. 2023-01-27 18:22:45 +00:00
Camille GILLOT
d45815eb4a Only consider a local to be SSA if assignment dominates all uses. 2023-01-27 18:22:45 +00:00
Camille GILLOT
6ed9f8f62e Implement SSA CopyProp pass. 2023-01-27 18:22:45 +00:00
Camille GILLOT
c4fe96c323 Allow to remove unused definitions without renumbering locals. 2023-01-27 18:22:45 +00:00
Camille GILLOT
982726cdc4 Consider CopyForDeref for DestProp. 2023-01-27 18:22:44 +00:00
Kyle Matsuda
a969c194d8 fix up subst_identity vs skip_binder; add some FIXMEs as identified in review 2023-01-26 20:28:31 -07:00
Kyle Matsuda
c2414dfaa4 change fn_sig query to use EarlyBinder; remove bound_fn_sig query; add EarlyBinder to fn_sig in metadata 2023-01-26 20:28:25 -07:00
Kyle Matsuda
e982971ff2 replace usages of fn_sig query with bound_fn_sig 2023-01-26 20:15:36 -07:00
Jakob Degen
f8aaf9aadb Disable ConstGoto opt in cleanup blocks 2023-01-26 03:50:37 -08:00
bors
885bf62887 Auto merge of #105582 - saethlin:instcombine-assert-inhabited, r=cjgillot
InstCombine away intrinsic validity assertions

This optimization (currently) fires 246 times on the standard library. It seems to fire hardly at all on the big crates in the benchmark suite. Interesting.
2023-01-26 03:10:52 +00:00
Jakob Degen
ad7393668f Delete SimplifyArmIdentity and SimplifyBranchSame mir opts 2023-01-24 04:13:52 -08:00
Ben Kimock
5bfad5cc85 Thread a ParamEnv down to might_permit_raw_init 2023-01-23 19:25:10 -05:00
Bryan Garza
1bbd655888 Improve efficiency of has_back_edge(...) 2023-01-24 00:01:37 +00:00
Bryan Garza
7618163a1c Add comments and remove unnecessary code 2023-01-23 23:56:22 +00:00
Bryan Garza
999d19d8aa Move CtfeLimit MirPass to inner_mir_for_ctfe 2023-01-23 23:56:22 +00:00
Bryan Garza
d3c13a0102 Revert "Move CtfeLimit to mir_const's set of passes"
This reverts commit 332542a92223b2800ed372d2d461921147f29477.
2023-01-23 23:56:22 +00:00
Bryan Garza
08de246cd7 Move CtfeLimit to mir_const's set of passes 2023-01-23 23:56:22 +00:00
Bryan Garza
8d99b0fc8d Abstract out has_back_edge fn 2023-01-23 23:56:22 +00:00
Bryan Garza
009beb00bc Change code to use map insead of for-loop 2023-01-23 23:56:22 +00:00
Bryan Garza
b763f9094f Remove debugging-related code 2023-01-23 23:56:22 +00:00
Bryan Garza
026a67377f Clean up CtfeLimit MirPass 2023-01-23 23:56:22 +00:00
Bryan Garza
360db516cc Create stable metric to measure long computation in Const Eval
This patch adds a `MirPass` that tracks the number of back-edges and
function calls in the CFG, adds a new MIR instruction to increment a
counter every time they are encountered during Const Eval, and emit a
warning if a configured limit is breached.
2023-01-23 23:56:22 +00:00
Tomasz Miąsko
955e7fbb16 Consistently use dominates instead of is_dominated_by
There is a number of APIs that answer dominance queries. Previously they
were named either "dominates" or "is_dominated_by". Consistently use the
"dominates" form.

No functional changes.
2023-01-21 12:15:02 +01:00
Guillaume Gomez
dee88e0fa2
Rollup merge of #107037 - tmiasko:rank, r=oli-obk
Fix Dominators::rank_partial_cmp to match documentation

The only use site is also updated accordingly and there is no change in end-to-end behaviour.
2023-01-19 11:19:36 +01:00
Arpad Borsos
96931a787a
Transform async ResumeTy in generator transform
- Eliminates all the `get_context` calls that async lowering created.
- Replace all `Local` `ResumeTy` types with `&mut Context<'_>`.

The `Local`s that have their types replaced are:
- The `resume` argument itself.
- The argument to `get_context`.
- The yielded value of a `yield`.

The `ResumeTy` hides a `&mut Context<'_>` behind an unsafe raw pointer, and the
`get_context` function is being used to convert that back to a `&mut Context<'_>`.

Ideally the async lowering would not use the `ResumeTy`/`get_context` indirection,
but rather directly use `&mut Context<'_>`, however that would currently
lead to higher-kinded lifetime errors.
See <https://github.com/rust-lang/rust/issues/105501>.

The async lowering step and the type / lifetime inference / checking are
still using the `ResumeTy` indirection for the time being, and that indirection
is removed here. After this transform, the generator body only knows about `&mut Context<'_>`.
2023-01-19 09:03:05 +01:00
Tomasz Miąsko
b8c5821ad8 Fix Dominators::rank_partial_cmp to match documentation
The only use site is also updated accordingly and there is no change in
end-to-end behaviour.
2023-01-18 17:11:43 +01:00
Matthias Krüger
68f12338af
Rollup merge of #104505 - WaffleLapkin:no-double-spaces-in-comments, r=jackh726
Remove double spaces after dots in comments

Most of the comments do not have double spaces, so I assume these are typos.
2023-01-17 20:21:25 +01:00
Maybe Waffle
6a28fb42a8 Remove double spaces after dots in comments 2023-01-17 08:09:33 +00:00
Tomasz Miąsko
d21696ae46 Remove ineffective run of SimplifyConstCondition
There are no constant conditions at this stage.
2023-01-16 00:00:00 +00:00
Ben Kimock
662199f125 InstCombine away intrinsic validity assertions 2023-01-15 16:51:42 -05:00
Camille GILLOT
389d52c1eb Remove visit_place. 2023-01-14 17:04:02 +00:00
Camille GILLOT
de9a5b076a Make the inlining destination a Local. 2023-01-14 12:09:06 +00:00
Albert Larsan
40ba0e84d5
Change src/test to tests in source files, fix tidy and tests 2023-01-11 09:32:13 +00:00
bors
89e0576bd3 Auto merge of #106340 - saethlin:propagate-operands, r=oli-obk
Always permit ConstProp to exploit arithmetic identities

Fixes https://github.com/rust-lang/rust/issues/72751

Initially, I thought I would need to enable operand propagation then do something else, but actually https://github.com/rust-lang/rust/pull/74491 already has the fix for the issue in question! It looks like this optimization was put under MIR opt level 3 due to possible soundness/stability implications, then demoted further to MIR opt level 4 when MIR opt level 2 became associated with `--release`.

Perhaps in the past we were doing CTFE on optimized MIR? We aren't anymore, so this optimization has no stability implications.

r? `@oli-obk`
2023-01-09 11:59:51 +00:00
bors
b1691f6413 Auto merge of #105323 - cjgillot:simplify-const-prop, r=davidtwco
Perform SimplifyLocals before ConstProp.

MIR before `ConstProp` may have a lot of dead writes, this makes `ConstProp` do unnecessary work.

r? `@ghost`
2023-01-07 16:13:18 +00:00
Tomasz Miąsko
3eabea9e2c Remove duplicated elaborate box derefs pass
The pass runs earlier as a part of `run_runtime_lowering_passes`.
2023-01-03 00:00:00 +00:00
Jakob Degen
ee6503a706 Reenable limited top-down MIR inlining 2023-01-01 22:01:29 -08:00
Ben Kimock
82f0973dd5 Always take advantage of arithmetic identities 2023-01-01 23:12:29 -05:00
Camille GILLOT
edc73f9719 Give the correct track-caller location with MIR inlining. 2022-12-25 18:48:13 +00:00
Camille GILLOT
e300abb593 Remove Nop in simplify_locals.
It's cheap and does not change anything.
2022-12-25 18:01:07 +00:00
Camille GILLOT
028b4745f4 Move SimplifyLocals before ConstProp. 2022-12-25 18:01:07 +00:00
Matthias Krüger
d23cb738d2
Rollup merge of #105975 - jeremystucki:rustc-remove-needless-lifetimes, r=eholk
rustc: Remove needless lifetimes
2022-12-24 00:31:41 +01:00
Jakob Degen
c359ab0b5d Retag argument to drop_in_place unconditionally 2022-12-21 14:59:55 -08:00
Jakob Degen
102040ce76 Retag as FnEntry on drop_in_place 2022-12-21 14:59:55 -08:00
Jeremy Stucki
42d100aad0
Add missing anonymous lifetime 2022-12-20 22:28:22 +01:00
Jeremy Stucki
3dde32ca97
rustc: Remove needless lifetimes 2022-12-20 22:10:40 +01:00
Matthias Krüger
c3af456d6d
Rollup merge of #105930 - JakobDegen:nal-unsound, r=oli-obk
Disable `NormalizeArrayLen`

cc #105929

r? mir-opt
2022-12-20 14:37:32 +01:00
Matthias Krüger
52fe5a1cc1
Rollup merge of #105835 - tmiasko:cleanup-post-borrowck, r=JakobDegen
Refactor post borrowck cleanup passes
2022-12-20 14:37:31 +01:00
bors
eb9e5e711d Auto merge of #105880 - Nilstrieb:make-newtypes-less-not-rust, r=oli-obk
Improve syntax of `newtype_index`

This makes it more like proper Rust and also makes the implementation a lot simpler.

Mostly just turns weird flags in the body into proper attributes.

It should probably also be converted to an attribute macro instead of function-like, but that can be done in a future PR.
2022-12-20 07:27:01 +00:00
Jakob Degen
4251289f27 Disable NormalizeArrayLen 2022-12-19 17:38:18 -08:00
Dylan DPC
a9005b6cc0
Rollup merge of #105864 - matthiaskrgr:compl, r=Nilstrieb
clippy::complexity fixes

filter_next
needless_question_mark
bind_instead_of_map
manual_find
derivable_impls
map_identity
redundant_slicing
skip_while_next
unnecessary_unwrap
needless_bool

r? `@compiler-errors`
2022-12-19 14:41:35 +05:30
Matthias Krüger
1da4a49912 clippy::complexity fixes
filter_next
needless_question_mark
bind_instead_of_map
manual_find
derivable_impls
map_identity
redundant_slicing
skip_while_next
unnecessary_unwrap
needless_bool
2022-12-19 00:04:28 +01:00
Nilstrieb
8bfd6450c7 A few small cleanups for newtype_index
Remove the `..` from the body, only a few invocations used it and it's
inconsistent with rust syntax.

Use `;` instead of `,` between consts. As the Rust syntax gods inteded.
2022-12-18 21:47:28 +01:00
Nilstrieb
d679764fb6 Make #[debug_format] an attribute in newtype_index
This removes the `custom` format functionality as its only user was
trivially migrated to using a normal format.

If a new use case for a custom formatting impl pops up, you can add it
back.
2022-12-18 21:37:38 +01:00
Matthias Krüger
8892698903
Rollup merge of #105870 - matthiaskrgr:useless_conv, r=oli-obk
avoid .into() conversion to identical types
2022-12-18 18:57:05 +01:00
Matthias Krüger
0aa4cde747 avoid .into() conversion to identical types 2022-12-18 16:20:32 +01:00
Tomasz Miąsko
62f9084dfa Remove false edges in CleanupPostBorrowck 2022-12-17 19:34:45 +01:00
Tomasz Miąsko
4c3efc7f1b Rename CleanupNonCodegenStatements to CleanupPostBorrowck 2022-12-17 19:34:45 +01:00
Tomasz Miąsko
2a8513d221 Replace visitor with a loop over blocks and statements 2022-12-17 18:23:37 +01:00
Tomasz Miąsko
adf53d4b06 Remove dead code after destination propagation 2022-12-16 00:00:00 +00:00
bors
ec56537c43 Auto merge of #105356 - JakobDegen:more-custom-mir, r=oli-obk
Custom MIR: Many more improvements

Commits are each atomic changes, best reviewed one at a time, with the exception that the last commit includes all the documentation.

### First commit

Unsafetyck was not correctly disabled before for `dialect = "built"` custom MIR. This is fixed and a regression test is added.

### Second commit

Implements `Discriminant`, `SetDiscriminant`, and `SwitchInt`.

### Third commit

Implements indexing, field, and variant projections.

### Fourth commit

Documents the previous commits and everything else.

There is some amount of weirdness here due to having to beat Rust syntax into cooperating with MIR concepts, but it hopefully should not be too much. All of it is documented.

r? `@oli-obk`
2022-12-15 19:59:48 +00:00
bors
4954a7ef5c Auto merge of #104616 - RalfJung:ctfe-alignment, r=oli-obk,RalfJung
always check alignment during CTFE

We originally disabled alignment checks because they got in the way -- there are some things we do with the interpreter during CTFE which does not correspond to actually running user-written code, but is purely administrative, and we didn't want alignment checks there, so we just disabled them entirely. But with `-Zextra-const-ub-checks` we anyway had to figure out how to disable those alignment checks while doing checks in regular code. So now it is easy to enable CTFE alignment checking by default. Let's see what the perf consequences of that are.

r? `@oli-obk`
2022-12-15 17:04:25 +00:00
Oli Scherer
d9d92ed7da Move alignment failure error reporting to machine 2022-12-15 16:07:35 +00:00
Oli Scherer
d66824dbc4 Make alignment checks a future incompat lint 2022-12-15 16:07:28 +00:00
Matthias Krüger
6cdc83b64e
Rollup merge of #105683 - JakobDegen:dest-prop-storage, r=tmiasko
Various cleanups to dest prop

This makes fixing the issues identified in #105577 easier. A couple changes

 - Use an enum with names instead of a bool
 - Only call `remove_candidates_if` from one place instead of two. Doing it from two places is far too fragile, since any divergence in the behavior between those callsites is likely to be unsound.
 - Remove `is_constant`. Right now we only merge locals, so this doesn't do anything, and the logic would be wrong if it did.

r? `@tmiasko`
2022-12-15 12:46:02 +01:00
Jakob Degen
a5beb7abb9 Various cleanups to dest prop 2022-12-14 23:11:52 -08:00
Oli Scherer
a5cd3bde95 Ensure no one constructs AliasTys themselves 2022-12-14 15:36:39 +00:00
Oli Scherer
1bf80249ae Remove many more cases of mk_substs_trait that can now use the iterator scheme` 2022-12-14 15:36:39 +00:00
Oli Scherer
0fe86aa977 Let mk_fn_def take an iterator instead to simplify some call sites 2022-12-14 15:36:39 +00:00
Jakob Degen
c1b27eea45 Fix unsafetyck disabling for custom MIR 2022-12-14 01:02:35 -08:00
bors
918d0ac38e Auto merge of #104986 - compiler-errors:opaques, r=oli-obk
Combine `ty::Projection` and `ty::Opaque` into `ty::Alias`

Implements https://github.com/rust-lang/types-team/issues/79.

This PR consolidates `ty::Projection` and `ty::Opaque` into a single `ty::Alias`, with an `AliasKind` and `AliasTy` type (renamed from `ty::ProjectionTy`, which is the inner data of `ty::Projection`) defined as so:

```
enum AliasKind {
  Projection,
  Opaque,
}

struct AliasTy<'tcx> {
  def_id: DefId,
  substs: SubstsRef<'tcx>,
}
```

Since we don't have access to `TyCtxt` in type flags computation, and because repeatedly calling `DefKind` on the def-id is expensive, these two types are distinguished with `ty::AliasKind`, conveniently glob-imported into `ty::{Projection, Opaque}`. For example:

```diff
  match ty.kind() {
-   ty::Opaque(..) =>
+   ty::Alias(ty::Opaque, ..) => {}
    _ => {}
  }
```

This PR also consolidates match arms that treated `ty::Opaque` and `ty::Projection` identically.

r? `@ghost`
2022-12-14 01:19:24 +00:00
Matthias Krüger
e0e9f3a7b7
Rollup merge of #105659 - JakobDegen:storage-live-borrow, r=davidtwco
Don't require owned data in `MaybeStorageLive`

Small improvement that avoids a clone. I don't expect this to have any noticeable perf effects, but better to have it than not to.

r? ``@tmiasko``
2022-12-13 19:57:12 +01:00
Michael Goulet
61adaf8187 Combine projection and opaque into alias 2022-12-13 17:48:55 +00:00
Michael Goulet
c13bd83528 squash OpaqueTy and ProjectionTy into AliasTy 2022-12-13 17:40:27 +00:00
Michael Goulet
7f3af72606 Use ty::OpaqueTy everywhere 2022-12-13 17:29:26 +00:00
Jakob Degen
3522d48112 Don't require owned data in MaybeStorageLive 2022-12-13 04:22:47 -08:00
Gary Guo
9342d1e73e Allow unsafe through inline const
This is handled similar to closures
2022-12-13 01:38:38 +00:00
Matthias Krüger
2daa3bcbc2
Rollup merge of #105537 - kadiwa4:remove_some_imports, r=fee1-dead
compiler: remove unnecessary imports and qualified paths

Some of these imports were necessary before Edition 2021, others were already in the prelude.

I hope it's fine that this PR is so spread-out across files :/
2022-12-11 09:51:57 +01:00
KaDiWa
9bc69925cb
compiler: remove unnecessary imports and qualified paths 2022-12-10 18:45:34 +01:00
Jakob Degen
9fb8da8f8f Remove unneeded field from SwitchTargets 2022-12-09 04:53:10 -08:00
Matthias Krüger
f1f7560598
Rollup merge of #105317 - RalfJung:retag-rework, r=oli-obk
make retagging work even with 'unstable' places

This is based on top of https://github.com/rust-lang/rust/pull/105301. Only the last two commits are new.

While investigating https://github.com/rust-lang/unsafe-code-guidelines/issues/381 I realized that we would have caught this issue much earlier if the add_retag pass wouldn't bail out on assignments of the form `*ptr = ...`.

So this PR changes our retag strategy:
- When a new reference is created via `Rvalue::Ref` (or a raw ptr via `Rvalue::AddressOf`), we do the retagging as part of just executing that address-taking operation.
- For everything else, we still insert retags -- these retags basically serve to ensure that references stored in local variables (and their fields) are always freshly tagged, so skipping this for assignments like `*ptr = ...` is less egregious.
r? ```@oli-obk```
2022-12-08 12:57:30 +01:00
bors
e60fbaf4ce Auto merge of #105229 - saethlin:zst-writes-to-unions, r=oli-obk
Re-enable removal of ZST writes to unions

This was previously disabled because Miri was lazily allocating unsized locals. But we aren't doing that anymore since  https://github.com/rust-lang/rust/pull/98831, so we can have this optimization back.
2022-12-06 15:35:55 +00:00
Ralf Jung
9397ea1368 make retagging work even with 'unstable' places 2022-12-06 10:33:34 +01:00
bors
ed61c139c2 Auto merge of #105220 - oli-obk:feeding, r=cjgillot
feed resolver_for_lowering instead of storing it in a field

r? `@cjgillot`

opening this as

* a discussion for `no_hash` + `feedable` queries. I think we'll want those, but I don't quite understand why they are rejected beyond a double check of the stable hashes for situations where the query is fed but also read from incremental caches.
* and a discussion on removing all untracked fields from TyCtxt and setting it up so that they are fed queries instead
2022-12-06 03:47:41 +00:00
bors
226202d902 Auto merge of #105119 - JakobDegen:inline-experiments, r=cjgillot
Disable top down MIR inlining

The current MIR inliner has exponential behavior in some cases: <https://godbolt.org/z/7jnWah4fE>. The cause of this is top-down inlining, where we repeatedly do inlining like `call_a() => { call_b(); call_b(); }`. Each decision on its own seems to make sense, but the result is exponential.

Disabling top-down inlining fundamentally prevents this. Each call site in the original, unoptimized source code is now considered for inlining exactly one time, which means that the total growth in MIR size is limited to number of call sites * inlining threshold.

Top down inlining may be worth re-introducing at some point, but it needs to be accompanied with a principled way to prevent this kind of behavior.
2022-12-06 00:53:01 +00:00
bors
fd02567705 Auto merge of #105121 - oli-obk:simpler-cheaper-dump_mir, r=nnethercote
Cheaper `dump_mir` take two

alternative to #105083

r? `@nnethercote`
2022-12-04 05:47:10 +00:00
Ben Kimock
74a270ac93 Re-enable removal of ZST writes to unions 2022-12-03 19:17:45 -05:00
Oli Scherer
c38ff3b385 Remove all but one call site of prepare_outputs and fetch the value from the TyCtxt instead 2022-12-03 12:28:01 +00:00
Tomasz Miąsko
b740cdcf43 Mark naked functions as never inline in codegen_fn_attrs
Use code generation attributes to ensure that naked functions are never
inline, replacing separate checks in MIR inliner and LLVM code
generation.
2022-12-03 01:04:42 +01:00
Oli Scherer
c7e94b0efd Use zero based indexing for pass_count 2022-12-02 15:55:24 +00:00
Oli Scherer
80dcc52934 Remove an impl and replace its only use with a method call 2022-12-02 15:43:36 +00:00
Jakob Degen
f4f777772e Disable top-down inlining 2022-12-01 18:32:45 -08:00
Oli Scherer
4f593ce5d8 Create format_args as late as possible 2022-12-01 08:49:51 +00:00
Oli Scherer
66797fa54f Remove needless Cow 2022-12-01 08:38:47 +00:00
Oli Scherer
c2166ec628 Don't go through the formatting infrastructure just to get the name of a phase 2022-12-01 08:31:54 +00:00
Dylan DPC
f90484df8a
Rollup merge of #104732 - WaffleLapkin:from_def_idn't, r=compiler-errors
Refactor `ty::ClosureKind` related stuff

I've tried to fix all duplication and weirdness, but if I missed something do tell :p

r? `@compiler-errors`
2022-11-28 15:42:10 +05:30
Maybe Waffle
1d42936b18 Prefer doc comments over //-comments in compiler 2022-11-27 11:19:04 +00:00
Maybe Waffle
881862ecb7 Rename fn_trait_kind_from_{from_lang=>def_id} to better convey meaning 2022-11-27 07:14:49 +00:00
Jakob Degen
245c60749a Rewrite dest prop.
This fixes a number of correctness issues from the previous version. Additionally, we use a new
strategy which has much better performance charactersitics and also finds more opportunities to
apply the optimization.
2022-11-26 18:04:54 -08:00
Matthias Krüger
4733312e09
Rollup merge of #104121 - Lokathor:mir-opt-when-instruction-set-missing-on-callee, r=tmiasko
Refine `instruction_set` MIR inline rules

Previously an exact match of the `instruction_set` attribute was required for an MIR inline to be considered. This change checks for an exact match *only* if the callee sets an `instruction_set` in the first place. When the callee does not declare an instruction set then it is considered to be platform agnostic code and it's allowed to be inline'd into the caller.

cc ``@oli-obk``

[Edit] Zulip Context: https://rust-lang.zulipchat.com/#narrow/stream/189540-t-compiler.2Fwg-mir-opt/topic/What.20exactly.20does.20the.20MIR.20optimizer.20do.3F
2022-11-26 10:39:10 +01:00
Lokathor
ea47943212 Refine instruction_set inline rules
Previously an exact match of the `instruction_set` attribute was required for an MIR inline to be considered. This change checks for an exact match *only* if the callee sets an `instruction_set` in the first place. When the callee does not declare an instruction set then it is considered to be platform agnostic code and it's allowed to be inline'd into the caller.
2022-11-25 15:19:16 -07:00
Santiago Pastorino
974e2837bb
Introduce PredicateKind::Clause 2022-11-25 00:04:54 -03:00
bors
b3bc6bf312 Auto merge of #103693 - HKalbasi:master, r=oli-obk
Make rustc_target usable outside of rustc

I'm working on showing type size in rust-analyzer (https://github.com/rust-lang/rust-analyzer/pull/13490) and I currently copied rustc code inside rust-analyzer, which works, but is bad. With this change, I would become able to use `rustc_target` and `rustc_index` directly in r-a, reducing the amount of copy needed.

This PR contains some feature flag to put nightly features behind them to make crates buildable on the stable compiler + makes layout related types generic over index type + removes interning of nested layouts.
2022-11-24 20:29:13 +00:00
hkalbasi
390a637e29 move things from rustc_target::abi to rustc_abi 2022-11-24 16:26:13 +03:30
Arpad Borsos
9f36f988ad
Avoid GenFuture shim when compiling async constructs
Previously, async constructs would be lowered to "normal" generators,
with an additional `from_generator` / `GenFuture` shim in between to
convert from `Generator` to `Future`.

The compiler will now special-case these generators internally so that
async constructs will *directly* implement `Future` without the need
to go through the `from_generator` / `GenFuture` shim.

The primary motivation for this change was hiding this implementation
detail in stack traces and debuginfo, but it can in theory also help
the optimizer as there is less abstractions to see through.
2022-11-24 10:04:27 +01:00
Oli Scherer
ec8d01fdcc Allow iterators instead of requiring slices that will get turned into iterators 2022-11-21 20:33:55 +00:00
Cameron Steffen
cc8dddbac9 Factor out conservative_is_privately_uninhabited 2022-11-20 19:04:11 -06:00
Matthias Krüger
820a41580e
Rollup merge of #104564 - RalfJung:either, r=oli-obk
interpret: use Either over Result when it is not representing an error condition

r? `@oli-obk`
2022-11-20 18:21:48 +01:00
Dylan DPC
00876c68c4
Rollup merge of #104411 - lcnr:bivariance-nll, r=compiler-errors
nll: correctly deal with bivariance

fixes #104409

when in a bivariant context, relating stuff should always trivially succeed. Also changes the mir validator to correctly deal with higher ranked regions.

r? types cc ``@RalfJung``
2022-11-19 11:54:44 +05:30
Ralf Jung
09a887cebf review feedback 2022-11-18 14:24:48 +01:00
Ralf Jung
4101889786 interpret: use Either over Result when it is not representing an error condition 2022-11-18 10:18:32 +01:00
bors
7c75fe4c85 Auto merge of #104170 - cjgillot:hir-def-id, r=fee1-dead
Record `LocalDefId` in HIR nodes instead of a side table

This is part of an attempt to remove the `HirId -> LocalDefId` table from HIR.
This attempt is a prerequisite to creation of `LocalDefId` after HIR lowering (https://github.com/rust-lang/rust/pull/96840), by controlling how `def_id` information is accessed.

This first part adds the information to HIR nodes themselves instead of a table.
The second part is https://github.com/rust-lang/rust/pull/103902
The third part will be to make `hir::Visitor::visit_fn` take a `LocalDefId` as last parameter.
The fourth part will be to completely remove the side table.
2022-11-17 07:42:27 +00:00
Ralf Jung
1115ec601a cleanup and dedupe CTFE and Miri error reporting 2022-11-16 10:13:29 +01:00
bors
79146baa9c Auto merge of #102570 - cjgillot:deagg-debuginfo, r=oli-obk
Perform simple scalar replacement of aggregates (SROA) MIR opt

This is a re-open of https://github.com/rust-lang/rust/pull/85796

I copied the debuginfo implementation (first commit) from `@eddyb's` own SROA PR.

This pass replaces plain field accesses by simple locals when possible.
To be eligible, the replaced locals:
- must not be enums or unions;
- must not be used whole;
- must not have their address taken.

The storage and deinit statements are duplicated on each created local.

cc `@tmiasko` who reviewed the former version of this PR.
2022-11-15 23:52:22 +00:00
Camille GILLOT
779007da06 Enable SROA by at mir-opt level 3. 2022-11-15 17:59:36 +00:00
Camille GILLOT
e4f343191a Flatten aggregates into locals. 2022-11-15 17:55:11 +00:00
Camille GILLOT
b550eabfa6 Introduce composite debuginfo. 2022-11-15 17:53:50 +00:00
bors
a00f8ba7fc Auto merge of #104054 - RalfJung:byte-provenance, r=oli-obk
interpret: support for per-byte provenance

Also factors the provenance map into its own module.

The third commit does the same for the init mask. I can move it in a separate PR if you prefer.

Fixes https://github.com/rust-lang/miri/issues/2181

r? `@oli-obk`
2022-11-15 17:37:15 +00:00
lcnr
6aa611a84c mv utility methods into separate module 2022-11-15 13:50:13 +01:00
lcnr
45f441a7b4 nll: correctly deal with bivariance 2022-11-15 13:34:08 +01:00
bors
357f660729 Auto merge of #101168 - jachris:dataflow-const-prop, r=oli-obk
Add new MIR constant propagation based on dataflow analysis

The current constant propagation in `rustc_mir_transform/src/const_prop.rs` fails to handle many cases that would be expected from a constant propagation optimization. For example:
```rust
let x = if true { 0 } else { 0 };
```
This pull request adds a new constant propagation MIR optimization pass based on the existing dataflow analysis framework. Since most of the analysis is not unique to constant propagation, a generic framework has been extracted. It works on top of the existing framework and could be reused for other optimzations.

Closes #80038. Closes #81605.

## Todo
### Essential
- [x] [Writes to inactive enum variants](https://github.com/rust-lang/rust/pull/101168#pullrequestreview-1089493974). Resolved by rejecting the registration of places with downcast projections for now. Could be improved by flooding other variants if mutable access to a variant is observed.
- [X] Handle [`StatementKind::CopyNonOverlapping`](https://github.com/rust-lang/rust/pull/101168#discussion_r957774914). Resolved by flooding the destination.
- [x] Handle `UnsafeCell` / `!Freeze` correctly.
- [X] Overflow propagation of `CheckedBinaryOp`: Decided to not propagate if overflow flag is `true` (`false` will still be propagated)
- [x] More documentation in general.
- [x] Arguments for correctness, documentation of necessary assumptions.
- [x] Better performance, or alternatively, require `-Zmir-opt-level=3` for now.

### Extra
- [x]  Add explicit unreachability, i.e. upgrading the lattice from $\mathbb{P} \to \mathbb{V}$ to $\set{\bot} \cup (\mathbb{P} \to \mathbb{V})$.
- [x] Use storage statements to improve precision.
- [ ] Consider opening issue for duplicate diagnostics: https://github.com/rust-lang/rust/pull/101168#issuecomment-1276609950
- [ ] Flood moved-from places with $\bot$ (requires some changes for places with tracked projections).
- [ ] Add downcast projections back in.
- [ ] [Algebraic simplifications](https://github.com/rust-lang/rust/pull/101168#discussion_r957967878) (possibly with a shared API; done by old const prop).
- [ ] Propagation through slices / arrays.
- [ ] Find other optimizations that are done by old `const_prop.rs`, but not by this one.
2022-11-15 09:38:05 +00:00
Ralf Jung
68af46c112 assert that we are (de)seiralizing ProvenanceMap correctly 2022-11-14 18:26:40 +01:00
Camille GILLOT
9d20aca983 Store a LocalDefId in hir::Variant & hir::Field. 2022-11-13 14:06:51 +00:00
Ralf Jung
c78021709a add is_sized method on Abi and Layout, and use it 2022-11-13 12:23:53 +01:00
Jannis Christopher Köhl
ea23585c91 Disable limits if mir-opt-level >= 4 2022-11-12 20:14:34 +01:00
Jannis Christopher Köhl
d66a00a7b1 Expand upon comment regarding self-assignment 2022-11-12 20:05:52 +01:00
Jannis Christopher Köhl
74d53ab912 Require -Zmir-opt-level >= 3 for now 2022-11-12 15:24:23 +01:00
Jannis Christopher Köhl
b3f648958d Add comment for guessed constants 2022-11-12 14:07:54 +01:00
Jannis Christopher Köhl
8ecb276735 Simplify creation of map 2022-11-10 19:12:10 +01:00
Michael Goulet
31157def1a Don't ICE when encountering ConstKind::Error in RequiredConstsVisitor 2022-11-10 05:14:04 +00:00
Jannis Christopher Köhl
9766ee0b20 Fix struct field tracking and add tests for it 2022-11-09 18:21:42 +01:00
Jannis Christopher Köhl
bfbca6c75c Completely remove tracking of references for now 2022-11-09 18:03:30 +01:00
Jakob Degen
ba359d8a51 Add support for custom MIR parsing 2022-11-08 23:13:15 -08:00
Jannis Christopher Köhl
72196ee666 Limit number of basic blocks and tracked places to 100 for now 2022-11-07 10:35:26 +01:00
Jannis Christopher Köhl
b478fcf270 Use new cast methods 2022-11-07 10:35:26 +01:00
Jannis Christopher Köhl
630e17d3e4 Limit number of tracked places, and some other perf improvements 2022-11-07 10:35:26 +01:00
Jannis Christopher Köhl
1f82a9f89e Move HasTop and HasBottom into lattice.rs 2022-11-07 10:35:25 +01:00
Jannis Christopher Köhl
f29533b4e0 Small documentation changes 2022-11-07 10:35:25 +01:00
Jannis Christopher Köhl
efc7ca8c7d Use ParamEnv consistently 2022-11-07 10:35:25 +01:00
Jannis Christopher Köhl
d86acdd72a Prevent propagation of overflow if overflow occured 2022-11-07 10:35:24 +01:00
Jannis Christopher Köhl
062053ba79 Fix unimplemented binary_ptr_op 2022-11-07 10:35:24 +01:00
Jannis Christopher Köhl
274a49132b Improve documentation, plus some small changes 2022-11-07 10:35:23 +01:00