| Age | Commit message (Collapse) | Author | Lines |
|---|
|
Allow shared access to `Exclusive<T>` when `T: Sync`
Addresses libs-api request in https://github.com/rust-lang/rust/issues/98407#issuecomment-3299348713.
Adds the following trait impls to `Exclusive<T>`, all bounded on `T: Sync`:
- `AsRef<T>`
- `Clone`
- `Copy`
- `PartialEq`
- `StructuralPartialEq`
- `Eq`
- `Hash`
- `PartialOrd`
- `Ord`
- `Fn`
``@rustbot`` label T-libs-api
|
|
TypeTree support in autodiff
# TypeTrees for Autodiff
## What are TypeTrees?
Memory layout descriptors for Enzyme. Tell Enzyme exactly how types are structured in memory so it can compute derivatives efficiently.
## Structure
```rust
TypeTree(Vec<Type>)
Type {
offset: isize, // byte offset (-1 = everywhere)
size: usize, // size in bytes
kind: Kind, // Float, Integer, Pointer, etc.
child: TypeTree // nested structure
}
```
## Example: `fn compute(x: &f32, data: &[f32]) -> f32`
**Input 0: `x: &f32`**
```rust
TypeTree(vec![Type {
offset: -1, size: 8, kind: Pointer,
child: TypeTree(vec![Type {
offset: -1, size: 4, kind: Float,
child: TypeTree::new()
}])
}])
```
**Input 1: `data: &[f32]`**
```rust
TypeTree(vec![Type {
offset: -1, size: 8, kind: Pointer,
child: TypeTree(vec![Type {
offset: -1, size: 4, kind: Float, // -1 = all elements
child: TypeTree::new()
}])
}])
```
**Output: `f32`**
```rust
TypeTree(vec![Type {
offset: -1, size: 4, kind: Float,
child: TypeTree::new()
}])
```
## Why Needed?
- Enzyme can't deduce complex type layouts from LLVM IR
- Prevents slow memory pattern analysis
- Enables correct derivative computation for nested structures
- Tells Enzyme which bytes are differentiable vs metadata
## What Enzyme Does With This Information:
Without TypeTrees (current state):
```llvm
; Enzyme sees generic LLVM IR:
define float ``@distance(ptr*`` %p1, ptr* %p2) {
; Has to guess what these pointers point to
; Slow analysis of all memory operations
; May miss optimization opportunities
}
```
With TypeTrees (our implementation):
```llvm
define "enzyme_type"="{[]:Float@float}" float ``@distance(``
ptr "enzyme_type"="{[]:Pointer}" %p1,
ptr "enzyme_type"="{[]:Pointer}" %p2
) {
; Enzyme knows exact type layout
; Can generate efficient derivative code directly
}
```
# TypeTrees - Offset and -1 Explained
## Type Structure
```rust
Type {
offset: isize, // WHERE this type starts
size: usize, // HOW BIG this type is
kind: Kind, // WHAT KIND of data (Float, Int, Pointer)
child: TypeTree // WHAT'S INSIDE (for pointers/containers)
}
```
## Offset Values
### Regular Offset (0, 4, 8, etc.)
**Specific byte position within a structure**
```rust
struct Point {
x: f32, // offset 0, size 4
y: f32, // offset 4, size 4
id: i32, // offset 8, size 4
}
```
TypeTree for `&Point` (internal representation):
```rust
TypeTree(vec![
Type { offset: 0, size: 4, kind: Float }, // x at byte 0
Type { offset: 4, size: 4, kind: Float }, // y at byte 4
Type { offset: 8, size: 4, kind: Integer } // id at byte 8
])
```
Generates LLVM:
```llvm
"enzyme_type"="{[]:Float@float}"
```
### Offset -1 (Special: "Everywhere")
**Means "this pattern repeats for ALL elements"**
#### Example 1: Array `[f32; 100]`
```rust
TypeTree(vec![Type {
offset: -1, // ALL positions
size: 4, // each f32 is 4 bytes
kind: Float, // every element is float
}])
```
Instead of listing 100 separate Types with offsets `0,4,8,12...396`
#### Example 2: Slice `&[i32]`
```rust
// Pointer to slice data
TypeTree(vec![Type {
offset: -1, size: 8, kind: Pointer,
child: TypeTree(vec![Type {
offset: -1, // ALL slice elements
size: 4, // each i32 is 4 bytes
kind: Integer
}])
}])
```
#### Example 3: Mixed Structure
```rust
struct Container {
header: i64, // offset 0
data: [f32; 1000], // offset 8, but elements use -1
}
```
```rust
TypeTree(vec![
Type { offset: 0, size: 8, kind: Integer }, // header
Type { offset: 8, size: 4000, kind: Pointer,
child: TypeTree(vec![Type {
offset: -1, size: 4, kind: Float // ALL array elements
}])
}
])
```
|
|
Allow `&raw [mut | const]` for union field in safe code
fixes rust-lang/rust#141264
r? ``@Veykril``
Unresolved questions:
- [x] Any edge cases?
- [x] How this works with rust-analyzer (because all I've did is prevent compiler from emitting error in `&raw` context) (rust-lang/rust-analyzer#19867)
- [x] Should we allow `addr_of!` and `addr_of_mut!` as well? In current version they both (`&raw` and `addr_of!`) are allowed (They are the same)
- [x] Is chain of union fields is a safe? (Yes)
|
|
|
|
Fix tracking issue number for feature(macro_attr)
The ability to define an attribute macro with `macro_rules!` is tracked at https://github.com/rust-lang/rust/issues/143547, not https://github.com/rust-lang/rust/issues/83527
|
|
fix rebasing cycle heads when not reaching a fixpoint
fixes https://github.com/rust-lang/trait-system-refactor-initiative/issues/232
annoyingly subtle, imagine the following proof tree
- A (no cycle head usages, final result Y)
- *ignored* B (depends on A with provisional result X)
- A (cycle, provisional result X)
- B (using the cache entry here incorrectly assumes A has final result X)
r? ``@BoxyUwU``
|
|
Clarified error note for usize range matching
Fixes rust-lang/rust#146476
This is kinda rough, but it gets the point across a little better and stays short.
|
|
Remove most `#[track_caller]` from allocating Vec methods
They cause significant binary size overhead while contributing little value.
closes rust-lang/rust#146963, see that issue for more details.
|
|
|
|
|
|
GuillaumeGomez:fix-doctest-compilation-time-display, r=lolbinarycat
Fix doctest compilation time display
Fixes rust-lang/rust#146960.
Small corner case that happened in case everything went fine and there was only merged doctests.
r? lolbinarycat
|
|
Fabian-Gruenbichler:mr/fix-linker-warning-test-normalization, r=jieyouxu
re-order normalizations in run-make linker-warning test
otherwise a buildroot containing `rustc[^/_-]*` or `libpanic_abort` would be mangled before being replaced by the build root placeholder value..
e.g., running `./x.py test --verbose tests/run-make/linker-warning` with rustc checked out in ~/ext/rustcfoobar will result in
```
running 1 tests
test [run-make] tests/run-make/linker-warning ... FAILED
failures:
---- [run-make] tests/run-make/linker-warning stdout ----
------rustc stdout------------------------------
------rustc stderr------------------------------
------------------------------------------
error: rmake recipe failed to complete
status: exit status: 101
command: cd "/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu/test/run-make/linker-warning/rmake_out" && env -u RUSTFLAGS -u __RUSTC_DEBUG_ASSERTIONS_ENABLED -u __STD_DEBUG_ASSERTIONS_ENABLED AR="ar" BUILD_ROOT="/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu" CC="cc" CC_DEFAULT_FLAGS="-ffunction-sections -fdata-sections -fPIC -m64" CXX="c++" CXX_DEFAULT_FLAGS="-ffunction-sections -fdata-sections -fPIC -m64" HOST_RUSTC_DYLIB_PATH="/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu/stage1/lib" LD_LIBRARY_PATH="/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu/bootstrap-tools/x86_64-unknown-linux-gnu/release/deps:/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu/stage0/lib/rustlib/x86_64-unknown-linux-gnu/lib" LD_LIB_PATH_ENVVAR="LD_LIBRARY_PATH" LLVM_BIN_DIR="/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu/ci-llvm/bin" LLVM_COMPONENTS="<..>" LLVM_FILECHECK="/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu/ci-llvm/bin/FileCheck" NODE="/usr/bin/node" PYTHON="/usr/bin/python3" RUSTC="/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu/stage1/bin/rustc" RUSTDOC="/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu/stage1/bin/rustdoc" SOURCE_ROOT="/home/user/ext/rustcfoobar" TARGET="x86_64-unknown-linux-gnu" TARGET_EXE_DYLIB_PATH="/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu/stage1/lib/rustlib/x86_64-unknown-linux-gnu/lib" "/home/user/ext/rustcfoobar/build/x86_64-unknown-linux-gnu/test/run-make/linker-warning/rmake"
stdout: none
--- stderr -------------------------------
thread 'main' panicked at /home/user/ext/rustcfoobar/tests/run-make/linker-warning/rmake.rs:74:14:
test failed: `short-error.txt` is different from `(linker error)`
--- short-error.txt
+++ (linker error)
``@@`` -1,6 +1,6 ``@@``
error: linking with `./fake-linker` failed: exit status: 1
|
- = note: "./fake-linker" "-m64" "/symbols.o" "<2 object files omitted>" "-Wl,--as-needed" "-Wl,-Bstatic" "/build-root/test/run-make/linker-warning/rmake_out/{libfoo,libbar}.rlib" "<sysroot>/lib/rustlib/x86_64-unknown-linux-gnu/lib/{libstd-*,libpanic_unwind-*,libobject-*,libmemchr-*,libaddr2line-*,libgimli-*,libcfg_if-*,librustc_demangle-*,libstd_detect-*,libhashbrown-*,librustc_std_workspace_alloc-*,libminiz_oxide-*,libadler2-*,libunwind-*,liblibc-*,librustc_std_workspace_core-*,liballoc-*,libcore-*,libcompiler_builtins-*}.rlib" "-Wl,-Bdynamic" "-lgcc_s" "-lutil" "-lrt" "-lpthread" "-lm" "-ldl" "-lc" "-L" "/raw-dylibs" "-Wl,--eh-frame-hdr" "-Wl,-z,noexecstack" "-L" "/build-root/test/run-make/linker-warning/rmake_out" "-L" "<sysroot>/lib/rustlib/x86_64-unknown-linux-gnu/lib" "-o" "main" "-Wl,--gc-sections" "-pie" "-Wl,-z,relro,-z,now" "-nodefaultlibs" "run_make_error"
+ = note: "./fake-linker" "-m64" "/symbols.o" "<2 object files omitted>" "-Wl,--as-needed" "-Wl,-Bstatic" "/home/user/ext/rustc/build/x86_64-unknown-linux-gnu/test/run-make/linker-warning/rmake_out/{libfoo,libbar}.rlib" "<sysroot>/lib/rustlib/x86_64-unknown-linux-gnu/lib/{libstd-*,libpanic_unwind-*,libobject-*,libmemchr-*,libaddr2line-*,libgimli-*,libcfg_if-*,librustc_demangle-*,libstd_detect-*,libhashbrown-*,librustc_std_workspace_alloc-*,libminiz_oxide-*,libadler2-*,libunwind-*,liblibc-*,librustc_std_workspace_core-*,liballoc-*,libcore-*,libcompiler_builtins-*}.rlib" "-Wl,-Bdynamic" "-lgcc_s" "-lutil" "-lrt" "-lpthread" "-lm" "-ldl" "-lc" "-L" "/raw-dylibs" "-Wl,--eh-frame-hdr" "-Wl,-z,noexecstack" "-L" "/home/user/ext/rustc/build/x86_64-unknown-linux-gnu/test/run-make/linker-warning/rmake_out" "-L" "<sysroot>/lib/rustlib/x86_64-unknown-linux-gnu/lib" "-o" "main" "-Wl,--gc-sections" "-pie" "-Wl,-z,relro,-z,now" "-nodefaultlibs" "run_make_error"
= note: some arguments are omitted. use `--verbose` to show all linker arguments
= note: error: baz
[..]
```
without this fix.
Note: this affected Debian's automated builds, since the buildroot contains the package name and version. while that particular issue got fixed in the meantime by accident by making the RE more strict in 77232fb9350dbe09821df27f184301475829f1e4 , other buildroot paths are still affected without a full fix.
|
|
tests: relax expectations after llvm change 902ddda120a5
LLVM 22 is able to drop assumes that seem to not help further optimizations, which actually seems to dramatically _help_ further optimizations in some of our small test cases.
I'm a little unclear how to fix the last failure, in `tests/codegen-llvm/issues/issue-122600-ptr-discriminant-update.rs`:
```
-; Function Attrs: mustprogress nofree norecurse nosync nounwind willreturn memory(argmem: readwrite, inaccessiblemem: write) uwtable
+; Function Attrs: mustprogress nofree norecurse nosync nounwind nonlazybind willreturn memory(argmem: readwrite, inaccessiblemem: write) uwtable
define void ``@update(ptr`` noundef captures(none) %s) unnamed_addr #0 {
start:
- %_3.sroa.0.0.copyload = load i8, ptr %s, align 1
- %0 = trunc nuw i8 %_3.sroa.0.0.copyload to i1
- %1 = xor i1 %0, true
- tail call void ``@llvm.assume(i1`` %1)
store i8 1, ptr %s, align 1
ret void
}
```
I'm just not conversant enough in LLVM IR to follow the changes here.
``@rustbot`` label llvm-main
r? nikic
|
|
Introduce CoerceShared lang item and trait, and basic Reborrow tests
Part of rust-lang/rust#145612: This introduces the `CoerceShared` trait which is the `Reborrow` equivalent of a `&mut T` -> `&T` coercion. The trait has a `Target` GAT which makes this (currently) unique in the `core/src/marker.rs`; I'm not sure if this can be considered problematic. Maybe this is not the way such things should be done at the marker trait level? Or maybe it is fine.
Improtantly, this PR introduces a battery of basic `Reborrow` and `CoerceShared` tests. These test the very basics of the feature; custom marker types intended to have exclusive semantics (`Custom<'a>(PhantomData<&'a mut ()>)`), custom exclusive reference wrappers, and standard library exclusive reference wrappers (`Pin<&mut T>` and `Option<&mut T>`). None of these of course work since the implementation for `Reborrow` and `CoerceShared` is entirely missing, but this is the first step towards making these work.
Future PRs will introduce more tests, such as "recursive" reborrowing (ie. reborrowing structs that contain multiple reborrowable fields) and checks around the lifetime semantics of reborrowing ie. that a reborrow produces a new type with the same lifetime as the original.
|
|
|
|
|
|
This wasn't caught by CI, because debug assertions aren't enabled there.
|
|
|
|
|
|
|
|
`#![feature(doc_cfg)]`
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
The ability to define an attribute macro with `macro_rules!` is tracked at https://github.com/rust-lang/rust/issues/143547, not https://github.com/rust-lang/rust/issues/83527
|
|
Bump bootstrap compiler to 1.91 beta
https://forge.rust-lang.org/release/process.html#default-branch-bootstrap-update-tuesday
|
|
Rollup of 10 pull requests
Successful merges:
- rust-lang/rust#145113 (resolve: Do not finalize shadowed bindings)
- rust-lang/rust#146523 (Demote both armebv7r-none-* targets.)
- rust-lang/rust#146704 (port `#[debugger_visualizer]` to the new attribute system)
- rust-lang/rust#146758 (Stop linking rs{begin,end} objects on x86_64-*-windows-gnu)
- rust-lang/rust#146778 (Use standard attribute logic for allocator shim)
- rust-lang/rust#146849 (Reduce some uses of `LegacyBang`)
- rust-lang/rust#147016 (fix doc comments to be more standard)
- rust-lang/rust#147027 (Add new `tyalias` intra-doc link disambiguator)
- rust-lang/rust#147031 (mbe: Simplify check_redundant_vis_repetition)
- rust-lang/rust#147058 (Ignore more failing ui tests for GCC backend)
Failed merges:
- rust-lang/rust#147046 (Rename `rust.use-lld` to `rust.bootstrap-override-lld`)
r? `@ghost`
`@rustbot` modify labels: rollup
|
|
|
|
|
|
also slightly minimized the test
|
|
|
|
|
|
|
|
|
|
|
|
Ignore more failing ui tests for GCC backend
Needed for https://github.com/rust-lang/rust/pull/146414.
r? ``@Kobzol``
|
|
Add new `tyalias` intra-doc link disambiguator
Fixes https://github.com/rust-lang/rust/issues/146855.
Alternative to rust-lang/rust#146866.
This adds support for a new disambiguator: `tyalias`. I think it's common enough to have type aliases nowaday, so no reason to not be able to have a disambiguator for them.
cc ``@fmease``
r? lolbinarycat
|
|
port `#[debugger_visualizer]` to the new attribute system
|
|
resolve: Do not finalize shadowed bindings
I.e. do not mark them as used, or non-speculatively loaded, or similar.
Previously they were sometimes finalized during early resolution, causing issues like https://github.com/rust-lang/rust/pull/144793#issuecomment-3168108005.
|
|
|
|
|
|
|
|
We'll still error due to the `opt_bad_ty` of `method_autoderef_steps`.
This slightly worsens the span of `infer_var.method()` which is now the
same as for `Box::new(infer_var).method()`.
Unlike `structurally_resolve_type`, `probe_op` does not check whether
the infcx is already tainted, so this results in 2 previously not emitted
errors.
|
|
|
|
|
|
|
