shared-generics: Do not share instantiations that contain local-only types
In Zed shared-generics loading takes up a significant chunk of time in incremental build, as rustc deserializes rmeta of all dependencies of a crate. I've recently realized that shared-generics includes all instantiations of some_generic_function in the following snippet:
```rs
pub fn some_generic_function(_: impl Fn()) {}
pub fn non_generic_function() {
some_generic_function(|| {});
some_generic_function(|| {});
some_generic_function(|| {});
some_generic_function(|| {});
some_generic_function(|| {});
some_generic_function(|| {});
some_generic_function(|| {});
}
```
even though none of these instantiations can actually be created from outside of `non_generic_function`. This is a dummy example, but we do rely on invoking callbacks with FnOnce a lot in our codebase.
This PR makes shared-generics account for visibilities of generic arguments; an item is only considered for exporting if it is reachable from the outside or if all of it's arguments are visible outside of the local crate.
This PR reduces incremental build time for Zed (touch editor.rs scenario) from 12.4s to 10.4s. I'd love to see a perf run if possible; per my checks this PR does not incur new instantiations in downstream crates, so if there'd be perf regressions, I'd expect them to come from newly-introduced visibility checks.
Issue an error when using `no_mangle` on language items
This pull requests adds the code to issue an error or a warning when using `no_mangle` on language items. This should detail why the `undefined symbol` error is issued for the code described in #139923.
The pull request adds two ui tests, one testing the error and the other one the warning.
I would love some feedback here, as I am not sure that the error and warning are issues using the right API.
add suggestion on how to add a panic breakpoint
Co-authored-by: Pat Pannuto <pat.pannuto@gmail.com>
delete no_mangle from ui/panic-handler/panic-handler-wrong-location test
issue an error for the usage of #[no_mangle] on internal language items
delete the comments
add newline
rephrase note
Co-authored-by: bjorn3 <17426603+bjorn3@users.noreply.github.com>
update error not to leak implementation details
delete no_mangle_span
Co-authored-by: bjorn3 <17426603+bjorn3@users.noreply.github.com>
delete commented code
Simplify `LazyAttrTokenStream`
`LazyAttrTokenStream` is an unpleasant type: `Lrc<Box<dyn ToAttrTokenStream>>`. Why does it look like that?
- There are two `ToAttrTokenStream` impls, one for the lazy case, and one for the case where we already have an `AttrTokenStream`.
- The lazy case (`LazyAttrTokenStreamImpl`) is implemented in `rustc_parse`, but `LazyAttrTokenStream` is defined in `rustc_ast`, which does not depend on `rustc_parse`. The use of the trait lets `rustc_ast` implicitly depend on `rustc_parse`. This explains the `dyn`.
- `LazyAttrTokenStream` must have a `size_of` as small as possible, because it's used in many AST nodes. This explains the `Lrc<Box<_>>`, which keeps it to one word. (It's required `Lrc<dyn _>` would be a fat pointer.)
This PR moves `LazyAttrTokenStreamImpl` (and a few other token stream things) from `rustc_parse` to `rustc_ast`. This lets us replace the `ToAttrTokenStream` trait with a two-variant enum and also remove the `Box`, changing `LazyAttrTokenStream` to `Lrc<LazyAttrTokenStreamInner>`. Plus it does a few cleanups.
r? `@petrochenkov`
This commit does the following.
- Changes it from `Lrc<Box<dyn ToAttrTokenStream>>` to
`Lrc<LazyAttrTokenStreamInner>`.
- Reworks `LazyAttrTokenStreamImpl` as `LazyAttrTokenStreamInner`, which
is a two-variant enum.
- Removes the `ToAttrTokenStream` trait and the two impls of it.
The recursion limit must be increased in some crates otherwise rustdoc
aborts.
Implement the internal feature `cfg_target_has_reliable_f16_f128`
Support for `f16` and `f128` is varied across targets, backends, and backend versions. Eventually we would like to reach a point where all backends support these approximately equally, but until then we have to work around some of these nuances of support being observable.
Introduce the `cfg_target_has_reliable_f16_f128` internal feature, which provides the following new configuration gates:
* `cfg(target_has_reliable_f16)`
* `cfg(target_has_reliable_f16_math)`
* `cfg(target_has_reliable_f128)`
* `cfg(target_has_reliable_f128_math)`
`reliable_f16` and `reliable_f128` indicate that basic arithmetic for the type works correctly. The `_math` versions indicate that anything relying on `libm` works correctly, since sometimes this hits a separate class of codegen bugs.
These options match configuration set by the build script at [1]. The logic for LLVM support is duplicated as-is from the same script. There are a few possible updates that will come as a follow up.
The config introduced here is not planned to ever become stable, it is only intended to replace the build scripts for `std` tests and `compiler-builtins` that don't have any way to configure based on the codegen backend.
MCP: https://github.com/rust-lang/compiler-team/issues/866
Closes: https://github.com/rust-lang/compiler-team/issues/866
[1]: 555e1d0386/library/std/build.rs (L84-L186)
---
The second commit makes use of this config to replace `cfg_{f16,f128}{,_math}` in `library/`. I omitted providing a `cfg(bootstrap)` configuration to keep things simpler since the next beta branch is in two weeks.
try-job: aarch64-gnu
try-job: i686-msvc-1
try-job: test-various
try-job: x86_64-gnu
try-job: x86_64-msvc-ext2
In Zed shared-generics loading takes up a significant chunk of time in incremental build, as rustc deserializes rmeta of all dependencies of a crate.
I've recently realized that shared-generics includes all instantiations of some_generic_function in the following snippet:
```rs
pub fn some_generic_function(_: impl Fn()) {}
pub fn non_generic_function() {
some_generic_function(|| {});
some_generic_function(|| {});
some_generic_function(|| {});
some_generic_function(|| {});
some_generic_function(|| {});
some_generic_function(|| {});
some_generic_function(|| {});
}
```
even though none of these instantiations can actually be created from outside of `non_generic_function`.
This PR makes shared-generics account for visibilities of generic arguments; an item is only considered for exporting if it is reachable from the outside or if all of it's arguments are visible outside of the local crate.
This PR reduces incremental build time for Zed (touch edito.rs scenario) from 12.4s to 10.4s.
Co-authored-by: bjorn3 <17426603+bjorn3@users.noreply.github.com>
Support for `f16` and `f128` is varied across targets, backends, and
backend versions. Eventually we would like to reach a point where all
backends support these approximately equally, but until then we have to
work around some of these nuances of support being observable.
Introduce the `cfg_target_has_reliable_f16_f128` internal feature, which
provides the following new configuration gates:
* `cfg(target_has_reliable_f16)`
* `cfg(target_has_reliable_f16_math)`
* `cfg(target_has_reliable_f128)`
* `cfg(target_has_reliable_f128_math)`
`reliable_f16` and `reliable_f128` indicate that basic arithmetic for
the type works correctly. The `_math` versions indicate that anything
relying on `libm` works correctly, since sometimes this hits a separate
class of codegen bugs.
These options match configuration set by the build script at [1]. The
logic for LLVM support is duplicated as-is from the same script. There
are a few possible updates that will come as a follow up.
The config introduced here is not planned to ever become stable, it is
only intended to replace the build scripts for `std` tests and
`compiler-builtins` that don't have any way to configure based on the
codegen backend.
MCP: https://github.com/rust-lang/compiler-team/issues/866
Closes: https://github.com/rust-lang/compiler-team/issues/866
[1]: 555e1d0386/library/std/build.rs (L84-L186)
If creating a temporary directory fails with permission denied then retry with backoff
On Windows, if creating a temporary directory fails with permission denied then use a retry/backoff loop. This hopefully fixes a recuring error in our CI.
cc ```@jieyouxu,``` https://github.com/rust-lang/rust/issues/133959
On Windows, if creating a temporary directory fails with permission denied then use a retry/backoff loop. This hopefully fixes a recuring error in our CI.
Make #![feature(let_chains)] bootstrap conditional in compiler/
Let chains have been stabilized recently in #132833, so we can remove the gating from our uses in the compiler (as the compiler uses edition 2024).
simd intrinsics with mask: accept unsigned integer masks, and fix some of the errors
It's not clear at all why the mask would have to be signed, it is anyway interpreted bitwise. The backend should just make sure that works no matter the surface-level type; our LLVM backend already does this correctly. The note of "the mask may be widened, which only has the correct behavior for signed integers" explains... nothing? Why can't the code do the widening correctly? If necessary, just cast to the signed type first...
Also while we are at it, fix the errors. For simd_masked_load/store, the errors talked about the "third argument" but they meant the first argument (the mask is the first argument there). They also used the wrong type for `expected_element`.
I have extremely low confidence in the GCC part of this PR.
See [discussion on Zulip](https://rust-lang.zulipchat.com/#narrow/channel/257879-project-portable-simd/topic/On.20the.20sign.20of.20masks)
I'm removing empty identifiers everywhere, because in practice they
always mean "no identifier" rather than "empty identifier". (An empty
identifier is impossible.) It's better to use `Option` to mean "no
identifier" because you then can't forget about the "no identifier"
possibility.
Some specifics:
- When testing an attribute for a single name, the commit uses the
`has_name` method.
- When testing an attribute for multiple names, the commit uses the new
`has_any_name` method.
- When using `match` on an attribute, the match arms now have `Some` on
them.
In the tests, we now avoid printing empty identifiers by not printing
the identifier in the `error:` line at all, instead letting the carets
point out the problem.
While LLVM is rather permissive in this regards, some other codegen
backends demand that once you declare a function for definition you
actually define contents of the function, which doesn't happen for naked
functions as we actually generate assembly for them.
Prepend temp files with per-invocation random string to avoid temp filename conflicts
https://github.com/rust-lang/rust/issues/139407 uncovered a very subtle unsoundness with incremental codegen, failing compilation sessions (due to assembler errors), and the "prefer hard linking over copying files" strategy we use in the compiler for file management.
Specifically, imagine we're building a single file 3 times, all with `-Csave-temps -Cincremental=...`. Let's call the object file we're building for the codegen unit for `main` "`XXX.o`" just for clarity since it's probably some gigantic hash name:
```
#[inline(never)]
#[cfg(any(rpass1, rpass3))]
fn a() -> i32 {
0
}
#[cfg(any(cfail2))]
fn a() -> i32 {
1
}
fn main() {
evil::evil();
assert_eq!(a(), 0);
}
mod evil {
#[cfg(any(rpass1, rpass3))]
pub fn evil() {
unsafe {
std::arch::asm!("/* */");
}
}
#[cfg(any(cfail2))]
pub fn evil() {
unsafe {
std::arch::asm!("missing");
}
}
}
```
Session 1 (`rpass1`):
* Type-check, borrow-check, etc.
* Serialize the dep graph to the incremental working directory `.../s-...-working/`.
* Codegen object file to a temp file `XXX.rcgu.o` which is spit out in the cwd.
* Hard-link[^1] `XXX.rcgu.o` to the incremental working directory `.../s-...-working/XXX.o`.
* Save-temps option means we don't delete `XXX.rgcu.o`.
* Link the binary and stuff.
* Finalize[^2] the working incremental session by renaming `.../s-...-working` to ` s-...-asjkdhsjakd` (some other finalized incr comp session dir name).
Session 2 (`cfail2`):
* Load artifacts from the previous *finalized* incremental session, namely the dep graph.
* Type-check, borrow-check, etc. since the file has changed, so most dep graph nodes are red.
* Serialize the dep graph to the incremental working directory `.../s-...-working/`.
* Codegen object file to a temp file `XXX.rcgu.o`. **HERE IS THE PROBLEM**: The hard-link is still set up to point to the inode from `XXX.o` from the first session, so this also modifies the `XXX.o` in the previous finalized session directory.
* Codegen emits an error b/c `missing` is not an instruction, so we abort before finalizing the incremental session. Specifically, this means that the *previous* session is the last finalized session.
Session 3 (`rpass3`):
* Load artifacts from the previous *finalized* incremental session, namely the dep graph. NOTE that this is from session 1.
* All the dep graph nodes are green since we are basically replaying session 1.
* codegen object file `XXX.o`, which is detected as *reused* from session 1 since dep nodes were green. That means we **reuse** `XXX.o` which had been dirtied from session 2.
* Link the binary and stuff.
This results in a binary which reuses some of the build artifacts from session 2, but thinks it's from session 1.
At this point, I hope it's clear to see that the incremental results from session 1 were dirtied from session 2, but we reuse them as if session 1 was the previous (finalized) incremental session we ran. This is at best really buggy, and at worst **unsound**.
This isn't limited to `-C save-temps`, since there are other combinations of flags that may keep around temporary files (hard linked) in the working directory (like `-C debuginfo=1 -C split-debuginfo=unpacked` on darwin, for example).
---
This PR implements a fix which is to prepend temp filenames with a random string that is generated per invocation of rustc. This string is not *deterministic*, but temporary files are transient anyways, so I don't believe this is a problem.
That means that temp files are now something like... `{crate-name}.{cgu}.{invocation_temp}.rcgu.o`, where `{invocation_temp}` is the new temporary string we generate per invocation of rustc.
Fixes https://github.com/rust-lang/rust/issues/139407
[^1]: 175dcc7773/compiler/rustc_fs_util/src/lib.rs (L60)
[^2]: 175dcc7773/compiler/rustc_incremental/src/persist/fs.rs (L1-L40)
Tell LLVM about impossible niche tags
I was trying to find a better way of emitting discriminant calculations, but sadly had no luck.
So here's a fairly small PR with the bits that did seem worth bothering:
1. As the [`TagEncoding::Niche` docs](https://doc.rust-lang.org/nightly/nightly-rustc/rustc_abi/enum.TagEncoding.html#variant.Niche) describe, it's possible to end up with a dead value in the input that's not already communicated via the range parameter attribute nor the range load metadata attribute. So this adds an `llvm.assume` in non-debug mode to tell LLVM about that. (That way it can tell that the sides of the `select` have disjoint possible values.)
2. I'd written a bunch more tests, or at least made them parameterized, in the process of trying things out, so this checks in those tests to hopefully help future people not trip on the same weird edge cases, like when the tag type is `i8` but yet there's still a variant index and discriminant of `258` which doesn't fit in that tag type because the enum is really weird.
Refactor Apple version handling in the compiler
Move various Apple version handling code in the compiler out `rustc_codegen_ssa` and into a place where it can be accessed by `rustc_attr_parsing`, which I found to be necessary when doing https://github.com/rust-lang/rust/pull/136867. Thought I'd split it out to make it easier to land, and to make further changes like https://github.com/rust-lang/rust/pull/131477 have fewer conflicts / PR dependencies.
There should be no functional changes in this PR.
`@rustbot` label O-apple
r? rust-lang/compiler
Autodiff batching
Enzyme supports batching, which is especially known from the ML side when training neural networks.
There we would normally have a training loop, where in each iteration we would pass in some data (e.g. an image), and a target vector. Based on how close we are with our prediction we compute our loss, and then use backpropagation to compute the gradients and update our weights.
That's quite inefficient, so what you normally do is passing in a batch of 8/16/.. images and targets, and compute the gradients for those all at once, allowing better optimizations.
Enzyme supports batching in two ways, the first one (which I implemented here) just accepts a Batch size,
and then each Dual/Duplicated argument has not one, but N shadow arguments. So instead of
```rs
for i in 0..100 {
df(x[i], y[i], 1234);
}
```
You can now do
```rs
for i in 0..100.step_by(4) {
df(x[i+0],x[i+1],x[i+2],x[i+3], y[i+0], y[i+1], y[i+2], y[i+3], 1234);
}
```
which will give the same results, but allows better compiler optimizations. See the testcase for details.
There is a second variant, where we can mark certain arguments and instead of having to pass in N shadow arguments, Enzyme assumes that the argument is N times longer. I.e. instead of accepting 4 slices with 12 floats each, we would accept one slice with 48 floats. I'll implement this over the next days.
I will also add more tests for both modes.
For any one preferring some more interactive explanation, here's a video of Tim's llvm dev talk, where he presents his work. https://www.youtube.com/watch?v=edvaLAL5RqU
I'll also add some other docs to the dev guide and user docs in another PR.
r? ghost
Tracking:
- https://github.com/rust-lang/rust/issues/124509
- https://github.com/rust-lang/rust/issues/135283
Rename `is_like_osx` to `is_like_darwin`
Replace `is_like_osx` with `is_like_darwin`, which more closely describes reality (OS X is the pre-2016 name for macOS, and is by now quite outdated; Darwin is the overall name for the OS underlying Apple's macOS, iOS, etc.).
``@rustbot`` label O-apple
r? compiler
