fusion-engineering/inline-python 393
Inline Python code directly in your Rust code
assert!() and check!() macros for Rust, inspired by Catch2
fusion-engineering/setup-git-credentials 24
GitHub action to clone private respositories.
fusion-engineering/rust-git-version 17
Compile the Git version (tag name, or hash otherwise) and dirty state into your Rust program.
The one and only perfect configuration for Vim, Zsh, Bash, Git, etc.
A minimalistic library to help making your x86 assembly program bootable.
Rust interface for Linux futexes, the fast user-space locking primitive.
Display git information next to every path component in your prompt.
Fast implementations of integer/floating point conversion operations.
C++ bindings for CloudABI.
push eventwg21link/db
commit sha f559e8a827805b0584faba318d57a2b4d6ec3428
Automatic update.
push time in 3 hours
pull request commentrust-lang/rust
Only use LOCAL_{STDOUT,STDERR} when set_{print/panic} is used.
cc @fasterthanlime whose blogpost inspired this change
comment created time in 5 hours
push eventfusion-engineering-forks/rust
commit sha dd7b8c85a6a2bffb2cce1c40ba72680a1d7be93b
Perform most diagnostic lookups in `resolution_failure` Previously, these were spread throughout the codebase. This had two drawbacks: 1. It caused the fast path to be slower: even if a link resolved, rustdoc would still perform various lookups for the error diagnostic. 2. It was inconsistent and didn't always give all diagnostics (https://github.com/rust-lang/rust/issues/76925) Now, diagnostics only perform expensive lookups in the error case. Additionally, the error handling is much more consistent, both in wording and behavior. - Remove `CannotHaveAssociatedItems`, `NotInScope`, `NoAssocItem`, and `NotAVariant` in favor of the more general `NotResolved` `resolution_failure` will now look up which of the four above categories is relevant, instead of requiring the rest of the code to be consistent and accurate in which it picked. - Remove unnecessary lookups throughout the intra-doc link pass. These are now done by `resolution_failure`. + Remove unnecessary `extra_fragment` argument to `variant_field()`; it was only used to do lookups on failure. + Remove various lookups related to associated items + Remove distinction between 'not in scope' and 'no associated item' - Don't perform unnecessary copies - Remove unused variables and code - Update tests - Note why looking at other namespaces is still necessary - 'has no inner item' -> 'contains no item' bless tests
commit sha 472e52e5a03790becdbe21be1002a90dd2d7d3d4
Fix intra-doc links for primitives - Add `PrimTy::name` and `PrimTy::name_str` - Use those new functions to distinguish between the name in scope and the canonical name - Fix diagnostics for primitive types - Add tests for primitives
commit sha 049d29bac580b958cdff6e898dad22de09a27958
Unify primitive errors with other intra-link errors Now that `PrimTy::name()` exists, there's no need to carry around the name of the primitive that failed to resolve. This removes the variants special-casing primitives in favor of `NotResolved`. - Remove `NoPrimitiveImpl` and `NoPrimitiveAssocItem` - Remove hacky `has_primitive` check in `resolution_failure()` - Fixup a couple tests that I forgot to `--bless` before
commit sha 6f881b3b7db6e08d5ca3a940befc7898aab1568a
might_permit_raw_init: also check aggregate fields
commit sha 71bdb8481760f7c6b8540d3dd176ed54a92c0f31
Auto merge of #76955 - jyn514:refactor-diagnostics, r=euclio Refactor and fix intra-doc link diagnostics, and fix links to primitives Closes https://github.com/rust-lang/rust/issues/76925, closes https://github.com/rust-lang/rust/issues/76693, closes https://github.com/rust-lang/rust/issues/76692. Originally I only meant to fix #76925. But the hack with `has_primitive` was so bad it was easier to fix the primitive issues than to try and work around it. Note that this still has one bug: `std::primitive::i32::MAX` does not resolve. However, this fixes the ICE so I'm fine with fixing the link in a later PR. This is part of a series of refactors to make #76467 possible. This is best reviewed commit-by-commit; it has detailed commit messages. r? `@euclio`
commit sha 3009c5260f292da6785c67af8c823d4fa0e870d9
update tokei and ripgrep in cargotest
commit sha b8363295d555494bbaa119eba8b16a3057e6728c
Auto merge of #71274 - RalfJung:raw-init-check-aggregate, r=petrochenkov might_permit_raw_init: also check aggregate fields This is the next step for https://github.com/rust-lang/rust/issues/66151: when doing `mem::zeroed`/`mem::uninitialized`, also recursively check fields of aggregates (except for arrays) for whether they permit zero/uninit initialization.
commit sha 07fd17f7014de6dc209c7dc4de159a2a5acea173
Only use LOCAL_{STDOUT,STDERR} when set_{print/panic} is used. The thread local LOCAL_STDOUT and LOCAL_STDERR are only used by the test crate to capture output from tests when running them in the same process in differen threads. However, every program will check these variables on every print, even outside of testing. This involves allocating a thread local key, and registering a thread local destructor. This can be somewhat expensive. This change keeps a global flag (LOCAL_STREAMS) which will be set to true when either of these local streams is used. (So, effectively only in test and benchmark runs.) When this flag is off, these thread locals are not even looked at and therefore will not be initialized on the first output on every thread, which also means no thread local destructors will be registered.
commit sha ed3ead013f40c65e2972c794a71b756237a31b46
Relax memory ordering of LOCAL_STREAMS and document it.
commit sha de597fca40d129435c53a69c6662d7bfac29771d
Optimize set_{panic,print}(None).
push time in 5 hours
PR opened rust-lang/rust
The thread local LOCAL_STDOUT and LOCAL_STDERR are only used by the test crate to capture output from tests when running them in the same process in differen threads. However, every program will check these variables on every print, even outside of testing.
This involves allocating a thread local key, and registering a thread local destructor. This can be somewhat expensive.
This change keeps a global flag (LOCAL_STREAMS) which will be set to true when either of these local streams is used. (So, effectively only in test and benchmark runs.) When this flag is off, these thread locals are not even looked at and therefore will not be initialized on the first output on every thread, which also means no thread local destructors will be registered.
pr created time in 6 hours
push eventfusion-engineering-forks/rust
commit sha d132cf6da07eb773d813f15338e4d8406db508dd
Optimize set_{panic,print}(None).
push time in 6 hours
push eventwg21link/db
commit sha de93521f411536520f903d5e95eeeba48cb79fea
Automatic update.
push time in 8 hours
push eventfusion-engineering-forks/rust
commit sha 73ada2d40429488aaaacf37b608bababc137b910
Explicitly document the size guarantees that Option makes. Triggered by a discussion on wg-unsafe-code-guidelines about which layouts of `Option<T>` one can guarantee are optimised to a single pointer.
commit sha f5118a525fcf9db4102d903650331039158eff11
Clarify and add guarantee about `transmute`.
commit sha 83f47aa11bd664ed8a15ef9833063833b7b3e71c
Be clear about the reverse `transmute` guarantees.
commit sha f3d7196caec3f54e572c7389b1cef9fd9e62c1ed
Be clearer about Some/None transmute.
commit sha 8cb8955d570c76631840bfc98825ca49c0dd8eea
Change notation. Co-authored-by: Ralf Jung <post@ralfj.de>
commit sha 55802e3bf3bf6d1db5c76aea581a7912bd752890
Add Rust function pointers. Co-authored-by: Ralf Jung <post@ralfj.de>
commit sha 68209c3fe4e0f5c3758f18e98efc175af31c2e51
Rename the types for clarity.
commit sha 9bac5774d7b452b2227c9fb77a4c6de3f432ee55
Grammar tweak.
commit sha c5975e9b6c5781b3b7300b7921c14b060086e1c1
Reduce duplicate in liballoc reserve error handling
commit sha 459969f88ff95c94b7b34043a7f0e13de91de4f8
added restriction lint that prohibits the usage of unimplemented, unreachable or panic in a function of type result or option
commit sha ceab1a9167655eba9f9556f8766f8702e49dfef3
removed unnecessary comment
commit sha 8462cce96081b87eba7a5bc89130a1a09fe1f6d0
edited documentation
commit sha b2d8ca9a766703469178ea37d4d46067bb6fa926
ran cargo dev update lints
commit sha b006522393a3c3c2656e1ccdfbb0076ff1bd7e99
added lint for todo and removed option
commit sha a424a2c1676a29c147252873037e8943d54941d3
changed check_impl_item to check_fn and added a few more test cases
commit sha 73a3288282e733bfc5893e9920d29f1de5a21591
uncommented fn
commit sha 27c90b881df93b53fd3f24dcbfed116379c2fc69
initial implementation of OpenOptions to c_int
commit sha eb3906be4ad375cc6b83cd6a6e0116817db22575
Fix typo get openoptions function name Co-authored-by: Ivan Tham <pickfire@riseup.net>
commit sha d3b9ece4c0028ff7e36e34df2d2b26366f9c0648
add tests related to tuple scalar layout with ZST in last field
commit sha e5d85f917b8965a5e62513c17cbb887366b152bc
allow reordering of the last field of a MaybeUnsized struct if it's a ZST
push time in 9 hours
push eventfusion-engineering-forks/rust
commit sha 73ada2d40429488aaaacf37b608bababc137b910
Explicitly document the size guarantees that Option makes. Triggered by a discussion on wg-unsafe-code-guidelines about which layouts of `Option<T>` one can guarantee are optimised to a single pointer.
commit sha f5118a525fcf9db4102d903650331039158eff11
Clarify and add guarantee about `transmute`.
commit sha 83f47aa11bd664ed8a15ef9833063833b7b3e71c
Be clear about the reverse `transmute` guarantees.
commit sha f3d7196caec3f54e572c7389b1cef9fd9e62c1ed
Be clearer about Some/None transmute.
commit sha 8cb8955d570c76631840bfc98825ca49c0dd8eea
Change notation. Co-authored-by: Ralf Jung <post@ralfj.de>
commit sha 55802e3bf3bf6d1db5c76aea581a7912bd752890
Add Rust function pointers. Co-authored-by: Ralf Jung <post@ralfj.de>
commit sha 68209c3fe4e0f5c3758f18e98efc175af31c2e51
Rename the types for clarity.
commit sha 9bac5774d7b452b2227c9fb77a4c6de3f432ee55
Grammar tweak.
commit sha c5975e9b6c5781b3b7300b7921c14b060086e1c1
Reduce duplicate in liballoc reserve error handling
commit sha 459969f88ff95c94b7b34043a7f0e13de91de4f8
added restriction lint that prohibits the usage of unimplemented, unreachable or panic in a function of type result or option
commit sha ceab1a9167655eba9f9556f8766f8702e49dfef3
removed unnecessary comment
commit sha 8462cce96081b87eba7a5bc89130a1a09fe1f6d0
edited documentation
commit sha b2d8ca9a766703469178ea37d4d46067bb6fa926
ran cargo dev update lints
commit sha b006522393a3c3c2656e1ccdfbb0076ff1bd7e99
added lint for todo and removed option
commit sha a424a2c1676a29c147252873037e8943d54941d3
changed check_impl_item to check_fn and added a few more test cases
commit sha 73a3288282e733bfc5893e9920d29f1de5a21591
uncommented fn
commit sha 27c90b881df93b53fd3f24dcbfed116379c2fc69
initial implementation of OpenOptions to c_int
commit sha eb3906be4ad375cc6b83cd6a6e0116817db22575
Fix typo get openoptions function name Co-authored-by: Ivan Tham <pickfire@riseup.net>
commit sha d3b9ece4c0028ff7e36e34df2d2b26366f9c0648
add tests related to tuple scalar layout with ZST in last field
commit sha e5d85f917b8965a5e62513c17cbb887366b152bc
allow reordering of the last field of a MaybeUnsized struct if it's a ZST
push time in 11 hours
push eventwg21link/db
commit sha 60bf2e50f23bcaecd8d9270e6d7bde6a1fec044b
Automatic update.
push time in 21 hours
push eventwg21link/db
commit sha 2862af8adce255a04d2da514b1d108c9453b384d
Automatic update.
push time in a day
push eventwg21link/db
commit sha b7a86249b1e935c0367a4b6c78cad2bcf871cd38
Automatic update.
push time in a day
Pull request review commentrust-lang/rust
Use futex-based thread::park/unpark on Linux.
+#![cfg(any(target_os = "linux", target_os = "android"))]++use crate::convert::TryInto;+use crate::ptr::null;+use crate::sync::atomic::AtomicI32;+use crate::time::Duration;++pub fn futex_wait(futex: &AtomicI32, expected: i32, timeout: Option<Duration>) {+ let timespec = timeout.and_then(|d| {+ Some(libc::timespec {+ // Sleep forever if the timeout is longer than fits in a timespec.
i32::MAX seconds is more than 68 years, and most systems use a i64 (2e+11 years) here, so I don't think it really matters. (Note this is a duration, not an absolute timestamp.) The implementations I've seen round up (in case the underlying call doesn't do nanosecond precision, e.g. on Windows), and overflow to infinity. So I just followed that.
comment created time in a day
Pull request review commentrust-lang/rust
Use futex-based thread::park/unpark on Linux.
+use crate::sync::atomic::AtomicI32;+use crate::sync::atomic::Ordering::{Acquire, Release};+use crate::sys::futex::{futex_wait, futex_wake};+use crate::time::Duration;++const PARKED: i32 = -1;+const EMPTY: i32 = 0;+const NOTIFIED: i32 = 1;++pub struct Parker {+ state: AtomicI32,+}++// Notes about memory ordering:+//+// Memory ordering is only relevant for the relative ordering of operations+// between different variables. Even Ordering::Relaxed guarantees a+// monotonic/consistent order when looking at just a single atomic variable.+//+// So, since this parker is just a single atomic variable, we only need to look+// at the ordering guarantees we need to provide to the 'outside world'.+//+// The only memory ordering guarantee that parking and unparking provide, is+// that things which happened before unpark() are visible on the thread+// returning from park() afterwards. Otherwise, it was effectively unparked+// before unpark() was called while still consuming the 'token'.+//+// In other words, unpark() needs to synchronize with the part of park() that+// consumes the token and returns.+//+// This is done with a release-acquire synchronization, by using+// Ordering::Release when writing NOTIFIED (the 'token') in unpark(), and using+// Ordering::Acquire when checking for this state in park().+impl Parker {+ #[inline]+ pub const fn new() -> Self {+ Parker { state: AtomicI32::new(EMPTY) }+ }++ // Assumes this is only called by the thread that owns the Parker,+ // which means that `self.state != PARKED`.+ pub unsafe fn park(&self) {+ // Change NOTIFIED=>EMPTY or EMPTY=>PARKED, and directly return in the+ // first case.+ if self.state.fetch_sub(1, Acquire) == NOTIFIED {+ return;+ }+ loop {+ // Wait for something to happen, assuming it's still set to PARKED.+ futex_wait(&self.state, PARKED, None);
Nope. This is the value expected when starting the wait. If it matches, it continues to put the thread to sleep and wait for a notification (all atomically). If it doesn't match, it doesn't wait and returns directly.
comment created time in a day
pull request commentrust-lang/rust
Remove std::io::lazy::Lazy in favour of SyncOnceCell
Maybe add comments explaining why this is ignored?
Done.
It might be the case that adding this in the test would also solve it:
#[cfg(target_os = "emscripten")]
#[link_args = "-s EXIT_RUNTIME=1"]
extern "C" {}
But I'm unable to test this locally. And I don't think this test really matters on emscripten.
comment created time in a day
push eventfusion-engineering-forks/rust
commit sha 6b8b9c4adbd7d8d53d730a3c01d154dec3bd179f
Disable stdout-during-shutdown test on emscripten.
push time in a day
push eventfusion-engineering-forks/rust
commit sha 120b26946bd4b7d8a074634367bc21694d5e6b02
Disable stdout-during-shutdown test on emscripten/wasm.
push time in a day
push eventfusion-engineering-forks/rust
commit sha c0ed0257609e469ede50f6784c8cdaed8270f15c
Disable stdout-during-shutdown test on emscripten/wasm.
push time in a day
push eventfusion-engineering-forks/rust
commit sha ea3ac18fd79b62d2c84bae9386b334503fd1f487
Disable stdout-during-shutdown test on wasm.
push time in a day
pull request commentrust-lang/rust
Remove std::io::lazy::Lazy in favour of SyncOnceCell
#77212 failed on this PR for auto (wasm32, ubuntu-latest-xl):
failures:
---- [ui] ui/stdout-during-shutdown.rs stdout ----
diff of run.stdout:
- hello, world!
The actual run.stdout differed from the expected run.stdout.
Actual run.stdout saved to /checkout/obj/build/x86_64-unknown-linux-gnu/test/ui/stdout-during-shutdown/stdout-during-shutdown.run.stdout
normalized run.stderr:
stdio streams had content in them that was not flushed. you should set EXIT_RUNTIME to 1 (see the FAQ), or make sure to emit a newline when you printf etc.
Apparently that target has trouble with atexit and output without a newline, which is exactly the point of the test. See this:
By default Emscripten sets EXIT_RUNTIME=0, which means that we don’t include code to shut down the runtime. That means that when main() exits, we don’t flush the stdio streams, or call the destructors of global C++ objects, or call atexit callbacks.
I'll disable the test for this platform.
comment created time in a day
push eventfusion-engineering-forks/rust
commit sha c5975e9b6c5781b3b7300b7921c14b060086e1c1
Reduce duplicate in liballoc reserve error handling
commit sha 459969f88ff95c94b7b34043a7f0e13de91de4f8
added restriction lint that prohibits the usage of unimplemented, unreachable or panic in a function of type result or option
commit sha ceab1a9167655eba9f9556f8766f8702e49dfef3
removed unnecessary comment
commit sha 8462cce96081b87eba7a5bc89130a1a09fe1f6d0
edited documentation
commit sha b2d8ca9a766703469178ea37d4d46067bb6fa926
ran cargo dev update lints
commit sha b006522393a3c3c2656e1ccdfbb0076ff1bd7e99
added lint for todo and removed option
commit sha a424a2c1676a29c147252873037e8943d54941d3
changed check_impl_item to check_fn and added a few more test cases
commit sha 73a3288282e733bfc5893e9920d29f1de5a21591
uncommented fn
commit sha 27c90b881df93b53fd3f24dcbfed116379c2fc69
initial implementation of OpenOptions to c_int
commit sha eb3906be4ad375cc6b83cd6a6e0116817db22575
Fix typo get openoptions function name Co-authored-by: Ivan Tham <pickfire@riseup.net>
commit sha d3b9ece4c0028ff7e36e34df2d2b26366f9c0648
add tests related to tuple scalar layout with ZST in last field
commit sha e5d85f917b8965a5e62513c17cbb887366b152bc
allow reordering of the last field of a MaybeUnsized struct if it's a ZST
commit sha e9bc3ddb073f2261ac46832d985efe8db863ed6a
test that we do not change the offset of ZST tuple fields when unsizing
commit sha 68217c9e0f1269d29b4c1c72d08fb1b95bf441cd
ignore zst offsets instead
commit sha 6fc1d8bc13124bb134d7ab54e56237821a55912e
add codegen test
commit sha 24e0913e37cc6fc363b37d13bf519db212f175a2
handle vector layout
commit sha 1bc0627607262cc60a7692b16e205f30fc88b89f
Add as_flag function to the OpenOptionsExt struct
commit sha 2c9e27b759a9e9feeb943fb855e32d7383d1dcc6
Merge branch 'convert-openoptions-cint' of github.com:FedericoPonzi/rust into convert-openoptions-cint
commit sha f9fcbbea03edb735c22311522b55d7b854bd6ac0
fixed bug
commit sha 7c1e5c1dcd25c945f619eda289f639dbe2b002da
Update OpenOptions::as_flags docs, and minor styling
push time in 2 days
push eventwg21link/db
commit sha 91785f70a0b16032f1d100a58f15bfc79e0e176f
Automatic update.
push time in 2 days
push eventwg21link/db
commit sha fa4572d373b0cc9eee608b74ed4ad0d42c82a320
Automatic update.
push time in 2 days
push eventwg21link/db
commit sha fecd0af10d1efca0760ac43803e4def9683d028e
Automatic update.
push time in 2 days
push eventwg21link/db
commit sha d29e6e969ec5e8a95767043a041b4487f5492f8b
Automatic update.
push time in 3 days
push eventwg21link/db
commit sha 3f4b35e3a296f15075c65687cb6176a182a73cf4
Automatic update.
push time in 3 days
push eventwg21link/db
commit sha 3284116f8a8bed2e2df27989bdbcb69693c5b825
Automatic update.
push time in 3 days
push eventwg21link/db
commit sha 27f42fd507402814c8caa10ccfe7cce954e5180e
Automatic update.
push time in 3 days
pull request commentrust-lang/rust
Fix FIXME in core::num test: Check sign of zero in min/max tests.
r? @nagisa
comment created time in 3 days
PR opened rust-lang/rust
r? nagisa
@rustbot modify labels: +C-cleanup
pr created time in 3 days
create barnchfusion-engineering-forks/rust
branch : fix-fixme-min-max-sign-test
created branch time in 3 days
issue commentrust-lang/rust
BufWriter should buffer up to its capacity
Maybe instead, writing "12" and then "34" into a BufWriter with a capacity of 4 should flush right after the "34", instead of waiting until the next call. Feels slightly more consistent.
comment created time in 3 days
issue commentrust-lang/rust
BufWriter should buffer up to its capacity
I think the current implementation makes sense. I'd say the goal of a BufWriter to reduce the number of write calls on the underlying stream. Buffering a write call of the full buffer capacity only serves the purpose of using the buffer as much as possible, as the number of write calls (and their sizes) remain the same. That doesn't seem useful to me.
comment created time in 3 days
push eventfusion-engineering-forks/rust
commit sha 95bbadb31564e8eb163e54cbe834d08a4ecabfc7
Split sys_common::Mutex in StaticMutex and MovableMutex. The (unsafe) Mutex from sys_common had a rather complicated interface. You were supposed to call init() manually, unless you could guarantee it was neither moved nor used reentrantly. Calling `destroy()` was also optional, although it was unclear if 1) resources might be leaked or not, and 2) if destroy() should only be called when `init()` was called. This allowed for a number of interesting (confusing?) different ways to use this Mutex, all captured in a single type. In practice, this type was only ever used in two ways: 1. As a static variable. In this case, neither init() nor destroy() are called. The variable is never moved, and it is never used reentrantly. It is only ever locked using the LockGuard, never with raw_lock. 2. As a Boxed variable. In this case, both init() and destroy() are called, it will be moved and possibly called reentrantly. No other combinations are used anywhere in `std`. This change simplifies things by splitting this Mutex type into two types matching the two use cases: StaticMutex and MovableMutex. The interface of both new types is now both safer and simpler. The first one does not call nor expose init/destroy, and the second one calls those automatically in its new() and Drop functions. Also, the locking functions of MovableMutex are no longer unsafe.
commit sha 21f2799183df40d116f993c82eb73ec7c5e78f6f
Add missing safety note about (internal) std::io::Lazy.
commit sha 6d2d2a67ba98f58f2b74be517813b67fa2f723fc
Fix ui test. This test checks if the compiler complains about accesing a private field before complaining (or crashing) about the private function on it not marked as stable/unstable. The interface of the internal type (sys_common's Mutex) used for this was changed. With this change, it uses another function to test for the same issue.
push time in 3 days
pull request commentrust-lang/rust
Split sys_common::Mutex in StaticMutex and MovableMutex.
@rustbot modify labels: +A-concurrency +C-cleanup
comment created time in 3 days
pull request commentrust-lang/rust
Use futex-based thread::park/unpark on Linux.
@rustbot modify labels: +T-libs +A-concurrency +O-linux
comment created time in 3 days
pull request commentrust-lang/rust
Remove std::io::lazy::Lazy in favour of SyncOnceCell
@rustbot modify labels: +T-libs +A-io
comment created time in 3 days
PR opened rust-lang/rust
The double underscores were used to work around issue #12808, which was solved in 2016.
pr created time in 3 days
create barnchfusion-engineering-forks/rust
branch : no-more-funny-underscores
created branch time in 3 days
pull request commentrust-lang/rust
Remove std::io::lazy::Lazy in favour of SyncOnceCell
Two interesting minor differences before/after:
1:
use std::{thread, io, mem, time::Duration};
fn main() {
print!("hello");
thread::spawn(|| mem::forget(io::stdout()));
thread::sleep(Duration::from_secs(1));
}
Before this change, this outputs nothing: The Arc count does not drop back to 0, the stdout buffer is not dropped/flushed.
After this change, this propefly flushes the stream, and outputs hello. Improvement!
2:
use std::{thread, io, mem, time::Duration};
fn main() {
print!("hello");
thread::spawn(|| mem::forget(io::stdout().lock()));
thread::sleep(Duration::from_secs(1));
}
Before this change, this outputs hello, while after this change, this no longer flushes the stream and outputs nothing. This seems like a regression. However: This only worked because the whole ReentrantMutex in the Arc was dropped (thus flushing the buffer), even though the ReentrantMutex inside there was still locked (since a guard was leaked). This might have been unsound: destroying a locked mutex can be undefined behaviour with posix pthreads.
comment created time in 3 days
push eventfusion-engineering-forks/rust
commit sha 6f9c1323a7a0fe162a5642e229d54afec7ccb299
Call ReentrantMutex::init() in stdout().
push time in 3 days
push eventwg21link/db
commit sha 60c2c08c98dda7f9d9648fb92879fa0192df6240
Automatic update.
push time in 3 days
push eventfusion-engineering-forks/rust
commit sha 45700a9d58679131a628ddc95dd8ce0fcf1f2430
Drop use of Arc from Stdin and Stdout.
push time in 3 days
push eventfusion-engineering-forks/rust
commit sha e9b25f520bd2e3687213aa1162e631b08b9bf7ed
Add test to check stdout flushing during shutdown.
push time in 3 days
push eventfusion-engineering-forks/rust
commit sha bab15f773afd5724023d9a065b5af276e2468ff5
Remove std::io::lazy::Lazy in favour of SyncOnceCell The (internal) std::io::lazy::Lazy was used to lazily initialize the stdout and stdin buffers (and mutexes). It uses atexit() to register a destructor to flush the streams on exit, and mark the streams as 'closed'. Using the stream afterwards would result in a panic. Stdout uses a LineWriter which contains a BufWriter that will flush the buffer on drop. This one is important to be executed during shutdown, to make sure no buffered output is lost. It also forbids access to stdout afterwards, since the buffer is already flushed and gone. Stdin uses a BufReader, which does not implement Drop. It simply forgets any previously read data that was not read from the buffer yet. This means that in the case of stdin, the atexit() function's only effect is making stdin inaccessible to the program, such that later accesses result in a panic. This is uncessary, as it'd have been safe to access stdin during shutdown of the program. --- This change removes the entire io::lazy module in favour of SyncOnceCell. SyncOnceCell's fast path is much faster (a single atomic operation) than locking a sys_common::Mutex on every access like Lazy did. However, SyncOnceCell does not use atexit() to drop the contained object during shutdown. As noted above, this is not a problem for stdin. It simply means stdin is now usable during shutdown. The atexit() call for stdout is moved to the stdio module. Unlike the now-removed Lazy struct, SyncOnceCell does not have a 'gone and unusable' state that panics. Instead of adding this again, this simply replaces the buffer with one with zero capacity. This effectively flushes the old buffer *and* makes any writes afterwards pass through directly without touching a buffer, making print!() available during shutdown without panicking.
commit sha 2fed486355c0177f955a700130770d258e425ea7
Add test to check stdout flushing during shutdown.
push time in 3 days
PR opened rust-lang/rust
The (internal) std::io::lazy::Lazy was used to lazily initialize the stdout and stdin buffers (and mutexes). It uses atexit() to register a destructor to flush the streams on exit, and mark the streams as 'closed'. Using the stream afterwards would result in a panic.
Stdout uses a LineWriter which contains a BufWriter that will flush the buffer on drop. This one is important to be executed during shutdown, to make sure no buffered output is lost. It also forbids access to stdout afterwards, since the buffer is already flushed and gone.
Stdin uses a BufReader, which does not implement Drop. It simply forgets any previously read data that was not read from the buffer yet. This means that in the case of stdin, the atexit() function's only effect is making stdin inaccessible to the program, such that later accesses result in a panic. This is uncessary, as it'd have been safe to access stdin during shutdown of the program.
This change removes the entire io::lazy module in favour of SyncOnceCell. SyncOnceCell's fast path is much faster (a single atomic operation) than locking a sys_common::Mutex on every access like Lazy did.
However, SyncOnceCell does not use atexit() to drop the contained object during shutdown.
As noted above, this is not a problem for stdin. It simply means stdin is now usable during shutdown.
The atexit() call for stdout is moved to the stdio module. Unlike the now-removed Lazy struct, SyncOnceCell does not have a 'gone and unusable' state that panics. Instead of adding this again, this simply replaces the buffer with one with zero capacity. This effectively flushes the old buffer and makes any writes afterwards pass through directly without touching a buffer, making print!() available during shutdown without panicking.
pr created time in 3 days
push eventwg21link/db
commit sha 8e7ed192935d382e649b15ff36833511a2ae7b13
Automatic update.
push time in 3 days
push eventfusion-engineering-forks/rust
commit sha 982fd1617521595b996e42fce54b3d39a1478e14
Fix ui test. This test checks if the compiler complains about accesing a private field before complaining (or crashing) about the private function on it not marked as stable/unstable. The interface of the internal type (sys_common's Mutex) used for this was changed. With this change, it uses another function to test for the same issue.
push time in 3 days
push eventfusion-engineering-forks/rust
commit sha 11b8eeb38306e8b193e2a174842ccea70f7dd4fa
Add missing safety note about (internal) std::io::Lazy.
push time in 3 days
PR opened rust-lang/rust
The (unsafe) Mutex from sys_common had a rather complicated interface. You were supposed to call init() manually, unless you could guarantee it was neither moved nor used reentrantly.
Also, wait_timeout was not guaranteed to use the monotonic clock when init() was not called, although that was not documented.
Calling destroy() was also optional, although it was unclear if 1) resources might be leaked or not, and 2) if destroy() should only be called when init() was called.
This allowed for a number of interesting (confusing?) different ways to use this Mutex, all captured in a single type.
In practice, this type was only ever used in two ways:
-
As a static variable. In this case, neither
init()nordestroy()are called. Locking with timeout is not used. The variable is never moved, and it is never used reentrantly. It is only ever locked using theLockGuard, never withraw_lock. -
As a
Boxed variable. In this case, bothinit()anddestroy()are called, locking with timeouts is used, it will be moved, etc.
No other combinations are used anywhere in std.
This change simplifies things by splitting this Mutex type into two types matching the two use cases: StaticMutex and MovableMutex.
The interface of both new types is now both safer and simpler. The first one does not call nor expose init/destroy, and the second one calls those automatically in its new() and Drop functions. Also, the locking functions of MovableMutex are no longer unsafe.
@rustbot modify labels: +T-libs
pr created time in 3 days
push eventwg21link/db
commit sha 73e5090e76bc7b46cdc7be33786eda25bd81aa22
Automatic update.
push time in 4 days
push eventwg21link/db
commit sha a55910ccd5dc83183c2406c1c876d35306327362
Automatic update.
push time in 4 days
push eventwg21link/db
commit sha f9553b823d5ba739de5cfd576513387932580659
Automatic update.
push time in 4 days
push eventwg21link/db
commit sha 63902378ecbd7bdc36919d9e379e2c6c75fb306c
Automatic update.
push time in 4 days
push eventwg21link/db
commit sha d460a2a52791eb0161005f140829e4b63d7c0eae
Automatic update.
push time in 4 days
pull request commentrust-lang/rust
@rustbot modify labels: +T-libs +A-raw-pointers +A-slice
comment created time in 4 days
PR opened rust-lang/rust
This has been unstable for almost a year now. Time to stabilize?
Closes #65807.
@rustbot modify labels: +T-libs +A-raw-pointers +A-slice +needs-fcp
pr created time in 4 days
create barnchfusion-engineering-forks/rust
branch : stabilize-slice-ptr-range
created branch time in 4 days
push eventwg21link/db
commit sha 53d4cb0b5e1b9475b8ebb2929345a796edc41906
Automatic update.
push time in 4 days
push eventwg21link/db
commit sha 4d20cb8127c5196bac82478431f4c88822ff065a
Automatic update.
push time in 4 days
issue commentrust-lang/rust
Tracking issue for slice_ptr_range
Note that with #71499 these functions can be made
const fn. The stabilization of these methods asconst fncan just be tracked as part of #71499.
Thanks! PR: #77097.
comment created time in 4 days
PR opened rust-lang/rust
Gated behind const_ptr_offset, as suggested by https://github.com/rust-lang/rust/issues/65807#issuecomment-697229404
This also marks [].as_mut_ptr() as const, because it's used by as_mut_ptr_range. I gated it behind the same feature, because I figured it's not worth adding a separate tracking issue for const as_mut_ptr.
pr created time in 4 days
create barnchfusion-engineering-forks/rust
branch : slice-ptr-range-const-fn
created branch time in 4 days
push eventwg21link/db
commit sha 2f1c45a50e29e0e8fb1ab89497d6477e8c071156
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha fa0c7f8350ce2211bef027498293a3ba55cbf1e2
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha 92447a73563fb26aed83c2d8bc2fdce50cba2445
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha 28df7edde553e7a48c16dbea9274e36ebc3d8c76
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha 9a4740c83eb6789888db5387b8df59d198acea5a
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha 069b49d76573f802c53486506c4137c51b716d5a
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha 3a0dc5868b3764c4ec7bf0235c506e98dc1f3b9e
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha eb62e7949c97353b284cb76bb91b0111cc3f6742
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha b97d0b81f44106a796fc9b442809c74725ac294c
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha 11f8f0367c1edfeabcb2633c40a90444ecedb86c
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha 9468bd6222ac3e7fe4ac834b6ed7a7ab9174e845
Automatic update.
push time in 5 days
pull request commentrust-lang/rust
Take 2: Add `take_...` functions to slices
The first commit is just the code of the previous PR.
Tip: You can use git commit --author to set the author on the original commit.
comment created time in 5 days
push eventwg21link/db
commit sha a47dd70163c3b4eca4933980315158ac46fb7499
Automatic update.
push time in 5 days
push eventwg21link/db
commit sha afdbbd65eb9a71bbc7ef23ed00d2f6fbba92ecd1
Automatic update.
push time in 6 days
push eventwg21link/db
commit sha 9403d09b17e34af331ce2b7fb1b90520554a2691
Automatic update.
push time in 6 days
push eventwg21link/db
commit sha 94a0e6e53e833116e1a44c34fc967e71c4c7bb5c
Automatic update.
push time in 6 days
push eventwg21link/db
commit sha f45786318552fa835788e0975a842783e596b62e
Automatic update.
push time in 6 days
push eventwg21link/db
commit sha 1a39b84186fcddc49c19f0d97f93faf788659ef9
Automatic update.
push time in 6 days
pull request commentrust-lang/rust
Add cfg(target_has_atomic_equal_alignment) and use it for Atomic::from_mut.
I've added the test. I made it specific for both x86 and linux, because some other x86 platforms do use 64 bit alignment for 64 bit integers:
target u64 alignment
i686-unknown-linux-gnu 32
i586-unknown-linux-gnu 32
i686-unknown-linux-musl 32
i586-unknown-linux-musl 32
i686-linux-android 32
i686-unknown-freebsd 32
i686-unknown-openbsd 32
i686-unknown-netbsd 32
i686-unknown-haiku 32
i686-apple-darwin 32
i686-pc-windows-gnu 64
i686-uwp-windows-gnu 64
i686-pc-windows-msvc 64
i686-uwp-windows-msvc 64
i586-pc-windows-msvc 64
i686-unknown-cloudabi 32
i686-unknown-uefi 64
i686-wrs-vxworks 32
But still, if this test ever happens to run on some x86 Linux target with fully aligned u64s, it'll fail. But that shouldn't affect people cross compiling libcore themselves (-Zbuld-std), as this is a ui test.
@rustbot modify labels: +S-waiting-on-review -S-waiting-on-author
comment created time in 6 days
push eventfusion-engineering-forks/rust
commit sha bcc1d56917fe061071ebc539b5d2e304a44a7b00
Test that AtomicU64::from_mut is not available on x86 linux.
push time in 6 days
push eventfusion-engineering-forks/rust
commit sha c9c35343073b6a389cd1e7118f59c16c229ae962
Test that AtomicU64::from_mut is not available on x86 linux.
push time in 6 days
push eventwg21link/db
commit sha 48d06bdacdea182df4155a8067e7897b8f476bc5
Automatic update.
push time in 6 days
pull request commentrust-lang/rust
Add cfg(target_has_atomic_equal_alignment) and use it for Atomic::from_mut.
Could you add a compile-fail ui test to ensure that
from_mutis not available on x86?
Sure. Working on it.
@rustbot modify labels: -S-waiting-on-review +S-waiting-on-author
comment created time in 6 days
push eventwg21link/db
commit sha 75c376ac575059b5eeb07c576fba23c0df008eaf
Automatic update.
push time in 6 days
Pull request review commentrust-lang/rust
Add cfg(target_has_atomic_equal_alignment) and use it for Atomic::from_mut.
use self::Ordering::*; use crate::cell::UnsafeCell; use crate::fmt; use crate::intrinsics;++#[allow(unused_imports)] use crate::mem::align_of;
Done!
comment created time in 6 days
push eventfusion-engineering-forks/rust
commit sha 5d6f1a1e32de4936e5b269b04781b2737f6d410b
Move `use align_of` in atomic.rs into the places where it is used.
push time in 6 days
push eventfusion-engineering-forks/rust
commit sha 518f1ccb728aa03665e51710c12973a74cc98df5
Stabilize some Result methods as const Stabilize the following methods of `Result` as const: - `is_ok` - `is_err` - `as_ref` Possible because of stabilization of #49146 (Allow if and match in constants).
commit sha 787b2707a77fa1a98f512a4905662a1a16daf13c
Move const tests for `Result` to `library\core` Part of #76268
commit sha 69ffed763d1540c387db3b578848184026da0484
Add error explanation for E0755
commit sha a06edda3ad9abd4f07d07bbe46cb488efeebbbd0
Fix segfault if pthread_getattr_np fails glibc destroys[1] the passed pthread_attr_t if pthread_getattr_np() fails. Destroying it again leads to a segfault. Fix it by only destroying it on success for glibc. [1]: https://sourceware.org/git/?p=glibc.git;a=blob;f=nptl/pthread_getattr_np.c;h=ce437205e41dc05653e435f6188768cccdd91c99;hb=HEAD#l205
commit sha a684153f2920729f9fc3ea27ddb77d7cc3543214
Only call pthread_attr_destroy() after getattr_np() succeeds on all libcs The calling convention of pthread_getattr_np() is to initialize the pthread_attr_t, so _destroy() is only necessary on success (and _init() isn't necessary beforehand). On the other hand, FreeBSD wants the attr_t to be initialized before pthread_attr_get_np(), and therefore it should always be destroyed afterwards.
commit sha fd79ed42254d7483569ccc48ef35a79635ed5881
Add docs for `BasicBlock`
commit sha a872ec47145ce3a82c39bedaca21b99e867d9be2
Clarify how branching works in a CFG
commit sha 176956c115c2b797471a3f59eef3e17789229007
Test and fix Sync & Send traits of BTreeMap artefacts
commit sha c051f61d3c078343db70e355feea808892deed75
Improve wording Co-authored-by: Joshua Nelson <joshua@yottadb.com>
commit sha 21b0c1286a1f01cb8a21bda0973833ca11889364
Extract some intrinsics out of rustc_codegen_llvm A significant amount of intrinsics do not actually need backend-specific behaviors to be implemented, instead relying on methods already in rustc_codegen_ssa. So, extract those methods out to rustc_codegen_ssa, so that each backend doesn't need to reimplement the same code.
commit sha c35ce3ff170333d11ccf89e75dc87c49f44a570a
Don't generate bootstrap usage unless it's needed Previously, `x.py` would unconditionally run `x.py build` to get the help message. After https://github.com/rust-lang/rust/issues/76165, when checking the CI stage was moved into `Config`, that would cause an assertion failure (but only only in CI!): ``` thread 'main' panicked at 'assertion failed: `(left == right)` left: `1`, right: `2`', src/bootstrap/config.rs:619:49 ``` This changes bootstrap to only generate a help message when it needs to (when someone passes `--help`).
commit sha 1d2290ffb0f8f85c8de56c17e6be748d84dcfb04
Update LLVM
commit sha 94dae6004035c356b00dac8764c1b4808d740928
simplfy condition in ItemLowerer::with_trait_impl_ref()
commit sha 57eb29cd2dc30c47b613ced135422a4120bc6a78
Update based on review suggestions
commit sha b7c8bea6cbf0aff4cbb36b127edc0bb4c66c4cf9
Fix wording in mir doc
commit sha f4a7149f499a4b91d296678002f0bce4ded85038
Don't compile regex at every function call. Use `SyncOnceCell` to only compile it once. I believe this still adds some kind of locking mechanism?
commit sha b47913962097874257ad10227b943ef7af85d49f
Remove intrinsics::arith_offset use from libarena The use of arith_offset was added in 803e9ae67b770d8500c4ab5862e988d29118356a before the stable wrapper of the intrinsic was available. https://doc.rust-lang.org/stable/std/intrinsics/fn.arith_offset.html
commit sha 4fe6ca37898bcd65488764f2679e27c936879ade
Replace const_generics feature gate with min_const_generics The latter is on the path to stabilization.
commit sha ebdea011436dfea810ff1bbd827636ac75f6f092
Remove redundant #![feature(...)] 's from compiler/
commit sha 5f58e00ca565c8964dbf89f510236f72951d0bab
fix array_windows docs
push time in 6 days
push eventwg21link/db
commit sha 51b3df2c282b205fc2b77356339c2300884433ab
Automatic update.
push time in 6 days
push eventfusion-engineering-forks/hx711
commit sha 2b6957bba9a67eacabf83e291c1d4e9e95f4f088
Add get_mode function.
push time in 6 days