PyJAGS: The Python Interface to JAGS
tmiasko/fsb-vorbis-extractor 3
FSB Vorbis Extractor
gobject-introspection ABI validator
Empowering everyone to build reliable and efficient software.
PR opened rust-lang/rust
- Move body_owner field from IrMaps to Liveness (the only user of the field).
- Use upvars instead of FnKind to check for closures (avoids FnKind, will be useful when checking all bodies, not just fns).
- Use visit_param to add variables corresponding to params.
- Store upvars_mentioned inside Liveness struct.
- Inline visitor implementation for IrMaps, avoiding unnecessary indirection.
- Test interaction with automatically_derived attribute (not covered by any of existing tests).
No functional changes intended.
Includes #77229.
r? @lcnr
pr created time in 2 hours
push eventtmiasko/rust
commit sha a37886b0cc6dda512768df6ea393ddf62048df9e
liveness: Move body_owner field from IrMaps to Liveness The Liveness struct is the only user of body_owner field. Move the field there.
commit sha 57ad7ad79ba8e5dde8fd5dbbc1e465159a74b292
liveness: Use upvars instead of FnKind to check for closures Avoiding FnKind will make it easier to run liveness analysis on all bodies in the future, not just fn-like things.
commit sha 3a3784e236d95e6f225f7c203b76843a4d7d22d4
liveness: Use visit_param to add variables corresponding to params
commit sha 9ba275603828375424be9f9508f441f024e3b217
liveness: Store upvars_mentioned inside Liveness struct
commit sha 6b73b7b8bc728a83eb85e704cd82cfaea09c4740
liveness: Inline visitor implementation for IrMaps
commit sha c2bb793257a9d66ff84ae5cbcf088c379b00b42a
liveness: Test interaction with automatically_derived attribute
push time in 2 hours
Pull request review commentrust-lang/rust
Small improvements in liveness pass
impl IrMaps<'tcx> { } fn lnk(&self, ln: LiveNode) -> LiveNodeKind {
Done.
comment created time in 11 hours
push eventtmiasko/rust
commit sha d68aa227c6b7ab9c1cfa003e752c3600a7269485
liveness: Access live nodes directly through self.lnks[ln]
push time in 11 hours
Pull request review commentrust-lang/rust
Small improvements in liveness pass
impl IrMaps<'tcx> { } fn add_live_node(&mut self, lnk: LiveNodeKind) -> LiveNode {- let ln = LiveNode(self.num_live_nodes as u32);+ let ln = LiveNode(self.lnks.len() as u32);
I used newtype_index and IndexVec in a few places. I will take a look if INVALID_NODE can be replaced with an Option, especially that newtype_index has a niche for that purpose, but that would be another PR.
comment created time in a day
push eventtmiasko/rust
commit sha 49d1ce00f30a10e607dc30506d3e890d1efb6309
liveness: Remove unnecessary local variable exit_ln
commit sha 57d38975cc638bcdeb7e9bcd10621615e8f9f3c0
liveness: Use newtype_index for Variable and LiveNode
push time in a day
PR opened rust-lang/rust
- Remove redundant debug logging (
add_variablealready contains logging). - Remove redundant fields for a number of live nodes and variables.
- Delay conversion from a symbol to a string until linting.
- Inline contents of specials struct.
No functional changes intended (except those related to the logging).
pr created time in a day
push eventtmiasko/rust
commit sha 9b5835ec7973a52f2e9d6f4ed9a2181bebfdc399
liveness: Remove redundant debug logging The IrMaps::add_variable already contains debug logging. Don't duplicate it.
commit sha 2fb1564457b47bd31087e2aba1b8eb6f15c000ef
liveness: Remove redundant fields for a number of live nodes and variables
commit sha 70f150b51e7d13e3bcd8977ff124a348057cf7ef
liveness: Delay conversion from a symbol to a string until linting
commit sha 141b91da6cb756ff5f36eebe9eee65922e295876
liveness: Inline contents of specials struct
push time in a day
pull request commentrust-lang/rust
Reopen standard file descriptors when they are missing on Unix
The child process spawned in the test case segfaults on arm-android. I don't know why yet, but it is unrelated to the changes in std, since the same issue occurs with them.
comment created time in a day
push eventtmiasko/rust
push time in a day
push eventtmiasko/rust
commit sha 9d974a9b4dcdfbb1838bc32ed59383c577c41412
DO NOT MERGE: Add test case
push time in a day
push eventtmiasko/rust
commit sha 08e356968ca9d9b3fdb7cda9606c47ec232d7a91
DO NOT MERGE: Enable arm-android on GitHub Actions CI
commit sha 3555eccafc9dbea8dce739ee5e2c144b6a1411fc
DO NOT MERGE: Revert "Reopen standard streams when they are closed on Unix" This reverts commit d2c3bb393047ebc5a0d2b746bd9c021955dd2c93.
commit sha 435798d4efbe7361e1a3e72b85bdff804d424ab7
DO NOT MERGE: Add test case
push time in a day
push eventtmiasko/rust
commit sha 8000f51225bebb6301554f476a46d505efa9d934
DO NOT MERGE: Close only stderr open in test
commit sha 029d6d94297016721412d7d138c0d9b751f39a96
DO NOT MERGE: Enable arm-android on GitHub Actions CI
push time in a day
push eventtmiasko/rust
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 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 f9fcbbea03edb735c22311522b55d7b854bd6ac0
fixed bug
commit sha e49a29933be3bd988ccb75b053f480d9c99a7ff5
Working map_err_ignore lint
commit sha 202a80c927412a548180eca5990ee764d76ed643
Added tests for map_err, ignored map_err lint on drop_ref tests
commit sha 337729137bdec31b55665653ef0cf7dc124b0eaa
Ran cargo dev update_lints
commit sha 2387f68e437bf2ff5f117f63936257ce64052cfa
Removed map_err suggestion in lint, and updated lint documentation example
commit sha c31d4730b0d40c62934839405d0c25e2ffa3fad1
update example to be more idiomatic
commit sha 4f1c4a99d4d98f1acea3c9c7cc55355aa46119aa
Fixed typo in lint and test
push time in a day
Pull request review commentrust-lang/rust
Add Linux-specific pidfd process extensions
+//! Linux-specific extensions to primitives in the `std::process` module.++#![unstable(feature = "linux_pidfd", issue = "none")]++use crate::process;+use crate::sys_common::AsInnerMut;+use crate::io::Result;++/// Os-specific extensions to [`process::Child`]+///+/// [`process::Child`]: crate::process::Child+pub trait ChildExt {+ /// Obtains the pidfd created for this child process, if available.+ ///+ /// A pidfd will only ever be available if `create_pidfd(true)` was called+ /// when the corresponding `Command` was created.+ ///+ /// Even if `create_pidfd(true)` is called, a pidfd may not be available+ /// due to an older version of Linux being in use, or if+ /// some other error occured.+ ///+ /// See `man pidfd_open` for more details about pidfds.+ fn pidfd(&self) -> Result<i32>;+}++/// Os-specific extensions to [`process::Command`]+///+/// [`process::Command`]: crate::process::Command+pub trait CommandExt {+ /// Sets whether or this `Command` will attempt to create a pidfd+ /// for the child. If this method is never called, a pidfd will+ /// not be crated.+ ///+ /// The pidfd can be retrieved from the child via [`ChildExt::pidfd`]+ ///+ /// A pidfd will only be created if it is possible to do so+ /// in a guaranteed race-free manner (e.g. if the `clone3` system call is+ /// supported). Otherwise, [`ChildExit::pidfd`] will return an error.+ fn create_pidfd(&mut self, val: bool) -> &mut process::Command;
The pidfd only exists in the parent process. Once clone3 succeeds, the pidfd is perfectly useable, regardless of what happens in the child process from that point on.
My point was that spawn returns an error in that case, so the user cannot access pidfd.
comment created time in 2 days
Pull request review commentrust-lang/rust
Add Linux-specific pidfd process extensions
+//! Linux-specific extensions to primitives in the `std::process` module.++#![unstable(feature = "linux_pidfd", issue = "none")]++use crate::process;+use crate::sys_common::AsInnerMut;+use crate::io::Result;++/// Os-specific extensions to [`process::Child`]+///+/// [`process::Child`]: crate::process::Child+pub trait ChildExt {+ /// Obtains the pidfd created for this child process, if available.+ ///+ /// A pidfd will only ever be available if `create_pidfd(true)` was called+ /// when the corresponding `Command` was created.+ ///+ /// Even if `create_pidfd(true)` is called, a pidfd may not be available+ /// due to an older version of Linux being in use, or if+ /// some other error occured.+ ///+ /// See `man pidfd_open` for more details about pidfds.+ fn pidfd(&self) -> Result<i32>;+}++/// Os-specific extensions to [`process::Command`]+///+/// [`process::Command`]: crate::process::Command+pub trait CommandExt {+ /// Sets whether or this `Command` will attempt to create a pidfd+ /// for the child. If this method is never called, a pidfd will+ /// not be crated.+ ///+ /// The pidfd can be retrieved from the child via [`ChildExt::pidfd`]+ ///+ /// A pidfd will only be created if it is possible to do so+ /// in a guaranteed race-free manner (e.g. if the `clone3` system call is+ /// supported). Otherwise, [`ChildExit::pidfd`] will return an error.+ fn create_pidfd(&mut self, val: bool) -> &mut process::Command;
- Without seeing this discussion, I would presume that Process owns the pidfd and will close it on drop.
- The pidfd should be closed when clone3 succeeds but exec fails, since a user has no ability to do so on their own, right?
comment created time in 2 days
Pull request review commentrust-lang/rust
Reopen standard file descriptors when they are missing on Unix
pub fn init() { reset_sigpipe(); } + // In the case when all file descriptors are open, the poll has been+ // observed to perform better than fcntl (on GNU/Linux).+ #[cfg(not(any(+ target_os = "emscripten",+ target_os = "fuchsia",+ // The poll on Darwin doesn't set POLLNVAL for closed fds.+ target_os = "macos",+ target_os = "ios",+ target_os = "redox",+ )))]+ unsafe fn sanitize_standard_fds() {+ use crate::sys::os::errno;+ let pfds: &mut [_] = &mut [+ libc::pollfd { fd: 0, events: 0, revents: 0 },+ libc::pollfd { fd: 1, events: 0, revents: 0 },+ libc::pollfd { fd: 2, events: 0, revents: 0 },+ ];+ while libc::poll(pfds.as_mut_ptr(), 3, 0) == -1 {
Miri should use empty implementation now.
comment created time in 2 days
push eventtmiasko/rust
commit sha 257f6e5f745c25b66e432f5222d1b746469fcdb0
Reopen standard streams when they are closed on Unix The syscalls returning a new file descriptors generally use lowest-numbered file descriptor not currently opened, without any exceptions for those corresponding to the standard streams. Previously when any of standard streams has been closed before starting the application, operations on std::io::{stderr,stdin,stdout} objects were likely to operate on other logically unrelated file resources opened afterwards. Avoid the issue by reopening the standard streams when they are closed.
push time in 2 days
Pull request review commentrust-lang/rust
Reopen standard file descriptors when they are missing on Unix
pub fn init() { reset_sigpipe(); } + // In the case when all file descriptors are open, the poll has been+ // observed to perform better than fcntl (on GNU/Linux).+ #[cfg(not(any(+ target_os = "emscripten",+ target_os = "fuchsia",+ // The poll on Darwin doesn't set POLLNVAL for closed fds.+ target_os = "macos",+ target_os = "ios",+ target_os = "redox",+ )))]+ unsafe fn sanitize_standard_fds() {+ use crate::sys::os::errno;+ let pfds: &mut [_] = &mut [+ libc::pollfd { fd: 0, events: 0, revents: 0 },+ libc::pollfd { fd: 1, events: 0, revents: 0 },+ libc::pollfd { fd: 2, events: 0, revents: 0 },+ ];+ while libc::poll(pfds.as_mut_ptr(), 3, 0) == -1 {
Which implementation would you like to use: fcntl(fd, F_GETFD) and open("/dev/null"), or empty one?
comment created time in 2 days
pull request commentrust-lang/rust
Reopen standard file descriptors when they are missing on Unix
I added a comment, but also revised it shortly afterwards so it might need another r+.
comment created time in 3 days
push eventtmiasko/rust
commit sha 44e083ec99ec904b170af9795d0112165553dedb
Reopen standard streams when they are closed on Unix The syscalls returning a new file descriptors generally use lowest-numbered file descriptor not currently opened, without any exceptions for those corresponding to the standard streams. Previously when any of standard streams has been closed before starting the application, operations on std::io::{stderr,stdin,stdout} objects were likely to operate on other logically unrelated file resources opened afterwards. Avoid the issue by reopening the standard streams when they are closed.
push time in 3 days
push eventtmiasko/rust
commit sha d6d59b17b6795531a70dd60cec69c9292ea1237f
Reopen standard streams when they are closed on Unix The syscalls returning a new file descriptors generally use lowest-numbered file descriptor not currently opened, without any exceptions for those corresponding to the standard streams. Previously when any of standard streams has been closed before starting the application, operations on std::io::{stderr,stdin,stdout} objects were likely to operate on other logically unrelated file resources opened afterwards. Avoid the issue by reopening the standard streams when they are closed.
push time in 3 days
pull request commentrust-lang/rust
Reopen standard file descriptors when they are missing on Unix
I changed the implementation to abort the process when reopening fails. In case of any bug reports with regards to this behaviour, I will revise the implementation to continue execution with Rust objects disabled.
comment created time in 3 days
push eventtmiasko/rust
commit sha c5975e9b6c5781b3b7300b7921c14b060086e1c1
Reduce duplicate in liballoc reserve error handling
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 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 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 7c1e5c1dcd25c945f619eda289f639dbe2b002da
Update OpenOptions::as_flags docs, and minor styling
commit sha 321b680fe66d1be04cd67fac75ff7f148fd117fe
Update docs of OpenOptions::as_flags
commit sha 28db5214d2c48e7a58cf79b9e272097260910a33
More implementations of Write for immutable refs Fixes #73836
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 ec7f9b927f1896c7f29c602d6b0f961c891d0941
Deduplicates io::Write implementations
commit sha 8f27e3cb1b4140d9124d60df0850ef734e73b884
Make some methods of `Pin` unstable const Make the following methods unstable const under the `const_pin` feature: - `new` - `new_unchecked` - `into_inner` - `into_inner_unchecked` - `get_ref` - `into_ref` Also adds tests for these methods in a const context. Tracking issue: #76654
commit sha e5447a22222ecc6a650e75282cb9931b910854b2
Fix #76432 Only insert StorageDeads if we actually removed one. Fixes an issue where we added StorageDead to a place with no StorageLive
commit sha c100e726c10fe9d212a58a5494126409e7d0e10b
Stabilize intra-doc links
commit sha 63d5beec43ff7721928821cd83f9790188b03276
Move intra-doc-links documentation out of unstable section
commit sha bc06674774e6457046e41a48dc3e8be8c5496f11
Mention super/crate/self in docs
commit sha f072e4a7322e8e2b16410e3225e1afc15d132e36
Mention URL fragments
push time in 3 days
pull request commentrust-lang/rust
Reopen standard file descriptors when they are missing on Unix
I am motivated by the cases where we can actually reopen them. I don't have strong opinion what we should do if that fails. Simply continuing or aborting, both seems fine to me.
comment created time in 3 days
pull request commentrust-lang/rust
Reopen standard file descriptors when they are missing on Unix
I implemented variant that detected closed fds and disabled corresponding Rust objects (without reopening them). That approached turned out to be insufficient to avoid problems while using various Rust applications in practice. I don't think it is worth the extra complexity.
I would again suggest the approach already taken in this PR.
comment created time in 3 days
Pull request review commentrust-lang/rust
Add regression test for issue #76042
+// run-pass
I would expect that this that would have a few extra compile flags. compile-flags: -Coverflow-checks=off -Ccodegen-units=1 -Copt-level=0 with --target i686-unknown-linux-gnu also being relevant but omitted.
comment created time in 3 days
Pull request review commentrust-lang/rust
MIR pass to remove unneeded drops on types not needing drop
+//! This pass replaces a drop of a type that does not need dropping, with a goto++use crate::transform::{MirPass, MirSource};+use rustc_hir::def_id::LocalDefId;+use rustc_middle::mir::visit::Visitor;+use rustc_middle::mir::*;+use rustc_middle::ty::TyCtxt;++use super::simplify::simplify_cfg;++pub struct RemoveUnneededDrops;++impl<'tcx> MirPass<'tcx> for RemoveUnneededDrops {+ fn run_pass(&self, tcx: TyCtxt<'tcx>, source: MirSource<'tcx>, body: &mut Body<'tcx>) {+ trace!("Running RemoveUnneededDrops on {:?}", source);+ let mut opt_finder = RemoveUnneededDropsOptimizationFinder {+ tcx,+ body,+ optimizations: vec![],+ def_id: source.def_id().expect_local(),+ };+ opt_finder.visit_body(body);+ let should_simplify = !opt_finder.optimizations.is_empty();+ for (loc, target) in opt_finder.optimizations {+ let terminator = body.basic_blocks_mut()[loc.block].terminator_mut();+ debug!("SUCCESS: replacing `drop` with goto({:?})", target);+ terminator.kind = TerminatorKind::Goto { target };+ }++ // if we applied optimizations, we potentially have some cfg to cleanup to+ // make it easier for further passes+ if should_simplify {+ simplify_cfg(body);+ }+ }+}++impl<'a, 'tcx> Visitor<'tcx> for RemoveUnneededDropsOptimizationFinder<'a, 'tcx> {+ fn visit_terminator(&mut self, terminator: &Terminator<'tcx>, location: Location) {+ match terminator.kind {+ TerminatorKind::Drop { place, target, .. }+ | TerminatorKind::DropAndReplace { place, target, .. } => {
The DropAndReplace terminator should never occur in an input to this pass, since it should be removed during drop elaboration. Additionally, its semantics is to drop and assign a new value, so the transformation wouldn't be correct without additional assignment. Either way, I would just remove it from here.
comment created time in 3 days
push eventtmiasko/rust
commit sha c100e726c10fe9d212a58a5494126409e7d0e10b
Stabilize intra-doc links
commit sha 63d5beec43ff7721928821cd83f9790188b03276
Move intra-doc-links documentation out of unstable section
commit sha bc06674774e6457046e41a48dc3e8be8c5496f11
Mention super/crate/self in docs
commit sha f072e4a7322e8e2b16410e3225e1afc15d132e36
Mention URL fragments
commit sha 2a98409634ec38547d03512898192b5bdce15f3d
Fill out docs on intra-doc resolution failure lint
commit sha 4e0eb0b73ba2defaf0f3c2152e9daa5c18d1603f
Update src/doc/rustdoc/src/linking-to-items-by-name.md Co-authored-by: Joshua Nelson <joshua@yottadb.com>
commit sha 175e30539d5ee9364a26f89f8a03a60b53690684
Update src/doc/rustdoc/src/linking-to-items-by-name.md Co-authored-by: Joshua Nelson <joshua@yottadb.com>
commit sha 51c1351f7b9366a295dce5d63c2fd387170d5c34
Resolve some conflicts
commit sha 6f1fa2b16376f586285453045788519df527a1b2
Fix lint name in docs
commit sha 792b2ea5812a06fe3d362ae36025998af66199ee
Update src/doc/rustdoc/src/lints.md Co-authored-by: Joshua Nelson <joshua@yottadb.com>
commit sha ff1a9e406bfccefedb6c4f2cabc0c4d59ac947d4
Fix underflow when calculating the number of no-op jumps folded When removing unwinds to no-op blocks and folding jumps to no-op blocks, remove the unwind target first. Otherwise we cannot determine if target has been already folded or not. Previous implementation incorrectly assumed that all resume targets had been folded already, occasionally resulting in an underflow: remove_noop_landing_pads: removed 18446744073709551613 jumps and 3 landing pads
commit sha 6928041c0a992093a7752ae3c04090caebcd4515
Update src/doc/rustdoc/src/linking-to-items-by-name.md Co-authored-by: Joshua Nelson <joshua@yottadb.com>
commit sha f05b47ccdfa63f8b4b9fb47a9aa92381801d3ff1
Don't download/sync llvm-project submodule if download-ci-llvm is set llvm-project takes > 1GB storage space and a long time to download. It's better to not download it unless needed.
commit sha d4039c55c9ef392261aeaba6c14ae81f5098139a
wip emit errors during AbstractConst building
commit sha 30cbc9729654c66b9f54779d18f023e47112cdf2
words
commit sha 16047d46a10813a25e956e659b8861cc67c20cd2
fix typo in docs and comments
commit sha db74e1f1e3e5c7497bb9af68814497c8df1fd0b8
Add cfg(target_has_atomic_equal_alignment). This is needed for Atomic::from_mut.
commit sha 9e3f94dabc334586d669f3aba0bded607ac782fc
Don't unwrap but report a fatal error for TargetDataLayout::parse.
commit sha af56ad7633668a790353f17440586e97b4e6a99b
Add feature gate ui test for cfg(target_has_atomic_load_store).
commit sha 54fdf54e06e3287958d906d2fb1b905b778d5e38
Add feature gate ui test for cfg(target_has_atomic_equal_alignment).
push time in 3 days
pull request commentrust-lang/rust
Reopen standard file descriptors when they are missing on Unix
OK, I see your point. In that case it should be fine to attempt reopening them, but disable when that fails? I generally feel that anything short of reopening them will likely break any real application, although admittedly it will involve unsafe code.
Hmm, it might be also possible go back to earlier suggestion of using pipes, with a slight twist where they are connected wrong way around, so that read / write operations return EBADF. Though, I couldn't really tell what other consequences it could have.
comment created time in 4 days
pull request commentrust-lang/rust
Reopen standard file descriptors when they are missing on Unix
Check if stdin, stdout, stderr are opened and provide no-op substitutes in std::io::{stdin,stdout,stderr} without reopening them directly.
I feel that this is a better solution if it is feasible.
It is definitely possible to detect closed descriptors using the same approach and disable corresponding std::io::{stderr,stdin,stdout} objects. It should avoid issues as far as access through those objects is concerned, but anything that uses file descriptors directly will still cause trouble. While an improvement, to me that still feel like incomplete solution of the problem.
@Amanieu I wonder why would you prefer that approach?
comment created time in 4 days
push eventtmiasko/rust
commit sha c5975e9b6c5781b3b7300b7921c14b060086e1c1
Reduce duplicate in liballoc reserve error handling
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 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 7c1e5c1dcd25c945f619eda289f639dbe2b002da
Update OpenOptions::as_flags docs, and minor styling
commit sha 321b680fe66d1be04cd67fac75ff7f148fd117fe
Update docs of OpenOptions::as_flags
commit sha 28db5214d2c48e7a58cf79b9e272097260910a33
More implementations of Write for immutable refs Fixes #73836
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 ec7f9b927f1896c7f29c602d6b0f961c891d0941
Deduplicates io::Write implementations
commit sha 8f27e3cb1b4140d9124d60df0850ef734e73b884
Make some methods of `Pin` unstable const Make the following methods unstable const under the `const_pin` feature: - `new` - `new_unchecked` - `into_inner` - `into_inner_unchecked` - `get_ref` - `into_ref` Also adds tests for these methods in a const context. Tracking issue: #76654
commit sha e5447a22222ecc6a650e75282cb9931b910854b2
Fix #76432 Only insert StorageDeads if we actually removed one. Fixes an issue where we added StorageDead to a place with no StorageLive
commit sha eede953c283c7bbe903a0e8abb44c923baf5cfac
Only get ImplKind::Impl once
commit sha 70292d45060ee2b8829e2af0be54d5a76696cd1d
Sets as_flags as unstable
commit sha 1d2290ffb0f8f85c8de56c17e6be748d84dcfb04
Update LLVM
commit sha ed6c7efd87f17a7d9282f4bc7341cb5cbda8db4d
Use enum for status of non-const ops
commit sha c3607bd7dd323a898b8cc9d2c603dfd14172c73c
Use helper function for searching `allow_internal_unstable`
commit sha e4edc161f20ccbc77fffb7ceb60ebb6102cbd747
Give name to extra `Span` in `LiveDrop` error
push time in 4 days
PR closed rust-lang/cargo
Add a new --kinds option to cargo doc, which allows to select what
kinds of dependencies should be documented: build dependencies, dev
dependencies, or normal dependencies. Multiple kinds of dependencies
can be specified, e.g., cargo doc --kinds=dev,normal.
When --kinds options is used, the default behaviour of documenting
normal dependencies is suppressed unless asked for explicitly with
--kinds=normal. The --kinds option also conflicts with --no-deps.
Follow up on proposal from #7077. Resolves #3475.
Open questions:
- Should
--kindsdocument normal dependencies by default, obviating the need for normal kind of dependencies? Current answer is no. - Should
--kindsinteract with--no-depsflag? Current answer is no, but presence or absence of--no-depscould control if dependencies of selected dependencies are also documented.
pr closed time in 5 days
pull request commentrust-lang/cargo
Document build dependencies and dev dependencies
After looking at this a bit more, it seems to me it will require completely different approach and more substantial changes. Thanks for reviewing this so far. I will close this for now.
comment created time in 5 days
pull request commentrust-lang/rust
unix: Extend UnixStream and UnixDatagram to send and receive file descriptors
@LinkTed cargo check might be feasible with something like env CC_x86_64-apple-darwin=/usr/bin/true ./x.py check --target x86_64-apple-darwin
comment created time in 5 days
issue commentrust-lang/rust
Miscompilation with mir-opt-level=2
The propagation is disabled only if both locals have their address taken. Changing that to be more conservative does fix this issue.
Minimized:
type M = [i64; 2];
fn f(a: &M) -> M {
let mut b: M = M::default();
b[0] = a[0] * a[0];
b
}
fn main() {
let mut a: M = [1, 1];
a = f(&a);
std::process::exit((a[0] != 1) as i32);
}
comment created time in 5 days
push eventtmiasko/cargo
commit sha 02fc16aece46abcd23d2ade2d969497f07fe26ab
Fix jobserver_exists test on single-cpu systems Closes #8595
commit sha 916b392cdbe4702c7b8f83e4f970cca7d722c553
doc: Qualify GNU licenses in example license field Since SPDX License list 3.0, unqualified GNU licenses are deprecated: https://www.gnu.org/licenses/identify-licenses-clearly.html We use version 3.6 of the list so we should not use the deprecated licenses.
commit sha ff7d328149878aec48f4b6366ada62f58eadf5dd
Fix small typo in reference/profiles.md
commit sha 3bf5ad8ca9d6255b153242ca6d270fe398933ef0
Auto merge of #8605 - louib:patch-1, r=ehuss Fix small typo in reference/profiles.md
commit sha 5305a3def63603b7e9d03233f216924ca38e2465
Fix typo Co-authored-by: Teymour Aldridge <42674621+teymour-aldridge@users.noreply.github.com>
commit sha 5de5915ca281b88cc3b8394823261ab33c2f9ab6
Auto merge of #8598 - alexcrichton:fix-test, r=ehuss Fix jobserver_exists test on single-cpu systems Closes #8595
commit sha 20d35ffeb6cea95aecae1ec85cdf36df6944f463
Auto merge of #8604 - jtojnar:patch-1, r=ehuss doc: Qualify GNU licenses in example license field Since SPDX License list 3.0, unqualified GNU licenses are deprecated: https://www.gnu.org/licenses/identify-licenses-clearly.html We use version 3.6 of the list so we should not use the deprecated licenses.
commit sha 624acc8e2be3cda315772c388feabf6bc6f69d2e
Build manpage archive deterministically Keep deterministic builds for Cargo! The changes here are: * Sort files being added to the archive to ensure they're added in the same order on all platforms. * Flag the archive builder as "deterministic mode" which means it won't pick up fields like mtime. Closes #8599
commit sha ab32ee88dade1b50c77347599e82ca2de3fb8a51
Auto merge of #8600 - alexcrichton:deterministic, r=ehuss Build manpage archive deterministically Keep deterministic builds for Cargo! Closes #8599
commit sha fd6a84aa83e6ce4431263a361bb52df12ddaa9dc
Add chapters on dependency resolution and SemVer compatibility.
commit sha 4d023b1ec5ffce5951d56182cd054bb22a69c02b
Don't check src/doc directory.
commit sha 1ee38a326e56723b54da2adb8dd7d9af43fae744
Try to fix CI for semver-check.
commit sha 006023c5827b2288230df51676b69b725b822e86
Update from review comments.
commit sha 79f351297501b16d3396b6c3b0bdb77a83d8958d
Renames SourceId::into_url -> SourceId::as_url While studying the source code, I am surprised to see that `into_url` does not actually consume the string when a function with such name usually does. Additionally, there are traits in `cargo::util::IntoUrl` with the same and does exactly what you expect, consumes T and yields a `Url`. I hope this is not too nitpicky. Thank you!
commit sha 3dbac42a57ae8914a9fec765276acac33360305d
Renamed SourceIdIntoUrl -> SourceIdAsUrl
commit sha ea0709e88a2553a130c5dd6189af6f5358db1e18
Fix typo in SIGQUIT description
commit sha 4de00d2c7fdd961e72eb189a31f8ae677833ed99
Auto merge of #8615 - davepacheco:sigquit-typo, r=ehuss Fix typo in SIGQUIT description This PR fixes a small typo in the SIGQUIT description. (edit: removed my notes about automated tests because the PR bot has a more complete summary.) In terms of manual testing: I had a test that raised `SIGQUIT` using the "coredump" crate. Previously, it emitted: ``` error: test failed, to rerun pass '--test the_test' Caused by: process didn't exit successfully: `/Users/dap/oxide/experiments/rust-coredump/target/debug/deps/the_test-a8a068307be90649 --nocapture` (signal: 3, SIGQUIT: terminal quite signal) ``` Now, it emits: ``` error: test failed, to rerun pass '--test the_test' Caused by: process didn't exit successfully: `/Users/dap/oxide/experiments/rust-coredump/target/debug/deps/the_test-20c8e90133861ece --nocapture` (signal: 3, SIGQUIT: terminal quit signal) ``` which looks correct.
commit sha 18bc90cd0ed28701e2b64ecfe8ad921ffc80320b
Show full error context on `cargo run` error.
commit sha a9a154f783fde8dad9f4ed437fc3e42f7cfe1b63
Fix bug with PathAndArg config values
commit sha ddc7090f495afaff48b39ccc74036a1e75c6368c
Remove debug only code
push time in 6 days
issue commentrust-lang/rust
[AVR] SIGSEGV on release build of core crate
llc: llvm/lib/CodeGen/LiveIntervals.cpp:536: bool llvm::LiveIntervals::computeDeadValues(llvm::LiveInterval&, llvm::SmallVectorImpl<llvm::MachineInstr*>*): Assertion `MI && "No instruction defining live value"' failed.
<details><summary>LLVM IR</summary>
; ModuleID = 'bugpoint-reduced-simplified.bc'
source_filename = "llvm-link"
target datalayout = "e-P1-p:16:8-i8:8-i16:8-i32:8-i64:8-f32:8-f64:8-n8-a:8"
target triple = "avr-unknown-unknown"
%0 = type { [0 x i8], i16, [0 x i8], [40 x i32], [0 x i8] }
; Function Attrs: argmemonly nounwind willreturn
declare void @llvm.memcpy.p0i8.p0i8.i16(i8* noalias nocapture writeonly, i8* noalias nocapture readonly, i16, i1 immarg) addrspace(1) #0
declare void @_ZN4core3num6bignum8Big32x408mul_pow217h8ca9be035823d302E() unnamed_addr addrspace(1) #1
define void @_ZN4core3num7flt2dec8strategy6dragon15format_shortest17h6b1eacf2bca21fe3E(i16 %0) unnamed_addr addrspace(1) #1 {
%2 = alloca %0, align 1
%3 = alloca %0, align 1
br label %4
4: ; preds = %1
%5 = load i8, i8* undef, align 1
br label %6
6: ; preds = %4
%7 = getelementptr inbounds %0, %0* %3, i16 0, i32 3
%8 = bitcast [40 x i32]* %7 to i8*
br label %9
9: ; preds = %6
%10 = icmp ult i16 undef, 40
br i1 %10, label %11, label %16
11: ; preds = %9
store i16 1, i16* undef, align 1
br label %12
12: ; preds = %11
call addrspace(1) void @llvm.memcpy.p0i8.p0i8.i16(i8* nonnull align 1 dereferenceable(160) undef, i8* nonnull align 1 dereferenceable(160) %8, i16 160, i1 false) #2
br i1 undef, label %16, label %13
13: ; preds = %12
%14 = icmp ugt i16 undef, 40
br i1 %14, label %16, label %15
15: ; preds = %13
br i1 undef, label %17, label %16
16: ; preds = %15, %13, %12, %9
unreachable
17: ; preds = %15
br i1 undef, label %21, label %18
18: ; preds = %17
br i1 undef, label %20, label %19
19: ; preds = %18
unreachable
20: ; preds = %18
br i1 undef, label %21, label %22
21: ; preds = %20, %17
unreachable
22: ; preds = %20
br i1 undef, label %36, label %24
23: ; preds = %24
br i1 undef, label %36, label %35
24: ; preds = %24, %22
%25 = phi i32 [ %32, %24 ], [ 0, %22 ]
%26 = load i32, i32* undef, align 1
%27 = zext i32 %26 to i64
%28 = mul nuw nsw i64 %27, 10
%29 = zext i32 %25 to i64
%30 = add nuw nsw i64 %28, %29
%31 = lshr i64 %30, 32
%32 = trunc i64 %31 to i32
%33 = trunc i64 %30 to i32
%34 = icmp eq i32* undef, undef
br i1 %34, label %23, label %24
35: ; preds = %23
br i1 undef, label %36, label %36
36: ; preds = %35, %35, %23, %22
call align 1 addrspace(1) void @_ZN4core3num6bignum8Big32x408mul_pow217h8ca9be035823d302E()
br i1 undef, label %39, label %37
37: ; preds = %36
%38 = load i16, i16* undef, align 1
br label %40
39: ; preds = %36
unreachable
40: ; preds = %128, %37
%41 = phi i16 [ 0, %37 ], [ %80, %128 ]
br label %42
42: ; preds = %43, %40
br i1 undef, label %44, label %43
43: ; preds = %42
br i1 false, label %42, label %44
44: ; preds = %43, %42
%45 = select i1 undef, i16 undef, i16 undef
%46 = icmp ugt i16 %45, 40
br i1 %46, label %47, label %48
47: ; preds = %44
unreachable
48: ; preds = %44
%49 = getelementptr inbounds %0, %0* %3, i16 0, i32 3, i16 %45
%50 = bitcast i32* %49 to i8*
br label %51
51: ; preds = %57, %48
%52 = phi i8* [ %50, %48 ], [ undef, %57 ]
%53 = phi i8* [ undef, %48 ], [ %58, %57 ]
%54 = icmp eq i8* %52, %8
br i1 %54, label %55, label %57
55: ; preds = %51
%56 = icmp ne i8* %53, undef
br label %60
57: ; preds = %51
%58 = getelementptr inbounds i8, i8* %53, i16 -4
%59 = icmp eq i8 undef, 0
br i1 %59, label %51, label %60
60: ; preds = %57, %55
br i1 undef, label %61, label %68
61: ; preds = %60
br i1 undef, label %67, label %62
62: ; preds = %62, %61
%63 = or i1 undef, undef
store i32 0, i32* undef, align 1
%64 = icmp ult i16 undef, %45
br i1 %64, label %62, label %65
65: ; preds = %62
br i1 %63, label %67, label %66
66: ; preds = %65
unreachable
67: ; preds = %65, %61
br label %68
68: ; preds = %67, %60
br i1 false, label %69, label %70
69: ; preds = %68
unreachable
70: ; preds = %68
%71 = select i1 undef, i16 undef, i16 %38
br i1 false, label %72, label %73
72: ; preds = %70
unreachable
73: ; preds = %70
br i1 undef, label %74, label %77
74: ; preds = %73
br i1 undef, label %77, label %75
75: ; preds = %75, %74
%76 = icmp ult i16 undef, %71
br i1 %76, label %75, label %77
77: ; preds = %75, %74, %73
%78 = icmp ult i16 %41, %0
br i1 %78, label %79, label %133
79: ; preds = %77
%80 = add nuw i16 %41, 1
%81 = load i16, i16* undef, align 1
%82 = icmp ugt i16 undef, 40
br i1 %82, label %83, label %84
83: ; preds = %79
unreachable
84: ; preds = %79
br label %85
85: ; preds = %87, %84
br i1 undef, label %86, label %87
86: ; preds = %85
br label %88
87: ; preds = %85
br i1 undef, label %85, label %88
88: ; preds = %87, %86
%89 = icmp slt i8 undef, %5
%90 = load i16, i16* undef, align 1
%91 = select i1 undef, i16 undef, i16 %90
%92 = icmp ugt i16 %91, 40
br i1 %92, label %93, label %94
93: ; preds = %102, %98, %88
unreachable
94: ; preds = %88
%95 = getelementptr inbounds %0, %0* %2, i16 0, i32 3, i16 %91
br i1 false, label %102, label %96
96: ; preds = %96, %94
br i1 undef, label %96, label %97
97: ; preds = %96
br i1 undef, label %98, label %102
98: ; preds = %97
%99 = icmp ult i16 %91, 40
br i1 %99, label %100, label %93
100: ; preds = %98
store i32 1, i32* %95, align 1
%101 = add nuw nsw i16 %91, 1
br label %102
102: ; preds = %100, %97, %94
%103 = phi i16 [ %101, %100 ], [ %91, %97 ], [ 0, %94 ]
br i1 false, label %93, label %104
104: ; preds = %102
%105 = getelementptr inbounds %0, %0* %2, i16 0, i32 3, i16 undef
%106 = bitcast i32* %105 to i8*
br label %107
107: ; preds = %110, %104
%108 = phi i8* [ %106, %104 ], [ %111, %110 ]
br i1 false, label %109, label %110
109: ; preds = %107
br label %114
110: ; preds = %107
%111 = getelementptr inbounds i8, i8* %108, i16 -4
%112 = bitcast i8* %111 to i32*
%113 = icmp eq i8 undef, 0
br i1 %113, label %107, label %114
114: ; preds = %110, %109
%115 = icmp slt i8 undef, %5
%116 = or i1 %89, %115
br i1 %116, label %129, label %117
117: ; preds = %114
%118 = icmp ugt i16 undef, 40
br i1 %118, label %120, label %119
119: ; preds = %117
br i1 undef, label %123, label %121
120: ; preds = %117
unreachable
121: ; preds = %119
br i1 undef, label %123, label %122
122: ; preds = %121
unreachable
123: ; preds = %121, %119
%124 = icmp ugt i16 %81, 40
br i1 %124, label %125, label %126
125: ; preds = %123
unreachable
126: ; preds = %123
%127 = icmp ugt i16 %90, 40
br i1 %127, label %133, label %128
128: ; preds = %126
br i1 false, label %130, label %40
129: ; preds = %114
br i1 %115, label %130, label %132
130: ; preds = %129, %128
br i1 %89, label %131, label %132
131: ; preds = %130
unreachable
132: ; preds = %130, %129
ret void
133: ; preds = %126, %77
unreachable
}
attributes #0 = { argmemonly nounwind willreturn }
attributes #1 = { "target-cpu"="atmega328" }
attributes #2 = { nounwind }
!llvm.module.flags = !{!0}
!0 = !{i32 7, !"PIC Level", i32 2}
</details>
comment created time in 6 days
PR opened rust-lang/rust
Allocating from the end of memory chunk simplifies the alignment code and reduces the number of checked arithmetic operations.
r? @Mark-Simulacrum
pr created time in 6 days
push eventtmiasko/rust
commit sha 73e27b3e18dcbbef3a36620c4a44306e2bbdcd13
deny(unsafe_op_in_unsafe_fn) in libstd/process.rs
commit sha 00d537dcd03f9ff5ebdf8b86e039dbdb0a7f850c
deny(unsafe_op_in_unsafe_fn) in libstd/path.rs
commit sha a7468705cbf0fb551b8b1d8b420123262f7d92b2
Use translated variable for test string Test should be educative, added english translation and pronounciation.
commit sha ba4c4988161abbe58e973b792c7e271785b4bc4d
Add more info for Vec Drain doc See its documentation for more
commit sha 71484121001b69aefdc41fd7192b7095250517a7
Vec slice example fix style and show type elision
commit sha 0f2bd56b29857453835e47abbe96a80b175632d1
lint/ty: move fns to avoid abstraction violation This commit moves `transparent_newtype_field` and `is_zst` to `LateContext` where they are used, rather than being on the `VariantDef` and `TyS` types. Signed-off-by: David Wood <david@davidtw.co>
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 ea5dc0909ea1ed4135f1bdecbaa3423051be578d
Make some Ordering methods const Constify the following methods of `core::cmp::Ordering`: - `reverse` - `then` Stabilizes these methods as const under the `const_ordering` feature. Also adds a test for these methods in a const context. Possible because of #49146 (Allow `if` and `match` in constants).
commit sha a80d39041e2d5cd58a846c9ef9e01ee9d691a7ed
Use inline(never) instead of cold inline(never) is better way to avoid optimizer to inline the function instead of cold.
commit sha 787b2707a77fa1a98f512a4905662a1a16daf13c
Move const tests for `Result` to `library\core` Part of #76268
commit sha 79d563c819483eaf6e67b6aaaef9d0ca6030337d
Move const tests for `Ordering` to `library\core` Part of #76268
commit sha b54386ab7a9da8a4f22db3a35a9ec7b0f2b98b6c
Detect overflow in proc_macro_server subspan
commit sha d98bac4e4e3cc87ec9b848c173d570ebe2aa30b6
Add tests for overflow in Vec::drain
commit sha f8cfb2f5ad847b871399dfef9b8b8ff4e84a75cb
Add tests for overflow in String / VecDeque operations using ranges
commit sha 6092828d1f432bb313818e7cfab961c0e494f69e
Add `[T; N]: TryFrom<Vec<T>>` This is very similar to the existing `Box<[T; N]>: TryFrom<Box<[T]>>`, but allows avoiding the `shrink_to_fit` if you have a vector and not a boxed slice.
commit sha f8376b59d1391a5191a561c44067355f1a99c7c0
shim: monomorphic `FnPtrShim`s during construction This commit adjusts MIR shim construction so that substitutions are applied to function pointer shims during construction, rather than during codegen (as determined by `substs_for_mir_body`) - as substitutions will no longer occur during codegen, function pointer shims can now be polymorphic without incurring double substitutions. Signed-off-by: David Wood <david@davidtw.co>
commit sha 6cb671628393e292d5e68e6367f80488ace46532
Add test for checking if-let or-patterns
commit sha 941dca8ed238a04a55741127165e9ad80671ed8a
Add Arith Tests in Library
commit sha dc37b553accd4fb2f8d0c59f69c701b524361cc2
Minor refactoring
commit sha 7d834c87d2ebb3d8dd4895bc5fabc4d44a1d2b52
Move Various str tests in library
push time in 7 days
PR closed tokio-rs/bytes
Replace atomic fences with atomic loads for compatibility with ThreadSanitizer.
pr closed time in 7 days
push eventtmiasko/rust
commit sha 65edf54c256e5369fdf9c0a972cd83d2414d5bb5
Add a regression test for copy propagation miscompilation
push time in 7 days
push eventtmiasko/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 ab4fa215b91491be0a7d2c3630efce7a6edb971d
Update Clippy testcases Update the test `redundant_pattern_matching`: check if `is_ok` and `is_err` are suggested within const contexts. Also removes the `redundant_pattern_matching_const_result` test, as it is no longer needed.
commit sha a59d48064931ab860fb065e4ef3a2f5cc4276f96
Remove `can_suggest` check for `is_ok` and `is_err`. `is_ok` and `is_err` are stabilized as const and can thus always be suggested.
commit sha bf70e21a7e48008a8c1d82c3b0376e574f9bc91b
Update src/tools/clippy/clippy_lints/src/matches.rs Co-authored-by: Ralf Jung <post@ralfj.de>
commit sha b873fa6d42cf305131d2583d03b84686e5e40f2e
Auto merge of #76136 - CDirkx:const-result, r=dtolnay Stabilize some Result methods as const Stabilize the following methods of Result as const: - `is_ok` - `is_err` - `as_ref` A test is also included, analogous to the test for `const_option`. These methods are currently const under the unstable feature `const_result` (tracking issue: #67520). I believe these methods to be eligible for stabilization because of the stabilization of #49146 (Allow if and match in constants) and the trivial implementations, see also: [PR#75463](https://github.com/rust-lang/rust/pull/75463) and [PR#76135](https://github.com/rust-lang/rust/pull/76135). Note: these methods are the only methods currently under the `const_result` feature, thus this PR results in the removal of the feature. Related: #76225
commit sha fe85a34fcfa867644f64526b058c56b3f6f6377f
Add a regression test for copy propagation miscompilation
push time in 7 days
pull request commentrust-lang/rust
Add a regression test for copy propagation miscompilation
r? @wesleywiser
comment created time in 7 days
push eventtmiasko/rust
commit sha 190fe462ab1166ced2d04bacf5c633e70f898418
Add a regression test for copy propagation miscompilation
push time in 7 days
push eventtmiasko/rust
commit sha 5af27adb4c9d0170631bf06a2df888cbf80aa15c
Add a regression test for copy propagation miscompilation
push time in 7 days
pull request commentrust-lang/rust
Make RawFd implement the RawFd traits
Thanks for correction. I thought that trait was unsafe to implemented. It is quite disappointing that it is otherwise. The search results from grep.app reveals that pretty much every single use of IntoRawFd as input bound is unsound, e.g., in wastime, nix, wayland-rs, duct.rs ...
comment created time in 7 days
pull request commentrust-lang/rust
Make RawFd implement the RawFd traits
I would generally assume that it is safe to take ownership of file descriptor returned from IntoRawFd. With changes here it would be unsafe, since RawFd is just an integer.
comment created time in 7 days
issue commentrust-lang/rust
Miscompilation introduced by CopyPropagation
The copy propagation incorrectly assumes that a use of a local in Operand::Move will not mutate it.
comment created time in 7 days
push eventtmiasko/rust
commit sha 812e0f40acd1f181309de0499a21201688ca8843
Reopen standard file descriptors when they are missing on Unix The syscalls returning a new file descriptors generally return lowest-numbered file descriptor not currently opened, without any exceptions for those corresponding to stdin, sdout, or stderr. Previously when any of standard file descriptors has been closed before starting the application, operations on std::io::{stderr,stdin,stdout} were likely to either succeed while being performed on unrelated file descriptor, or fail with EBADF which is silently ignored. Avoid the issue by using /dev/null as a replacement when the standard file descriptors are missing. The implementation is based on the one found in musl. It was selected among a few others on the basis of the lowest overhead in the case when all descriptors are already present (measured on GNU/Linux).
push time in 7 days
push eventtmiasko/rust
commit sha fb64e6dcf0a7ea0518c8f2d499327a2b6974f859
Add test for `Ipv4Addr` methods in a const context
commit sha 130359cb05246fcacdde61baa2613419ef6570c7
Auto merge of #76010 - Aaron1011:fix/cfg-generic-param, r=petrochenkov Run cfg-stripping on generic parameters before invoking derive macros Fixes #75930 This changes the tokens seen by a proc-macro. However, ising a `#[cfg]` attribute on a generic paramter is unusual, and combining it with a proc-macro derive is probably even more unusual. I don't expect this to cause any breakage.
commit sha c86d249e94ae45fff1c882c98d85aaf49133f0b6
Use "Fira Sans" for crate list font Fira Sans is what's used for module lists and other item lists. Previously, the default body font, "Source Serif Pro", was used for crate lists, which didn't visually match other item lists.
commit sha d0af12560c9672546a6434e2e1e9ec72ad8f8e70
Rename and expose LoopState as ControlFlow
commit sha bd49eec3d76d5894b539a28309c2fe24f915ee94
interface: use `OnceCell` from standard library
commit sha 1b650d0fea209e0b69407a0a1e04feeb57c4b2d4
datastructures: replace `lazy_static` by `SyncLazy` from std
commit sha 67b8f9491c30cba0e4892f6b5b8bda8b312428da
hir: replace `lazy_static` by `SyncLazy` from std
commit sha 73a7204983a28177b296a3dcd8632540a6fcee1a
feature: replace `lazy_static` by `SyncLazy` from std
commit sha 99c96c5bfe67ccbf7fc379458b8bcf9513bf9b7d
driver: replace `lazy_static` by `SyncLazy` from std
commit sha b1491eacfc2f2ff05ff2a1d3b557c30e15c5f81f
lexer: Tiny improvement to shebang detection Lexer now discerns between regular comments and doc comments, so use that. The change only affects the choice of reported errors.
commit sha 0c77257e56a20a81f5b4e4452cae5e460ad30140
Make all remaining methods of `std::net::Ipv4Addr` const Makes the following methods of `std::net::Ipv4Addr` unstable const under the `const_ipv4` feature: - `is_global` - `is_reserved` - `is_broadcast` - `to_ipv6_compatible` - `to_ipv6_mapped` This results in all methods of `Ipv4Addr` being const. Also adds tests for these methods in a const context.
commit sha 07cd4c8778a219f681f78dfb3c8ee123f630b06d
Use intra-doc links
commit sha 48248af6307643072b337db3fe991fd6e638bfdd
Auto merge of #5999 - rail-rain:fix_fp_transmute_ptr_to_ptr_in_consts, r=ebroto Fix a fp in `transmute_ptr_to_ptr` fixes #5959 changelog: Fix a false positive in `transmute_ptr_to_ptr` that the lint fires when `transmute` is used to cast a reference in const contexts although dereferencing raw pointers in consts is unstable.
commit sha e5c17bff3514cf951ffb304ddd1d5fe1887cca65
Clean up header in `iter` docs for `for` loops
commit sha 6c51ec96bb5b6679645be1651d4ff1d079412af6
inliner: Avoid query cycles when optimizing generators The HIR Id trick is insufficient to prevent query cycles when optimizing generators, since merely requesting a layout of a generator also computes its `optimized_mir`. Make no attempts to inline functions into generators within the same crate to avoid query cycles.
commit sha 54a4fd1eb9418e2ea8313e66cde525930cc191ae
Minor improvements Co-authored-by: Joshua Nelson <joshua@yottadb.com>
commit sha 383da5e1e670c5028731aa3e477578e5e97c6e52
Rollup merge of #74880 - chrisduerr:fix_matches, r=dtolnay Add trailing comma support to matches macro
commit sha 5f28831a40477a0d564368e11fe6c5d6a26cd56f
Rollup merge of #76074 - richkadel:llvm-coverage-map-gen-6b.5.1, r=wesleywiser Add new `-Z dump-mir-spanview` option Similar to `-Z dump-mir-graphviz`, this adds the option to write HTML+CSS files that allow users to analyze the spans associated with MIR elements (by individual statement, just terminator, or overall basic block). This PR was split out from PR #76004, and exposes an API for spanview HTML+CSS files that is also used to analyze code regions chosen for coverage instrumentation (in a follow-on PR). Rust compiler MCP rust-lang/compiler-team#278 Relevant issue: #34701 - Implement support for LLVMs code coverage instrumentation r? @tmandry FYI @wesleywiser
commit sha 7c1c7de85f47cd527d1935295e68218857d2d969
Rollup merge of #76088 - hbina:add_example, r=LukasKalbertodt Add more examples to lexicographic cmp on Iterators. Given two arrays of T1 and T2, the most important rule of lexicographical comparison is that two arrays of equal length will be compared until the first difference occured. The examples provided only focuses on the second rule that says that the shorter array will be filled with some T2 that is less than every T1. Which is only possible because of the first rule.
commit sha 34c8b7a92cd2ced1b4aee960a84f69f6b3a56bb6
Rollup merge of #76099 - camelid:patch-8, r=jyn514 Add info about `!` and `impl Trait` Fixes #76094. @rustbot modify labels: T-doc C-enhancement
push time in 7 days
pull request commentrust-lang/rust
Reopen standard file descriptors when they are missing on Unix
r? @Amanieu
comment created time in 8 days
issue commentrust-lang/rust
Deadlocks on NetBSD around fork()
I can reproduce the last issue on daily snapshot from 2020-09-18, usually after just a few seconds of executing program in the loop. If you do reproduce it at some point, I would appreciate if you could report this upstream, since I generally don't use NetBSD.
Simplified reproducer, since it turns out neither fork nor pthread_getattr_np is required, except for the fact that latter is allocating memory:
#include <assert.h>
#include <pthread.h>
#include <stdlib.h>
#define N 4
static pthread_t threads[N];
static void *run(void *arg) {
return malloc(1024);
}
int main() {
for (int i = 0; i != N; ++i) assert(pthread_create(&threads[i], NULL, run, NULL) == 0);
for (int i = 0; i != N; ++i) assert(pthread_join(threads[i], NULL) == 0);
}
comment created time in 8 days
pull request commentrust-lang/rust
[mir-opt] Introduce a new flag to enable experimental/unsound mir opts
Is this option intended only for unsound passes or also that are buggy in other ways e.g., ICE? If the latter I would also include inlining pass. Recently when I evaluated mir-opt-level=2 on a different crates, rustc generally ICEd due to normalization issues in inliner or other issues exposed indirectly by inlining. The attempt to enable inlining in #75495 didn't fare any better.
comment created time in 8 days
issue openedrust-lang/rust
unreachable_patterns false negative with or_patterns
The match remains exhaustive with highlighted pattern removed, but it does not produce any warnings:
#![feature(or_patterns)]
pub struct A;
pub enum B {
X,
Y,
}
pub enum E {
A(A),
B(B),
}
pub fn f(e: E) -> () {
match e {
E::B(B::X) => (),
E::B(B::Y | B::X)
// ~~~~
| E::A(A) => (),
}
}
created time in 10 days
issue commentrust-lang/rust
starting from nightly-2020-08-18 rustls can't connect to some websites
It minimized code from rustls <HandshakeMessagePayload as Codec>::encode.
comment created time in 11 days
issue commentrust-lang/rust
starting from nightly-2020-08-18 rustls can't connect to some websites
#[derive(Copy, Clone, Debug, PartialEq, Eq)]
#[repr(u8)]
pub enum Type {
A = 1,
B = 7,
}
impl Type {
fn encode(&self) -> Type {
match self {
Type::A => Type::B,
_ => *self,
}
}
}
fn main() {
assert_eq!(Type::A.encode(), Type::B);
}
$ rustc -Zmir-opt-level=0 b.rs && ./b
$ rustc -Zmir-opt-level=1 b.rs && ./b
thread 'main' panicked at 'assertion failed: `(left == right)`
left: `A`,
right: `B`', b.rs:18:5
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
comment created time in 11 days
push eventtmiasko/rust
commit sha ff1a9e406bfccefedb6c4f2cabc0c4d59ac947d4
Fix underflow when calculating the number of no-op jumps folded When removing unwinds to no-op blocks and folding jumps to no-op blocks, remove the unwind target first. Otherwise we cannot determine if target has been already folded or not. Previous implementation incorrectly assumed that all resume targets had been folded already, occasionally resulting in an underflow: remove_noop_landing_pads: removed 18446744073709551613 jumps and 3 landing pads
push time in 11 days
Pull request review commentrust-lang/rust
Collect statistics about MIR optimizations
+//! # Support for collecting simple statistics+//!+//! Statistics are useful for collecting metrics from optimization passes, like+//! the number of simplifications performed. To avoid introducing overhead, the+//! collection of statistics is enabled only when rustc is compiled with+//! debug-assertions.+//!+//! Statistics are static variables defined in the module they are used, and+//! lazy registered in the global collector on the first use. Once registered,+//! the collector will obtain their values at the end of compilation process+//! when requested with -Zmir-opt-stats option.++use parking_lot::{const_mutex, Mutex};+use std::fmt::Write as _;+use std::io::{self, Write as _};+use std::sync::atomic::{AtomicBool, AtomicUsize, Ordering};++static COLLECTOR: Collector = Collector::new();++/// Enables the collection of statistics.+/// To be effective it has to be called before the first use of a statistics.+pub fn try_enable() -> Result<(), ()> {+ COLLECTOR.try_enable()+}++/// Prints all statistics collected so far to the stderr.+pub fn print() {+ COLLECTOR.print();+}++pub struct Statistic {+ category: &'static str,+ name: &'static str,+ initialized: AtomicBool,+ value: AtomicUsize,+}++struct Collector(Mutex<State>);++struct State {+ enabled: bool,+ stats: Vec<&'static Statistic>,+}++#[derive(Eq, PartialEq, Ord, PartialOrd)]+struct Snapshot {+ category: &'static str,+ name: &'static str,+ value: usize,+}++impl Statistic {+ pub const fn new(category: &'static str, name: &'static str) -> Self {+ Statistic {+ category,+ name,+ initialized: AtomicBool::new(false),+ value: AtomicUsize::new(0),+ }+ }++ pub fn name(&self) -> &'static str {+ self.name+ }++ pub fn category(&self) -> &'static str {+ self.category.rsplit("::").next().unwrap()+ }++ #[inline]+ pub fn register(&'static self) {+ if cfg!(debug_assertions) {+ if !self.initialized.load(Ordering::Acquire) {+ COLLECTOR.register(self);+ }+ }+ }++ #[inline]+ pub fn increment(&'static self, value: usize) {+ if cfg!(debug_assertions) {+ self.value.fetch_add(value, Ordering::Relaxed);
That would give a wrong impression that some form of synchronization is required here. Atomicity of RMW operations is sufficient for correctness.
comment created time in 11 days
Pull request review commentrust-lang/rust
Collect statistics about MIR optimizations
+//! # Support for collecting simple statistics+//!+//! Statistics are useful for collecting metrics from optimization passes, like+//! the number of simplifications performed. To avoid introducing overhead, the+//! collection of statistics is enabled only when rustc is compiled with+//! debug-assertions.+//!+//! Statistics are static variables defined in the module they are used, and+//! lazy registered in the global collector on the first use. Once registered,+//! the collector will obtain their values at the end of compilation process+//! when requested with -Zmir-opt-stats option.++use parking_lot::{const_mutex, Mutex};+use std::fmt::Write as _;+use std::io::{self, Write as _};+use std::sync::atomic::{AtomicBool, AtomicUsize, Ordering};++static COLLECTOR: Collector = Collector::new();
I experimented with different implementation approaches:
- Statistics definitions are in the context together with their values.
- Statistics definitions are in MIR passes, their values are stored in the context.
- Statistics definitions and values are stored together with MIR passes.
I strongly prefer variants that put statistics definitions together with MIR passes that uses them. In that approach, adding and removing counters is trivial, additionally any unused counters are detected as such during compilation.
Values can be stored in the context if we want to support multiple rustc sessions in a single process, although is requires passing context to all those places where counters are used. But is there a any real use-case? Why would you use rustc with debug-assertions and MIR optimizations counters inside the RLS?
comment created time in 11 days
Pull request review commentrust-lang/rust
Collect statistics about MIR optimizations
fn validate_commandline_args_with_session_available(sess: &Session) { break; } }++ if sess.opts.debugging_opts.mir_opt_stats {+ if let Err(_) = rustc_data_structures::statistics::try_enable() {+ sess.warn(&format!(+ "-Zmir-opt-stats is only available when rustc is compiled with debug-assertions"+ ));
Replaced with an error.
comment created time in 11 days
push eventtmiasko/rust
commit sha b0a81f4868dcbf4f4b58fd828ad8defe33b23d75
Collect statistics about MIR optimizations
push time in 11 days
PR opened rust-lang/rust
For example -Zmir-opt-level=3 -Zmir-opt-stats:
1813 const_prop Number of const propagations
313 copy_prop Maximum number of copy propagations in a function
6424 copy_prop Number of constants copy propagated
927 copy_prop Number of locals copy propagated
3345 inline Number of functions inlined
21 instcombine Number of array length operations simplified
4109 instcombine Number of deref-ref operations simplified
72 instcombine Number of equality comparisons with a const bool simplified
75 match_branches Number of similar basic blocks merged together
38 nrvo Number of functions eligible for named return value optimization
207 remove_noop_landing_pads Number of no-op jumps folded
4182 remove_noop_landing_pads Number of no-op landing pads removed
15334 simplify Number of basic blocks merged with their only predecessor
36697 simplify Number of dead basic blocks removed
844 simplify Number of switch int terminators with identical successors replaced with a goto
1213 simplify_branches Number of false edges simplified to a goto
67 simplify_branches Number of false unwinds simplified to a goto
21 simplify_branches Number of true assertions simplified to a goto
62 simplify_comparison_integral Number of switches on a result of integer comparison simplified to switches on integers
5 simplify_try Number of arm identities simplified to a move
0 uninhabited_enum_branching Number of branches corresponding to uninhabited enum variants removed
526 unreachable_prop Number of unreachable terminators propagated
pr created time in 12 days
pull request commentrust-lang/rust
Detect overflow in proc_macro_server subspan
@rustbot modify labels: +S-waiting-on-review -S-waiting-on-author
comment created time in 12 days
PR opened rust-lang/rust
When removing unwinds to no-op blocks and folding jumps to no-op blocks, remove the unwind target first, as otherwise we cannot determine if target has been already folded or not, and the calculation might underflow:
remove_noop_landing_pads: removed 18446744073709551613 jumps and 3 landing pads
pr created time in 12 days
pull request commentrust-lang/rust
Online def-use analysis for copy propagation pass
Does the existing const prop pass handle those propagations?
Last time I looked into it, I was under the impression that constant propagation is in a large part a lint, and has to be quite conservative to avoid introducing spurious warnings, e.g., in a code that is otherwise unreachable.
comment created time in 12 days
PR closed rust-lang/rust
Implement an online def-use analysis for the copy propagation pass. The analysis is updated in an incremental fashion as changes are made to the MIR body instead of being recomputed from scratch after each change.
The analysis is intended to be functionally equivalent to the previous one, but it also offers a new methods to: remove specific statement, remove storage markers, and replace the uses of a local with another local or a constant, all while keeping the analysis up-to date.
There are two functional changes to the copy propagation pass, but they should have no effect in practice:
-
Each locals is examined only once as a potential destination for copy propagation, since propagation does not introduce new optimization opportunities.
-
The code for self-assignment elimination has been removed, since it not necessary (assignments that are used for the propagation are removed separately).
For def-use analysis changes, I would recommend reading the new version and the old version separately, rather than examining the diff. For the copy propagation changes, I would recommend looking at the diff that ignores whitespace during comparison.
Closes #36673. Closes #73717.
pr closed time in 12 days
pull request commentrust-lang/rust
Online def-use analysis for copy propagation pass
There is a pre-existing issue in the copy propagation pass, I opened #76740 to track it.
@jonas-schievink regarding those copy propagations performed after destination propagation, my notes show that roughly 12000 of those were arguments (they could be propagated by dest-prop if it was a little bit more flexible with regards to propagation direction, I suppose?). There were also substantial number of constant propagated as noted before.
Given that copy propagation pass is working incorrectly right now and has a candidate for a replacement, I am going to close this now.
comment created time in 12 days
push eventtmiasko/rust
push time in 12 days
push eventtmiasko/rust
commit sha b4935e07269429e04abe0d6d25f7e3211f4fa3f6
use sort_unstable to sort primitive types It's not important to retain original order if we have &[1, 1, 2, 3] for example. clippy::stable_sort_primitive
commit sha 21f8326cec03848368e02936a032103aa24cf6d2
Provide suggestion for missing fields in patterns
commit sha 458aaba08467dc6b8d04b79f955a46be779d94f1
Add Atomic*::from_mut. The atomic equivalent of Cell::from_mut.
commit sha 3be40b22c83491e4a4d34e2ad06aa81b804b9fad
Fix AtomicPtr::from_mut align check: Avoid generic arg in const expr. See #76200.
commit sha 9914c3beed0f4a829e78eafb36bac39d418b71f5
Supress unused_macros error on architectures with no atomics.
commit sha 245f69ad3c16c1f7355c3901e131fa01b588bf36
Refactor `resolve_link` into a separate function
commit sha 7dc0d335bcc819c770320635a055c9cfe076339a
Refactor `resolve_with_disambiguator` into a separate function
commit sha 8a13fc494d4365e57d8f343219f0201458b68591
Require `module_id` param to `resolve` to be non-empty Previously, `resolve` would immediately check that `module_id` was non-empty and give an error if not. This had two downsides: - It introduced `Option`s everywhere, even if the calling function knew it had a valid module, and - It checked the module on each namespace, which is unnecessary: it only needed to be checked once. This makes the caller responsible for checking the module exists, making the code a lot simpler.
commit sha 79aa9b15d7403ce2dc40b525a1d16e6c4ad1973c
Optimize behavior of vec.split_off(0) (take all) Optimization improvement to `split_off()` so the performance meets the intuitively expected behavior when `at == 0`, avoiding the current behavior of copying the entire vector. The change honors documented behavior that the method leaves the original vector's "previous capacity unchanged". This improvement better supports the pattern for building and flushing a buffer of elements, such as the following: ```rust let mut vec = Vec::new(); loop { vec.push(something); if condition_is_met { process(vec.split_off(0)); } } ``` `Option` wrapping is the first alternative I thought of, but is much less obvious and more verbose: ```rust let mut capacity = 1; let mut vec: Option<Vec<Stuff>> = None; loop { vec.get_or_insert_with(|| Vec::with_capacity(capacity)).push(something); if condition_is_met { capacity = vec.capacity(); process(vec.take().unwrap()); } } ``` Directly applying `mem::replace()` could work, but `mem::` functions are typically a last resort, when a developer is actively seeking better performance than the standard library provides, for example. The benefit of the approach to this change is it does not change the existing API contract, but improves the peformance of `split_off(0)` for `Vec`, `String` (which delegates `split_off()` to `Vec`), and any other existing use cases. This change adds tests to validate the behavior of `split_off()` with regard to capacity, as originally documented, and confirm that behavior still holds, when `at == 0`. The change is an implementation detail, and does not require a documentation change, but documenting the new behavior as part of its API contract may benefit future users. (Let me know if I should make that documentation update.) Note, for future consideration: I think it would be helpful to introduce an additional method to `Vec` (if not also to `String`): ``` pub fn take_all(&mut self) -> Self { self.split_off(0) } ``` This would make it more clear how `Vec` supports the pattern, and make it easier to find, since the behavior is similar to other `take()` methods in the Rust standard library.
commit sha 62effcbd5bfaf74e99def3e9a660dba9728b0b47
Detect turbofish with multiple type params missing leading `::` Fix #76072.
commit sha 9b4154193e8471f36b1a9e781f1ef7d492fc6a6c
Auto merge of #76541 - matthiaskrgr:unstable_sort, r=davidtwco use sort_unstable to sort primitive types It's not important to retain original order if we have &[1, 1, 2, 3] for example. clippy::stable_sort_primitive
commit sha 255ceeb5ff9875b7f525aa101c8adc155f3e0ef8
Auto merge of #76612 - estebank:pat-missing-fields-suggestion, r=davidtwco Provide suggestion for missing fields in patterns
commit sha 715e9340a1006e37aed381e320ddf30311c2c2a6
Auto merge of #74532 - fusion-engineering-forks:atomic-from-mut, r=KodrAus Add Atomic*::from_mut. The atomic equivalent of [`Cell::from_mut`](https://doc.rust-lang.org/stable/std/cell/struct.Cell.html#method.from_mut).
commit sha 6cae28165f0450fd3f100374b26841e458b8cfef
Auto merge of #76682 - richkadel:vec-take, r=Mark-Simulacrum Optimize behavior of vec.split_off(0) (take all) Optimization improvement to `split_off()` so the performance meets the intuitively expected behavior when `at == 0`, avoiding the current behavior of copying the entire vector. The change honors documented behavior that the original vector's "previous capacity unchanged". This improvement better supports the pattern for building and flushing a buffer of elements, such as the following: ```rust let mut vec = Vec::new(); loop { vec.push(something); if condition_is_met { process(vec.split_off(0)); } } ``` `Option` wrapping is the first alternative I thought of, but is much less obvious and more verbose: ```rust let mut capacity = 1; let mut vec: Option<Vec<Stuff>> = None; loop { vec.get_or_insert_with(|| Vec::with_capacity(capacity)).push(something); if condition_is_met { capacity = vec.capacity(); process(vec.take().unwrap()); } } ``` Directly using `mem::replace()` (instead of calling`split_off()`) could work, but `mem::replace()` is a more advanced tool for Rust developers, and in this case, I believe developers would assume the standard library should be sufficient for the purpose described here. The benefit of the approach to this change is it does not change the existing API contract, but improves the peformance of `split_off(0)` for `Vec`, `String` (which delegates `split_off()` to `Vec`), and any other existing use cases. This change adds tests to validate the behavior of `split_off()` with regard to capacity, as originally documented, and confirm that behavior still holds, when `at == 0`. The change is an implementation detail, and does not require a documentation change, but documenting the new behavior as part of its API contract may benefit future users. (Let me know if I should make that documentation update.) Note, for future consideration: I think it would be helpful to introduce an additional method to `Vec` (if not also to `String`): ``` pub fn take_all(&mut self) -> Self { self.split_off(0) } ``` This would make it more clear how `Vec` supports the pattern, and make it easier to find, since the behavior is similar to other `take()` methods in the Rust standard library. r? `@wesleywiser` FYI: `@tmandry`
commit sha c1589cc819ba7cf289c3ccbab30c215f0a6ba7d7
Auto merge of #76684 - jyn514:refactor-intra-links, r=manishearth Refactor intra doc link code I got tired of `fold_item` being 500 lines long. This is best reviewed one commit at a time with whitespace changes hidden. There are no logic changes other than the last commit making a parameter checked by the caller instead of the callee. r? `@Manishearth`
commit sha 90b1f5ae59291dd69d72fad41a22277df19dc953
Auto merge of #76171 - estebank:turbofish-the-revenge, r=davidtwco Detect turbofish with multiple type params missing leading `::` Fix #76072.
commit sha 63f3dcefb950d9d9f63b8a71de39d990a742b675
copy-prop: Add test case for missed optimization due const prop failure Currently a failure to completely propagate a constant inhibits propagation of any subsequent locals. Add a test case demonstrating the problem before resolving it.
commit sha 2ee57d752f3bd1ce3bf386465de9e60b6ebbfa52
copy-prop: Attempt to propagate each local only once The copy propagation as currently implemented does not introduce new copy propagation opportunities. Examine each local for possible optimization only once. Additionally, this also fixes the issue where copy propagation optimization would stop without examining all locals after failing to completely propagate a constant.
commit sha 35c8b0e78a02dc1f42a27cf009003be9b86df314
copy-prop: Remove unused self-assignment elimination The assignments that are starting point for the copy propagation are removed separately. The copy propagation does not introduce self-assignments otherwise, so the code for self-assignments elimination is functionally inert. Remove it. No functional changes intended.
commit sha 232adee6169ed779faddcf5ddf491c058d93c782
copy-prop: Online def-use analysis for copy propagation pass Implement an online def-use analysis for the copy propagation pass. The analysis is updated in an incremental fashion as changes are made to the MIR body instead of being recomputed from scratch after each change. The analysis is intended to be equivalent to the previous one, but it also offers a new methods to: remove specific statement, remove storage markers, and replace the uses of a local with another local or a constant, all while keeping the analysis up-to date. No functional changes intended.
push time in 12 days
issue openedrust-lang/rust
Miscompilation introduced by CopyPropagation
To reproduce the issue:
- Checkout 41dc3942eb33e8e882b6e4782a9bd9d2b8970647 (for reproducibility, but might be unnecessary).
- Enable copy propagation so that it runs on mir-opt-level=1.
- ./x.py test --stage 2 src/test/ui (disable incremental compilation, so that rustc is rebuild completely).
- Observe tests failures: "not all trait items implemented, missing ...".
created time in 12 days