summary refs log tree commit diff
path: root/library/std/src/sys/windows
AgeCommit message (Collapse)AuthorLines
2023-10-28Rollup merge of #116816 - ChrisDenton:api.rs, r=workingjubileeJubilee-53/+212
Create `windows/api.rs` for safer FFI FFI is inherently unsafe. For memory safety we need to assert that some contract is being upheld on both sides of the FFI, though of course we can only ever check our side. In Rust, `unsafe` blocks are used to assert safety and `// SAFETY` comments describing why it is safe. Currently in sys/windows we have a lot of this unsafety spread all over the place, with variations on the same unsafe patterns repeated. And because of the repitition and frequency, we're a bit lax with the safety comments. This PR aims to fix this and to make FFI safety more auditable by creating an `api` module with the goal of centralising and consolidating this unsafety. It contains thin wrappers around the Windows API that make most functions safe to call or, if that's not possible, then at least safer. Note that its goal is *only* to address safety. It does not stray far from the Windows API and intentionally does not attempt to make higher lever wrappers around, for example, file handles. This is better left to the existing modules. The windows/api.rs file has a top level comment to help future contributors understand the intent of the module and the design decisions made. I chose two functions as a first tentative step towards the above goal: - [`GetLastError`](https://learn.microsoft.com/en-us/windows/win32/api/errhandlingapi/nf-errhandlingapi-getlasterror) is trivially safe. There's no reason to wrap it in an `unsafe` block every time. So I simply created a safe `get_last_error` wrapper. - [`SetFileInformationByHandle`](https://learn.microsoft.com/en-us/windows/win32/api/fileapi/nf-fileapi-setfileinformationbyhandle) is more complex. It essentially takes a generic type but over a C API which necessitates some amount of ceremony. Rather than implementing similar unsafe patterns in multiple places, I provide a safe `set_file_information_by_handle` that takes a Rusty generic type and handles converting that to the form required by the C FFI. r? libs
2023-10-24Auto merge of #116461 - ChrisDenton:sleep, r=thomccbors-1/+88
Windows: Support sub-millisecond sleep Use `CreateWaitableTimerExW` with `CREATE_WAITABLE_TIMER_HIGH_RESOLUTION`. Does not work before Windows 10, version 1803 so in that case we fallback to using `Sleep`. I've created a `WaitableTimer` type so it can one day be adapted to also support waiting to an absolute time (which has been talked about). Note though that it currently returns `Err(())` because we can't do anything with the errors other than fallback to the old `Sleep`. Feel free to tell me to do errors properly. It just didn't seem worth constructing an `io::Error` if we're never going to surface it to the user. And it *should* all be infallible anyway unless the OS is too old to support it. Closes #43376
2023-10-23Auto merge of #116606 - ChrisDenton:empty, r=dtolnaybors-0/+8
On Windows make `read_dir` error on the empty path This makes Windows consistent with other platforms. Note that this should not be taken to imply any decision on #114149 has been taken. However it was felt that while there is a lack of libs-api consensus, we should be consistent across platforms in the meantime. This is a change in behaviour for Windows so will also need an fcp before merging. r? libs-api
2023-10-19Auto merge of #116132 - darthunix:connect_poll, r=cuviperbors-5/+7
Make TCP connect handle EINTR correctly According to the [POSIX](https://pubs.opengroup.org/onlinepubs/009695399/functions/connect.html) standard, if connect() is interrupted by a signal that is caught while blocked waiting to establish a connection, connect() shall fail and set errno to EINTR, but the connection request shall not be aborted, and the connection shall be established asynchronously. When the connection has been established asynchronously, select() and poll() shall indicate that the file descriptor for the socket is ready for writing. The previous implementation differs from the recomendation: in a case of the EINTR we tried to reconnect in a loop and sometimes get EISCONN error (this problem was originally detected on MacOS). 1. More details about the problem in an [article](http://www.madore.org/~david/computers/connect-intr.html). 2. The original [issue](https://git.picodata.io/picodata/picodata/tarantool-module/-/issues/157).
2023-10-19Auto merge of #116402 - joboet:global_alloc_tls_unsoundness, ↵bors-4/+15
r=thomcc,workingjubilee Panic when the global allocator tries to register a TLS destructor Using a `RefCell` avoids the undefined behaviour encountered in #116390 and reduces the amount of `unsafe` code in the codebase.
2023-10-16Create `windows/api.rs` for safer FFIChris Denton-53/+212
2023-10-13Make TCP connect() handle EINTR correctlyDenis Smirnov-5/+7
According to the POSIX standard, if connect() is interrupted by a signal that is caught while blocked waiting to establish a connection, connect() shall fail and set errno to EINTR, but the connection request shall not be aborted, and the connection shall be established asynchronously. If asynchronous connection was successfully established after EINTR and before the next connection attempt, OS returns EISCONN that was handled as an error before. This behavior is fixed now and we handle it as success. The problem affects MacOS users: Linux doesn't return EISCONN in this case, Windows connect() can not be interrupted without an old-fashoin WSACancelBlockingCall function that is not used in the library. So current solution gives connect() as OS specific implementation.
2023-10-13Test that unix sockets exist on WindowsChris Denton-0/+14
2023-10-13Make try_exists return Ok(true) for Windows UDSChris Denton-0/+7
`fs::try_exists` currently fails on Windows if encountering a Unix Domain Socket (UDS). Fix this by checking for an error code that's returned when there's a failure to use a reparse point. A reparse point is a way to invoke a filesystem filter on a file instead of the file being opened normally. This is used to implement symbolic links (by redirecting to a different path) but also to implement other types of special files such as Unix domain sockets. If the reparse point is not a link type then opening it with `CreateFileW` may fail with `ERROR_CANT_ACCESS_FILE` because the filesystem filter does not implement that operation. This differs from resolving links which may fail with errors such as `ERROR_FILE_NOT_FOUND` or `ERROR_CANT_RESOLVE_FILENAME`. So `ERROR_CANT_ACCESS_FILE` means that the file exists but that we can't open it normally. Still, the file does exist so `try_exists` should report that as `Ok(true)`.
2023-10-10On Windows make readdir error on the empty pathChris Denton-0/+8
2023-10-09std: explain unconventional choice of let-else binding over while-let loopjoboet-0/+3
2023-10-07std: fix registering of Windows TLS destructorsjoboet-1/+4
2023-10-06Windows: Support sub-millisecond sleepChris Denton-1/+88
Use `CreateWaitableTimerExW` with `CREATE_WAITABLE_TIMER_HIGH_RESOLUTION`. Does not work before Windows 10, version 1803 so in that case we fallback to using `Sleep`.
2023-10-06Remove libcChris Denton-13/+18
We don't use much libc on Windows.
2023-10-04std: abort instead of panicking if the global allocator uses TLSjoboet-1/+5
2023-10-03std: panic when the global allocator tries to register a TLS destructorjoboet-4/+5
2023-09-28Auto merge of #114882 - ChrisDenton:riddle-me, r=dtolnaybors-121/+112
Update windows ffi bindings Bump `windows-bindgen` to version 0.51.1. This brings with it some changes to the generated FFI bindings, but little that affects the code. One change that does have more of an impact is `SOCKET` being `usize` instead of either `u64` or `u32` (as is used in std's public `SOCKET` type). However, it's now easy enough to abstract over that difference. Finally I added a few new bindings that are likely to be used in pending PRs, mostly to make sure they're ok with the new metadata. r? libs
2023-09-28Auto merge of #98704 - vthib:impl-from-raw-for-childstd-structs, r=dtolnaybors-1/+7
Implement From<OwnedFd/Handle> for ChildStdin/out/err object ## Summary Comments in `library/std/src/process.rs` ( https://github.com/rust-lang/rust/commit/ab08639e5950f5c8a42a2870c9636181308c3686 ) indicates that `ChildStdin`, `ChildStdout`, `ChildStderr` implements some traits that are not actually implemented: `FromRawFd`, `FromRawHandle`, and the `From<OwnedFd>/From<OwnedHandle>` from the io_safety feature. In this PR I implement `FromRawHandle` and `FromRawFd` for those 3 objects. ## Usecase I have a usecase where those implementations are basically needed. I want to customize in the `Command::spawn` API how the pipes for the parent/child communications are created (mainly to strengthen the security attributes on them). I can properly setup the pipes, and the "child" handles can be provided to `Child::spawn` easily using `Stdio::from_raw_handle`. However, there is no way to generate the `ChildStd*` objects from the raw handle of the created name pipe, which would be very useful to still expose the same API than in other OS (basically a `spawn(...) -> (Child, ChildStdin, ChildStdout, ChildSterr)`, where on windows this is customized), and to for example use `tokio::ChildStdin::from_std` afterwards. ## Questions * Are those impls OK to add? I have searched to see if those impls were missing on purpose, or if it was just never implemented because never needed. I haven't found any indication on why they couldn't be added, although the user clearly has to be very careful that the handle provided makes sense (i think, mainly that it is in overlapped mode for windows). * If this change is ok, adding the impls for the io_safety feature would probably be best, or should it be done in another PR? * I just copy-pasted the `#[stable(...)]` attributes, but the `since` value has to be updated, I'm not sure to which value.
2023-09-09Auto merge of #114590 - ijackson:stdio-stdio-2, r=dtolnaybors-10/+25
Allow redirecting subprocess stdout to our stderr etc. (redux) This is the code from #88561, tidied up, including review suggestions, and with the for-testing-only CI commit removed. FCP for the API completed in #88561. I have made a new MR to facilitate review. The discussion there is very cluttered and the branch is full of changes (in many cases as a result of changes to other Rust stdlib APIs since then). Assuming this MR is approvedl we should close that one. ### Reviewer doing a de novo review Just code review these four commits.. FCP discussion starts here: https://github.com/rust-lang/rust/pull/88561#issuecomment-1640527595 Portability tests: you can see that this branch works on Windows too by looking at the CI results in #88561, which has the same code changes as this branch but with an additional "DO NOT MERGE" commit to make the Windows tests run. ### Reviewer doing an incremental review from some version of #88561 Review the new commits since your last review. I haven't force pushed the branch there. git diff the two branches (eg `git diff 176886197d6..0842b69c219`). You'll see that the only difference is in gitlab CI files. You can also see that *this* MR doesn't touch those files.
2023-09-03Command::spawn: Fix STARTUPINFOW.cb being initialized with the address of ↵Fulgen301-1/+1
size_of
2023-09-01fix(std): Rename os_str_bytes to encoded_bytesEd Page-22/+22
2023-08-28Remove old bindings that have been movedChris Denton-7/+1
2023-08-28Add GetActiveProcessorCount and process attributesChris Denton-0/+7
2023-08-28Abstract over internal `SOCKET` typeChris Denton-16/+22
This allows `SOCKET` to be `usize` internally
2023-08-28Update windows ffi bindingsChris Denton-98/+82
2023-08-28Auto merge of #114848 - michaelvanstraten:spawn_with_attributes, r=ChrisDentonbors-2/+152
Add ability to spawn Windows process with Proc Thread Attributes | Take 2 This is the second attempt to merge pull request #88193 into the standard library. This PR implements the ability to add arbitrary attributes to a command on Windows targets using a new `raw_attribute` method on the [`CommandExt`](https://doc.rust-lang.org/stable/std/os/windows/process/trait.CommandExt.html) trait. `@TyPR124` and my main motivation behind adding this feature is to enable the support of pseudo terminals in the std library, but there are many more applications. A good starting point to get into this topic is to head over to the [`Win32 API documentation`](https://learn.microsoft.com/en-us/windows/win32/api/processthreadsapi/nf-processthreadsapi-updateprocthreadattribute).
2023-08-25Add a new helper to avoid calling io::Error::kindBen Kimock-0/+5
2023-08-25Added option to set ProcThreadAttributes for Windows processesMichael van Straten-2/+152
This implements the ability to add arbitrary attributes to a command on Windows targets using a new `raw_attribute` method on the [`CommandExt`](https://doc.rust-lang.org/stable/std/os/windows/process/trait.CommandExt.html) trait. Setting these attributes provides extended configuration options for Windows processes. Co-authored-by: Tyler Ruckinger <t.ruckinger@gmail.com>
2023-08-19Usage zero as language id for FormatMessageW()Charles Lew-6/+2
2023-08-12Rollup merge of #114132 - tamird:better-env-debug-impls, r=AmanieuMatthias Krüger-5/+49
Better Debug for Vars and VarsOs Display actual vars instead of two dots. The same was done for Args and ArgsOs in 275f9a04af6191e3aee3852a5a1713.
2023-08-09Auto merge of #99747 - ankane:float_gamma, r=workingjubileebors-1/+5
Add gamma function to f32 and f64 Adds the [gamma function](https://en.wikipedia.org/wiki/Gamma_function) to `f32` and `f64` (`tgamma` and `tgammaf` from C). Refs: - https://github.com/rust-lang/rfcs/issues/864 - https://github.com/rust-lang/rust/issues/18271
2023-08-08Rollup merge of #106425 - ijackson:exit-status-default, r=dtolnayMatthias Krüger-1/+1
Make ExitStatus implement Default And, necessarily, make it inhabited even on platforms without processes. I noticed while preparing https://github.com/rust-lang/rfcs/pull/3362 that there was no way for anyone to construct an `ExitStatus`. This would be insta-stable so needs an FCP.
2023-08-07Better Debug for Vars and VarsOsTamir Duberstein-5/+49
Display actual vars instead of two dots. The same was done for Args and ArgsOs in 275f9a04af6191e3aee3852a5a1713.
2023-08-07impl Default for ExitStatusIan Jackson-1/+1
2023-08-07std::process (windows): Implement From<io::Stdout> etc. for imp::StdioIan Jackson-0/+14
This involves a new variant `imp;::Stdio::InheritSpecific`. Signed-off-by: Ian Jackson <ijackson@chiark.greenend.org.uk>
2023-08-07std::process (windows): refactor Stdio::to_handle slightlyIan Jackson-10/+11
We're going to want to reuse this bit of code.
2023-07-31Add gamma and ln_gamma functions to f32 and f64Andrew Kane-1/+5
2023-07-24remove additional [allow(unused_unsafe)]James Dietz-4/+1
2023-07-22Rollup merge of #113442 - epage:osstring, r=cuviperMatthias Krüger-0/+10
Allow limited access to `OsString` bytes This extends #109698 to allow no-cost conversion between `Vec<u8>` and `OsString` as suggested in feedback from `os_str_bytes` crate in #111544.
2023-07-20Minor improvements to Windows TLS dtorsChris Denton-28/+49
2023-07-07Allow limited access to `OsString` bytesEd Page-0/+10
This extends #109698 to allow no-cost conversion between `Vec<u8>` and `OsString` as suggested in feedback from `os_str_bytes` crate in #111544.
2023-07-06Rollup merge of #113334 - fmease:revert-lexing-c-str-lits, r=compiler-errorsfee1-dead-7/+7
Revert the lexing of `c"…"` string literals Fixes \[after beta-backport\] #113235. Further progress is tracked in #113333. This PR *manually* reverts parts of #108801 (since a git-revert would've been too coarse-grained & messy) and git-reverts #111647. CC `@fee1-dead` (#108801) `@klensy` (#111647) r? `@compiler-errors` `@rustbot` label F-c_str_literals beta-nominated
2023-07-05Revert "use c literals in library"León Orell Valerian Liehr-7/+7
This reverts commit f212ba6d6d60963c8101bb24fc3e53fca80c046f.
2023-07-05Auto merge of #112594 - ChrisDenton:process=-kill, r=Amanieubors-1/+10
Return `Ok` on kill if process has already exited This will require an FCP from `@rust-lang/libs-api.` Fixes #112423. See that issue for more details.
2023-07-01Return `Ok` on kill if process has already exitedChris Denton-1/+10
2023-06-25Auto merge of #113001 - ChrisDenton:win-arm32-shim, r=thomccbors-49/+44
Move windows-sys arm32 shim to c.rs This moves the arm32 shim in to c.rs instead of appending to the generated file itself. This makes it simpler to change these workarounds if/when needed. The downside is we need to exclude a couple of functions from being generated (see the comment). A metadata solution could help here but they'll be easy enough to add back if that happens.
2023-06-24Remove unnecessary `path` attributeChris Denton-1/+0
2023-06-24Move arm32 shim to c.rsChris Denton-49/+44
2023-06-18Fix windows `Socket::connect_timeout` overflowEval EXEC-1/+1
Signed-off-by: Eval EXEC <execvy@gmail.com>
2023-06-14Rollup merge of #98202 - aticu:impl_tryfrom_osstr_for_str, r=AmanieuMatthias Krüger-1/+1
Implement `TryFrom<&OsStr>` for `&str` Recently when trying to work with `&OsStr` I was surprised to find this `impl` missing. Since the `to_str` method already existed the actual implementation is fairly non-controversial, except for maybe the choice of the error type. I chose an opaque error here instead of something like `std::str::Utf8Error`, since that would already make a number of assumption about the underlying implementation of `OsStr`. As this is a trait implementation, it is insta-stable, if I'm not mistaken? Either way this will need an FCP. I chose "1.64.0" as the version, since this is unlikely to land before the beta cut-off. `@rustbot` modify labels: +T-libs-api API Change Proposal: rust-lang/rust#99031 (accepted)
generated by cgit-pink 1.4.1-3-g733a5 (git 2.37.1) at 2026-03-01 18:46:15 +0000