| Age | Commit message (Collapse) | Author | Lines |
|---|
|
|
|
|
|
|
|
file size
This solves several problems
- race conditions where a file is truncated while copying from it. if we blindly trusted
the file size this would lead to an infinite loop
- proc files appearing empty to copy_file_range but not to read/write
https://github.com/coreutils/coreutils/commit/4b04a0c
- copy_file_range returning 0 for some filesystems (overlay? bind mounts?)
inside docker, again leading to an infinite loop
|
|
Fix wasi::fs::OpenOptions to imply write when append is on
This PR fixes a bug in `OpenOptions` of `wasi` platform that it currently doesn't imply write mode when only `append` is enabled.
As explained in the [doc of OpenOptions#append](https://doc.rust-lang.org/std/fs/struct.OpenOptions.html#method.append), calling `.append(true)` should imply `.write(true)` as well.
## Reproduce
Given below simple Rust program:
```rust
use std::fs::OpenOptions;
use std::io::Write;
fn main() {
let mut file = OpenOptions::new()
.write(true)
.create(true)
.open("foo.txt")
.unwrap();
writeln!(file, "abc").unwrap();
}
```
it can successfully compiled into wasm and execute by `wasmtime` runtime:
```sh
$ rustc --target wasm32-wasi write.rs
$ ~/wasmtime/target/debug/wasmtime run --dir=. write.wasm
$ cat foo.txt
abc
```
However when I change `.write(true)` to `.append(true)`, it fails to execute by the error "Capabilities insufficient":
```sh
$ ~/wasmtime/target/debug/wasmtime run --dir=. append.wasm
thread 'main' panicked at 'called `Result::unwrap()` on an `Err` value: Os { code: 76, kind: Other, message: "Capabilities insufficient" }', append.rs:10:5
note: run with `RUST_BACKTRACE=1` environment variable to display a backtrace
Error: failed to run main module `append.wasm`
...
```
This is because of lacking "rights" on the opened file:
```sh
$ RUST_LOG=trace ~/wasmtime/target/debug/wasmtime run --dir=. append.wasm 2>&1 | grep validate_rights
TRACE wasi_common::entry > | validate_rights failed: required rights = HandleRights { base: fd_write (0x40), inheriting: empty (0x0) }; actual rights = HandleRights { base: fd_seek|fd_fdstat_set_flags|fd_sync|fd_tell|fd_advise|fd_filestat_set_times|poll_fd_readwrite (0x88000bc), inheriting: empty (0x0) }
```
|
|
|
|
on a NFS mount under RHEL/CentOS 7.
The syscall is supposed to return ENOSYS in most cases but when calling it on NFS it may leak through
EOPNOTSUPP even though that's supposed to be handled by the kernel and not returned to userspace.
Since it returns ENOSYS in some cases anyway this will trip the HAS_COPY_FILE_RANGE
detection anyway, so treat EOPNOTSUPP as if it were a ENOSYS.
https://access.redhat.com/documentation/en-us/red_hat_enterprise_linux/7/html/7.8_release_notes/deprecated_functionality#the_literal_copy_file_range_literal_call_has_been_disabled_on_local_file_systems_and_in_nfs
https://bugzilla.redhat.com/show_bug.cgi?id=1783554
|
|
|
|
|
|
adjust remaining targets
- fix commit 7dc3886
- previous commit doesn't adjust all targets
|
|
Remove some redundant parts from `unrolled_find_u16s`
See each commit message for details.
r? @wesleywiser from old PR #67705 .
|
|
|
|
|
|
All #[cfg(unix)] platforms follow the POSIX standard and define _SC_IOV_MAX so
that we rely purely on POSIX semantics to determine the limits on I/O vector
count.
|
|
|
|
Keep the I/O vector count limit in a `SyncOnceCell` to avoid the overhead of
repeatedly calling `sysconf` as these limits are guaranteed to not change during
the lifetime of a process by POSIX.
|
|
Both Linux and MacOS enforce limits on the vector count when performing vectored
I/O via the readv and writev system calls and return EINVAL when these limits
are exceeded. This changes the standard library to handle those limits as short
reads and writes to avoid forcing its users to query these limits using
platform specific mechanisms.
|
|
|
|
|
|
Previously `std::fs::File::metadata` on wasm32-wasi would call `fd_filestat_get`
to get metadata associated with fd, but that fd is opened without
RIGHTS_FD_FILESTAT_GET right, so it will failed on correctly implemented WASI
environment.
This change instead to add the missing rights when opening an fd.
|
|
This commit updates the src/stdarch submodule primarily to include
rust-lang/stdarch#874 which updated and revamped WebAssembly SIMD
intrinsics and renamed WebAssembly atomics intrinsics. This is all
unstable surface area of the standard library so the changes should be
ok here. The SIMD updates also enable SIMD intrinsics to be used by any
program any any time, yay!
cc #74372, a tracking issue I've opened for the stabilization of SIMD
intrinsics
|
|
- fix commit 7dc3886
- previous commit doesn't adjust all targets
|
|
|
|
|
|
|
|
|
