syscall issueshttps://gitlab.redox-os.org/redox-os/syscall/-/issues2024-03-16T06:46:15Zhttps://gitlab.redox-os.org/redox-os/syscall/-/issues/35Has non-latest major version of bitflags crate as dependency2024-03-16T06:46:15ZShahar OrHas non-latest major version of bitflags crate as dependencyI'd like to ask that it be updated. That's all. Thank you.I'd like to ask that it be updated. That's all. Thank you.https://gitlab.redox-os.org/redox-os/syscall/-/issues/34Update `bitflags` to v22023-12-07T15:29:58ZdaxpeddaUpdate `bitflags` to v2As in the title.
See https://github.com/bitflags/bitflags/releases/tag/2.0.0.
This is a breaking change as it modifies the exposed API.As in the title.
See https://github.com/bitflags/bitflags/releases/tag/2.0.0.
This is a breaking change as it modifies the exposed API.https://gitlab.redox-os.org/redox-os/syscall/-/issues/31support Rust nightly 1.54.02021-06-16T08:08:26Zpoly000support Rust nightly 1.54.0https://github.com/rust-lang/rust/pull/84310
```
error[E0557]: feature has been removed
--> /home/poly/.cargo/registry/src/github.com-1ecc6299db9ec823/redox_syscall-0.2.8/src/lib.rs:3:12
|
3 | #![feature(const_fn)] // see https://git...https://github.com/rust-lang/rust/pull/84310
```
error[E0557]: feature has been removed
--> /home/poly/.cargo/registry/src/github.com-1ecc6299db9ec823/redox_syscall-0.2.8/src/lib.rs:3:12
|
3 | #![feature(const_fn)] // see https://github.com/rust-lang/rfcs/pull/2632
| ^^^^^^^^ feature has been removed
|
= note: split into finer-grained feature gates
```https://gitlab.redox-os.org/redox-os/syscall/-/issues/30New asm!() syntax?2021-03-31T09:12:49Zcoolreader18New asm!() syntax?On the current nightly, building redox_syscall fails due to RFC 2873 renaming the `asm!()` macro to `llvm_asm!()` and moving it to be replaced by a new inline assembly syntax. Here's an example of the error message (I get about 6 of thes...On the current nightly, building redox_syscall fails due to RFC 2873 renaming the `asm!()` macro to `llvm_asm!()` and moving it to be replaced by a new inline assembly syntax. Here's an example of the error message (I get about 6 of these total):
```
error: the legacy LLVM-style asm! syntax is no longer supported
--> /home/coolreader18/.cargo/registry/src/github.com-1ecc6299db9ec823/redox_syscall-0.1.56/src/io/pio.rs:31:13
|
31 | asm!("in $0, $1" : "={al}"(value) : "{dx}"(self.port) : "memory" : "intel", "volatile");
| ----^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
| |
| help: replace with: `llvm_asm!`
|
= note: consider migrating to the new asm! syntax specified in RFC 2873
= note: alternatively, switch to llvm_asm! to keep your code working as it is
```
What would be the best course of action? I understand that redox stays on somewhat up-to-date version of nightly, so would it be ok to update the rust-toolchain in the redox repo and just use either the new asm syntax or the new llvm_asm feature gate? Or maybe have a build.rs check whether the nightly version calls the llvm asm macro llvm_asm!() or asm!() and do conditional compilation off of that?https://gitlab.redox-os.org/redox-os/syscall/-/issues/28WASM WASI compatibility2023-10-15T09:06:40ZDaniel OlanoWASM WASI compatibilityHello! Just wanted to bring some discussion about this topic.
There is this big fancy train call WASM that will surely get far and seems to me that people jumping on board might benefit form its momentum. And what better that a new fre...Hello! Just wanted to bring some discussion about this topic.
There is this big fancy train call WASM that will surely get far and seems to me that people jumping on board might benefit form its momentum. And what better that a new fresh, secure and lightweight OS? It might be already to late to make Redox completely Web Assembly centric but some adaptations could be made to make it play better with wasm apps using the Web Assembly System Interface? The plan of WASI is to have a modular set of non web system APIs so WASM apps can run outside of the web in a secure way, currently they started with the [core API system calls](https://github.com/CraneStation/wasmtime/blob/master/docs/WASI-api.md) to implement basic IO functionality.
I don't know much about anything but in my simple way to see things I would seem that if Redox syscalls heppen to be the same as WASI and a WASM app targeting this kind of runtime is run using a JIT compiler built-in or compiled ahead of time at "installation" time, the app just "just work"? What do you guys think about being more web assembly friendly? does it deviate too much from the design choices of Redox? ultimately I just want to see this project to succeed, been a patreon for some time and see an opportunity there to gain back momentum that has been lost over time. :sweat\_smile:
As web technologies supporter I would definitely be very exited to see a WASM centric OS whether is Redox based or not, WASI runtimes will be popular and Operating systems built around it will be used, for sure there will be linux distros just for that that will see a big success in the cloud for example but I rather see Redox there ;) on the plus side all this wasm stuff compilers and runtimes are being written in Rust so they would integrate much nicely with Redox than with any other OS.
What do you think? makes some sense?https://gitlab.redox-os.org/redox-os/syscall/-/issues/27What is the purpose of `buf` in arguments of `dup`?2018-06-15T05:04:53ZJeremy SollerWhat is the purpose of `buf` in arguments of `dup`?*Created by: equal-l2*
`dup` has `buf: &[u8]` as an argument.
What is the purpose of this argument?
I saw it was used in [termion in the implementation of `isatty`](https://github.com/ticki/termion/blob/master/src/sys/redox/tty.rs#L8).*Created by: equal-l2*
`dup` has `buf: &[u8]` as an argument.
What is the purpose of this argument?
I saw it was used in [termion in the implementation of `isatty`](https://github.com/ticki/termion/blob/master/src/sys/redox/tty.rs#L8).https://gitlab.redox-os.org/redox-os/syscall/-/issues/5Do not compile if the platform is not redox2018-06-14T03:00:11ZJeremy SollerDo not compile if the platform is not redox*Created by: P-E-Meunier*
So, I wrote an alternative build system for Rust, based on Nix, to reduce compilation times when deploying using NixOps.
See https://nest.pijul.com/pmeunier/nix-rust
It works fine, except for one thing: the...*Created by: P-E-Meunier*
So, I wrote an alternative build system for Rust, based on Nix, to reduce compilation times when deploying using NixOps.
See https://nest.pijul.com/pmeunier/nix-rust
It works fine, except for one thing: the `time` crate depends on `redox-syscall` only on when the platform is Redox. Unfortunately, that information is not written to `Cargo.lock` (for a good reason: portability of `Cargo.lock`).
So, I'd love it if this entire crate could compile on all platforms, but just compile nothing on non-Redox platforms.https://gitlab.redox-os.org/redox-os/syscall/-/issues/7Should `Stat` et al be repr(C)?2018-06-15T05:04:53ZIan Douglas ScottShould `Stat` et al be repr(C)?As far as I am aware, the current layout of this stuct and the others is undefined, which makes it impossible to reliably use from C. With Rust code in libc, it could be copied to a C struct, but that would be inefficient.
It is `repr...As far as I am aware, the current layout of this stuct and the others is undefined, which makes it impossible to reliably use from C. With Rust code in libc, it could be copied to a C struct, but that would be inefficient.
It is `repr(packed)`, but I think that still allows Rust to reorder the elements. I don't know if this is currently an issue, but it is in principle undefined behavior. (And it could theoretically change in a future Rust version too, breaking the kernel ABI, though we may not care about that.)