Files
rust/library/std/src/sys/wasi/fs.rs

715 lines
21 KiB
Rust
Raw Normal View History

#![deny(unsafe_op_in_unsafe_fn)]
2021-02-24 10:58:21 -08:00
use super::fd::WasiFd;
use crate::ffi::{CStr, CString, OsStr, OsString};
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
use crate::fmt;
use crate::io::{self, IoSlice, IoSliceMut, SeekFrom};
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
use crate::iter;
use crate::mem::{self, ManuallyDrop};
use crate::os::raw::c_int;
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
use crate::os::wasi::ffi::{OsStrExt, OsStringExt};
use crate::os::wasi::io::{AsFd, AsRawFd, BorrowedFd, FromRawFd, IntoRawFd, RawFd};
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
use crate::path::{Path, PathBuf};
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
use crate::ptr;
use crate::sync::Arc;
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
use crate::sys::time::SystemTime;
use crate::sys::unsupported;
use crate::sys_common::{AsInner, FromInner, IntoInner};
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
pub use crate::sys_common::fs::{remove_dir_all, try_exists};
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub struct File {
fd: WasiFd,
}
#[derive(Clone)]
pub struct FileAttr {
meta: wasi::Filestat,
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub struct ReadDir {
inner: Arc<ReadDirInner>,
cookie: Option<wasi::Dircookie>,
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
buf: Vec<u8>,
offset: usize,
cap: usize,
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
struct ReadDirInner {
root: PathBuf,
dir: File,
}
pub struct DirEntry {
2019-08-19 09:47:58 +03:00
meta: wasi::Dirent,
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
name: Vec<u8>,
inner: Arc<ReadDirInner>,
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
#[derive(Clone, Debug, Default)]
pub struct OpenOptions {
read: bool,
write: bool,
append: bool,
dirflags: wasi::Lookupflags,
fdflags: wasi::Fdflags,
oflags: wasi::Oflags,
2019-08-19 09:47:58 +03:00
rights_base: Option<wasi::Rights>,
rights_inheriting: Option<wasi::Rights>,
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
#[derive(Clone, PartialEq, Eq, Debug)]
pub struct FilePermissions {
readonly: bool,
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
#[derive(PartialEq, Eq, Hash, Debug, Copy, Clone)]
pub struct FileType {
bits: wasi::Filetype,
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
#[derive(Debug)]
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub struct DirBuilder {}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
impl FileAttr {
pub fn size(&self) -> u64 {
self.meta.size
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn perm(&self) -> FilePermissions {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
// not currently implemented in wasi yet
FilePermissions { readonly: false }
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn file_type(&self) -> FileType {
FileType { bits: self.meta.filetype }
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn modified(&self) -> io::Result<SystemTime> {
Ok(SystemTime::from_wasi_timestamp(self.meta.mtim))
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn accessed(&self) -> io::Result<SystemTime> {
Ok(SystemTime::from_wasi_timestamp(self.meta.atim))
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn created(&self) -> io::Result<SystemTime> {
Ok(SystemTime::from_wasi_timestamp(self.meta.ctim))
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn as_wasi(&self) -> &wasi::Filestat {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
&self.meta
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
}
impl FilePermissions {
pub fn readonly(&self) -> bool {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.readonly
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub fn set_readonly(&mut self, readonly: bool) {
self.readonly = readonly;
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
}
impl FileType {
pub fn is_dir(&self) -> bool {
2019-08-19 09:47:58 +03:00
self.bits == wasi::FILETYPE_DIRECTORY
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn is_file(&self) -> bool {
2019-08-19 09:47:58 +03:00
self.bits == wasi::FILETYPE_REGULAR_FILE
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn is_symlink(&self) -> bool {
2019-08-19 09:47:58 +03:00
self.bits == wasi::FILETYPE_SYMBOLIC_LINK
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn bits(&self) -> wasi::Filetype {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.bits
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
}
impl fmt::Debug for ReadDir {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
f.debug_struct("ReadDir").finish_non_exhaustive()
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
}
impl Iterator for ReadDir {
type Item = io::Result<DirEntry>;
fn next(&mut self) -> Option<io::Result<DirEntry>> {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
loop {
// If we've reached the capacity of our buffer then we need to read
// some more from the OS, otherwise we pick up at our old offset.
let offset = if self.offset == self.cap {
let cookie = self.cookie.take()?;
match self.inner.dir.fd.readdir(&mut self.buf, cookie) {
Ok(bytes) => self.cap = bytes,
Err(e) => return Some(Err(e)),
}
self.offset = 0;
self.cookie = Some(cookie);
// If we didn't actually read anything, this is in theory the
// end of the directory.
if self.cap == 0 {
self.cookie = None;
return None;
}
0
} else {
self.offset
};
let data = &self.buf[offset..self.cap];
// If we're not able to read a directory entry then that means it
// must have been truncated at the end of the buffer, so reset our
// offset so we can go back and reread into the buffer, picking up
// where we last left off.
2019-08-19 09:47:58 +03:00
let dirent_size = mem::size_of::<wasi::Dirent>();
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
if data.len() < dirent_size {
assert!(self.cookie.is_some());
assert!(self.buf.len() >= dirent_size);
self.offset = self.cap;
continue;
}
let (dirent, data) = data.split_at(dirent_size);
let dirent = unsafe { ptr::read_unaligned(dirent.as_ptr() as *const wasi::Dirent) };
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
// If the file name was truncated, then we need to reinvoke
// `readdir` so we truncate our buffer to start over and reread this
// descriptor. Note that if our offset is 0 that means the file name
// is massive and we need a bigger buffer.
if data.len() < dirent.d_namlen as usize {
if offset == 0 {
let amt_to_add = self.buf.capacity();
self.buf.extend(iter::repeat(0).take(amt_to_add));
}
assert!(self.cookie.is_some());
self.offset = self.cap;
continue;
}
self.cookie = Some(dirent.d_next);
self.offset = offset + dirent_size + dirent.d_namlen as usize;
let name = &data[..(dirent.d_namlen as usize)];
// These names are skipped on all other platforms, so let's skip
// them here too
if name == b"." || name == b".." {
continue;
}
return Some(Ok(DirEntry {
meta: dirent,
name: name.to_vec(),
inner: self.inner.clone(),
}));
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
}
impl DirEntry {
pub fn path(&self) -> PathBuf {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
let name = OsStr::from_bytes(&self.name);
self.inner.root.join(name)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn file_name(&self) -> OsString {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
OsString::from_vec(self.name.clone())
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn metadata(&self) -> io::Result<FileAttr> {
metadata_at(&self.inner.dir.fd, 0, OsStr::from_bytes(&self.name).as_ref())
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn file_type(&self) -> io::Result<FileType> {
Ok(FileType { bits: self.meta.d_type })
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
2019-08-19 09:47:58 +03:00
pub fn ino(&self) -> wasi::Inode {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.meta.d_ino
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
}
impl OpenOptions {
pub fn new() -> OpenOptions {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
let mut base = OpenOptions::default();
base.dirflags = wasi::LOOKUPFLAGS_SYMLINK_FOLLOW;
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
return base;
}
pub fn read(&mut self, read: bool) {
self.read = read;
}
pub fn write(&mut self, write: bool) {
self.write = write;
}
pub fn truncate(&mut self, truncate: bool) {
self.oflag(wasi::OFLAGS_TRUNC, truncate);
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn create(&mut self, create: bool) {
self.oflag(wasi::OFLAGS_CREAT, create);
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn create_new(&mut self, create_new: bool) {
self.oflag(wasi::OFLAGS_EXCL, create_new);
self.oflag(wasi::OFLAGS_CREAT, create_new);
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn directory(&mut self, directory: bool) {
self.oflag(wasi::OFLAGS_DIRECTORY, directory);
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
fn oflag(&mut self, bit: wasi::Oflags, set: bool) {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
if set {
self.oflags |= bit;
} else {
self.oflags &= !bit;
}
}
pub fn append(&mut self, append: bool) {
self.append = append;
self.fdflag(wasi::FDFLAGS_APPEND, append);
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn dsync(&mut self, set: bool) {
self.fdflag(wasi::FDFLAGS_DSYNC, set);
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn nonblock(&mut self, set: bool) {
self.fdflag(wasi::FDFLAGS_NONBLOCK, set);
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn rsync(&mut self, set: bool) {
self.fdflag(wasi::FDFLAGS_RSYNC, set);
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn sync(&mut self, set: bool) {
self.fdflag(wasi::FDFLAGS_SYNC, set);
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
fn fdflag(&mut self, bit: wasi::Fdflags, set: bool) {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
if set {
self.fdflags |= bit;
} else {
self.fdflags &= !bit;
}
}
2019-08-19 09:47:58 +03:00
pub fn fs_rights_base(&mut self, rights: wasi::Rights) {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.rights_base = Some(rights);
}
2019-08-19 09:47:58 +03:00
pub fn fs_rights_inheriting(&mut self, rights: wasi::Rights) {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.rights_inheriting = Some(rights);
}
2019-08-19 09:47:58 +03:00
fn rights_base(&self) -> wasi::Rights {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
if let Some(rights) = self.rights_base {
return rights;
}
// If rights haven't otherwise been specified try to pick a reasonable
// set. This can always be overridden by users via extension traits, and
// implementations may give us fewer rights silently than we ask for. So
// given that, just look at `read` and `write` and bucket permissions
// based on that.
let mut base = 0;
if self.read {
base |= wasi::RIGHTS_FD_READ;
base |= wasi::RIGHTS_FD_READDIR;
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
if self.write || self.append {
base |= wasi::RIGHTS_FD_WRITE;
base |= wasi::RIGHTS_FD_DATASYNC;
base |= wasi::RIGHTS_FD_ALLOCATE;
base |= wasi::RIGHTS_FD_FILESTAT_SET_SIZE;
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
// FIXME: some of these should probably be read-only or write-only...
base |= wasi::RIGHTS_FD_ADVISE;
base |= wasi::RIGHTS_FD_FDSTAT_SET_FLAGS;
base |= wasi::RIGHTS_FD_FILESTAT_GET;
base |= wasi::RIGHTS_FD_FILESTAT_SET_TIMES;
base |= wasi::RIGHTS_FD_SEEK;
base |= wasi::RIGHTS_FD_SYNC;
base |= wasi::RIGHTS_FD_TELL;
base |= wasi::RIGHTS_PATH_CREATE_DIRECTORY;
base |= wasi::RIGHTS_PATH_CREATE_FILE;
base |= wasi::RIGHTS_PATH_FILESTAT_GET;
base |= wasi::RIGHTS_PATH_LINK_SOURCE;
base |= wasi::RIGHTS_PATH_LINK_TARGET;
base |= wasi::RIGHTS_PATH_OPEN;
base |= wasi::RIGHTS_PATH_READLINK;
base |= wasi::RIGHTS_PATH_REMOVE_DIRECTORY;
base |= wasi::RIGHTS_PATH_RENAME_SOURCE;
base |= wasi::RIGHTS_PATH_RENAME_TARGET;
base |= wasi::RIGHTS_PATH_SYMLINK;
base |= wasi::RIGHTS_PATH_UNLINK_FILE;
base |= wasi::RIGHTS_POLL_FD_READWRITE;
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
return base;
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
2019-08-19 09:47:58 +03:00
fn rights_inheriting(&self) -> wasi::Rights {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.rights_inheriting.unwrap_or_else(|| self.rights_base())
}
pub fn lookup_flags(&mut self, flags: wasi::Lookupflags) {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.dirflags = flags;
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
impl File {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub fn open(path: &Path, opts: &OpenOptions) -> io::Result<File> {
let (dir, file) = open_parent(path)?;
open_at(&dir, &file, opts)
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn open_at(&self, path: &Path, opts: &OpenOptions) -> io::Result<File> {
open_at(&self.fd, path, opts)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn file_attr(&self) -> io::Result<FileAttr> {
2019-08-21 17:57:22 +03:00
self.fd.filestat_get().map(|meta| FileAttr { meta })
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn metadata_at(&self, flags: wasi::Lookupflags, path: &Path) -> io::Result<FileAttr> {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
metadata_at(&self.fd, flags, path)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn fsync(&self) -> io::Result<()> {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.fd.sync()
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn datasync(&self) -> io::Result<()> {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.fd.datasync()
}
pub fn truncate(&self, size: u64) -> io::Result<()> {
self.fd.filestat_set_size(size)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub fn read(&self, buf: &mut [u8]) -> io::Result<usize> {
self.read_vectored(&mut [IoSliceMut::new(buf)])
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn read_vectored(&self, bufs: &mut [IoSliceMut<'_>]) -> io::Result<usize> {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.fd.read(bufs)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
#[inline]
2020-03-11 18:02:52 -07:00
pub fn is_read_vectored(&self) -> bool {
true
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub fn write(&self, buf: &[u8]) -> io::Result<usize> {
self.write_vectored(&[IoSlice::new(buf)])
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn write_vectored(&self, bufs: &[IoSlice<'_>]) -> io::Result<usize> {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.fd.write(bufs)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
#[inline]
2020-03-11 18:02:52 -07:00
pub fn is_write_vectored(&self) -> bool {
true
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
pub fn flush(&self) -> io::Result<()> {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
Ok(())
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub fn seek(&self, pos: SeekFrom) -> io::Result<u64> {
self.fd.seek(pos)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn duplicate(&self) -> io::Result<File> {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
// https://github.com/CraneStation/wasmtime/blob/master/docs/WASI-rationale.md#why-no-dup
unsupported()
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn set_permissions(&self, _perm: FilePermissions) -> io::Result<()> {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
// Permissions haven't been fully figured out in wasi yet, so this is
// likely temporary
unsupported()
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn read_link(&self, file: &Path) -> io::Result<PathBuf> {
read_link(&self.fd, file)
}
}
impl AsInner<WasiFd> for File {
fn as_inner(&self) -> &WasiFd {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
&self.fd
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
impl IntoInner<WasiFd> for File {
fn into_inner(self) -> WasiFd {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
self.fd
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
impl FromInner<WasiFd> for File {
fn from_inner(fd: WasiFd) -> File {
File { fd }
}
}
impl AsFd for File {
fn as_fd(&self) -> BorrowedFd<'_> {
self.fd.as_fd()
}
}
impl AsRawFd for File {
fn as_raw_fd(&self) -> RawFd {
self.fd.as_raw_fd()
}
}
impl IntoRawFd for File {
fn into_raw_fd(self) -> RawFd {
self.fd.into_raw_fd()
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
}
impl FromRawFd for File {
unsafe fn from_raw_fd(raw_fd: RawFd) -> Self {
unsafe { Self { fd: FromRawFd::from_raw_fd(raw_fd) } }
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
impl DirBuilder {
pub fn new() -> DirBuilder {
DirBuilder {}
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub fn mkdir(&self, p: &Path) -> io::Result<()> {
let (dir, file) = open_parent(p)?;
dir.create_directory(osstr2str(file.as_ref())?)
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
impl fmt::Debug for File {
fn fmt(&self, f: &mut fmt::Formatter<'_>) -> fmt::Result {
f.debug_struct("File").field("fd", &self.as_raw_fd()).finish()
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
}
pub fn readdir(p: &Path) -> io::Result<ReadDir> {
let mut opts = OpenOptions::new();
opts.directory(true);
opts.read(true);
let dir = File::open(p, &opts)?;
Ok(ReadDir {
cookie: Some(0),
buf: vec![0; 128],
offset: 0,
cap: 0,
inner: Arc::new(ReadDirInner { dir, root: p.to_path_buf() }),
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
})
}
pub fn unlink(p: &Path) -> io::Result<()> {
let (dir, file) = open_parent(p)?;
dir.unlink_file(osstr2str(file.as_ref())?)
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn rename(old: &Path, new: &Path) -> io::Result<()> {
let (old, old_file) = open_parent(old)?;
let (new, new_file) = open_parent(new)?;
old.rename(osstr2str(old_file.as_ref())?, &new, osstr2str(new_file.as_ref())?)
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
pub fn set_perm(_p: &Path, _perm: FilePermissions) -> io::Result<()> {
// Permissions haven't been fully figured out in wasi yet, so this is
// likely temporary
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
unsupported()
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub fn rmdir(p: &Path) -> io::Result<()> {
let (dir, file) = open_parent(p)?;
dir.remove_directory(osstr2str(file.as_ref())?)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub fn readlink(p: &Path) -> io::Result<PathBuf> {
let (dir, file) = open_parent(p)?;
read_link(&dir, &file)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
fn read_link(fd: &WasiFd, file: &Path) -> io::Result<PathBuf> {
// Try to get a best effort initial capacity for the vector we're going to
// fill. Note that if it's not a symlink we don't use a file to avoid
// allocating gigabytes if you read_link a huge movie file by accident.
// Additionally we add 1 to the initial size so if it doesn't change until
// when we call `readlink` the returned length will be less than the
// capacity, guaranteeing that we got all the data.
let meta = metadata_at(fd, 0, file)?;
let initial_size = if meta.file_type().is_symlink() {
(meta.size() as usize).saturating_add(1)
} else {
1 // this'll fail in just a moment
};
// Now that we have an initial guess of how big to make our buffer, call
// `readlink` in a loop until it fails or reports it filled fewer bytes than
// we asked for, indicating we got everything.
let file = osstr2str(file.as_ref())?;
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
let mut destination = vec![0u8; initial_size];
loop {
let len = fd.readlink(file, &mut destination)?;
if len < destination.len() {
destination.truncate(len);
destination.shrink_to_fit();
return Ok(PathBuf::from(OsString::from_vec(destination)));
}
let amt_to_add = destination.len();
destination.extend(iter::repeat(0).take(amt_to_add));
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
2020-11-14 14:40:08 -08:00
pub fn symlink(original: &Path, link: &Path) -> io::Result<()> {
let (link, link_file) = open_parent(link)?;
link.symlink(osstr2str(original.as_ref())?, osstr2str(link_file.as_ref())?)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
2020-11-14 14:40:08 -08:00
pub fn link(original: &Path, link: &Path) -> io::Result<()> {
let (original, original_file) = open_parent(original)?;
let (link, link_file) = open_parent(link)?;
// Pass 0 as the flags argument, meaning don't follow symlinks.
2021-02-19 07:31:01 -08:00
original.link(0, osstr2str(original_file.as_ref())?, &link, osstr2str(link_file.as_ref())?)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub fn stat(p: &Path) -> io::Result<FileAttr> {
let (dir, file) = open_parent(p)?;
metadata_at(&dir, wasi::LOOKUPFLAGS_SYMLINK_FOLLOW, &file)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
pub fn lstat(p: &Path) -> io::Result<FileAttr> {
let (dir, file) = open_parent(p)?;
metadata_at(&dir, 0, &file)
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
fn metadata_at(fd: &WasiFd, flags: wasi::Lookupflags, path: &Path) -> io::Result<FileAttr> {
let meta = fd.path_filestat_get(flags, osstr2str(path.as_ref())?)?;
Ok(FileAttr { meta })
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn canonicalize(_p: &Path) -> io::Result<PathBuf> {
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
// This seems to not be in wasi's API yet, and we may need to end up
// emulating it ourselves. For now just return an error.
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
unsupported()
}
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
fn open_at(fd: &WasiFd, path: &Path, opts: &OpenOptions) -> io::Result<File> {
let fd = fd.open(
opts.dirflags,
osstr2str(path.as_ref())?,
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
opts.oflags,
opts.rights_base(),
opts.rights_inheriting(),
opts.fdflags,
)?;
Ok(File { fd })
}
/// Attempts to open a bare path `p`.
///
/// WASI has no fundamental capability to do this. All syscalls and operations
/// are relative to already-open file descriptors. The C library, however,
2020-07-07 20:48:15 -04:00
/// manages a map of pre-opened file descriptors to their path, and then the C
/// library provides an API to look at this. In other words, when you want to
/// open a path `p`, you have to find a previously opened file descriptor in a
/// global table and then see if `p` is relative to that file descriptor.
///
/// This function, if successful, will return two items:
///
2020-07-07 20:48:15 -04:00
/// * The first is a `ManuallyDrop<WasiFd>`. This represents a pre-opened file
/// descriptor which we don't have ownership of, but we can use. You shouldn't
/// actually drop the `fd`.
///
/// * The second is a path that should be a part of `p` and represents a
/// relative traversal from the file descriptor specified to the desired
/// location `p`.
///
/// If successful you can use the returned file descriptor to perform
/// file-descriptor-relative operations on the path returned as well. The
/// `rights` argument indicates what operations are desired on the returned file
/// descriptor, and if successful the returned file descriptor should have the
/// appropriate rights for performing `rights` actions.
///
/// Note that this can fail if `p` doesn't look like it can be opened relative
2020-07-07 20:48:15 -04:00
/// to any pre-opened file descriptor.
fn open_parent(p: &Path) -> io::Result<(ManuallyDrop<WasiFd>, PathBuf)> {
let p = CString::new(p.as_os_str().as_bytes())?;
let mut buf = Vec::<u8>::with_capacity(512);
loop {
unsafe {
let mut relative_path = buf.as_ptr().cast();
let mut abs_prefix = ptr::null();
let fd = __wasilibc_find_relpath(
p.as_ptr(),
&mut abs_prefix,
&mut relative_path,
buf.capacity(),
);
if fd == -1 {
if io::Error::last_os_error().raw_os_error() == Some(libc::ENOMEM) {
// Trigger the internal buffer resizing logic of `Vec` by requiring
// more space than the current capacity.
let cap = buf.capacity();
buf.set_len(cap);
buf.reserve(1);
continue;
}
let msg = format!(
"failed to find a pre-opened file descriptor \
through which {:?} could be opened",
p
);
return Err(io::Error::new(io::ErrorKind::Uncategorized, msg));
}
let relative = CStr::from_ptr(relative_path).to_bytes().to_vec();
return Ok((
ManuallyDrop::new(WasiFd::from_raw_fd(fd as c_int)),
PathBuf::from(OsString::from_vec(relative)),
));
wasi: Fill out `std::fs` module for WASI This commit fills out the `std::fs` module and implementation for WASI. Not all APIs are implemented, such as permissions-related ones and `canonicalize`, but all others APIs have been implemented and very lightly tested so far. We'll eventually want to run a more exhaustive test suite! For now the highlights of this commit are: * The `std::fs::File` type is now backed by `WasiFd`, a raw WASI file descriptor. * All APIs in `std::fs` (except permissions/canonicalize) have implementations for the WASI target. * A suite of unstable extension traits were added to `std::os::wasi::fs`. These traits expose the raw filesystem functionality of WASI, namely `*at` syscalls (opening a file relative to an already opened one, for example). Additionally metadata only available on wasi is exposed through these traits. Perhaps one of the most notable parts is the implementation of path-taking APIs. WASI actually has no fundamental API that just takes a path, but rather everything is relative to a previously opened file descriptor. To allow existing APIs to work (that only take a path) WASI has a few syscalls to learn about "pre opened" file descriptors by the runtime. We use these to build a map of existing directory names to file descriptors, and then when using a path we try to anchor it at an already-opened file. This support is very rudimentary though and is intended to be shared with C since it's likely to be so tricky. For now though the C library doesn't expose quite an API for us to use, so we implement it for now and will swap it out as soon as one is available.
2019-03-29 15:57:14 -07:00
}
}
extern "C" {
pub fn __wasilibc_find_relpath(
path: *const libc::c_char,
abs_prefix: *mut *const libc::c_char,
relative_path: *mut *const libc::c_char,
relative_path_len: libc::size_t,
) -> libc::c_int;
}
Add a new wasm32-unknown-wasi target This commit adds a new wasm32-based target distributed through rustup, supported in the standard library, and implemented in the compiler. The `wasm32-unknown-wasi` target is intended to be a WebAssembly target which matches the [WASI proposal recently announced.][LINK]. In summary the WASI target is an effort to define a standard set of syscalls for WebAssembly modules, allowing WebAssembly modules to not only be portable across architectures but also be portable across environments implementing this standard set of system calls. The wasi target in libstd is still somewhat bare bones. This PR does not fill out the filesystem, networking, threads, etc. Instead it only provides the most basic of integration with the wasi syscalls, enabling features like: * `Instant::now` and `SystemTime::now` work * `env::args` is hooked up * `env::vars` will look up environment variables * `println!` will print to standard out * `process::{exit, abort}` should be hooked up appropriately None of these APIs can work natively on the `wasm32-unknown-unknown` target, but with the assumption of the WASI set of syscalls we're able to provide implementations of these syscalls that engines can implement. Currently the primary engine implementing wasi is [wasmtime], but more will surely emerge! In terms of future development of libstd, I think this is something we'll probably want to discuss. The purpose of the WASI target is to provide a standardized set of syscalls, but it's *also* to provide a standard C sysroot for compiling C/C++ programs. This means it's intended that functions like `read` and `write` are implemented for this target with a relatively standard definition and implementation. It's unclear, therefore, how we want to expose file descriptors and how we'll want to implement system primitives. For example should `std::fs::File` have a libc-based file descriptor underneath it? The raw wasi file descriptor? We'll see! Currently these details are all intentionally hidden and things we can change over time. A `WasiFd` sample struct was added to the standard library as part of this commit, but it's not currently used. It shows how all the wasi syscalls could be ergonomically bound in Rust, and they offer a possible implementation of primitives like `std::fs::File` if we bind wasi file descriptors exactly. Apart from the standard library, there's also the matter of how this target is integrated with respect to its C standard library. The reference sysroot, for example, provides managment of standard unix file descriptors and also standard APIs like `open` (as opposed to the relative `openat` inspiration for the wasi ssycalls). Currently the standard library relies on the C sysroot symbols for operations such as environment management, process exit, and `read`/`write` of stdio fds. We want these operations in Rust to be interoperable with C if they're used in the same process. Put another way, if Rust and C are linked into the same WebAssembly binary they should work together, but that requires that the same C standard library is used. We also, however, want the `wasm32-unknown-wasi` target to be usable-by-default with the Rust compiler without requiring a separate toolchain to get downloaded and configured. With that in mind, there's two modes of operation for the `wasm32-unknown-wasi` target: 1. By default the C standard library is statically provided inside of `liblibc.rlib` distributed as part of the sysroot. This means that you can `rustc foo.wasm --target wasm32-unknown-unknown` and you're good to go, a fully workable wasi binary pops out. This is incompatible with linking in C code, however, which may be compiled against a different sysroot than the Rust code was previously compiled against. In this mode the default of `rust-lld` is used to link binaries. 2. For linking with C code, the `-C target-feature=-crt-static` flag needs to be passed. This takes inspiration from the musl target for this flag, but the idea is that you're no longer using the provided static C runtime, but rather one will be provided externally. This flag is intended to also get coupled with an external `clang` compiler configured with its own sysroot. Therefore you'll typically use this flag with `-C linker=/path/to/clang-script-wrapper`. Using this mode the Rust code will continue to reference standard C symbols, but the definition will be pulled in by the linker configured. Alright so that's all the current state of this PR. I suspect we'll definitely want to discuss this before landing of course! This PR is coupled with libc changes as well which I'll be posting shortly. [LINK]: [wasmtime]:
2019-02-13 10:02:22 -08:00
}
pub fn osstr2str(f: &OsStr) -> io::Result<&str> {
f.to_str()
.ok_or_else(|| io::Error::new_const(io::ErrorKind::Uncategorized, &"input must be utf-8"))
}
pub fn copy(from: &Path, to: &Path) -> io::Result<u64> {
use crate::fs::File;
let mut reader = File::open(from)?;
let mut writer = File::create(to)?;
io::copy(&mut reader, &mut writer)
}