rust/tests/ui/mir/mir_misc_casts.rs

Ignoring revisions in .git-blame-ignore-revs. Click here to bypass and see the normal blame view.

321 lines
9.4 KiB
Rust
Raw Normal View History

// run-pass
fn func(){}
const STR: &'static str = "hello";
const BSTR: &'static [u8; 5] = b"hello";
fn from_ptr()
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, *const ()) {
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let f = 1_usize as *const String;
let c1 = f as isize;
let c2 = f as usize;
let c3 = f as i8;
let c4 = f as i16;
let c5 = f as i32;
let c6 = f as i64;
let c7 = f as u8;
let c8 = f as u16;
let c9 = f as u32;
let c10 = f as u64;
let c11 = f as *const ();
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11)
}
fn from_1()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1 as isize;
let c2 = 1 as usize;
let c3 = 1 as i8;
let c4 = 1 as i16;
let c5 = 1 as i32;
let c6 = 1 as i64;
let c7 = 1 as u8;
let c8 = 1 as u16;
let c9 = 1 as u32;
let c10 = 1 as u64;
let c11 = 1 as f32;
let c12 = 1 as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1 as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_1usize()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1_usize as isize;
let c2 = 1_usize as usize;
let c3 = 1_usize as i8;
let c4 = 1_usize as i16;
let c5 = 1_usize as i32;
let c6 = 1_usize as i64;
let c7 = 1_usize as u8;
let c8 = 1_usize as u16;
let c9 = 1_usize as u32;
let c10 = 1_usize as u64;
let c11 = 1_usize as f32;
let c12 = 1_usize as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1_usize as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_1isize()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1_isize as isize;
let c2 = 1_isize as usize;
let c3 = 1_isize as i8;
let c4 = 1_isize as i16;
let c5 = 1_isize as i32;
let c6 = 1_isize as i64;
let c7 = 1_isize as u8;
let c8 = 1_isize as u16;
let c9 = 1_isize as u32;
let c10 = 1_isize as u64;
let c11 = 1_isize as f32;
let c12 = 1_isize as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1_isize as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_1u8()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1_u8 as isize;
let c2 = 1_u8 as usize;
let c3 = 1_u8 as i8;
let c4 = 1_u8 as i16;
let c5 = 1_u8 as i32;
let c6 = 1_u8 as i64;
let c7 = 1_u8 as u8;
let c8 = 1_u8 as u16;
let c9 = 1_u8 as u32;
let c10 = 1_u8 as u64;
let c11 = 1_u8 as f32;
let c12 = 1_u8 as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1_u8 as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_1i8()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1_i8 as isize;
let c2 = 1_i8 as usize;
let c3 = 1_i8 as i8;
let c4 = 1_i8 as i16;
let c5 = 1_i8 as i32;
let c6 = 1_i8 as i64;
let c7 = 1_i8 as u8;
let c8 = 1_i8 as u16;
let c9 = 1_i8 as u32;
let c10 = 1_i8 as u64;
let c11 = 1_i8 as f32;
let c12 = 1_i8 as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1_i8 as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_1u16()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1_u16 as isize;
let c2 = 1_u16 as usize;
let c3 = 1_u16 as i8;
let c4 = 1_u16 as i16;
let c5 = 1_u16 as i32;
let c6 = 1_u16 as i64;
let c7 = 1_u16 as u8;
let c8 = 1_u16 as u16;
let c9 = 1_u16 as u32;
let c10 = 1_u16 as u64;
let c11 = 1_u16 as f32;
let c12 = 1_u16 as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1_u16 as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_1i16()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1_i16 as isize;
let c2 = 1_i16 as usize;
let c3 = 1_i16 as i8;
let c4 = 1_i16 as i16;
let c5 = 1_i16 as i32;
let c6 = 1_i16 as i64;
let c7 = 1_i16 as u8;
let c8 = 1_i16 as u16;
let c9 = 1_i16 as u32;
let c10 = 1_i16 as u64;
let c11 = 1_i16 as f32;
let c12 = 1_i16 as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1_i16 as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_1u32()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1_u32 as isize;
let c2 = 1_u32 as usize;
let c3 = 1_u32 as i8;
let c4 = 1_u32 as i16;
let c5 = 1_u32 as i32;
let c6 = 1_u32 as i64;
let c7 = 1_u32 as u8;
let c8 = 1_u32 as u16;
let c9 = 1_u32 as u32;
let c10 = 1_u32 as u64;
let c11 = 1_u32 as f32;
let c12 = 1_u32 as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1_u32 as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_1i32()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1_i32 as isize;
let c2 = 1_i32 as usize;
let c3 = 1_i32 as i8;
let c4 = 1_i32 as i16;
let c5 = 1_i32 as i32;
let c6 = 1_i32 as i64;
let c7 = 1_i32 as u8;
let c8 = 1_i32 as u16;
let c9 = 1_i32 as u32;
let c10 = 1_i32 as u64;
let c11 = 1_i32 as f32;
let c12 = 1_i32 as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1_i32 as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_1u64()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1_u64 as isize;
let c2 = 1_u64 as usize;
let c3 = 1_u64 as i8;
let c4 = 1_u64 as i16;
let c5 = 1_u64 as i32;
let c6 = 1_u64 as i64;
let c7 = 1_u64 as u8;
let c8 = 1_u64 as u16;
let c9 = 1_u64 as u32;
let c10 = 1_u64 as u64;
let c11 = 1_u64 as f32;
let c12 = 1_u64 as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1_u64 as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_1i64()
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64, *const String) {
let c1 = 1_i64 as isize;
let c2 = 1_i64 as usize;
let c3 = 1_i64 as i8;
let c4 = 1_i64 as i16;
let c5 = 1_i64 as i32;
let c6 = 1_i64 as i64;
let c7 = 1_i64 as u8;
let c8 = 1_i64 as u16;
let c9 = 1_i64 as u32;
let c10 = 1_i64 as u64;
let c11 = 1_i64 as f32;
let c12 = 1_i64 as f64;
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let c13 = 1_i64 as *const String;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12, c13)
}
fn from_bool()
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64) {
let c1 = true as isize;
let c2 = true as usize;
let c3 = true as i8;
let c4 = true as i16;
let c5 = true as i32;
let c6 = true as i64;
let c7 = true as u8;
let c8 = true as u16;
let c9 = true as u32;
let c10 = true as u64;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10)
}
fn from_1f32()
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64) {
let c1 = 1.0_f32 as isize;
let c2 = 1.0_f32 as usize;
let c3 = 1.0_f32 as i8;
let c4 = 1.0_f32 as i16;
let c5 = 1.0_f32 as i32;
let c6 = 1.0_f32 as i64;
let c7 = 1.0_f32 as u8;
let c8 = 1.0_f32 as u16;
let c9 = 1.0_f32 as u32;
let c10 = 1.0_f32 as u64;
let c11 = 1.0_f32 as f32;
let c12 = 1.0_f32 as f64;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12)
}
fn from_1f64()
-> (isize, usize, i8, i16, i32, i64, u8, u16, u32, u64, f32, f64) {
let c1 = 1.0f64 as isize;
let c2 = 1.0f64 as usize;
let c3 = 1.0f64 as i8;
let c4 = 1.0f64 as i16;
let c5 = 1.0f64 as i32;
let c6 = 1.0f64 as i64;
let c7 = 1.0f64 as u8;
let c8 = 1.0f64 as u16;
let c9 = 1.0f64 as u32;
let c10 = 1.0f64 as u64;
let c11 = 1.0f64 as f32;
let c12 = 1.0f64 as f64;
(c1, c2, c3, c4, c5, c6, c7, c8, c9, c10, c11, c12)
}
fn other_casts()
-> (*const u8, *const isize, *const u8, *const u8) {
let c1 = func as *const u8;
let c2 = c1 as *const isize;
let r = &42u32;
let _ = r as *const u32;
// fat-ptr -> fat-ptr -> fat-raw-ptr -> thin-ptr
let c3 = STR as &str as *const str as *const u8;
let c4 = BSTR as *const [u8] as *const [u16] as *const u8;
(c1, c2, c3, c4)
}
pub fn assert_eq_13(l: (isize, usize, i8, i16, i32, i64, u8,
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
u16, u32, u64, f32, f64, *const String),
r: (isize, usize, i8, i16, i32, i64, u8,
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
u16, u32, u64, f32, f64, *const String)) -> bool {
let (l1, l2, l3, l4, l5, l6, l7, l8, l9, l10, l11, l12, l13) = l;
let (r1, r2, r3, r4, r5, r6, r7, r8, r9, r10, r11, r12, r13) = r;
l1 == r1 && l2 == r2 && l3 == r3 && l4 == r4 && l5 == r5 && l6 == r6 && l7 == r7 &&
l8 == r8 && l9 == r9 && l10 == r10 && l11 == r11 && l12 == r12 && l13 == r13
}
pub fn main() {
std: Add a new wasm32-unknown-unknown target This commit adds a new target to the compiler: wasm32-unknown-unknown. This target is a reimagining of what it looks like to generate WebAssembly code from Rust. Instead of using Emscripten which can bring with it a weighty runtime this instead is a target which uses only the LLVM backend for WebAssembly and a "custom linker" for now which will hopefully one day be direct calls to lld. Notable features of this target include: * There is zero runtime footprint. The target assumes nothing exists other than the wasm32 instruction set. * There is zero toolchain footprint beyond adding the target. No custom linker is needed, rustc contains everything. * Very small wasm modules can be generated directly from Rust code using this target. * Most of the standard library is stubbed out to return an error, but anything related to allocation works (aka `HashMap`, `Vec`, etc). * Naturally, any `#[no_std]` crate should be 100% compatible with this new target. This target is currently somewhat janky due to how linking works. The "linking" is currently unconditional whole program LTO (aka LLVM is being used as a linker). Naturally that means compiling programs is pretty slow! Eventually though this target should have a linker. This target is also intended to be quite experimental. I'm hoping that this can act as a catalyst for further experimentation in Rust with WebAssembly. Breaking changes are very likely to land to this target, so it's not recommended to rely on it in any critical capacity yet. We'll let you know when it's "production ready". --- Currently testing-wise this target is looking pretty good but isn't complete. I've got almost the entire `run-pass` test suite working with this target (lots of tests ignored, but many passing as well). The `core` test suite is still getting LLVM bugs fixed to get that working and will take some time. Relatively simple programs all seem to work though! --- It's worth nothing that you may not immediately see the "smallest possible wasm module" for the input you feed to rustc. For various reasons it's very difficult to get rid of the final "bloat" in vanilla rustc (again, a real linker should fix all this). For now what you'll have to do is: cargo install --git https://github.com/alexcrichton/wasm-gc wasm-gc foo.wasm bar.wasm And then `bar.wasm` should be the smallest we can get it! --- In any case for now I'd love feedback on this, particularly on the various integration points if you've got better ideas of how to approach them!
2017-10-23 03:01:00 +00:00
let f = 1_usize as *const String;
let t13 = (1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1.0, 1.0, f);
let t12 = (1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1.0, 1.0);
assert_eq_13(from_1(), t13);
assert_eq_13(from_1usize(), t13);
assert_eq_13(from_1isize(), t13);
assert_eq_13(from_1u8(), t13);
assert_eq_13(from_1i8(), t13);
assert_eq_13(from_1u16(), t13);
assert_eq_13(from_1i16(), t13);
assert_eq_13(from_1u32(), t13);
assert_eq_13(from_1i32(), t13);
assert_eq_13(from_1u64(), t13);
assert_eq_13(from_1i64(), t13);
assert_eq!(from_1f32(), t12);
assert_eq!(from_1f64(), t12);
assert_eq!(from_ptr(), (1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1 as *const ()));
assert_eq!(from_bool(), (1, 1, 1, 1, 1, 1, 1, 1, 1, 1));
assert_eq!(other_casts(), (func as *const u8, func as *const isize,
STR as *const str as *const u8, BSTR as *const [u8] as *const u8));
}