std: Cut down #[inline] annotations where not necessary
This PR cuts down on a large number of `#[inline(always)]` and `#[inline]`
annotations in libcore for various core functions. The `#[inline(always)]`
annotation is almost never needed and is detrimental to debug build times as it
forces LLVM to perform inlining when it otherwise wouldn't need to in debug
builds. Additionally `#[inline]` is an unnecessary annoation on almost all
generic functions because the function will already be monomorphized into other
codegen units and otherwise rarely needs the extra "help" from us to tell LLVM
to inline something.
Overall this PR cut the compile time of a [microbenchmark][1] by 30% from 1s to
0.7s.
[1]: https://gist.github.com/alexcrichton/a7d70319a45aa60cf36a6a7bf540dd3a
Pass debugging arguments to emcc
Tells emcc to enable assertions and debugging information for
wasm32-experimental-emscripten. This makes the codegen issues caused by
LLVM bug 33824 manifest more frequently at runtime and improves the wasm
debugging experience.
Less verbose output for unused arguments
Closes#37718
This is my first contribution to rust, so sorry if I'm missing anything!
The output now looks like this:
<img width="831" alt="screen shot 2017-07-18 at 5 01 32 pm" src="https://user-images.githubusercontent.com/12972285/28347566-dbfa9962-6c05-11e7-8730-c2e8062a04cc.png">
It's not the prettiest, but whenever #41850 gets resolved, this should be able to be improved.
**EDIT:** This also does not seem
r? @Mark-Simulacrum
make JSON error byte position start at top of file
The `hi` and `lo` offsets in a span are relative to a `CodeMap`, but this
doesn't seem to be terribly useful for tool consumers who don't have the
codemap, but might want the byte offset within an actual file?
I couldn't get @killercup's [example](https://github.com/rust-lang/rust/issues/35164#issuecomment-301436519) to run, perhaps due to the limitations of the merely-stage-1 compiler that I built (error was `libproc_macro-456500c7095d8fbe.so: cannot open shared object file: No such file or directory`)??—but a dummy project confirms that the byte offsets have successfully been changed to be file-relative—
**Before:**
```
$ cargo run --message-format json
Compiling byte_json v0.1.0 (file:///home/ubuntu/byte_json)
{"message":{"children":[{"children":[],"code":null,"level":"note","message":"#[warn(dead_code)] on by default","rendered":null,"spans":[]}],"code":null,"level":"warning","message":"function is never used: `rah`","rendered":null,"spans":[{"byte_end":100,"byte_start":67,"column_end":2,"column_start":1,"expansion":null,"file_name":"src/foo.rs","is_primary":true,"label":null,"line_end":5,"line_start":3,"suggested_replacement":null,"text":[{"highlight_end":11,"highlight_start":1,"text":"fn rah() {"},{"highlight_end":21,"highlight_start":1,"text":" println!(\"rah!\")"},{"highlight_end":2,"highlight_start":1,"text":"}"}]}]},"package_id":"byte_json 0.1.0 (path+file:///home/ubuntu/byte_json)","reason":"compiler-message","target":{"crate_types":["bin"],"kind":["bin"],"name":"byte_json","src_path":"/home/ubuntu/byte_json/src/main.rs"}}
{"message":{"children":[{"children":[],"code":null,"level":"note","message":"#[warn(dead_code)] on by default","rendered":null,"spans":[]}],"code":null,"level":"warning","message":"function is never used: `alas`","rendered":null,"spans":[{"byte_end":137,"byte_start":102,"column_end":2,"column_start":1,"expansion":null,"file_name":"src/bar.rs","is_primary":true,"label":null,"line_end":3,"line_start":1,"suggested_replacement":null,"text":[{"highlight_end":12,"highlight_start":1,"text":"fn alas() {"},{"highlight_end":22,"highlight_start":1,"text":" println!(\"alas\");"},{"highlight_end":2,"highlight_start":1,"text":"}"}]}]},"package_id":"byte_json 0.1.0 (path+file:///home/ubuntu/byte_json)","reason":"compiler-message","target":{"crate_types":["bin"],"kind":["bin"],"name":"byte_json","src_path":"/home/ubuntu/byte_json/src/main.rs"}}
{"features":[],"filenames":["/home/ubuntu/byte_json/target/debug/byte_json"],"fresh":false,"package_id":"byte_json 0.1.0 (path+file:///home/ubuntu/byte_json)","profile":{"debug_assertions":true,"debuginfo":2,"opt_level":"0","overflow_checks":true,"test":false},"reason":"compiler-artifact","target":{"crate_types":["bin"],"kind":["bin"],"name":"byte_json","src_path":"/home/ubuntu/byte_json/src/main.rs"}}
Finished dev [unoptimized + debuginfo] target(s) in 0.36 secs
Running `target/debug/byte_json`
Hello, world!
```
**After:**
```
$ RUSTC=../rust/build/x86_64-unknown-linux-gnu/stage1/bin/rustc cargo run --message-format json
Compiling byte_json v0.1.0 (file:///home/ubuntu/byte_json)
{"message":{"children":[{"children":[],"code":null,"level":"note","message":"#[warn(dead_code)] on by default","rendered":null,"spans":[]}],"code":null,"level":"warning","message":"function is never used: `rah`","rendered":null,"spans":[{"byte_end":35,"byte_start":2,"column_end":2,"column_start":1,"expansion":null,"file_name":"src/foo.rs","is_primary":true,"label":null,"line_end":5,"line_start":3,"suggested_replacement":null,"text":[{"highlight_end":11,"highlight_start":1,"text":"fn rah() {"},{"highlight_end":21,"highlight_start":1,"text":" println!(\"rah!\")"},{"highlight_end":2,"highlight_start":1,"text":"}"}]}]},"package_id":"byte_json 0.1.0 (path+file:///home/ubuntu/byte_json)","reason":"compiler-message","target":{"crate_types":["bin"],"kind":["bin"],"name":"byte_json","src_path":"/home/ubuntu/byte_json/src/main.rs"}}
{"message":{"children":[{"children":[],"code":null,"level":"note","message":"#[warn(dead_code)] on by default","rendered":null,"spans":[]}],"code":null,"level":"warning","message":"function is never used: `alas`","rendered":null,"spans":[{"byte_end":35,"byte_start":0,"column_end":2,"column_start":1,"expansion":null,"file_name":"src/bar.rs","is_primary":true,"label":null,"line_end":3,"line_start":1,"suggested_replacement":null,"text":[{"highlight_end":12,"highlight_start":1,"text":"fn alas() {"},{"highlight_end":22,"highlight_start":1,"text":" println!(\"alas\");"},{"highlight_end":2,"highlight_start":1,"text":"}"}]}]},"package_id":"byte_json 0.1.0 (path+file:///home/ubuntu/byte_json)","reason":"compiler-message","target":{"crate_types":["bin"],"kind":["bin"],"name":"byte_json","src_path":"/home/ubuntu/byte_json/src/main.rs"}}
{"features":[],"filenames":["/home/ubuntu/byte_json/target/debug/byte_json"],"fresh":false,"package_id":"byte_json 0.1.0 (path+file:///home/ubuntu/byte_json)","profile":{"debug_assertions":true,"debuginfo":2,"opt_level":"0","overflow_checks":true,"test":false},"reason":"compiler-artifact","target":{"crate_types":["bin"],"kind":["bin"],"name":"byte_json","src_path":"/home/ubuntu/byte_json/src/main.rs"}}
Finished dev [unoptimized + debuginfo] target(s) in 2.59 secs
Running `target/debug/byte_json`
Hello, world!
```
Resolves#35164.
r? @jonathandturner
The `hi` and `lo` offsets in a span are relative to a `CodeMap`, but this
doesn't seem to be terribly useful for tool consumers who don't have the
codemap, but might want the byte offset within an actual file?
Resolves#35164.
[LLVM] Fix an assertion when a weak symbol is defined in global_asm.
This change will fix the issue from
https://github.com/japaric/svd2rust/pull/130
cc @japaric
r? @alexcrichton
Fix in weak_count in Arc in the case the weak count is locked.
In the case the weak count was locked, the weak_count function could
return usize::MAX. We need to test this condition manually.
trans: Internalize symbols without relying on LLVM
This PR makes the compiler use the information gather by the trans collector in order to determine which symbols/trans-items can be made internal. This has the advantages:
+ of being LLVM independent,
+ of also working in incremental mode, and
+ of allowing to not keep all LLVM modules in memory at the same time.
This is in preparation for fixing issue #39280.
cc @rust-lang/compiler
This PR cuts down on a large number of `#[inline(always)]` and `#[inline]`
annotations in libcore for various core functions. The `#[inline(always)]`
annotation is almost never needed and is detrimental to debug build times as it
forces LLVM to perform inlining when it otherwise wouldn't need to in debug
builds. Additionally `#[inline]` is an unnecessary annoation on almost all
generic functions because the function will already be monomorphized into other
codegen units and otherwise rarely needs the extra "help" from us to tell LLVM
to inline something.
Overall this PR cut the compile time of a [microbenchmark][1] by 30% from 1s to
0.7s.
[1]: https://gist.github.com/alexcrichton/a7d70319a45aa60cf36a6a7bf540dd3a
Tidy: allow common lang+lib features
This allows changes to the Rust language that have both library
and language components share one feature gate.
The feature gates need to be "about the same change", so that both
library and language components must either be both unstable, or
both stable, and share the tracking issue.
Removes the ugly "proc_macro" exception added by #40939.
Closes#43089
Tells emcc to enable assertions and debugging information for
wasm32-experimental-emscripten. This makes the codegen issues caused by
LLVM bug 33824 manifest more frequently at runtime and improves the wasm
debugging experience.
Attempt to fix appveyor
This will fix the problem, I think, but I don't know that this is a good idea (potentially leaving ourselves open to attackers, I guess, if a cert was revoked...). Of course, it may not. I don't actually have windows to check on..
r? @alexcrichton
Add support for full RELRO
This commit adds support for full RELRO, and enables it for the
platforms I know have support for it.
Full RELRO makes the PLT+GOT data read-only on startup, preventing it
from being overwritten.
http://tk-blog.blogspot.com/2009/02/relro-not-so-well-known-memory.htmlFixesrust-lang/rust#29877.
---
I'm not entirely certain if this is the best way to do it, but I figured mimicking the way it's done for PIE seemed like a good start at least. I'm not sure whether we want to have it enabled by default globally and then disabling it explicitly for targets that don't support it though. I'm also not sure whether the `full_relro` function should call `bug!()` or something like it for linkers that don't support it rather than no-opping.
Slew of builtin-attribute gating tests
Slew of builtin-attribute "gating" tests for issue #43106.
Some stray observations:
* I don't know if its a good thing that so many attributes allow inputs which are silently discarded. (I made heavy use of that in writing my tests, but that was more out of curiosity than necessity.)
* The difference between crate-level and non-crate-level behavior is quite significant in some cases. Definitely worth making sure one has tests for both cases. (Not as clear whether it was worthwhile trying the various other AST forms like `fn f()` vs `struct S;`)
* `#[no_builtins]` and `#[no_mangle]` occur twice on the `BUILTIN_ATTRIBUTES` list. Thats almost certainly a bug. (Filed as #43148)
* We are maximally liberal in what we allow for `#[test]` and `#[bench]` when one compiles without `--test`.
* We allow `#[no_mangle]` on arbitrary AST nodes, but only warn about potential misuse on `fn`
* We allow `#[cold]`, `#[must_use]`, `#[windows_subsystem]`, and `#[no_builtins]` on arbitrary AST nodes. I don't know off-hand what the semantics are for e.g. a `#[cold] type T = ...;`
* We allow crate-level `#![inline]`. That's probably a bug since its otherwise restricted to `fn` items
