Implement lazy decoding of DefPathTable during incremental compilation
PR https://github.com/rust-lang/rust/pull/75813 implemented lazy decoding of the `DefPathTable` from crate metadata. However, it requires decoding the entire `DefPathTable` when incremental compilation is active, so that we can map a decoded `DefPathHash` to a `DefId` from an arbitrary crate.
This PR adds support for lazy decoding of dependency `DefPathTable`s when incremental compilation si active.
When we load the incremental cache and dep
graph, we need the ability to map a `DefPathHash` to a `DefId` in the
current compilation session (if the corresponding definition still
exists).
This is accomplished by storing the old `DefId` (that is, the `DefId`
from the previous compilation session) for each `DefPathHash` we need to
remap. Since a `DefPathHash` includes the owning crate, the old crate is
guaranteed to be the right one (if the definition still exists). We then
use the old `DefIndex` as an initial guess, which we validate by
comparing the expected and actual `DefPathHash`es. In most cases,
foreign crates will be completely unchanged, which means that we our
guess will be correct. If our guess is wrong, we fall back to decoding
the entire `DefPathTable` for the foreign crate. This still represents
an improvement over the status quo, since we can skip decoding the
entire `DefPathTable` for other crates (where all of our guesses were
correct).
Rollup of 11 pull requests
Successful merges:
- #79038 (Change ui test that are run-pass and that do not test the compiler to library tests)
- #79184 (Stop adding '*' at the end of slice and str typenames for MSVC case)
- #79227 (Remove const_fn_feature_flags test)
- #79444 (Move const ip in ui test to unit test)
- #79522 (Validate lint docs separately.)
- #79525 (Add -Z normalize-docs and enable it for compiler docs)
- #79527 (Move intra-doc link tests into a subdirectory)
- #79548 (Show since when a function is const in stdlib)
- #79568 (update Miri)
- #79573 (Update with status for various NetBSD ports.)
- #79583 (Update books)
Failed merges:
r? `@ghost`
`@rustbot` modify labels: rollup
Update books
## nomicon
1 commits in 23c49f1d5ce4720bc5b7e3a920f47eccc8da6b63..d8383b65f7948c2ca19191b3b4bd709b403aaf45
2020-11-05 13:30:53 +0900 to 2020-11-22 10:24:42 -0500
- Clarify that any alignment is valid for ZSTs
## reference
5 commits in a7de763c213292f5b44bf10acb87ffa38724814d..a8afdca5d0715b2257b6f8b9a032fd4dd7dae855
2020-11-11 19:13:21 -0800 to 2020-11-30 06:44:46 -0800
- Describe relationship between reference and optimizers (rust-lang/reference#902)
- A simple missing space (rust-lang/reference#909)
- Cleanup formatting (rust-lang/reference#907)
- Use `doc`, not `test` for fn item inner attr example (rust-lang/reference#906)
- Update where our chat is. (rust-lang/reference#903)
## book
1 commits in 13e1c05420bca86ecc79e4ba5b6d02de9bd53c62..a190438d77d28041f24da4f6592e287fab073a61
2020-10-20 14:57:32 -0500 to 2020-11-16 10:44:08 -0600
- Change SipHash not found (404) link to SipHash in Wikipedia (rust-lang/book#2503)
## rust-by-example
4 commits in 1886fda6981b723e4de637074455558f8bc1e83c..236c734a2cb323541b3394f98682cb981b9ec086
2020-10-28 13:46:54 -0500 to 2020-11-30 14:05:49 -0300
- Update old invalid link (rust-lang/rust-by-example#1392)
- Moved "See also" link to the correct page. (rust-lang/rust-by-example#1389)
- Fix some markdown lint warnings (rust-lang/rust-by-example#1388)
- Minor grammar suggestion (rust-lang/rust-by-example#1386)
## embedded-book
1 commits in ca8169e69b479f615855d0eece7e318138fcfc00..ba34b8a968f9531d38c4dc4411d5568b7c076bfe
2020-10-15 15:06:35 +0000 to 2020-11-17 00:20:43 +0000
- Clarify CAS availability (rust-embedded/book#273)
Update with status for various NetBSD ports.
The NetBSD ports of rust to aarch64, armv7*, i686, and powerpc**
all both build and run. Status is as of rust 1.47.0.
*) Natively requires repeated successive build attempts (`rustc` is
such a resource pig VM-consumption-wise), or run in a chroot
on an aarch64 host where the available VM space is 4GB instead
of the native 2GB.
**) Powerpc either requires `-latomic` in a directory searched by
default by `ld` or to be built within pkgsrc which has a patch and
support package to tackle this issue.
Validate lint docs separately.
This addresses some concerns raised in https://github.com/rust-lang/rust/pull/76549#issuecomment-727638552 about errors with the lint docs being confusing and cumbersome. Errors from validating the lint documentation were being generated during `x.py doc` (and `x.py dist`), since extraction and validation are being done in a single step. This changes it so that extraction and validation are separated, so that `x.py doc` will not error if there is a validation problem, and tests are moved to `x.py test src/tools/lint-docs`.
This includes the following changes:
* Separate validation to `x.py test`.
* Added some more documentation on how to more easily modify and test the docs.
* Added more help to the error messages to hopefully provide more information on how to fix things.
The first commit just moves the code around, so you may consider looking at the other commits for a smaller diff.
Remove const_fn_feature_flags test
## Overview
Helps with #76268
I found `const_fn_feature_flags` is targeting feature-gate and remove it.
r? ``@matklad``
Stop adding '*' at the end of slice and str typenames for MSVC case
When computing debug info for MSVC debuggers, Rust compiler emits C++ style type names for compatibility with .natvis visualizers. All Ref types are treated as equivalences of C++ pointers in this process, and, as a result, their type names end with a '\*'. Since Slice and Str are treated as Ref by the compiler, their type names also end with a '\*'. This causes the .natvis engine for WinDbg fails to display data of Slice and Str objects. We addressed this problem simply by removing the '*' at the end of type names for Slice and Str types.
Debug info in WinDbg before the fix:
Debug info in WinDbg after the fix:
This change has also been tested with debuggers for Visual Studio, VS Code C++ and VS Code LLDB to make sure that it does not affect the behavior of other kinds of debugger.
Make keyboard interactions in the settings menu more pleasant
#78868 improved the keyboard interactions with the settings page. This PR goes a bit further by allowing more than just "space" to toggle the checkboxes.
r? `@jyn514`
Add built-in support for the armv5te-unknown-linux-uclibcgnueabi target
Hi!
I'd like to add built-in support for the `armv5te-unknown-linux-uclibcgnueabi` target. It's a pretty common target used by many devices like routers and IP cameras. It's mostly a copy-paste of `armv5te-unknown-linux-gnueabi`. I've tested it on a quite complex application that uses tokio, openssl and a lot of other stuff and everything seems to be working fine.
I'm not sure about the `post_link_args` but the point is that my linker fails when `-ldl` isn't specified. Maybe there is a better place where to put this option...
It's my first contribution to Rust itself, so feel free to wash my head 😄
_Note: The app mentioned above was built with this in my `.cargo/config`:_
```
[unstable]
build-std = ["core", "std", "alloc", "proc_macro", "panic_abort"]
build-std-features = ["panic_immediate_abort"]
```
More intra doc links
Helps with #75080.
I did a commit by group of file, I can squash if wanted.
`@rustbot` modify labels: T-doc, A-intra-doc-links
r? `@jyn514`
The NetBSD ports of rust to aarch64, armv7*, i686, and powerpc**
all both build and run.
*) Natively requires repeated successive build attempts (rustc is
such a resource pig VM-consumption-wise), or run in a chroot
on an aarch64 host where the available VM space is 4GB instead
of the native 2GB.
**) Powerpc either requires -latomic in a directory searched by
default by 'ld' or to be built within pkgsrc which has a patch
to tackle this.
Fix intra-doc links for `Self` on cross-crate items and primitives
- Remove the difference between `parent_item` and `current_item`; these
should never have been different.
- Remove `current_item` from `resolve` and `variant_field` so that
`Self` is only substituted in one place at the very start.
- Resolve the current item as a `DefId`, not a `HirId`. This is what
actually fixed the bug.
Hacks:
- `clean` uses `TypedefItem` when it _really_ should be
`AssociatedTypeItem`. I tried fixing this without success and hacked
around it instead (see comments)
- This second-guesses the `to_string()` impl since it wants
fully-qualified paths. Possibly there's a better way to do this.
Update error to reflect that integer literals can have float suffixes
For example, `1` is parsed as an integer literal, but it can be turned
into a float with the suffix `f32`. Now the error calls them "numeric
literals" and notes that you can add a float suffix since they can be
either integers or floats.
Avoid panic_bounds_check in fmt::write.
Writing any fmt::Arguments would trigger the inclusion of usize formatting and padding code in the resulting binary, because indexing used in fmt::write would generate code using panic_bounds_check, which prints the index and length.
These bounds checks are not necessary, as fmt::Arguments never contains any out-of-bounds indexes.
This change replaces them with unsafe get_unchecked, to reduce the amount of generated code, which is especially important for embedded targets.
---
Demonstration of the size of and the symbols in a 'hello world' no_std binary:
<details>
<summary>Source code</summary>
```rust
#![feature(lang_items)]
#![feature(start)]
#![no_std]
use core::fmt;
use core::fmt::Write;
#[link(name = "c")]
extern "C" {
#[allow(improper_ctypes)]
fn write(fd: i32, s: &str) -> isize;
fn exit(code: i32) -> !;
}
struct Stdout;
impl fmt::Write for Stdout {
fn write_str(&mut self, s: &str) -> fmt::Result {
unsafe { write(1, s) };
Ok(())
}
}
#[start]
fn main(_argc: isize, _argv: *const *const u8) -> isize {
let _ = writeln!(Stdout, "Hello World");
0
}
#[lang = "eh_personality"]
fn eh_personality() {}
#[panic_handler]
fn panic(_: &core::panic::PanicInfo) -> ! {
unsafe { exit(1) };
}
```
</details>
Before:
```
text data bss dec hex filename
6059 736 8 6803 1a93 before
```
```
0000000000001e00 T <T as core::any::Any>::type_id
0000000000003dd0 D core::fmt::num::DEC_DIGITS_LUT
0000000000001ce0 T core::fmt::num:👿:<impl core::fmt::Display for u64>::fmt
0000000000001ce0 T core::fmt::num:👿:<impl core::fmt::Display for usize>::fmt
0000000000001370 T core::fmt::write
0000000000001b30 t core::fmt::Formatter::pad_integral::write_prefix
0000000000001660 T core::fmt::Formatter::pad_integral
0000000000001350 T core::ops::function::FnOnce::call_once
0000000000001b80 t core::ptr::drop_in_place
0000000000001120 t core::ptr::drop_in_place
0000000000001c50 t core::iter::adapters::zip::Zip<A,B>::new
0000000000001c90 t core::iter::adapters::zip::Zip<A,B>::new
0000000000001b90 T core::panicking::panic_bounds_check
0000000000001c10 T core::panicking::panic_fmt
0000000000001130 t <&mut W as core::fmt::Write>::write_char
0000000000001200 t <&mut W as core::fmt::Write>::write_fmt
0000000000001250 t <&mut W as core::fmt::Write>::write_str
```
After:
```
text data bss dec hex filename
3068 600 8 3676 e5c after
```
```
0000000000001360 T core::fmt::write
0000000000001340 T core::ops::function::FnOnce::call_once
0000000000001120 t core::ptr::drop_in_place
0000000000001620 t core::iter::adapters::zip::Zip<A,B>::new
0000000000001660 t core::iter::adapters::zip::Zip<A,B>::new
0000000000001130 t <&mut W as core::fmt::Write>::write_char
0000000000001200 t <&mut W as core::fmt::Write>::write_fmt
0000000000001250 t <&mut W as core::fmt::Write>::write_str
```
Fix overlap detection of `usize`/`isize` range patterns
`usize` and `isize` are a bit of a special case in the match usefulness algorithm, because the range of values they contain depends on the platform. Specifically, we don't want `0..usize::MAX` to count as an exhaustive match (see also [`precise_pointer_size_matching`](https://github.com/rust-lang/rust/issues/56354)). The way this was initially implemented is by treating those ranges like float ranges, i.e. with limited cleverness. This means we didn't catch the following as unreachable:
```rust
match 0usize {
0..10 => {},
10..20 => {},
5..15 => {}, // oops, should be detected as unreachable
_ => {},
}
```
This PRs fixes this oversight. Now the only difference between `usize` and `u64` range patterns is in what ranges count as exhaustive.
r? `@varkor`
`@rustbot` label +A-exhaustiveness-checking
- Remove the difference between `parent_item` and `current_item`; these
should never have been different.
- Remove `current_item` from `resolve` and `variant_field` so that
`Self` is only substituted in one place at the very start.
- Resolve the current item as a `DefId`, not a `HirId`. This is what
actually fixed the bug.
Hacks:
- `clean` uses `TypedefItem` when it _really_ should be
`AssociatedTypeItem`. I tried fixing this without success and hacked
around it instead (see comments)
- This stringifies DefIds, then resolves them a second time. This is
really silly and rustdoc should just use DefIds throughout. Fixing
this is a larger task than I want to take on right now.
Bump dependencies invalidly assuming memory layout of SocketAddr
Bumps net2, socket2 and miow.
Helps unblock #78802
Done as separate PR since frequent lockfile collisions is a thing... And since the main PR can't be merged until large parts of the ecosystem uses the newer crates only, so we have to start somewhere.
