Commit graph

142 commits

Author SHA1 Message Date
woppopo
5e97fc9aa2 Make NonNull::new const 2022-01-23 23:04:39 +09:00
Ralf Jung
bb1423e019 fix const_ptr_offset_from tracking issue 2022-01-16 15:21:42 -05:00
Martin Habovstiak
1a96623513 Implemented const casts of raw pointers
This adds `as_mut()` method for `*const T` and `as_const()` for `*mut T`
which are intended to make casting of consts safer. This was discussed
in the [internals discussion][discussion].

[discussion]: https://internals.rust-lang.org/t/casting-constness-can-be-risky-heres-a-simple-fix/15933
2022-01-08 18:04:51 +01:00
Matthias Krüger
4d840a6e45
Rollup merge of #91823 - woppopo:const_ptr_as_ref, r=lcnr
Make `PTR::as_ref` and similar methods `const`.

Tracking issue: #91822
Feature gate: `#![feature(const_ptr_as_ref)]`

```rust
// core::ptr
impl<T: ?Sized> *const T {
    pub const unsafe fn as_ref<'a>(self) -> Option<&'a T>;
    pub const unsafe fn as_uninit_ref<'a>(self) -> Option<&'a MaybeUninit<T>>
    where
        T: Sized;
    pub const unsafe fn as_uninit_slice<'a>(self) -> Option<&'a [MaybeUninit<T>]>;
}

impl<T: ?Sized> *mut T {
    pub const unsafe fn as_ref<'a>(self) -> Option<&'a T>;
    pub const unsafe fn as_uninit_ref<'a>(self) -> Option<&'a MaybeUninit<T>>
    where
        T: Sized;
    pub const unsafe fn as_mut<'a>(self) -> Option<&'a mut T>;
    pub const unsafe fn as_uninit_mut<'a>(self) -> Option<&'a mut MaybeUninit<T>>
    where
        T: Sized;
    pub const unsafe fn as_uninit_slice<'a>(self) -> Option<&'a [MaybeUninit<T>]>;
    pub const unsafe fn as_uninit_slice_mut<'a>(self) -> Option<&'a mut [MaybeUninit<T>]>;
}

impl<T: Sized> NonNull<T> {
    pub const unsafe fn as_uninit_ref<'a>(&self) -> &'a MaybeUninit<T>;
    pub const unsafe fn as_uninit_mut<'a>(&mut self) -> &'a mut MaybeUninit<T>;
}

impl<T: ?Sized> NonNull<T> {
    pub const unsafe fn as_ref<'a>(&self) -> &'a T;
    pub const unsafe fn as_mut<'a>(&mut self) -> &'a mut T;
    pub const unsafe fn as_uninit_slice<'a>(&self) -> &'a [MaybeUninit<T>];
    pub const unsafe fn as_uninit_slice_mut<'a>(&self) -> &'a mut [MaybeUninit<T>];
}
```
2021-12-21 08:33:40 +01:00
Matthias Krüger
9e662d0c03
Rollup merge of #91824 - woppopo:const_ptr_write_bytes, r=oli-obk
Make `(*mut T)::write_bytes` `const`

Tracking issue: #86302
2021-12-13 00:20:10 +01:00
Matthias Krüger
955e552d31
Rollup merge of #91814 - japm48:spelling-fix, r=RalfJung
doc: fix typo in comments

`dereferencable -> dereferenceable`

Fixes #91802.
2021-12-12 07:45:30 +01:00
woppopo
7f5dc0f609 Make (*mut T)::write_bytes const 2021-12-12 14:02:53 +09:00
woppopo
a4b3fe0887 Make PTR::as_ref and similar methods const. 2021-12-12 13:45:27 +09:00
japm48
0d7b830139 doc: fix typo in comments
dereferencable -> dereferenceable
2021-12-12 00:27:27 +01:00
woppopo
34eaf52829 Make Uniques methods const 2021-12-12 04:27:43 +09:00
Scott McMurray
348a25044b Intra-doc links apparently don't like pointers? 2021-11-22 02:40:56 -08:00
Scott McMurray
875e01e616 Add <*{const|mut} T>::{to|from}_bits
Named based on the floating-point methods of the same name, as those are also about returning the *representation* of the value.
2021-11-22 02:08:59 -08:00
Maybe Waffle
f926c0e0d9 Fill in tracking issue for feature const_align_offset 2021-11-16 23:58:40 +03:00
Maybe Waffle
8f5f094432 Mark <*const _>::align_offset and <*mut _>::align_offset as const fn 2021-11-16 23:03:28 +03:00
est31
eeaa2f16aa Extend the const_swap feature
This makes the inherent method ptr::swap unstably
const, as well as slice::swap{,_unchecked}.
2021-11-09 20:09:56 +01:00
Matthias Krüger
88e5ae2dd3
Rollup merge of #89786 - jkugelman:must-use-len-and-is_empty, r=joshtriplett
Add #[must_use] to len and is_empty

Parent issue: #89692

r? `@joshtriplett`
2021-10-31 13:20:05 +01:00
Matthias Krüger
e79e9f5e2a
Rollup merge of #89839 - jkugelman:must-use-mem-ptr-functions, r=joshtriplett
Add #[must_use] to mem/ptr functions

There's a lot of low-level / unsafe stuff here. Are there legit use cases for ignoring any of these return values?

* No regressions in `./x.py test --stage 1 library/std src/tools/clippy`.
* One regression in `./x.py test --stage 1 src/test/ui`. Fixed.
* I am unable to run `./x.py doc` on my machine so I'll need to wait for the CI to verify doctests pass. I eyeballed all the adjacent tests and they all look okay.

Parent issue: #89692

r? ```@joshtriplett```
2021-10-31 09:20:25 +01:00
John Kugelman
6745e8da06 Add #[must_use] to len and is_empty 2021-10-30 19:25:12 -04:00
John Kugelman
887503ad14 Add #[must_use] to mem/ptr functions 2021-10-30 18:54:48 -04:00
woppopo
7936ecff48 Make more From impls const 2021-10-18 19:19:28 +09:00
Yuki Okushi
c1bde6e4b6
Rollup merge of #89794 - jkugelman:must-use-to_value-conversions, r=joshtriplett
Add #[must_use] to to_value conversions

`NonNull<T>::cast` snuck in when I wasn't looking. What a scamp!

Parent issue: #89692

r? ````@joshtriplett````
2021-10-13 21:55:13 +09:00
John Kugelman
0cf84c8c19 Add #[must_use] to to_value conversions 2021-10-11 19:37:16 -04:00
John Kugelman
06e625f7d5 Add #[must_use] to as_type conversions 2021-10-11 13:57:38 -04:00
John Kugelman
b115781bcd Add #[must_use] to conversions that move self 2021-10-10 19:50:52 -04:00
ast-ral
9da8e2a2fa remove redundant / misplaced sentence from docs 2021-09-01 20:52:30 -07:00
Frank Steffahn
2f9ddf3bc7 Fix typos “an”→“a” and a few different ones that appeared in the same search 2021-08-22 18:15:49 +02:00
bors
4e886d6876 Auto merge of #87827 - eddyb:wrapperless-mem-replace, r=m-ou-se
Avoid using the `copy_nonoverlapping` wrapper through `mem::replace`.

This is a much simpler way to achieve the pre-#86003 behavior of `mem::replace` not needing dynamically-sized `memcpy`s (at least before inlining), than re-doing #81238 (which needs #86699 or something similar).

I didn't notice it until recently, but `ptr::write` already explicitly avoided using the wrapper, while `ptr::read` just called the wrapper (and was the reason for us observing any behavior change from #86003 in Rust-GPU).

<hr/>

The codegen test I've added fails without the change to `core::ptr::read` like this (ignore the `v0` mangling, I was using a worktree with it turned on by default, for this):
```llvm
       13: ; core::intrinsics::copy_nonoverlapping::<u8>
       14: ; Function Attrs: inlinehint nonlazybind uwtable
       15: define internal void `@_RINvNtCscK5tvALCJol_4core10intrinsics19copy_nonoverlappinghECsaS4X3EinRE8_25mem_replace_direct_memcpy(i8*` %src, i8* %dst, i64 %count) unnamed_addr #0 {
       16: start:
       17:  %0 = mul i64 %count, 1
       18:  call void `@llvm.memcpy.p0i8.p0i8.i64(i8*` align 1 %dst, i8* align 1 %src, i64 %0, i1 false)
not:17      !~~~~~~~~~~~~~~~~~~~~~                                                                     error: no match expected
       19:  ret void
       20: }
```
With the `core::ptr::read` change, `core::intrinsics::copy_nonoverlapping` doesn't get instantiated and the test passes.

<hr/>

r? `@m-ou-se` cc `@nagisa` (codegen test) `@oli-obk` / `@RalfJung` (miri diagnostic changes)
2021-08-08 13:11:09 +00:00
Eduard-Mihai Burtescu
a1d014bdbc Avoid using the copy_nonoverlapping wrapper through mem::replace. 2021-08-08 13:59:36 +03:00
Max Fan
89a8ba46b8 Fix typo in the ptr documentation 2021-08-05 22:50:56 -04:00
Cameron Steffen
7fc26e9665
Rollup merge of #87690 - sharnoff:mut-ptr-allocated-obj-link, r=Mark-Simulacrum
Add missing "allocated object" doc link to `<*mut T>::add`

The portion of the documentation expecting the link was already there, but it was rendered as "[allocated object]". The added reference is just copied from the documentation for `<*const T>::add`.
2021-08-02 09:36:55 -05:00
sharnoff
12d199be77
Add missing "allocated object" doc link 2021-08-01 19:48:26 -07:00
Ralf Jung
6aaa8327f9 mark a UB doctest as no_run 2021-07-31 11:37:48 +02:00
Yuki Okushi
f7a2a22815
Rollup merge of #87547 - GuillaumeGomez:nonnull-examples, r=kennytm
Add missing examples for NonNull
2021-07-31 04:09:26 +09:00
Guillaume Gomez
1bbe6188d2 Add missing examples for NonNull 2021-07-30 11:30:34 +02:00
Ali Malik
e43254aad1 Fix may not to appropriate might not or must not 2021-07-29 01:15:20 -04:00
Albin Hedman
22fe76d97d
Add reference to tracking issue #86302 for const_ptr_write 2021-06-27 12:05:19 +02:00
Albin Hedman
1aa032f506
Add reference to issue for const_intrinsic_copy in ptr::write 2021-06-27 12:05:19 +02:00
Albin Hedman
5fbb1354ce
Revert "Revert effects of PRs 81167 and 83091."
This reverts commit 9d96b0ed8c.
2021-06-27 12:05:16 +02:00
Eric Huss
6235e6f93f Fix a few misspellings. 2021-06-25 13:18:56 -07:00
Michael Snoyman
770e8cc01e
Typo correction: s/is/its 2021-06-16 19:20:15 +03:00
Felix S. Klock II
5f6016f125 Revert PRs 81238 and 82967 (which made copy and copy_nonoverlapping intrinsics).
This is to address issue 84297.
2021-06-04 16:44:28 -04:00
Felix S. Klock II
9d96b0ed8c Revert effects of PRs 81167 and 83091.
This is preparation for reverting 81238 for short-term resolution of issue 84297.
2021-06-04 16:03:45 -04:00
Frank Steffahn
f9752b4930 Fix UB in documented example for ptr::swap 2021-05-19 14:09:23 +02:00
Kornel
3773740719 #[inline(always)] on basic pointer methods 2021-05-12 10:10:28 +01:00
Ralf Jung
722bebf163
Rollup merge of #83553 - jfrimmel:addr-of, r=m-ou-se
Update `ptr` docs with regards to `ptr::addr_of!`

This updates the documentation since `ptr::addr_of!` and `ptr::addr_of_mut!` are now stable. One might remove the distinction between the sections `# On packed structs` and `# Examples`, as the old section on packed structs was primarily to prevent users of doing undefined behavior, which is not necessary anymore.

Technically there is now wrong/outdated documentation on stable, but I don't think this is worth a point release 😉

Fixes #83509.

``````````@rustbot`````````` modify labels: T-doc
2021-05-05 17:52:18 +02:00
Julian Frimmel
389333a21c Update ptr docs with regards to ptr::addr_of!
This updates the documentation since `ptr::addr_of!` and
`ptr::addr_of_mut!` are now stable. One might remove the distinction
between the sections `# On packed structs` and `# Examples`, as the old
section on packed structs was primarily to prevent users of doing unde-
fined behavior, which is not necessary anymore.
There is also a new section in "how to obtain a pointer", which referen-
ces the `ptr::addr_of!` macros.

This commit contains squashed commits from code review.

Co-authored-by: Joshua Nelson <joshua@yottadb.com>
Co-authored-by: Mara Bos <m-ou.se@m-ou.se>
Co-authored-by: Soveu <marx.tomasz@gmail.com>
Co-authored-by: Ralf Jung <post@ralfj.de>
2021-05-03 23:14:17 +02:00
Brent Kerby
6679f5ceb1 Change 'NULL' to 'null' 2021-05-02 17:46:00 -06:00
bors
83ca4b7e60 Auto merge of #84061 - AngelicosPhosphoros:issue-75598-add-inline-always-arithmetic, r=nagisa
Add some #[inline(always)] to arithmetic methods of integers

I tried to add it only to methods which return results of intrinsics and don't have any branching.
Branching could made performance of debug builds (`-Copt-level=0`) worse.
Main goal of changes is allowing wider optimizations in `-Copt-level=1`.

Closes: https://github.com/rust-lang/rust/issues/75598

r? `@nagisa`
2021-04-17 23:31:10 +00:00
bors
7537b20626 Auto merge of #83948 - ABouttefeux:lint-nullprt-deref, r=RalfJung
add lint deref_nullptr detecting when a null ptr is dereferenced

fixes #83856
changelog: add lint that detect code like
```rust
unsafe {
      &*core::ptr::null::<i32>()
 };
unsafe {
     addr_of!(std::ptr::null::<i32>())
};
let x: i32 = unsafe {*core::ptr::null()};
let x: i32 = unsafe {*core::ptr::null_mut()};
unsafe {*(0 as *const i32)};
unsafe {*(core::ptr::null() as *const i32)};
```
```
warning: Dereferencing a null pointer causes undefined behavior
 --> src\main.rs:5:26
  |
5 |     let x: i32 = unsafe {*core::ptr::null()};
  |                          ^^^^^^^^^^^^^^^^^^
  |                          |
  |                          a null pointer is dereferenced
  |                          this code causes undefined behavior when executed
  |
  = note: `#[warn(deref_nullptr)]` on by default
```

Limitation:
It does not detect code like
```rust
const ZERO: usize = 0;
unsafe {*(ZERO as *const i32)};
```
or code where `0` is not directly a literal
2021-04-14 18:04:22 +00:00
AngelicosPhosphoros
f8a12c6311 Add some #[inline(always)] to arithmetic methods of integers
I tried to add it only to methods which return results of intrinsics and don't have any branching.
Branching could made performance of debug builds (`-Copt-level=0`) worse.
Main goal of changes is allowing wider optimizations in `-Copt-level=1`.

Closes: https://github.com/rust-lang/rust/issues/75598
2021-04-11 21:19:39 +03:00