Make &Slice a thin pointer #50612
Merged
Make &Slice a thin pointer #50612
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
rust-highfive
added
the
S-waiting-on-review
|
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
|
At least the linker errors are reproducible |
src/librustc/ty/mod.rs
Outdated
|
|
||
| impl<T> PartialEq for Slice<T> { | ||
| #[inline] | ||
| fn eq(&self, other: &Slice<T>) -> bool { | ||
| (&self.0 as *const [T]) == (&other.0 as *const [T]) | ||
| (self as *const _) == (other as *const _) |
There was a problem hiding this comment.
PartialEq and Hash don't agree. Is it possible that we ever have two slices with the same address but different lengths?
There was a problem hiding this comment.
That's not possible as the length is stored at that address.
There was a problem hiding this comment.
The Hash implementation should just hash the pointer then. Should be faster too.
There was a problem hiding this comment.
The HashStable impl in impls_ty.rs should also just use the pointer as the cache key.
src/librustc/ty/mod.rs
Outdated
| pub fn empty<'a>() -> &'a Slice<T> { | ||
| unsafe { | ||
| mem::transmute(slice::from_raw_parts(0x1 as *const T, 0)) | ||
| &*(EMPTY_SLICE as *const _) |
There was a problem hiding this comment.
I'm wondering if this can trigger some kind of undefined behavior? Maybe mem::transmute(EMPTY_SLICE) would be better.
There was a problem hiding this comment.
I can get rid of the optimization for the empty slice and see if that helps...
There was a problem hiding this comment.
Removing it does make things work.
|
☔ The latest upstream changes (presumably #50332) made this pull request unmergeable. Please resolve the merge conflicts. |
TimNN
added
S-waiting-on-author
|
Ping from triage @Zoxc! It's been a while since we heard from you, will you have time to work on this again soon? |
|
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
|
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
|
@bors try |
bors
added a commit
that referenced
this pull request
May 22, 2018
Make &Slice a thin pointer Split out from #50395 r? @michaelwoerister
|
☀️ Test successful - status-travis |
|
@Mark-Simulacrum I'd like a perf run here |
|
This will be interesting |
kennytm
added
S-waiting-on-perf
eddyb
reviewed
May 24, 2018
src/librustc/ty/mod.rs
Outdated
| /// A wrapper for slices with the additional invariant | ||
| /// that the slice is interned and no other slice with | ||
| /// the same contents can exist in the same context. | ||
| /// This means we can use pointer + length for both | ||
| /// equality comparisons and hashing. | ||
| #[derive(Debug, RustcEncodable)] | ||
| pub struct Slice<T>([T]); | ||
| pub struct Slice<T>(PhantomData<T>, OpaqueSliceContents); |
There was a problem hiding this comment.
If you write this as:
#[repr(C)]
pub struct Slice<T> {
pub len: usize,
data: [T; 0],
_opaque: OpaqueSliceContents,
}Then you can read/write the len, and get a pointer to the start of the data by .data.as_ptr(), both of which work in safe code.
|
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
eddyb
reviewed
May 24, 2018
src/librustc/ty/mod.rs
Outdated
| let align = mem::align_of::<T>(); | ||
| let align_mask = align - 1; | ||
| let offset = mem::size_of::<usize>(); | ||
| (offset + align_mask) & !align_mask |
There was a problem hiding this comment.
Since this is only used to compute the size to allocate, could you inline it into from_arena?
|
☔ The latest upstream changes (presumably #50930) made this pull request unmergeable. Please resolve the merge conflicts. |
bors
added
the
S-waiting-on-author
|
The job Click to expand the log.I'm a bot! I can only do what humans tell me to, so if this was not helpful or you have suggestions for improvements, please ping or otherwise contact |
|
@eddyb's suggestions have been applied. |
|
📌 Commit fb4e3b6 has been approved by |
bors
added
S-waiting-on-bors
bors
added a commit
that referenced
this pull request
May 28, 2018
Make &Slice a thin pointer Split out from #50395 r? @michaelwoerister
|
☀️ Test successful - status-appveyor, status-travis |
kennytm
removed
the
S-waiting-on-perf
|
This looks like a slight performance win. The previous benchmark run hashed the length of slices and it had an assertion in the |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Split out from #50395
r? @michaelwoerister