Even more thread-safety changes #49558
Merged
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
|
Your PR failed on Travis. Through arcane magic we have determined that the following fragments from the build log may contain information about the problem. 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 |
|
Your PR failed on Travis. Through arcane magic we have determined that the following fragments from the build log may contain information about the problem. 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 |
|
Your PR failed on Travis. Through arcane magic we have determined that the following fragments from the build log may contain information about the problem. 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 |
Zoxc
force-pushed
the
sync-misc
branch
2 times, most recently
from
d873356 to
0b4571c
Compare
|
Your PR failed on Travis. Through arcane magic we have determined that the following fragments from the build log may contain information about the problem. 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 |
|
Your PR failed on Travis. Through arcane magic we have determined that the following fragments from the build log may contain information about the problem. 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 |
|
Your PR failed on Travis. Through arcane magic we have determined that the following fragments from the build log may contain information about the problem. 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 |
Zoxc
changed the title
|
Your PR failed on Travis. Through arcane magic we have determined that the following fragments from the build log may contain information about the problem. 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 |
|
Your PR failed on Travis. Through arcane magic we have determined that the following fragments from the build log may contain information about the problem. 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 |
|
Your PR failed on Travis. Through arcane magic we have determined that the following fragments from the build log may contain information about the problem. 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 |
pietroalbini
added
the
S-waiting-on-review
|
Ping from triage, @michaelwoerister ! |
|
☔ The latest upstream changes (presumably #49714) made this pull request unmergeable. Please resolve the merge conflicts. |
src/librustc_data_structures/sync.rs
Outdated
| @@ -241,6 +242,75 @@ impl<K: Eq + Hash, V: Eq, S: BuildHasher> HashMapExt<K, V> for HashMap<K, V, S> | |||
| } | |||
| } | |||
|
|
|||
| /// A type whose inner value can be written once and then will stay read-only | |||
| pub struct Once<T>(Lock<Option<T>>, PhantomData<T>); | |||
There was a problem hiding this comment.
What's the PhantomData for?
There was a problem hiding this comment.
Once conceptually owns a T outside the lock once it is initialized. This ensures we get the right impl for Sync: impl<T: Sync> Sync for Once<T>
There was a problem hiding this comment.
Good catch! Could you add a comment explaining that?
There was a problem hiding this comment.
Plus maybe an example of how one could exploit Once<T> if the PhantomData was not there?
| @@ -154,13 +154,13 @@ fn get_llvm_opt_size(optimize: config::OptLevel) -> llvm::CodeGenOptSize { | |||
| } | |||
| } | |||
|
|
|||
| pub fn create_target_machine(sess: &Session) -> TargetMachineRef { | |||
| target_machine_factory(sess)().unwrap_or_else(|err| { | |||
| pub fn create_target_machine(sess: &Session, find_features: bool) -> TargetMachineRef { | |||
There was a problem hiding this comment.
What's the find_features parameter about?
There was a problem hiding this comment.
create_target_machine calls is_pie_binary which requires crate_types. However when we are discovering the target CPU features, crate_types is uninitialized. So we pass in find_features to let create_target_machine assume that we aren't looking at a pie binary.
There was a problem hiding this comment.
Ah, OK. Could you add a comment explaining that?
src/librustc/traits/select.rs
Outdated
| // This may overwrite the cache with the same value | ||
| self.tcx().evaluation_cache | ||
| .hashmap.borrow_mut() | ||
| .insert(trait_ref, WithDepNode::new(dep_node, result)); |
There was a problem hiding this comment.
Should this use insert_same()?
There was a problem hiding this comment.
The type inserted doesn't implement Eq, so it cannot use insert_same.
There was a problem hiding this comment.
For this one, implementing Eq should be easy. Let's do it.
src/librustc/traits/select.rs
Outdated
| // This may overwrite the cache with the same value | ||
| tcx.selection_cache | ||
| .hashmap.borrow_mut() | ||
| .insert(trait_ref, WithDepNode::new(dep_node, candidate)); |
There was a problem hiding this comment.
This might also be a candidate for insert_same().
There was a problem hiding this comment.
Same thing, the type inserted doesn't implement Eq, so it cannot use insert_same.
There was a problem hiding this comment.
Yeah, this one looks more complicated. Pity.
src/librustc_data_structures/sync.rs
Outdated
| } | ||
|
|
||
| #[inline] | ||
| pub fn try_set(&self, value: T) -> Option<T> { |
There was a problem hiding this comment.
Could you document this method, especially what the return value is.
src/librustc_data_structures/sync.rs
Outdated
| } | ||
|
|
||
| #[inline(always)] | ||
| pub fn try_get(&self) -> Option<&T> { |
There was a problem hiding this comment.
This method could use documentation too.
src/librustc_data_structures/sync.rs
Outdated
| } | ||
|
|
||
| #[inline] | ||
| pub fn init_locking<F: FnOnce() -> T>(&self, f: F) { |
There was a problem hiding this comment.
init_locking and init_nonlocking should be documented to. I needed a while to see what these are about but the difference can be explained rather quickly.
|
Looks great, thanks @Zoxc! |
|
@bors r+ |
|
📌 Commit 006f9b2 has been approved by |
bors
added
S-waiting-on-bors
bors
added a commit
that referenced
this pull request
Apr 12, 2018
Even more thread-safety changes r? @michaelwoerister
|
☀️ Test successful - status-appveyor, status-travis |
This was referenced Apr 12, 2018
bors
added a commit
that referenced
this pull request
May 9, 2018
Make SelectionCache and EvaluationCache thread-safe Split out from #49558 r? @nikomatsakis
37 tasks
nnethercote
added a commit
to nnethercote/rust
that referenced
this pull request
Nov 10, 2023
The included measurements have varied over the years. At one point there were quite a few more, but rust-lang#49558 deleted a lot that were no longer used. Today there's just four, and it's a motley collection that doesn't seem particularly valuable. I think it has been well and truly subsumed by self-profiling, which collects way more data.
Merged
nnethercote
added a commit
to nnethercote/rust
that referenced
this pull request
Nov 12, 2023
The included measurements have varied over the years. At one point there were quite a few more, but rust-lang#49558 deleted a lot that were no longer used. Today there's just four, and it's a motley collection that doesn't seem particularly valuable. I think it has been well and truly subsumed by self-profiling, which collects way more data.
bors
added a commit
to rust-lang-ci/rust
that referenced
this pull request
Nov 14, 2023
…wiser Remove `-Zperf-stats`. The included measurements have varied over the years. At one point there were quite a few more, but rust-lang#49558 deleted a lot that were no longer used. Today there's just four, and it's a motley collection that doesn't seem particularly valuable. I think it has been well and truly subsumed by self-profiling, which collects way more data. r? `@wesleywiser`
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.
r? @michaelwoerister