-
Notifications
You must be signed in to change notification settings - Fork 12.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Use min_specialization in the remaining rustc crates #72707
Conversation
r? @eddyb (rust_highfive has picked a reviewer for you, use r? to override) |
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 |
9217821
to
dc4d551
Compare
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 |
dc4d551
to
21eab5b
Compare
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 |
21eab5b
to
cefb703
Compare
☔ The latest upstream changes (presumably #72778) made this pull request unmergeable. Please resolve the merge conflicts. |
cefb703
to
596afe0
Compare
☔ The latest upstream changes (presumably #72927) made this pull request unmergeable. Please resolve the merge conflicts. |
Is there anyone in @rust-lang/compiler willing to review this? |
@@ -228,8 +228,13 @@ impl<'tcx> SpecializedEncoder<LocalDefId> for EncodeContext<'tcx> { | |||
} | |||
} | |||
|
|||
impl<'tcx> SpecializedEncoder<Ty<'tcx>> for EncodeContext<'tcx> { | |||
fn specialized_encode(&mut self, ty: &Ty<'tcx>) -> Result<(), Self::Error> { | |||
impl<'a, 'b, 'tcx> SpecializedEncoder<&'a ty::TyS<'b>> for EncodeContext<'tcx> |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
are the two separate lifetimes 'a
and 'b
really necessary here? Couldn't they just be one and use Ty<'a>
?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Unfortunately not. Specializing impls can't repeat type/lifetime parameters.
&'a ty::TyS<'b>: UseSpecializedEncodable, | ||
{ | ||
fn specialized_encode(&mut self, ty: &&'a ty::TyS<'b>) -> Result<(), Self::Error> { | ||
debug_assert!(self.tcx.lift(ty).is_some()); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is there a measurable perf loss to just doing lift
and unwrap and skipping the transmute?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Ok, looks like there are some serious losses here. Let's benchmark against the transmute variant and if that has no perf losses, let's merge it.
) -> Result<(), Self::Error> { | ||
ty_codec::encode_spanned_predicates(self, predicates, |ecx| &mut ecx.predicate_shorthands) | ||
debug_assert!(predicates.iter().all(|(p, _)| self.tcx.lift(p.kind()).is_some())); |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think you can just lift the entire borrowed slice instead of each element
I agree. r? @oli-obk r=me with the comments resolved in some way |
596afe0
to
09a2e80
Compare
@bors try @rust-timer queue |
Awaiting bors try build completion |
⌛ Trying commit 09a2e80cc4266f81c7d60516f8eff5eb92093b7f with merge 70c770a6c3a5241ab4f65e6841ecd641c983a4fb... |
☀️ Try build successful - checks-azure |
Queued 70c770a6c3a5241ab4f65e6841ecd641c983a4fb with parent feb3536, future comparison URL. |
Finished benchmarking try commit (70c770a6c3a5241ab4f65e6841ecd641c983a4fb): comparison url. |
Awaiting bors try build completion |
⌛ Trying commit 88ea7e5 with merge 3b49dc19913ecb228b5ce64932615370ac1ebc35... |
One thing I want to mention: The specialization hack that we use was meant to be replaced with a custom trait for serialization, but that was blocked on the ability to use custom derive -- now that we have custom derive, couldn't we make the encodable trait carry a Also, how much is any of this documented? I cast my eye over the diff but I admit I've always found the encoding system rather confusing. |
💥 Test timed out |
@bors try @rust-timer queue |
Awaiting bors try build completion |
⌛ Trying commit 88ea7e5 with merge d5dd610915fff92af707dd35f4a864570eaa8520... |
I've started working on safe serialization traits, but it's going to be a large change affecting a lot of code, so I think it makes sense to have it separated. |
☀️ Try build successful - checks-azure |
Queued d5dd610915fff92af707dd35f4a864570eaa8520 with parent bb86748, future comparison URL. |
Finished benchmarking try commit (d5dd610915fff92af707dd35f4a864570eaa8520): comparison url. |
@bors r=oli-obk |
📌 Commit 88ea7e5 has been approved by |
@bors r- r=oli-obk This wasn't showing in the queue at https://buildbot2.rust-lang.org/homu/queue/rust |
📌 Commit 88ea7e5 has been approved by |
…-obk Use min_specialization in the remaining rustc crates This adds a lot of `transmute` calls to replace the unsound uses of specialization. It's ugly, but at least it's honest about what's going on. cc rust-lang#71420, @RalfJung
Rollup of 10 pull requests Successful merges: - rust-lang#72707 (Use min_specialization in the remaining rustc crates) - rust-lang#72740 (On recursive ADT, provide indirection structured suggestion) - rust-lang#72879 (Miri: avoid tracking current location three times) - rust-lang#72938 (Stabilize Option::zip) - rust-lang#73086 (Rename "cyclone" to "apple-a7" per changes in upstream LLVM) - rust-lang#73104 (Example about explicit mutex dropping) - rust-lang#73139 (Add methods to go from a nul-terminated Vec<u8> to a CString) - rust-lang#73296 (Remove vestigial CI job msvc-aux.) - rust-lang#73304 (Revert heterogeneous SocketAddr PartialEq impls) - rust-lang#73331 (extend network support for HermitCore) Failed merges: r? @ghost
This adds a lot of
transmute
calls to replace the unsound uses of specialization.It's ugly, but at least it's honest about what's going on.
cc #71420, @RalfJung