rustc_target: use Integer instead of Primitive for discriminants and niches. #63450
Conversation
This file contains hidden or 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
|
r? @cramertj (rust_highfive has picked a reviewer for you, use r? to override) |
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 |
RalfJung
reviewed
Aug 10, 2019
| @@ -619,18 +619,23 @@ impl<'mir, 'tcx, M: Machine<'mir, 'tcx>> InterpCx<'mir, 'tcx, M> { | |||
| Err(_) => | |||
| throw_unsup!(InvalidDiscriminant(raw_discr.erase_tag())), | |||
| }; | |||
| let real_discr = if discr_val.layout.ty.is_signed() { | |||
| let (discr_ty_signed, discr_ty_size) = match &rval.layout.ty.sty { | |||
| ty::Adt(adt, _) => { | |||
There was a problem hiding this comment.
This seems fishy. In my experience it is basically never correct to look at the Ty of something in Miri, with very few exceptions.
There was a problem hiding this comment.
OTOH, this looks like you basically inlined old discr_type? Why can't we use that any more / extend that to also return signedness?
There was a problem hiding this comment.
What old discr_type? I hoisted the discr_type call, and note that it only made sense for ADTs, whereas generators can also have discriminants (but they're not signed, so they wouldn't hit the signed logic previously).
There was a problem hiding this comment.
Oh, discr_type only works for ADTs, of course.
Maybe this match here should move to Ty, so that we can just do layout.ty.discr_type()? Its return type IntType (yet another one, next to layout::Integer, IntTy and UintTy) seems to have just the right amount of freedom.
There was a problem hiding this comment.
I think we already have something for variant names, so this seems... okay-ish?
Maybe the generator case can just be a _ => so we really only special-case ADTs with signed discriminants.
I don't know, I really don't like signed integers :(.
oli-obk
reviewed
Aug 11, 2019
oli-obk
approved these changes
Aug 11, 2019
|
Note that this is broken right now, likely because of how signedness is handled. |
|
r? @oli-obk |
|
Ping from triage, @eddyb any update on this? Some check are failing. Thanks. |
|
@gagan0723 See #63450 (comment). Overall, I think I want this discussed by the compiler-team. We should figure out how precisely we want to keep track of the discriminant encoding (tag/niche) "type", we have 3 options AFAICT:
Going for 2. instead of 3. would fix the signedness problems more directly, and result in a more straight-forward patch, but it doesn't feel like it goes far enough. If we want 3., it should be possible to fix all the signedness issues, since the information can be extracted from the |
eddyb
added
I-nominated
T-compiler
|
☔ The latest upstream changes (presumably #63873) made this pull request unmergeable. Please resolve the merge conflicts. |
|
Ping from triage |
|
(adding self to assignee list; i want to better acquaint myself with the issued here) |
|
Wow, somehow I'd missed that using just pub enum Foo {
A = 0,
B = 255,
}
pub enum Bar {
A = 0,
B = -128,
C = 127,
} |
|
Given #63450 (comment) I'll just abandon this PR for now. |
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.
Instead of signed/unsigned integers or pointers, discriminants (tag/niche-filling) and niches are now always unsigned integers (signedness is handled when reading them as signed values).
This does seems like a simplification, but only slightly (perhaps I haven't cleaned up everything that had to special-case signed integers or pointers).
To avoid some duplication,
Scalarwas made generic over the type of itsvaluefield.See #62138 (comment) for some of the discussion that led to me trying this out.
cc @RalfJung @oli-obk @nagisa
(As I wrote this, just noticed 3 tests failed, AFAICT all because of signedness)