macro future proofing rules.
(We may want to think about what this test was actually testing and
figure out a way to test it without running afoul of macro future
proofing. I spent some time trying to do this, e.g. by inserting
parenthesis in the macro input pattern, but I could not quickly get it
working, so I took this tack instead.)
It was recently realized that we accept defaulted type parameters everywhere, without feature gate, even though the only place that we really *intended* to accept them were on types. This PR adds a lint warning unless the "type-parameter-defaults" feature is enabled. This should eventually become a hard error.
This is a [breaking-change] in that new feature gates are required (or simply removing the defaults, which is probably a better choice as they have little effect at this time). Results of a [crater run][crater] suggest that approximately 5-15 crates are affected. I didn't do the measurement quite right so that run cannot distinguish "true" regressions from "non-root" regressions, but even the upper bound of 15 affected crates seems relatively minimal.
[crater]: https://gist.github.com/nikomatsakis/760c6a67698bd24253bf
cc @rust-lang/lang
r? @pnkfelix
The motivation (other than removing boilerplate) is that this is a baby step towards a parser with error recovery.
[breaking-change] if you use any of the changed functions, you'll need to remove a try! or panictry!
This is roughly the same as my previous PR that created a dependency graph, but that:
1. The dependency graph is only optionally constructed, though this doesn't seem to make much of a difference in terms of overhead (see measurements below).
2. The dependency graph is simpler (I combined a lot of nodes).
3. The dependency graph debugging facilities are much better: you can now use `RUST_DEP_GRAPH_FILTER` to filter the dep graph to just the nodes you are interested in, which is super help.
4. The tests are somewhat more elaborate, including a few known bugs I need to fix in a second pass.
This is potentially a `[breaking-change]` for plugin authors. If you are poking about in tcx state or something like that, you probably want to add `let _ignore = tcx.dep_graph.in_ignore();`, which will cause your reads/writes to be ignored and not affect the dep-graph.
After this, or perhaps as an add-on to this PR in some cases, what I would like to do is the following:
- [x] Write-up a little guide to how to use this system, the debugging options available, and what the possible failure modes are.
- [ ] Introduce read-only and perhaps the `Meta` node
- [x] Replace "memoization tasks" with node from the map itself
- [ ] Fix the shortcomings, obviously! Notably, the HIR map needs to register reads, and there is some state that is not yet tracked. (Maybe as a separate PR.)
- [x] Refactor the dep-graph code so that the actual maintenance of the dep-graph occurs in a parallel thread, and the main thread simply throws things into a shared channel (probably a fixed-size channel). There is no reason for dep-graph construction to be on the main thread. (Maybe as a separate PR.)
Regarding performance: adding this tracking does add some overhead, approximately 2% in my measurements (I was comparing the build times for rustdoc). Interestingly, enabling or disabling tracking doesn't seem to do very much. I want to poke at this some more and gather a bit more data -- in some tests I've seen that 2% go away, but on others it comes back. It's not entirely clear to me if that 2% is truly due to constructing the dep-graph at all.
The next big step after this is write some code to dump the dep-graph to disk and reload it.
r? @michaelwoerister
So far `librustc::trans::base::trans_fn()` and `trans_closure()` have been passed the list of attributes on the function being translated *only* if the function was local and non-generic. For generic functions, functions inlined from other crates, functions with foreign ABI and for closures, only an empty list of attributes was ever passed to `trans_fn()`.
This led to the case that generic functions marked with `#[rustc_mir]` where not actually translated via MIR but via the legacy translation path.
This PR makes function/closure attributes always be passed to `trans_fn()` and disables the one test where this makes a difference.
If there is an actual reason why attributes were not passed along in these cases, let me know.
cc @rust-lang/compiler
cc @luqmana regarding the test case
Add OpAssign to Wrapping<T>, plus fix some problems in core::num::wrapping
including, but not limited to:
* Testing Wrapping<T>
* Pull out a lot of broken code that doesn't need to be there with the new stage0 compiler
* Adding Rem and RemAssign to Wrapping<T>
* Removed 3 (assumed accidental) re-exports, which is a minor [breaking-change].
* Change shl and shr to take all integer types, instead of a usize; this is a more major [breaking-change], because of values that were inferred before, but brings us in line with the integer shifts.
Fixes#30524 and #30523
This is not a fix to checks themselves per se (though we still use `Eq` MIR test instead of calling `PartialEq::eq`), but rather how we handle items we encounter in pattern position.
Previously we would just call `PartialEq` with the constant and the matchee, but now we essentially inline the constant instead. E.g. these two snippets are functionally equivalent at MIR level:
```
match val { Some(42) => true, _ => false }
```
and
```
const SECRET: Option<u8> = Some(42);
match val { SECRET => true, _ => false }
```
This approach also allows for more optimizations of matches. I.e. It can now exploit `SwitchInt` to switch on number inside a `Some` regardless of whether the value being an item or not.
This is based on @tsion’s already approved PR so I could reuse the file for more tests.
r? @eddyb
cc @nikomatsakis @tsion
This moves back (essentially reverts #30265) into MIR-specific translation code, but keeps the
funcition split out, since it is expected to eventually become recursive.
Fixes https://github.com/rust-lang/rust/issues/29572
cc @oli-obk
Since `darwin` is really `apple-darwin`, the valgrind-rpass tests were not actually being run with valgrind on mac before. Also, the `HOST` check was completely wrong.
r? @alexcrichton
The previous version using `PartialOrd::le` was broken since it passed `T` arguments where `&T` was expected.
It makes sense to use primitive comparisons since range patterns can only be used with chars and numeric types.
r? @eddyb
The current help message is too much about "normal" macros to be used
as general message. Keep it for normal macros, and add custom help and
error messages for macro definitions.
The current help message is too much about "normal" macros to be used
as general message. Keep it for normal macros, and add custom help and
error messages for macro definitions.
