operand types are not normalized in MIR shims #45940
Labels
T-compiler
Relevant to the compiler team, which will review and decide on the PR/issue.
Comments
nikomatsakis
added
the
T-compiler
|
cc @Nashenas88 |
|
I don't think this belongs in the milestone. This isn't really a problem, it's just causing us to run some extra normalizations -- or normalizations at an unexpected time. |
ritiek
added a commit
to ritiek/rust
that referenced
this issue
Nov 26, 2017
MIR: adopt borrowck test Fix trailing whitespace span_bug! on unexpected action Make RegionVid use newtype_index! Closes rust-lang#45843 Check rvalue aggregates during check_stmt in tycheck, add initial, (not passing) test Fix failing test Remove attributes and test comments accidentally left behind, add in span_mirbugs Normalize LvalueTy for ops and format code to satisfy tidy check only normalize operand types when in an ADT constructor avoid early return handle the active field index in unions normalize types in ADT constructor Fixes rust-lang#45940 Fix borrowck compiler errors for upvars contain "spurious" dereferences Fixes rust-lang#46003 added associated function Box::leak Box::leak - improve documentation Box::leak - fixed bug in documentation Box::leak - relaxed constraints wrt. lifetimes Box::leak - updated documentation Box::leak - made an oops, fixed now =) Box::leak: update unstable issue number (46179). Add test for rust-lang#44953 Add missing Debug impls to std_unicode Also adds #![deny(missing_debug_implementations)] so they don't get missed again. Amend RELEASES for 1.22.1 and fix the date for 1.22.0 Rename param in `[T]::swap_with_slice` from `src` to `other`. The idea of ‘source’ and ‘destination’ aren’t very applicable for this operation since both slices can both be considered sources and destinations. Clarify stdin behavior of `Command::output`. Fixes rust-lang#44929. Add hints for the case of confusing enum with its variants Add failing testcases Add module population and case of enum in place of expression Use for_each_child_stable in find_module Use multiline text for crate conflict diagnostics Make float::from_bits transmute (and update the documentation to reflect this). The current implementation/documentation was made to avoid sNaN because of potential safety issues implied by old/bad LLVM documentation. These issues aren't real, so we can just make the implementation transmute (as permitted by the existing documentation of this method). Also the documentation didn't actually match the behaviour: it said we may change sNaNs, but in fact we canonicalized *all* NaNs. Also an example in the documentation was wrong: it said we *always* change sNaNs, when the documentation was explicitly written to indicate it was implementation-defined. This makes to_bits and from_bits perfectly roundtrip cross-platform, except for one caveat: although the 2008 edition of IEEE-754 specifies how to interpet the signaling bit, earlier editions didn't. This lead to some platforms picking the opposite interpretation, so all signaling NaNs on x86/ARM are quiet on MIPS, and vice-versa. NaN-boxing is a fairly important optimization, while we don't even guarantee that float operations properly preserve signalingness. As such, this seems like the more natural strategy to take (as opposed to trying to mangle the signaling bit on a per-platform basis). This implementation is also, of course, faster. Simplify an Iterator::fold to Iterator::any This method of once-diagnostics doesn't allow nesting UI tests extract the regular output from the 'rendered' field in json Merge cfail and ui tests into ui tests Add a MIR pass to lower 128-bit operators to lang item calls Runs only with `-Z lower_128bit_ops` since it's not hooked into targets yet. Include tuple projections in MIR tests Add type checking for the lang item As part of doing so, add more lang items instead of passing u128 to the i128 ones where it doesn't matter in twos-complement. Handle shifts properly * The overflow-checking shift items need to take a full 128-bit type, since they need to be able to detect idiocy like `1i128 << (1u128 << 127)` * The unchecked ones just take u32, like the `*_sh?` methods in core * Because shift-by-anything is allowed, cast into a new local for every shift incr.comp.: Make sure we don't lose unused green results from the query cache. rustbuild: Update LLVM and enable ThinLTO This commit updates LLVM to fix rust-lang#45511 (https://reviews.llvm.org/D39981) and also reenables ThinLTO for libtest now that we shouldn't hit rust-lang#45768. This also opportunistically enables ThinLTO for libstd which was previously blocked (rust-lang#45661) on test failures related to debuginfo with a presumed cause of rust-lang#45511. Closes rust-lang#45511 std: Flag Windows TLS dtor symbol as #[used] Turns out ThinLTO was internalizing this symbol and eliminating it. Worse yet if you compiled with LTO turns out no TLS destructors would run on Windows! The `#[used]` annotation should be a more bulletproof implementation (in the face of LTO) of preserving this symbol all the way through in LLVM and ensuring it makes it all the way to the linker which will take care of it. Add enum InitializationRequiringAction Fix tidy tests
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
In normal MIR, the types of local variables and things are fully normalized. However, in the MIR shims we construct, this is not always the case. We encountered this when fixing #45889, where it is necessary to normalize the types of operands in some cases to make the existing MIR typeck.
I'm not sure if this is a bug per se. Certainly once we move to lazy norm it will be fine. =) But it was surprising. For now we are normalizing (in the case of ADT constructors) during typeck.
cc @arielb1
The text was updated successfully, but these errors were encountered: