Fix const eval bug breaking run-pass tests in Miri
PR #63580 broke miri's ability to run the run-pass test suite with MIR
optimizations enabled. The issue was that we weren't properly handling
the substs and DefId associated with a Promoted value. This didn't break
anything in rustc because in rustc this code runs before the Inliner
pass which is where the DefId and substs can diverge from their initial
values. It broke Miri though because it ran this code again after
running the optimization pass.
r? @oli-obk
cc @RalfJung
Add x86_64-linux-kernel target
This adds a target specification for Linux kernel modules on x86_64, as well as base code that can be shared with other architectures.
I wasn't totally sure about what the best name for this was.
There's one open question on whether we should use the LLVM generic x86_64-elf target, or the same one used for the Linux userspace.
r? @joshtriplett
Fix unlock ordering in SGX synchronization primitives
Avoid holding spinlocks during usercalls. This should avoid deadlocks in certain pathological scheduling cases.
cc @mzohreva @parthsane
r? @alexcrichton
place: Passing `align` = `layout.align.abi`, when also passing `layout`
Of the calls changed:
7/12 use `align` = `layout.align.abi`.
`from_const_alloc` uses `alloc.align`, but that is `assert_eq!` to `layout.align.abi`.
only 4/11 use something interesting for `align`.
Use unicode-xid crate instead of libcore
This PR proposes to remove `char::is_xid_start` and `char::is_xid_continue` functions from `libcore` and use `unicode_xid` crate from crates.io (note that this crate is already present in rust-lang/rust's Cargo.lock).
Reasons to do this:
* removing rustc-binary-specific stuff from libcore
* making sure that, across the ecosystem, there's a single definition of what rust identifier is (`unicode-xid` has almost 10 million downs, as a `proc_macro2` dependency)
* making it easier to share `rustc_lexer` crate with rust-analyzer: no need to `#[cfg]` if we are building as a part of the compiler
Reasons not to do this:
* increased maintenance burden: we'll need to upgrade unicode version both in libcore and in unicode-xid. However, this shouldn't be a too heavy burden: just running `./unicode.py` after new unicode version. I (@matklad) am ready to be a t-compiler side maintainer of unicode-xid. Moreover, given that xid-unicode is an important dependency of syn, *someone* needs to maintain it anyway.
* xid-unicode implementation is significantly slower. It uses a more compact table with binary search, instead of a trie. However, this shouldn't matter in practice, because we have fast-path for ascii anyway, and code size savings is a plus. Moreover, in #59706 not using libcore turned out to be *faster*, presumably beacause checking for whitespace with match is even faster.
<details>
<summary>old description</summary>
Followup to #59706
r? @eddyb
Note that this doesn't actually remove tables from libcore, to avoid conflict with https://github.com/rust-lang/rust/pull/62641.
cc https://github.com/unicode-rs/unicode-xid/pull/11
</details>
Extend Polonius fact generation for (some) move tracking
This PR will extend rustc to emit facts used for tracking moves and initialization in Polonius. It is most likely the final part of my master's thesis work.
Previously the build would take the presence of the LLVM_CONFIG envvar to
mean that the sanitizers should be built, but this is a common envvar that
could be set for reasons unrelated to the rustc sanitizers.
This commit adds a new envvar RUSTC_BUILD_SANITIZERS and uses it instead.
Rollup of 15 pull requests
Successful merges:
- #62860 (Stabilize checked_duration_since for 1.38.0)
- #63549 (Rev::rposition counts from the wrong end)
- #63985 (Stabilize pin_into_inner in 1.39.0)
- #64005 (Add a `Place::is_indirect` method to determine whether a `Place` contains a `Deref` projection)
- #64031 (Harden `param_attrs` test wrt. usage of a proc macro `#[attr]`)
- #64038 (Check impl trait substs when checking for recursive types)
- #64043 (Add some more tests for underscore imports)
- #64092 (Update xLTO compatibility table in rustc book.)
- #64110 (Refer to "`self` type" instead of "receiver type")
- #64120 (Move path parsing earlier)
- #64123 (Added warning around code with reference to uninit bytes)
- #64128 (unused_parens: account for or-patterns and `&(mut x)`)
- #64141 (Minimize uses of `LocalInternedString`)
- #64142 (Fix doc links in `std::cmp` module)
- #64148 (fix a few typos in comments)
Failed merges:
r? @ghost
Minimize uses of `LocalInternedString`
`LocalInternedString` is described as "An alternative to `Symbol` and `InternedString`, useful when the chars within the symbol need to be accessed. It is best used for temporary values."
This PR makes the code match that comment, by removing all non-local uses of `LocalInternedString`. This allows the removal of a number of operations on `LocalInternedString` and a couple of uses of `unsafe`.
Added warning around code with reference to uninit bytes
Officially, uninitialized integers, and therefore, Rust references to them are _invalid_ (note that this may evolve into official defined behavior (_c.f._, https://github.com/rust-lang/unsafe-code-guidelines/issues/71)).
However, `::std` uses references to uninitialized integers when working with the `Read::initializer` feature (#42788), since it relies on this unstably having defined behavior with the current implementation of the compiler (IIUC).
Hence the comment to disincentivize people from using this pattern outside the standard library.
