Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Disable alignment checks on i686-pc-windows-msvc #112684

Merged
merged 2 commits into from
Jun 17, 2023
Merged

Disable alignment checks on i686-pc-windows-msvc #112684

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c54672e
Disable alignment checks on i686-pc-windows-msvc
saethlin Jun 15, 2023
4ef316f
Update compiler/rustc_mir_transform/src/check_alignment.rs
wesleywiser Jun 16, 2023
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 4 additions & 0 deletions compiler/rustc_mir_transform/src/check_alignment.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -15,6 +15,10 @@ pub struct CheckAlignment;

impl<'tcx> MirPass<'tcx> for CheckAlignment {
fn is_enabled(&self, sess: &Session) -> bool {
// FIXME(#112480) MSVC and rustc disagree on minimum stack alignment on x86 Windows
if sess.target.llvm_target == "i686-pc-windows-msvc" {
wesleywiser marked this conversation as resolved.
Show resolved Hide resolved
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Is this potentially going to make some of the MIR tests for this fail on i686-msvc? Are there any tests that should get something like a ignore-i686-pc-windows-msvc?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think just the test for the alignment check itself. The mir-opt tests have debug assertions disabled and I think I was otherwise pretty careful to modify the other tests so that they are not perturbed by them, but that was some time ago.

I'd prefer to be able to run the msvc test suite locally but I don't think that's supported on a Linux host?

wesleywiser marked this conversation as resolved.
Show resolved Hide resolved
return false;
}
sess.opts.debug_assertions
}

Expand Down
1 change: 1 addition & 0 deletions tests/ui/mir/mir_alignment_check.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,5 +1,6 @@
// run-fail
// ignore-wasm32-bare: No panic messages
// ignore-i686-pc-windows-msvc: #112480
// compile-flags: -C debug-assertions
// error-pattern: misaligned pointer dereference: address must be a multiple of 0x4 but is

Expand Down
21 changes: 21 additions & 0 deletions tests/ui/mir/mir_alignment_check_i686-pc-windows-msvc.rs
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -0,0 +1,21 @@
// run-pass
// only-i686-pc-windows-msvc
// compile-flags: -Copt-level=0 -Cdebug-assertions=yes

// MSVC isn't sure if on 32-bit Windows its u64 type is 8-byte-aligned or 4-byte-aligned.
// So this test ensures that on i686-pc-windows-msvc, we do not insert a runtime check
// that will fail on dereferencing of a pointer to u64 which is not 8-byte-aligned but is
// 4-byte-aligned.

#![feature(strict_provenance)]

fn main() {
let mut x = [0u64; 2];
let ptr: *mut u8 = x.as_mut_ptr().cast::<u8>();
unsafe {
let misaligned = ptr.add(4).cast::<u64>();
assert!(misaligned.addr() % 8 != 0);
assert!(misaligned.addr() % 4 == 0);
*misaligned = 42;
}
}