diff options
| author | Jubilee <46493976+workingjubilee@users.noreply.github.com> | 2023-12-12 18:48:51 -0800 |
|---|---|---|
| committer | GitHub <noreply@github.com> | 2023-12-12 18:48:51 -0800 |
| commit | a33f1a3d3af81aad8215dc3fc0f958e8c636da0e (patch) | |
| tree | 7d59f6d01682f5331b2d8866cf63920f24650288 /compiler/rustc_codegen_llvm/src | |
| parent | 2d1d443d7f868ea5bf3880d929ea1c66172f243d (diff) | |
| parent | 95b5a80f47fa30a594985f36eab87d1ccbd0435a (diff) | |
| download | rust-a33f1a3d3af81aad8215dc3fc0f958e8c636da0e.tar.gz rust-a33f1a3d3af81aad8215dc3fc0f958e8c636da0e.zip | |
Rollup merge of #118864 - farnoy:masked-load-store-fixes, r=workingjubilee
Fix alignment passed down to LLVM for simd_masked_load Follow up to #117953 The alignment for a masked load operation should be that of the element/lane, not the vector as a whole It can produce miscompilations after the LLVM optimizer notices the higher alignment and promotes this to an unmasked, aligned load followed up by blend/select - https://rust.godbolt.org/z/KEeGbevbb
Diffstat (limited to 'compiler/rustc_codegen_llvm/src')
| -rw-r--r-- | compiler/rustc_codegen_llvm/src/intrinsic.rs | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/compiler/rustc_codegen_llvm/src/intrinsic.rs b/compiler/rustc_codegen_llvm/src/intrinsic.rs index 23b424f25ba..58e68a64907 100644 --- a/compiler/rustc_codegen_llvm/src/intrinsic.rs +++ b/compiler/rustc_codegen_llvm/src/intrinsic.rs @@ -1568,7 +1568,7 @@ fn generic_simd_intrinsic<'ll, 'tcx>( // Alignment of T, must be a constant integer value: let alignment_ty = bx.type_i32(); - let alignment = bx.const_i32(bx.align_of(values_ty).bytes() as i32); + let alignment = bx.const_i32(bx.align_of(values_elem).bytes() as i32); // Truncate the mask vector to a vector of i1s: let (mask, mask_ty) = { |