Repository navigation
aarch64 neon intrinsics broken after #90621 #95002
Copy link
Copy link
Closed
Labels
A-SIMDArea: SIMD (Single Instruction Multiple Data)Area: SIMD (Single Instruction Multiple Data)C-bugCategory: This is a bug.Category: This is a bug.O-AArch64Armv8-A or later processors in AArch64 modeArmv8-A or later processors in AArch64 modeT-compilerRelevant to the compiler team, which will review and decide on the PR/issue.Relevant to the compiler team, which will review and decide on the PR/issue.regression-from-stable-to-nightlyPerformance or correctness regression from stable to nightly.Performance or correctness regression from stable to nightly.
Description
Activity
- addedA-SIMDArea: SIMD (Single Instruction Multiple Data)Area: SIMD (Single Instruction Multiple Data)O-ArmTarget: 32-bit Arm processors (armv6, armv7, thumb...), including 64-bit Arm in AArch32 stateTarget: 32-bit Arm processors (armv6, armv7, thumb...), including 64-bit Arm in AArch32 stateC-bugCategory: This is a bug.Category: This is a bug.
on Mar 16, 2022 Working on a fix for this.
Reacted by Hans Kratz, Tony Arcieri, unageek, David Koloski and Andrii Savytskyi- addedT-compilerRelevant to the compiler team, which will review and decide on the PR/issue.Relevant to the compiler team, which will review and decide on the PR/issue.regression-from-stable-to-nightlyPerformance or correctness regression from stable to nightly.Performance or correctness regression from stable to nightly.
on Mar 17, 2022 - addedI-prioritizeIssue needs a team member to assess the impact. Will be replaced by P-{low,medium,high,critical}Issue needs a team member to assess the impact. Will be replaced by P-{low,medium,high,critical}
on Mar 17, 2022 - addedO-AArch64Armv8-A or later processors in AArch64 modeArmv8-A or later processors in AArch64 modeand removedO-ArmTarget: 32-bit Arm processors (armv6, armv7, thumb...), including 64-bit Arm in AArch32 stateTarget: 32-bit Arm processors (armv6, armv7, thumb...), including 64-bit Arm in AArch32 state
on Mar 18, 2022 - added a commit that references this issue
on Mar 22, 2022 - added a commit that references this issue
on Mar 23, 2022 - removedI-prioritizeIssue needs a team member to assess the impact. Will be replaced by P-{low,medium,high,critical}Issue needs a team member to assess the impact. Will be replaced by P-{low,medium,high,critical}
on Mar 23, 2022
Metadata
Metadata
Assignees
Labels
A-SIMDArea: SIMD (Single Instruction Multiple Data)Area: SIMD (Single Instruction Multiple Data)C-bugCategory: This is a bug.Category: This is a bug.O-AArch64Armv8-A or later processors in AArch64 modeArmv8-A or later processors in AArch64 modeT-compilerRelevant to the compiler team, which will review and decide on the PR/issue.Relevant to the compiler team, which will review and decide on the PR/issue.regression-from-stable-to-nightlyPerformance or correctness regression from stable to nightly.Performance or correctness regression from stable to nightly.
#90621 causes a compile-time error on every aarch64 function which just has the neon feature explicitly enabled via
#[target_feature(enable = "neon"), even iffpis enabled implicitly. This will cause a good amount of breakage across the ecosystem - Github Code Search. After this change even runningTARGET=aarch64-apple-darwin cargo testonstdarchfails with lots of compile-time errors.Apparently the stdarch build problems also cause any usage of aarch64 intrinsics to fail in current nightly. Trying to compile this:
results in:
(I found out because of a sudden CI failure for simdutf8 here.)
Originally posted by @hkratz in #90621 (comment)
@rustbot label O-arm A-simd C-bug