Skip to content

SIMD implementation for XMFloatLoad/Store3SE - #308

Merged
Chuck Walbourn (walbourn) merged 5 commits into
mainfrom
float3seopts
May 15, 2026
Merged

SIMD implementation for XMFloatLoad/Store3SE#308
Chuck Walbourn (walbourn) merged 5 commits into
mainfrom
float3seopts

Conversation

@walbourn

@walbourn Chuck Walbourn (walbourn) commented May 14, 2026

Copy link
Copy Markdown
Collaborator

The original implementation of these was purely scalar. This adds ARM-NEON and SIMD versions.

While ARM32 is deprecated and no longer supported by Windows SDKs after 22621, I added a basic GHA compile validation of the ARM32 code-path just to keep it 'as is'. Eventually I'll need to remove all ARM32 support from the library.

@jenatali Jesse Natalie (jenatali) left a comment

Copy link
Copy Markdown
Member

Choose a reason for hiding this comment

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

Readability suggestions but overall seems fine.

Comment thread Inc/DirectXPackedVector.inl Outdated
Comment thread Inc/DirectXPackedVector.inl Outdated
@walbourn

Copy link
Copy Markdown
Collaborator Author

More test coverage walbourn/directxmathtest@488f2a5

@walbourn

Copy link
Copy Markdown
Collaborator Author

Verified on an ARM64 device locally, and of course x86/x64..

@walbourn
Chuck Walbourn (walbourn) merged commit 50822d3 into main May 15, 2026
238 checks passed
@walbourn
Chuck Walbourn (walbourn) deleted the float3seopts branch May 15, 2026 19:28
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

arm64 Related to Windows on ARM64 optimization

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Revisit XMLoad/StoreFloat3SE

2 participants