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.

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

ARM64 SVE: Zip/Unzip should use predicate variant when all inputs are masks #101598

Open
a74nh opened this issue Apr 26, 2024 · 3 comments
Open
Labels
area-System.Runtime.Intrinsics arm-sve Work related to arm64 SVE/SVE2 support

Comments

@a74nh
Copy link
Contributor

a74nh commented Apr 26, 2024

See #101294 (review)

There is a version of zip/unzip that takes predicates as inputs and output.

In lowering (?) we should check the inputs, if they are all convert-from-masks then we should remove the masks and then in codegen detect somehow (how if the masks have been removed? maybe we contain the mask conversions??)

By itself this is a low priority optimisation, but we should do it as soon as other instructions will also probably fit the same pattern. (Add them to this ticket as we find them)

@dotnet-issue-labeler dotnet-issue-labeler bot added the needs-area-label An area label is needed to ensure this gets routed to the appropriate area owners label Apr 26, 2024
@dotnet-policy-service dotnet-policy-service bot added the untriaged New issue has not been triaged by the area owner label Apr 26, 2024
@kunalspathak kunalspathak added arm-sve Work related to arm64 SVE/SVE2 support and removed untriaged New issue has not been triaged by the area owner labels Apr 26, 2024
Copy link
Contributor

Tagging subscribers to this area: @dotnet/area-system-io-compression
See info in area-owners.md if you want to be subscribed.

@vcsjones vcsjones added area-System.Runtime.Intrinsics and removed area-System.IO.Compression needs-area-label An area label is needed to ensure this gets routed to the appropriate area owners labels May 7, 2024
Copy link
Contributor

Tagging subscribers to this area: @dotnet/area-system-runtime-intrinsics
See info in area-owners.md if you want to be subscribed.

@tannergooding
Copy link
Member

In lowering (?) we should check the inputs, if they are all convert-from-masks then we should remove the masks and then in codegen detect somehow (how if the masks have been removed? maybe we contain the mask conversions??)

We notably do this in morph for x64 to catch various early cases, but rechecking in lowering may be able to catch others as well.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
area-System.Runtime.Intrinsics arm-sve Work related to arm64 SVE/SVE2 support
Projects
None yet
Development

No branches or pull requests

5 participants