GitLab

Revert "[Kernel] Marlin_24: Ensure the mma.sp instruction is using the ::ordered_metadata modifier (introduced with PTX 8.5)" (#5149)

Co-authored-by: Zhuohan Li <zhuohan123@gmail.com>

[Kernel] Marlin_24: Ensure the mma.sp instruction is using the ::ordered_metadata modifier (introduced with PTX 8.5) (#5136)

Co-authored-by: Roger Wang <ywang@roblox.com>

Co-authored-by: Breno Faria <breno.faria@intrafind.com>

Co-authored-by: Alexey Kondratiev <alexey.kondratiev@amd.com>
Co-authored-by: Alexei-V-Ivanov-AMD <156011006+Alexei-V-Ivanov-AMD@users.noreply.github.com>
Co-authored-by: Alexei V. Ivanov <alexei.ivanov@amd.com>
Co-authored-by: omkarkakarparthi <okakarpa>

Co-authored-by: Roger Wang <136131678+ywang96@users.noreply.github.com>

[Core] Cross-attention KV caching and memory-management (towards eventual encoder/decoder model support) (#4837)

Signed-off-by: pandyamarut <pandyamarut@gmail.com>

Co-authored-by: Roger Wang <ywang@roblox.com>

This PR adds Triton kernel configs for the MoE kernel for MI300X

Co-authored-by: Ruth Evans <ruthevans@Ruths-MacBook-Pro.local>

Co-authored-by: rsnm2 <rshaw@neuralmagic.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-neuralmagic@users.noreply.github.com>