GitLab

This reverts commit 658350b3.

This reverts commit 09486ebf.

Merge branch 'feature/fmha-fwd-async-splitkv' into feature/support-vllm-kcache-layout-add-splitkv-instance

Merge branch 'feature/add-splitkv-instance' into feature/support-vllm-kcache-layout-add-splitkv-instance

Merge branch 'feature/fmha-fwd-async-splitkv' into feature/support-vllm-kcache-layout-add-splitkv-instance

Merge branch 'feature/add-splitkv-instance' into feature/support-vllm-kcache-layout-add-splitkv-instance

* Gemm Kernel Refactor part1

* Gemm Kernel Refactor common gemm pipeline part2

* [CK TILE] Refactor batched gemm to reuse GemmKernel

* [CK TILE] Refactor GemmKernel - review changes part1

* [CK TILE] Refactor GemmKernel - references fix

* [CK TILE] Refactor GemmKernel - naming changes, add problem

* [CK_TILE] Refactor GemmKernel - update tests

* [CK_TILE] Refactor GemmKernel - review changes

* [CK_TILE] Refactor GemmKernel - update test

* [CK_TILE] Refactor GemmKernel - constness fixes

* [CK_TILE] Refactor GemmKernel - update tests

Adding namespace to disambiguate with std::bit_cast
Co-authored-by: Po Yen Chen <PoYen.Chen@amd.com>

* [CK_TILE] Move hipmalloc/memcpy calls out of gpu reference gemm

* [CK_TILE] Move hipmalloc/memcpy calls out of gpu reference gemm - review changes

* [CK_TILE] Move hipmalloc/memcpy calls out of gpu reference gemm - review fix

* updated fp16 instances to be on parity with universal gemm instances

* corrected instance name to streamk instance

* pass the build flags to config.h

* fix clang format