[3/N][warmup][DSv4] Migrate FA4 MLA and shared CuTeDSL kernels (#53565)

Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Co-authored-by: Codex <codex@openai.com>
Co-authored-by: mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
This commit is contained in:
Roberto L. Castro
2026-09-07 09:36:35 -04:00
committed by GitHub
co-authored by Codex mergify[bot] <37929162+mergify[bot]@users.noreply.github.com>
parent 1713b9866a
commit 42801b3a6b
8 changed files with 52 additions and 67 deletions
@@ -20,9 +20,6 @@ from vllm.model_executor.warmup.deep_gemm_warmup import deep_gemm_warmup
from vllm.model_executor.warmup.deepseek_v4_mhc_warmup import (
deepseek_v4_mhc_warmup,
)
from vllm.model_executor.warmup.fa4_cutedsl_warmup import (
fa4_cutedsl_warmup,
)
from vllm.model_executor.warmup.flashinfer_autotune_cache import (
resolve_flashinfer_autotune_file,
write_flashinfer_autotune_cache,
@@ -169,7 +166,6 @@ def kernel_warmup(worker: "Worker", *, process_local_only: bool = False):
# Run next so input-prep kernels JIT against pristine runner state.
if worker.vllm_config.kernel_config.enable_jit_warmup:
kimi_k3_triton_warmup(worker)
fa4_cutedsl_warmup(worker)
spec_decode_rejection_warmup(worker)
qwen4_exp_qsa_triton_warmup(worker)