Skip to content

[Bugfix][ROCm] Only run FP8 AITER MLA prefill when using FP8 KV #48712

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.

Sign up for GitHub

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

Open
simondanielsson wants to merge 1 commit into
base: main
Choose a base branch
from
+8 −4
Open

[Bugfix][ROCm] Only run FP8 AITER MLA prefill when using FP8 KV #48712

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
12 changes: 8 additions & 4 deletions vllm/v1/attention/backends/mla/rocm_aiter_mla.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -241,7 +241,9 @@ def __init__(
device=device,
)

self._fp8_prefill_enabled = _fp8_mla_prefill_supported()
self._fp8_prefill_enabled = (
_fp8_mla_prefill_supported() and kv_cache_dtype_str == "fp8"
)
if self._fp8_prefill_enabled:
max_prefill_qlen = min(
vllm_config.model_config.max_model_len,
Expand Down Expand Up @@ -724,9 +726,11 @@ def __init__(

self.flash_attn_varlen_func = flash_attn_varlen_func

# FP8 MLA prefill kernel imports (lazy, only when enabled).
# Auto-enabled on gfx950 when AITER ships the kernels.
self._fp8_prefill_enabled = _fp8_mla_prefill_supported()
from vllm.utils.torch_utils import is_quantized_kv_cache

self._fp8_prefill_enabled = _fp8_mla_prefill_supported() and (
is_quantized_kv_cache(kv_cache_dtype)
)
if self._fp8_prefill_enabled:
from aiter import mla_prefill_ps_asm_fwd, mla_reduce_v1

Expand Down
Loading