|
|
@@ -40,7 +40,7 @@ FLASHINFER_VERSION=0.6.17
|
|
|
# parser hardening, prefill fairness engines, KDA prefill backend, FlashKDA
|
|
|
# stable-inverse fix, #649 logits top-k single exchange, GLM5Next decode
|
|
|
# defaults + LM-head/metadata reuse. No requirements/setup.py drift.
|
|
|
-# Staged tree on the build host: sm121/r18 @ 343661fe63dd64fe4125cfe1897c0c6c21673bb2
|
|
|
+# Staged tree on the build host: sm121/r18 @ 67e7e6e2ae204f40d6b60b96326bdcdc50a61b96
|
|
|
# (base + pr582 + pr646 merges + staged-cutlass cmake patch; /tmp/vllm-src.tar.gz).
|
|
|
VLLM_REPO=https://github.com/local-inference-lab/vllm.git
|
|
|
VLLM_BASE_REF=a8c796f3af74106b2d8d441e9ec54588936a5388
|
|
|
@@ -62,6 +62,14 @@ B12X_REPO=https://github.com/local-inference-lab/b12x.git
|
|
|
B12X_BASE_REF=b58f34eaf978277621efced6678e6713fd7122e4
|
|
|
B12X_PRS="284"
|
|
|
|
|
|
+# Deps staging: deps-src.tar.gz (in this dir) carries the pre-fetched build
|
|
|
+# sub-deps. 2026-09-06 re-stage: flash-attention/csrc/cutlass was an empty
|
|
|
+# stub (staging skipped the submodule) — the r18 vllm tree dropped the
|
|
|
+# implicit cutlass include wiring for the FA targets, so the stub broke the
|
|
|
+# FA3 build (sm90_get_smem_store_op_for_accumulator missing in cutlass 4.4.2).
|
|
|
+# Now populated with FA's own pin (vllm pins flash-attention @ f3e1a4f, whose
|
|
|
+# csrc/cutlass gitlink = 62750a2b75c802660e4894434dc55e839f322277, nvidia/cutlass).
|
|
|
+
|
|
|
# DeepGEMM — fetched by vllm's cmake; pre-cloned (raw-SHA fetches of a
|
|
|
# non-default-branch tip 404 over GitHub's git protocol)
|
|
|
DEEPGEMM_REPO=https://github.com/deepseek-ai/DeepGEMM.git
|
