Size the VMM graph-input exchange by the widest input across ranks (#42125)

Co-authored-by: merrymercy <merrymercy@users.noreply.github.com>
Co-authored-by: oulgen <oulgen@users.noreply.github.com>
This commit is contained in:
metamergebot
2026-10-02 13:45:32 -07:00
committed by GitHub
co-authored by merrymercy oulgen
parent 4a4dad1cf1
commit 354ed2a7b6
2 changed files with 102 additions and 10 deletions
+11 -9
View File
@@ -989,7 +989,6 @@ class VmmGraphInputManager:
"""
FABRIC_HANDLE_BYTES = 64
MAX_VMM_BASES = 4096
MAX_CHUNKS_PER_INPUT = 16
t0 = time.perf_counter()
@@ -1025,12 +1024,15 @@ class VmmGraphInputManager:
local_input_chunks = [
[int(idx) for idx in indices] for indices in input_chunk_indices
]
for chunks in local_input_chunks:
if len(chunks) > MAX_CHUNKS_PER_INPUT:
raise RuntimeError(
"Too many VMM chunks for graph input: "
f"{len(chunks)} > {MAX_CHUNKS_PER_INPUT}"
)
# Every rank must pack the same input struct, so the per-input span
# width is the widest input on any rank (floored at 16 slots).
widest_input = torch.tensor(
[max((len(c) for c in local_input_chunks), default=0)],
dtype=torch.int64,
device="cpu",
)
dist.all_reduce(widest_input, op=dist.ReduceOp.MAX, group=self.group)
max_chunks_per_input = max(16, int(widest_input.item()))
# All-gather base metadata and per-input VMM spans. A captured tensor
# can cross expandable-segment allocation boundaries, so peer mappings
@@ -1041,7 +1043,7 @@ class VmmGraphInputManager:
base_struct = struct.Struct(
f"<QQ{FABRIC_HANDLE_BYTES}s" if use_fabric else "<QQ"
)
input_struct = struct.Struct(f"<QQ{MAX_CHUNKS_PER_INPUT}Q")
input_struct = struct.Struct(f"<QQ{max_chunks_per_input}Q")
base_offset = header_struct.size
input_offset = base_offset + MAX_VMM_BASES * base_struct.size
payload_size = input_offset + new_count * input_struct.size
@@ -1067,7 +1069,7 @@ class VmmGraphInputManager:
for i, (chunks, offset) in enumerate(
zip(local_input_chunks, input_offsets)
):
padded_chunks = chunks + [0] * (MAX_CHUNKS_PER_INPUT - len(chunks))
padded_chunks = chunks + [0] * (max_chunks_per_input - len(chunks))
input_struct.pack_into(
local_payload,
input_offset + i * input_struct.size,