whn09 2836e365ea Neuron engram shared tables: one file for the whole group (#28)
The name came from the parent pid, which CP followers do not share with their
driver (they are its children): followers created a second table, never
unlinked (622 GB of /dev/shm leaked over five runs, then a host OOM), and each
side read zeros for the other's shards. Rank 0 now publishes the path through
the distributed store, and opening asserts the size so it can never create a
file.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-24 07:08:37 +08:00
..
__init__.py Neuron basic mode — part 2: deep_ep, sampling kernels, fixes 2026-10-20 18:57:27 +08:00
allocator.py Neuron basic mode — part 2: deep_ep, sampling kernels, fixes 2026-10-20 18:57:27 +08:00
base.py Neuron basic mode — part 2: deep_ep, sampling kernels, fixes 2026-10-20 18:57:27 +08:00
bucket.py Neuron basic mode — part 2: deep_ep, sampling kernels, fixes 2026-10-20 18:57:27 +08:00
engram.py Neuron engram shared tables: one file for the whole group (#28) 2026-10-24 07:08:37 +08:00
ep.py Neuron deep_ep: MoE dispatch/combine via all-gather + reduce-scatter (issue #12) 2026-10-22 04:01:28 +08:00