The official source reports this update: b11047: cuda : fix CUB argsort corruption caused by in-place keys. cuda : fix CUB argsort corruption caused by in-place keys ( #28389 ) argsort_f32_i32_cuda_cub called the one-shot DeviceRadixSort::SortPairs API with d_keys_in == d_keys_out (temp_keys, temp_keys). CUB's internal double-buffer ping-pong requires distinct key buffers: with aliased buffers the sort partially overwrites its own input mid-pass and emits a corrupted permutation, surfacing as intermittent garbage indices (e.g. backend top_k over a 248k-column vocab on Maxwell/CUDA 12.5/CCCL 2.x, which then triggered out-of-bounds gathers in downstream get_rows).
Canonical AI event
Nvidia fixes CUB argsort corruption in CUDA 12.5.
A fix for CUB argsort corruption in CUDA addresses in-place key issues, preventing intermittent garbage indices in top_k operations.
19 Sept 20261 verified claims1 sources1 observations
This official update documents a development concerning b11047: cuda : fix CUB argsort corruption caused by in-place keys. Its practical significance depends on the scope and evidence stated by the source.
Read the official source update and verify its stated scope, evidence and timing before acting on it.
Connected knowledge
Entities affected by this event
Continue this topic
Related verified updates
Evidence trail
Sources behind the event
Editorial presentation
Open story