The official source reports this update: b10728: CUDA: XOR swizzle flash attn K,V smem fp16 tiles.
Canonical AI event
Nvidia updates CUDA to use 64-bit pointers for flash attention
Nvidia's Yashankani fixed shared memory race in flash attention on DGX Spark.
1 Sept 20261 verified claims1 sources1 observations
This official update documents a development concerning b10728: CUDA: XOR swizzle flash attn K,V smem fp16 tiles. 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
Evidence trail
Sources behind the event
Editorial presentation
Open story