mtl: reduction: Bypass group reduction on SLM by sparing workload to other SGs when SIMD=8 · intel/torch-xpu-ops@a9de4ae

Triggered via pull request August 5, 2024 07:36

xytintel

opened #692

xyt/reduce

Status Success

Total duration 5h 25m 9s

Artifacts 1

pull.yml

on: pull_request

Produced during runtime

Name	Size
Inductor-XPU-E2E-Data-692 Expired	65.5 KB