PR #24360 CUDA: Fix ssm_scan_f32 data-races

CUDA: Fix ssm_scan_f32 data-races #24360

ORippler merged 3 commits into ggml-org:master from ORippler:osimons/fix_ssm_scan_f32

Add missing syncthreads before resuing cub_temp_storage

a941c2cb

Add one more missing __syncthreads

4fbecf73

Remove unused smem from ssm_scan_f32

814e106f

ORippler requested a review 70 days ago

github-actions added Nvidia GPU

github-actions added ggml

IMbackK approved these changes on 2026-06-09

ORippler requested a review from

gaugarg-nv 69 days ago

gaugarg-nv approved these changes on 2026-06-10

ORippler merged fb83cc9a into master 69 days ago

ORippler deleted the osimons/fix_ssm_scan_f32 branch 69 days ago

Reviewers

gaugarg-nv

IMbackK

Assignees

No one assigned

Labels

Nvidia GPU ggml

Milestone

No milestone