Skip to content

Pull requests: ROCm/spur

New pull request New
52 Open 530 Closed
52 Open 530 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

feat(spurd): containerized srun steps on bare-metal nodes (streaming + review fixes)
#833 opened Sep 5, 2026 by sajmera-pensando Collaborator Loading…
3 tasks
1
1
feat: adopt compute-node SSH sessions into the allocation (pam_exec)
#832 opened Sep 5, 2026 by powderluv Collaborator Loading…
fix(spurd): sweep the allocation cgroup on cancel
#831 opened Sep 5, 2026 by powderluv Collaborator Loading…
fix(spurd): confine step and interactive jobs to the job cgroup
#830 opened Sep 5, 2026 by powderluv Collaborator Loading…
1
6
fix(spur-core): parse --time minutes:seconds and days-hours correctly
#829 opened Sep 5, 2026 by powderluv Collaborator Loading…
1
3
feat(spurd): run containerized srun steps and cancel them on scancel
#821 opened Sep 5, 2026 by powderluv Collaborator Loading…
1
8
feat(spur-cli): honor --pty for srun steps in an allocation
#820 opened Sep 5, 2026 by powderluv Collaborator Loading…
1
13
feat(spurd): implement containerized srun job steps on bare-metal nodes
#818 opened Sep 5, 2026 by sajmera-pensando Collaborator Loading…
3 tasks
1
17
test(e2e): cover fair-share preemption, same-QOS allow-list, and exempt-time precedence
#817 opened Sep 3, 2026 by yansun1996 Member Loading…
12
fix(deploy): give the operator readiness probe time to answer
#816 opened Sep 3, 2026 by nikhilsk Contributor • Draft
1
fix(spur-cli): render squeue %L (TIME_LEFT) instead of ?
#815 opened Sep 3, 2026 by shiv-tyagi Member Loading…
3
fix(spurd): kernel-enforced GPU device isolation across every launch path
#814 opened Sep 3, 2026 by biluriuday Collaborator Loading…
6
fix(spur-core): honor pod/service CIDR under kuberouter, not just calico
#813 opened Sep 2, 2026 by yansun1996 Member Loading…
4
fix(spurctld): make cni=calico work without the WireGuard mesh
#812 opened Sep 2, 2026 by yansun1996 Member Loading…
10
fix(core)!: read -t MM:SS as minutes:seconds and accept bare days-hours
#811 opened Sep 2, 2026 by sunxxuns Loading…
1
8
fix(spurctld): stop serving when the raft core is dead
#810 opened Sep 2, 2026 by pre Loading…
1 task done
1
fix(k8s): recover a node the controller has forgotten
#809 opened Sep 2, 2026 by pre Loading…
1 task done
fix(spur-k8s): keep the cancel of a deleted SpurJob
#808 opened Sep 2, 2026 by pre Loading…
1 task done
fix(spur-k8s): give SPUR_PEER_NODES resolvable pod names
#807 opened Sep 2, 2026 by pre Loading…
1 task done
fix(spurctld)!: refuse to bootstrap a second raft cluster
#806 opened Sep 2, 2026 by pre Loading…
1 task done
feat(spurctld): surface per-account submit, group wall, and per-job TRES caps in assoc_mgr
#798 opened Sep 2, 2026 by nikhilsk Contributor • Draft
1
perf(spurctld): aggregate assoc_mgr scope usage in a single pass
#797 opened Sep 2, 2026 by nikhilsk Contributor • Draft
1
fix(spur-net): persist WireGuard peer mutations to the interface config
#792 opened Sep 2, 2026 by yansun1996 Member Loading…
8
fix(spur-cli): let salloc wait for pending allocations
#790 opened Sep 1, 2026 by jamesETsmith Contributor Loading…
1
1
docs(accounting): say that a TRES dimension of 0 is ignored, not blocking
#788 opened Sep 1, 2026 by nikhilsk Contributor Loading…
3
ProTip! Exclude everything labeled bug with -label:bug.