Skip to content

Pull requests: hw-native-sys/simpler

New pull request New
31 Open 896 Closed
31 Open 896 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

Fix: atomic acquire/release for sim register handshake gate (a2a3 + a5)
#1081 opened Jun 17, 2026 by ChaoZheng109 Collaborator Loading…
10
feat(a2a3/runtime): speculative early-dispatch (pre-stage + doorbell)
#1079 opened Jun 17, 2026 by poursoul Collaborator Loading…
2 tasks
36
Rename dump tensor surface to dump args
#1072 opened Jun 16, 2026 by vegetabledoww Contributor Loading…
2 tasks done
1
8
Refactor: share dump arg selection state
#1069 opened Jun 16, 2026 by zmnobug Contributor Loading…
1
20
Fix: fanin producer dedup lookup cost
#1066 opened Jun 16, 2026 by sunkaixuan2018 Contributor Loading…
1
22
docs(args): fix stale cap literals missed by #1056
#1064 opened Jun 16, 2026 by ChaoZheng109 Collaborator Loading…
2
Refactor: drop per-run AICPU init launch; per-thread run-wall capture
#1061 opened Jun 16, 2026 by hw-native-sys-bot Collaborator Loading…
3 of 4 tasks
9
perf(l3): remove useless exchange buffer from ring allreduce kernel
#1059 opened Jun 15, 2026 by georgebisbas Contributor Loading…
4
Fix: make a5 L2 swimlane work onboard (host-shadow alloc + payload buffer channel)
#1058 opened Jun 15, 2026 by indigo1973 Contributor Loading…
11
docs(hardware): MMIO performance reference + 2 cann-example probe tools
#1057 opened Jun 15, 2026 by hw-native-sys-bot Collaborator Loading…
1 of 3 tasks
21
[WIP] feat(dfx): add l0_swimlane intra-core pipeline trace tool
#1053 opened Jun 15, 2026 by indigo1973 Contributor Loading…
12
[Fix] [Performance] Fix AICPU thread affinity assignment to fit in a single AICPU package (a.k.a NUMA domain)
#1046 opened Jun 12, 2026 by noabauma Contributor Loading…
7
Fix: keep HCCL weak-link dependency
#1032 opened Jun 11, 2026 by sunkaixuan2018 Contributor Loading…
1
4
fix(runtime): allow zero-size view at any offset (Tensor::view bounds-check)
#1023 opened Jun 10, 2026 by csy0225 Loading…
2 of 3 tasks
1
Add: L3-L2 orchestration communication design
#1015 opened Jun 8, 2026 by ccyywwen Contributor Loading…
10
feat: add remote l3 python session runtime
#1011 opened Jun 8, 2026 by puddingfjz Contributor Loading…
19
Add: AICore receive_time DFX field — split head_OH into NoC + dcci/ack
#1004 opened Jun 8, 2026 by hw-native-sys-bot Collaborator Loading…
4 of 6 tasks
5
Fix L3 worker child-exit hangs
#1003 opened Jun 8, 2026 by ndleslx • Draft
6
Paged Attention benchmark vs pypto
#986 opened Jun 4, 2026 by MirkoDeVita98 Contributor • Draft
5
Add: per-task and scope-filter granularity to scope_stats
#976 opened Jun 3, 2026 by doraemonmj Contributor • Draft
10
[WIP] A5 performance opt
#906 opened May 30, 2026 by poursoul Collaborator Loading…
18
Add remote L3 worker
#866 opened May 27, 2026 by puddingfjz Contributor Loading…
7
Add: L3/L2 host-device mapped region design
#861 opened May 26, 2026 by ccyywwen Contributor Loading…
13
Triangular Inverse Kernel (continuation of Zouzias' impl)
#830 opened May 20, 2026 by MirkoDeVita98 Contributor Loading…
22
[Performance Analysis] Adding intra-kernel timing runs
#829 opened May 20, 2026 by SergioMartin86 Loading…
11
ProTip! Type g i on any issue or pull request to go back to the issue listing page.