-
Notifications
You must be signed in to change notification settings - Fork 72
Pull requests: hw-native-sys/simpler
Author
Label
Projects
Milestones
Reviews
Assignee
Sort
Pull requests list
Add: run Qwen3-14B decode with host graph execution
#1819
opened Aug 13, 2026 by
doraemonmj
Contributor
Loading…
2 tasks done
feat(a5): port qwen3_14b_decode to tensormap_and_ringbuffer (WIP, attention SyncAll blocked)
#1818
opened Aug 13, 2026 by
ChaoZheng109
Collaborator
•
Draft
test: add manual cross-runtime SPMD paged attention coverage
#1817
opened Aug 13, 2026 by
doraemonmj
Contributor
Loading…
Update: classify TMR platform differences
#1814
opened Aug 13, 2026 by
yanghaoran29
Contributor
Loading…
Refactor: share HBG Graph Execution across architectures
#1813
opened Aug 12, 2026 by
zmnobug
Contributor
Loading…
Fix: preserve A2A3 early dispatch on queue overflow
#1812
opened Aug 12, 2026 by
sjduan
Contributor
Loading…
Fix: track reader anti-dependencies in TensorMap
#1810
opened Aug 12, 2026 by
yanghaoran29
Contributor
•
Draft
10 of 15 tasks
Fix: preserve args dump selection in host-built graphs
#1808
opened Aug 12, 2026 by
vegetabledoww
Contributor
Loading…
Fix A2A3 repeated-run AICore stream regression
#1807
opened Aug 12, 2026 by
Crane-Liu
Contributor
Loading…
tensormap_and_ringbuffer: split fanin edges into orthogonal WAIT/RETAIN flags (#1375)
#1806
opened Aug 12, 2026 by
ChaoZheng109
Collaborator
Loading…
CI: reduce packaging, profiling, and unit-test overhead
#1805
opened Aug 12, 2026 by
doraemonmj
Contributor
Loading…
[Performance] host_build_graph: activate graph nodes incrementally as they materialize
#1804
opened Aug 12, 2026 by
SergioMartin86
Contributor
Loading…
Add: benchmark host_build_graph runtime
#1797
opened Aug 12, 2026 by
doraemonmj
Contributor
Loading…
Add: host_build_graph copy of the qwen3_14b_decode example
#1666
opened Aug 3, 2026 by
noabauma
Contributor
Loading…
ci: bump pinned pto-isa to e8f558d5
#1660
opened Aug 3, 2026 by
chenshengxin2026
Contributor
Loading…
Refactor: reuse native execution thread per runner
#1654
opened Aug 3, 2026 by
Crane-Liu
Contributor
Loading…
Add: dispatch MPI worker groups through mailbox
#1624
opened Aug 1, 2026 by
sunkaixuan2018
Contributor
Loading…
2 of 5 tasks
Add: support mpirun-launched multi-host L3 workers
#1623
opened Aug 1, 2026 by
sunkaixuan2018
Contributor
Loading…
4 of 5 tasks
Refactor: host build graph, lazy completion watermark, 4 AICPU schedulers
#1619
opened Jul 31, 2026 by
raphael-s-steiner
Contributor
Loading…
Refactor: host build graph, eager completion watermark, 4 AICPU schedulers
#1618
opened Jul 31, 2026 by
raphael-s-steiner
Contributor
Loading…
a5: batch sync_to_sm every K=16 local advances
#1575
opened Jul 29, 2026 by
yanghaoran29
Contributor
•
Draft
feat: support pipelined orchestration and scheduling across tokens and requests
#1379
opened Jul 16, 2026 by
TaoZQY
Contributor
Loading…
[Optimization] Replace wiring with polling-based task readiness test (~17% median device speedup)
#1137
opened Jun 24, 2026 by
SergioMartin86
Contributor
Loading…
6 of 7 tasks
ProTip!
Type g i on any issue or pull request to go back to the issue listing page.