Skip to content

Pull requests: hw-native-sys/pypto-lib

New pull request New
30 Open 704 Closed
30 Open 704 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

Perf: fuse DSv4-flash indexer score scope, pack SWA output store
#831 opened Jul 24, 2026 by Hzfengsy Member Loading…
4
perf(qwen3-a8w8): skip padded attention rows
#830 opened Jul 24, 2026 by vegetabledoww Contributor Loading…
1
perf(dsv4-flash): restore on-core seed for decode split-K accumulators
#829 opened Jul 24, 2026 by zhangqi-chen Collaborator Loading…
4
fix(deepseek-v4-flash): fit decode buffers for PTOAS planning
#827 opened Jul 24, 2026 by Little-oil Contributor Loading…
1
Add: dynamic cache block sizing for DeepSeek V4
#826 opened Jul 24, 2026 by high-cloud Collaborator Loading…
4
Perf: fuse DeepSeek V4 MoE preprocessing kernels
#825 opened Jul 24, 2026 by Little-oil Contributor Loading…
1
CI: disable all jobs, add A5 runner bring-up smoke job
#821 opened Jul 23, 2026 by luohuan19 Contributor Loading…
1
dsv4 decode_fwd eplb test
#814 opened Jul 22, 2026 by zhangqi-chen Collaborator • Draft
4
perf(dsv4/moe): predicate empty routed-expert stages
#809 opened Jul 21, 2026 by wangqin1723-max Contributor Loading…
5
fix(qwen3-14b): guard negative slot_mapping KV writes
#806 opened Jul 21, 2026 by Leaf-Salix Contributor • Draft
16
perf(dsv4 hc_pre): widen linear split-K LINEAR_OK 4->8 (align with hc_head)
#804 opened Jul 20, 2026 by wangqin1723-max Contributor Loading…
3
feat(deepseek-v4): Define dynamic prefill attention contract
#793 opened Jul 17, 2026 by hashiqiqixian Contributor Loading…
20
Fuse QKV prologue into Qwen3-14B paged attention
#789 opened Jul 16, 2026 by Little-oil Contributor Loading…
7
test ci
#788 opened Jul 16, 2026 by high-cloud Collaborator Loading…
1
perf(dsv4 hca decode): early-resolve gather_kv/merge_norm + L2 swimlane level flag
#785 opened Jul 15, 2026 by wangqin1723-max Contributor Loading…
5
PoC: fold dsv4/moe dispatch/combine wait into syncall barriers enhancement New feature or request
#768 opened Jul 13, 2026 by lwDavid Collaborator • Draft
@lwDavid
4
refactor(dsv4): port #734 rope_cs collapse to HCA/SWA decode sparse kernels
#763 opened Jul 13, 2026 by wangqin1723-max Contributor Loading…
4
Add direct CANN FAI decode attention for Qwen3
#755 opened Jul 12, 2026 by ChaoWao Collaborator Loading…
10
fix(dsv4): zero inactive prefill MoE tails
#744 opened Jul 10, 2026 by high-cloud Collaborator Loading…
3
Merge DSV4 compressor ratio modules
#730 opened Jul 8, 2026 by zhaozhaozz Contributor Loading…
12
Fix Qwen3 attention mask NaNs
#714 opened Jul 7, 2026 by superxf Contributor Loading…
1
6
perf(dsv4): qproj wq_b transpose + serving contract; sparse_attn/indexer tuning
#676 opened Jul 2, 2026 by Hzfengsy Member Loading…
6
perf(dsv4 qkv): precompute q-head rope factors
#648 opened Jun 30, 2026 by wangqin1723-max Contributor Loading…
4
perf(dsv4 prefill): avoid indexer pool token scan
#618 opened Jun 25, 2026 by zhaozhaozz Contributor Loading…
2
Add: vector (AIV) row_sum variant for qwen3-14b K/V projection
#586 opened Jun 23, 2026 by Inspiron-st Collaborator Loading…
7
ProTip! Adding no:label will show everything without a label.