cjy-dystopia #3

gh0s7 · 2026-03-02T21:36:18+08:00

gh0s7 commented

2026-03-02 21:36:18 +08:00

Make a fortune quitely ~

gh0s7 added 13 commits 2026-03-02 21:36:18 +08:00

对prolong3做访存优化 23a82d063b

修改prolong 75dd5353b0

prolong3：提升cache命中率 f70e90f694

Optimize fdderivs: skip redundant 2nd-order work in 4th-order overlap e11363e06e

prolong3: 减少Z-pass 冗余计算 61ccef9f97

fix prolong/restrict index bounds after cherry-pick 12e1f63 466b084a58

optimize fdderivs core-region branch elimination for ghost_width=3 95220a05c8

refactor(Parallel): streamline MPI communication by consolidating request handling and memory management 42c69fab24

perf(MPatch): 用空间 bin 索引加速 Interp_Points 的 block 归属查找 7543d3e8c7

- 为 Patch::Interp_Points 三个重载引入 BlockBinIndex（候选筛选 + 全扫回退）
  - 保持原 point-in-block 判定与后续插值/通信流程不变
  - 将逐点线性扫块从 O(N_points*N_blocks) 降为近似 O(N_points*k)
  - 测试：bin 上限如果太大，会引入不必要的索引构建开销。将 bins 上限设为 16。

Co-authored-by: gpt-5.3-codex

perf(restrict3): shrink X-pass ii sweep to required overlap window e73911f292

- compute fi_min/fi_max from output i-range and derive ii_lo/ii_hi
 - replace full ii sweep (-1:extf(1)) with windowed sweep in Z/Y precompute passes
 - keep stencil math unchanged; add bounds sanity check for ii window

prolong3 改为先算实际 stencil 窗口；只有窗口触及对称边界时才走全域 symmetry_bd，否则只复制必需窗口。restrict3 同样改成窗口判定，无触边时仅填 ii/jj/kk 必需窗口。 b3c367f15b

perf(RestrictProlong): 用 Restrict_cached/OutBdLow2Hi_cached 替换非缓存版本，Sync_finish 改为渐进式解包 4012e9d068

- RestrictProlong/RestrictProlong_aux 中的 Restrict() 和 OutBdLow2Hi() 替换为 _cached 版本，
  复用 gridseg 列表和 MPI 缓冲区，避免每次调用重新分配
- 新增 sync_cache_restrict/sync_cache_outbd 两组 per-level 缓存
- Sync_finish 从 MPI_Waitall 改为 MPI_Waitsome 渐进式解包，降低尾延迟
- AsyncSyncState 扩展 req_node/req_is_recv/pending_recv 字段支持渐进解包

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

perf(transfer_cached): 将 per-call new/delete 的 req_node/req_is_recv/completed 数组移入 SyncCache 复用 5070134857

避免 transfer_cached 每次调用分配释放 3 个临时数组，减少堆操作开销。

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

gh0s7 merged commit 688bdb6708 into main

2026-03-02 21:36:26 +08:00

gh0s7 referenced this issue from a commit

2026-03-02 21:36:27 +08:00

Merge pull request 'cjy-dystopia' (#3) from cjy-dystopia into main

Sign in to join this conversation.

1 Participants

Notifications

Due Date

No due date set.

Dependencies

No dependencies set.

Reference: 64-BitBrainstorm_2026/AMSS-NCKU#3