| 文件 | 最后提交记录 | 最后更新时间 |
|---|---|---|
fix: 排序Reduce tiling的归一化轴，消除乱序输入的结果不一致 Co-authored-by: wulinyu<wulinyu4@huawei.com> # message auto-generated for no-merge-commit merge: !62 merge fix/reduce-tiling-unsorted-axes into master fix: 排序Reduce tiling的归一化轴，消除乱序输入的结果不一致 Created-by: wulinyu Commit-by: wulinyu Merged-by: cann-robot Description: ## 描述 修复 CalcReduceTiling 对同一约减轴集合不同输入顺序产生不同 tiling 结果的问题。 **复现**（独立Host程序，CANN 9.3.0，未修改源码）： shape = [2, 3, 4, 5, 6, 7, 8] axes {0, 3, 4, 6} -> policy={51,20,0} outSize=84 axes {4, 3, 6, 0} -> policy={40,20,0} outSize=672 与 issue 报告完全一致。 **根因**：EliminateOne 对 axes 就地 +1 后按 shape 顺序遍历 erase；MergeAxis 假设轴索引升序——连续 R 轴合并依赖 iter0/iter1 的位置关系，乱序输入时 erase 可能命中错误元素并留下失效迭代器（libstdc++ iterator checks 下解引用失效迭代器即暴露）。 **修复**（include/reduce/host/tiling/reduce_tiling.h，+3 行）：在 Run() 中 MakeWrapDim 归一化与 IsAxesValid 校验之后、EliminateOne 之前，对约减轴执行 std::sort——消除后续算法对输入顺序的隐含假设，任意输入顺序得到一致结果： diff if (IsAxesValid(opInput_.reduceShape, opInput_.reduceDim) == -1) { return -1; } +// EliminateOne/MergeAxis assume ascending axis indices; sort the normalized axes so +// that the same axis set produces identical tiling regardless of input order. +std::sort(opInput_.reduceDim.begin(), opInput_.reduceDim.end()); std::vector<uint64_t> newShape(ATVC::MAX_DIM, 1); 排序在归一化（负轴转正）与校验（范围/数量合法性）之后执行，不影响原有错误检查语义。 ## 关联的Issue Fixes #23 ## 测试 新增 Host 回归 TestAtvcTiling.TestAtvcTilingReduceSumUnsortedAxesCase：同一轴集合 {0,3,4,6} 以升序/乱序两种形式调用，逐项比较： - policy 三元组（patternID / loopARCount / loopInnerARCount）； - workspaceSize 与标量 tiling 字段（outSize / meanVar / groupR）； - 全部 MAX_DIM 个 shape / stride / dstStride 槽位（比较前对 param 值初始化清零，排除未写入槽位的脏值干扰）。 验证记录（CANN 9.3.0 Host 环境）： 1. **修复前**：回归用例失败——policy={51,20,0} vs {40,20,0}、outSize=84 vs 672（独立复现程序与 gtest 用例双重确认）； 2. **修复后**：回归用例通过，同一轴集合两种顺序结果完全一致（policy={51,20,0}、outSize=84）； 3. **存量回归**：tiling 全部 8 个用例（EleWise×2 / ReduceSum×3 / Broadcast×2 / 本用例）全部通过。 说明：与 issue 一致，本验证为 Host 侧 tiling 路径；缺陷发生在 kernel launch 之前，无需 Device 侧数值运行。 ## 文档更新 N/A ## 类型标签 - [x] 🐛 fix: Bug 修复 - [ ] ✨ feat: 新功能 - [ ] ⚡ perf: 性能优化 - [ ] ♻️ refactor: 代码重构 - [ ] 🧪 test: 新增或修改测试 - [ ] 📝 docs: 文档更新 - [ ] 🔧 ci: CI/CD 配置修改 - [ ] ↩️ revert: 回退 - [ ] 🧹 chore: 其他，请具体描述 🤖 Generated with [Claude Code](https://claude.com/claude-code) See merge request: cann/atvc!62 | 10 天前 | |
Fix multi-translation-unit links for ATVC host headers Co-authored-by: KaranocaVe<karanocave@qq.com> # message auto-generated for no-merge-commit merge: !51 merge fix/atvc-public-headers-multi-tu-odr-01 into master Fix multi-translation-unit links for ATVC host headers Created-by: KaranocaVe Commit-by: KaranocaVe Merged-by: cann-robot Description: # Fix multi-translation-unit links for ATVC host headers ## 描述 Several non-template helpers included by atvc.h have external linkage but are not inline. Linking two operator host translation units that include the public header therefore fails with duplicate definitions. Mark the five affected helpers inline; their behavior is unchanged. ## 测试 The host fixture reproduces the duplicate definitions on the baseline and links and runs with the fix in both single- and multi-translation-unit cases. ## 文档更新 无。 ## 类型标签 - [x] Bug修复 - [ ] 新特性 - [ ] 性能优化 - [ ] 文档更新 - [ ] 其他，请描述： 关联 Issue: https://gitcode.com/cann/atvc/issues/26 See merge request: cann/atvc!51 | 18 天前 |