文件最后提交记录最后更新时间
feat(examples): add udma_perftest under shmem_perftest Co-authored-by: suqwe<sujianjia@huawei.com> # message auto-generated for no-merge-commit merge: !367 merge feat/udma-perftest into master feat(examples): add udma_perftest under shmem_perftest Created-by: suqwe Commit-by: suqwe Merged-by: cann-robot Description: ## 描述 新增 UDMA 引擎专用的性能测试样例 examples/shmem_perftest/udma_perftest/,平行于已有的 mte_perftest: - **新增样例 udma_perftest**:覆盖 aclshmemx_udma_put_nbi / aclshmemx_udma_get_nbi / aclshmemx_udma_put_signal_nbi 三个 UDMA 低阶接口,模式包含 put / bi_put / get / bi_get / put_signal,配套 main.cpp / udma_perftest_kernel.cpp / run.sh / CMakeLists.txt / README.md。 - **目录重命名**:mte_perftest -> shmem_perftest、内层 inner -> mte_perftest,把 MTE / UDMA / AscendC 三个样例统一收编到 shmem_perftest 下。 - **HBM-only 范围**:UDMA 引擎当前未对 Host 侧 DRAM 提供 RMA 路径,本样例仅测 HBM (DEVICE_SIDE),不再支持 D2H / HOST_SIDE。 - **强制单核**:UDMA 不允许同 peer 多核并发,block_dim=1 固定,-b/--block-size--block-range 仅形式上保留。 - **BW / Latency 双口径**: - --metric bw(默认):prof_start → loop(*_nbi) → quiet → prof_end,窗口含 quiet。 - --metric latprof_start → loop(put_nbi) → prof_end → quietquiet 移到窗口外,仅测下发。 - 两种口径都是单 SHMEMI_PROF_START/END 包整段循环,再除以 loop_count,避免每次循环打点把 pipe_barrier 开销叠到延时数字上。 - **--batch 提交粒度(仅 BW 路径)**: - --batch 0(默认)等价于 --batch <loop_count>,全异步,仅末尾一次 quiet,反映稳态吞吐; - --batch 1 每次 *_nbi 后立刻 quiet,等价于同步提交,反映"提交+完成"端到端开销; - --batch N1 < N < loop_count)每 N 次 *_nbi 一次 quiet,可观察 batch size 与吞吐的关系;loop_count % N != 0prof_end 之前补一次 quiet。 - --metric lat 不受 --batch 影响。 - **put_signal 行为**:测试启动分配对称信号缓冲并初始化为 0,每次 put_signal_nbi 信号值线性递增;数据点结束后 host 端读回并校验 signal_base + warmup + loop_count - 1。 - **配套修复**:put_signal_nbi 内部传给 write_notify 的指针改为 typed pointer,避免 SOC 上 ABI 错位(src/device/gm2gm/engine/shmem_device_udma.hpp)。 ## 关联的Issue Fixes #311 ## 测试 - 在 Ascend950 上 bash scripts/build.sh -examples -soc_type Ascend950 通过;以 PR 当前 head 重新构建 udma_perftest target 链接成功。 - ./run.sh -t put -d float --exponent-range 8 17 --loop-count 1000 跑通;-t bi_put / get / bi_get / put_signal 各跑通,--metric lat -t put 跑通。 - --batch 1 / --batch 16 / --batch 1000 三档跑通,CSV 列与默认 bw 一致;CLI 校验:--batch -1 被 binary 拒绝,--batch abc 被 run.sh 正则拒绝,--metric lat + -t get 组合直接报错退出。 - put_signal 的远端 signal 槽校验通过(signal_base + warmup + loop_count - 1)。 - 非 Ascend950 SOC 上 device kernel 内置 aclshmemi_kernel_abort,按预期退出。 ## 文档更新 - 新增 examples/shmem_perftest/udma_perftest/README.md(含 CLI 参数表、metric 口径说明、--batch 章节、put_signal 行为说明、CSV 输出说明、已知约束)。 - 更新 examples/shmem_perftest/README.md 增加 udma 子目录索引。 - mte_perftest/README.md / ascendc_perftest/README.md 同步路径调整。 ## 类型标签 - [ ] Bug修复 - [x] 新特性 - [ ] 性能优化 - [x] 文档更新 - [ ] 其他,请描述: See merge request: cann/shmem!3676 小时前
原子加高阶接口移除quiet并添加文档说明 Co-authored-by: zhangyunqi<zhangyunqi5@huawei.com> # message auto-generated for no-merge-commit merge: !392 merge fix-opt-atomic-add into master 原子加高阶接口移除quiet并添加文档说明 Created-by: zhangyunqi Commit-by: zhangyunqi Merged-by: cann-robot Description: ## 描述 <!--在这里详细描述你的改动,包括改动的原因和所采取的方法。--> 原子加移除同步并添加文档说明 ## 关联的Issue <!-- 如果这个PR是为了解决特定的Issue,请在这里提供Issue链接。例如:关联Issue #123--> https://gitcode.com/cann/shmem/issues/276 ## 测试 <!--描述进行了哪些测试来验证你的改动。包括但不限于二级冒烟、算子泛化等。--> ![image.png](https://raw.gitcode.com/user-images/assets/8546182/5bcead74-91bb-42b6-bb96-14cf32546fdc/image.png 'image.png') ## 文档更新 <!--如果这个PR包含文档的更新,请在这里指出。例如:更新了README.md文件。--> ## 类型标签 <!-- [x] 表示选中 --> - [x] Bug修复 - [ ] 新特性 - [ ] 性能优化 - [x] 文档更新 - [ ] 其他,请描述: See merge request: cann/shmem!3921 天前
文件名aclshmem改为shmem Co-authored-by: caixilong<caixilong2@h-partners.com> 5 个月前
Added memory detection information reporting Co-authored-by: huangxiaolan<huangxiaolan7@huawei.com> # message auto-generated for no-merge-commit merge: !272 merge add_message_report into master 补充 kernel 侧卡间 barrier 和 signal 语义上报 以及 host 侧共享内存信息上报 Created-by: huangxiaolan Commit-by: huangxiaolan Merged-by: cann-robot Description: ## 描述 <!--在这里详细描述你的改动,包括改动的原因和所采取的方法。--> 补充 kernel 侧卡间 barrier 和 signal 语义上报 以及 host 侧共享内存信息上报 ## 关联的Issue <!-- 如果这个PR是为了解决特定的Issue,请在这里提供Issue链接。例如:关联Issue #123--> <!-- 如果这个PR是为了解决特定的问题单,请在这里描述问题单单号。--> 关联issue #213 ## 测试 <!--描述进行了哪些测试来验证你的改动。包括但不限于二级冒烟、算子泛化等。--> ## 文档更新 <!--如果这个PR包含文档的更新,请在这里指出。例如:更新了README.md文件。--> 不涉及 ## 类型标签 <!-- [x] 表示选中 --> - [ ] Bug修复 - [ ] 新特性 - [ ] 性能优化 - [ ] 文档更新 - [x] 其他,请描述: 补充工具信息上报 See merge request: cann/shmem!2721 个月前
UDMA场景补齐AMO接口 Co-authored-by: YeZZzzz1<yezhenni1@huawei.com> # message auto-generated for no-merge-commit merge: !284 merge master into master UDMA场景补齐AMO接口 Created-by: YeZZzzz1 Commit-by: YeZZzzz1 Merged-by: cann-robot Description: ## 描述 UMDA场景补齐AMO接口,包括:fetch, set, swap, fetch inc, inc, fetch and, and, fetch or, or, fetch xor, xor ## 关联的Issue https://gitcode.com/cann/shmem/issues/199 https://gitcode.com/cann/shmem/issues/205 ## 测试 examples: put ![image.png](https://raw.gitcode.com/user-images/assets/8546182/a65b141e-6e0f-4aea-ba60-861b0352df91/image.png 'image.png') put signal ![image.png](https://raw.gitcode.com/user-images/assets/8546182/92f0405e-5489-4397-b60f-1946aedfa12f/image.png 'image.png') ![image.png](https://raw.gitcode.com/user-images/assets/8546182/1d1eec28-7f76-4239-b6eb-7198df80ea39/image.png 'image.png') FAA ![image.png](https://raw.gitcode.com/user-images/assets/8546182/cab44fdc-54d8-4399-9238-3ac61b2eb719/image.png 'image.png') CAS ![image.png](https://raw.gitcode.com/user-images/assets/8546182/822cc3ea-a87a-4227-aa43-d256a5055713/image.png 'image.png') fetch ![image.png](https://raw.gitcode.com/user-images/assets/8546182/bcd2662d-78f1-4750-8875-3b1db8cf0f5e/image.png 'image.png') set ![image.png](https://raw.gitcode.com/user-images/assets/8546182/1f2e5f4c-a4c2-46a7-a122-1772219b4867/image.png 'image.png') swap ![image.png](https://raw.gitcode.com/user-images/assets/8546182/9f9ee4e1-1916-4c97-be3b-ed89846168e6/image.png 'image.png') fetch inc ![image.png](https://raw.gitcode.com/user-images/assets/8546182/73a0aa4c-a13f-4419-9621-c77438aaa06f/image.png 'image.png') fetch and ![image.png](https://raw.gitcode.com/user-images/assets/8546182/80973f9a-4c74-4e47-9670-0f647faacbb9/image.png 'image.png') fetch or ![image.png](https://raw.gitcode.com/user-images/assets/8546182/b792e1d8-b3db-425a-a2ad-50e2ab5ef7f3/image.png 'image.png') fetch xor ![image.png](https://raw.gitcode.com/user-images/assets/8546182/ab98986f-cafe-467a-a29c-6ede9739f6e8/image.png 'image.png') uttests: ![image.png](https://raw.gitcode.com/user-images/assets/8546182/0b0defe1-0542-424e-b8c7-ea72e0fcf94f/image.png 'image.png') ![image.png](https://raw.gitcode.com/user-images/assets/8546182/082c6f2f-f7a2-484a-bd83-8c05801fa682/image.png 'image.png') ![image.png](https://raw.gitcode.com/user-images/assets/8546182/263d295a-b00d-4c81-bc25-f1badc2e523e/image.png 'image.png') ![image.png](https://raw.gitcode.com/user-images/assets/8546182/32b440df-c37c-4b48-b25f-cc8838c45823/image.png 'image.png') ![image.png](https://raw.gitcode.com/user-images/assets/8546182/ba0784ae-0a64-461f-bd04-36122f9c3ee3/image.png 'image.png') ![image.png](https://raw.gitcode.com/user-images/assets/8546182/754c52fd-385c-4234-98e1-c7fda364d97d/image.png 'image.png') ![image.png](https://raw.gitcode.com/user-images/assets/8546182/8dd015c8-7898-4fbf-ada7-d03a9b9b3699/image.png 'image.png') ![image.png](https://raw.gitcode.com/user-images/assets/8546182/88650c7c-4178-4d85-9b11-7e55ed2ef86f/image.png 'image.png') ## 文档更新 <!--如果这个PR包含文档的更新,请在这里指出。例如:更新了README.md文件。--> ## 类型标签 <!-- [x] 表示选中 --> - [ ] Bug修复 - [x] 新特性 - [ ] 性能优化 - [ ] 文档更新 - [ ] 其他,请描述: See merge request: cann/shmem!2841 个月前
支持Ascend950 mte atomic Co-authored-by: vector5<caobingjie@huawei.com> Co-authored-by: zhangyunqi<zhangyunqi5@huawei.com> Co-authored-by: QK_25415<zhuzhiming1@huawei.com> # message auto-generated for no-merge-commit merge: !283 merge mteatomic into master 支持Ascend950 mte atomic Created-by: vector5 Commit-by: zhangyunqi;vector5;QK_25415 Merged-by: cann-robot Description: ## 描述 <!--在这里详细描述你的改动,包括改动的原因和所采取的方法。--> 支持Ascend950 mte atomic的以下接口和UT T aclshmemx_mte_atomic_fetch(gm T *src, int32_t pe); void aclshmemx_mte_atomic_set(gm T *dst, T value, int32_t pe); T aclshmemx_mte_atomic_compare_swap(gm T *dst, T cond, T value, int32_t pe); T aclshmemx_mte_atomic_swap(gm T *dst, T value, int32_t pe); void aclshmemx_mte_atomic_inc(gm T *dst, int32_t pe); void aclshmemx_mte_atomic_add(gm T *dst, T value, int32_t pe); T aclshmemx_mte_atomic_fetch_inc(gm T *dst, int32_t pe); T aclshmemx_mte_atomic_fetch_add(gm T *dst, T value, int32_t pe); ## 关联的Issue <!-- 如果这个PR是为了解决特定的Issue,请在这里提供Issue链接。例如:关联Issue #123--> <!-- 如果这个PR是为了解决特定的问题单,请在这里描述问题单单号。--> https://gitcode.com/cann/shmem/issues/220 ## 测试 <!--描述进行了哪些测试来验证你的改动。包括但不限于二级冒烟、算子泛化等。--> A2验证ShmemAtomic ![image.png](https://raw.gitcode.com/user-images/assets/8546182/546614a6-b5e6-4124-97ea-b94023431719/image.png 'image.png') MteAtomic ![image.png](https://raw.gitcode.com/user-images/assets/8546182/c6eef612-3641-4bf6-a588-e3647ebbc7d0/image.png 'image.png') A5验证ShmemAtomic ![image.png](https://raw.gitcode.com/user-images/assets/8546182/0eb0a34a-4814-45b2-af42-7b467caeab4d/image.png 'image.png') MteAtomic ![image.png](https://raw.gitcode.com/user-images/assets/8546182/d8da0f93-294a-4fcd-a5ba-593aceaa7a06/image.png 'image.png') ## 类型标签 <!-- [x] 表示选中 --> - [ ] Bug修复 - [x] 新特性 - [ ] 性能优化 - [ ] 文档更新 - [ ] 其他,请描述: See merge request: cann/shmem!2833 天前
feat(issue-280): [Task|任务]: namespace 整改 Co-authored-by: nino888<yinqiran1@huawei.com> # message auto-generated for no-merge-commit merge: !387 merge autodev/issue-280 into master feat(issue-280): [Task|任务]: namespace 整改 Created-by: nino888 Commit-by: nino888 Merged-by: cann-robot Description: ## Summary - Implement issue #280: [Task|任务]: namespace 整改 - Source issue: https://gitcode.com/cann/shmem/issues/280 - Branch: autodev/issue-280 (nino888/shmem -> cann/shmem) ## Changes - examples/dispatch_gmm_combine/include/dispatch_gmm_combine.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_init_routing_quant_v2.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_common.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_expert_token_out.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_fullload_dynamic_quant.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_fullload_quant.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_fullload_quant_base.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_gather_dynamic_quant.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_gather_quant.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_mrgsort.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_mrgsort_out.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_sort_base.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_sort_multi_core.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_sort_one_core.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_src_to_dst_and_gather.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_src_to_dst_op.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_src_to_dst_with_capacity.h - examples/dispatch_gmm_combine/include/moe_token_unpermute.h - examples/dispatch_gmm_combine/include/select_helper.h - examples/dispatch_gmm_combine/include/sync_util.h - examples/dynamic_tiling/impl/kernel/allgather_matmul.h - examples/dynamic_tiling/impl/kernel/allgather_matmul_padding.h - examples/dynamic_tiling/impl/kernel/allgather_matmul_with_gather_result.h - examples/dynamic_tiling/impl/kernel/matmul_allreduce.h - examples/dynamic_tiling/impl/kernel/matmul_reduce_scatter.h - examples/dynamic_tiling/impl/kernel/matmul_reduce_scatter_padding_a.h - examples/dynamic_tiling/impl/kernel/matmul_reduce_scatter_padding_ab.h - examples/dynamic_tiling/impl/kernel/matmul_reduce_scatter_padding_b.h - examples/matmul_allreduce/epilogue/block/epilogue_allreduce.hpp - src/device/gm2gm/shmemi_device_rma.cpp - src/host/bootstrap/shmemi_bootstrap_config_store.cpp - src/host/data_plane/shmem_host_rma.cpp - src/host/entity/mem_entity_default.cpp - src/host/entity/mem_entity_entry.cpp - src/host/init/shmem_init.cpp - src/host/mem/heap/hybm_vmm_based_segment.cpp - src/host/mem/shmem_rma.cpp - src/host/team/shmem_team.cpp - src/host/transport/transport_manager.cpp ## Local Validation - echo 'TODO: replace with real tests, e.g. pytest -q': passed See merge request: cann/shmem!38716 小时前
移除必须0核执行限制 && 修复当前加上-enable_ascendc_dump编译报错的问题 Co-authored-by: vector5<caobingjie@huawei.com> # message auto-generated for no-merge-commit merge: !220 merge fixato into master 移除必须0核执行限制 && 修复当前加上-enable_ascendc_dump编译报错的问题 Created-by: vector5 Commit-by: vector5 Merged-by: cann-robot Description: ## 描述 <!--在这里详细描述你的改动,包括改动的原因和所采取的方法。--> 移除必须0核执行限制,8.3.RC1版本支持[[bisheng::core_ratio(0,1)]] ## 测试 <!--描述进行了哪些测试来验证你的改动。包括但不限于二级冒烟、算子泛化等。--> ![image.png](https://raw.gitcode.com/user-images/assets/8546182/732c8973-1e1b-4cbc-96cd-75d2e33ec5ee/image.png 'image.png') ![image.png](https://raw.gitcode.com/user-images/assets/8546182/a07f1c77-1495-4247-9284-fd16d82e92c7/image.png 'image.png') 8.2.RC1.alp003版本不支持__global__ __aicore__ ![image.png](https://raw.gitcode.com/user-images/assets/8546182/e7679005-0391-4cb8-aac5-0488b319dd72/image.png 'image.png') 8.2.RC1.alp003版本支持[[bisheng::core_ratio(1,1)]] 但不支持[[bisheng::core_ratio(0,1)]] ![image.png](https://raw.gitcode.com/user-images/assets/8546182/4620a99c-4177-4997-b7b8-c34d11cc9170/image.png 'image.png') ## 类型标签 <!-- [x] 表示选中 --> - [x] Bug修复 See merge request: cann/shmem!2202 个月前
aclshmem p2p sync interface Co-authored-by: Liwansi<liwansi@huawei.com> # message auto-generated for no-merge-commit merge: !79 merge p2p_sync into master aclshmem p2p sync interface Created-by: liwansi Commit-by: Liwansi Merged-by: cann-robot Description: ## 描述 对标nvshmem,支持P2P同步接口:wait类接口和test类接口 ## 关联的Issue Issue https://gitcode.com/cann/shmem/issues/57 ## 测试 新增UT: ![image.png](https://raw.gitcode.com/user-images/assets/8546182/6c2c5630-de27-48b7-842f-bf4665e7d644/image.png 'image.png') 所有UT: ![image.png](https://raw.gitcode.com/user-images/assets/8546182/6b8219c2-82c2-41fc-80f1-d32657fe8efd/image.png 'image.png') ## 文档更新 /docs/pythonAPI.md ## 类型标签 <!-- [x] 表示选中 --> - [ ] Bug修复 - [x] 新特性 - [ ] 性能优化 - [ ] 文档更新 - [ ] 其他,请描述: See merge request: cann/shmem!794 个月前
文件名aclshmem改为shmem Co-authored-by: caixilong<caixilong2@h-partners.com> 5 个月前
include 目录下 aclshmemi_ 内容整改 Co-authored-by: zhu-wangyi<zhuwangyi@huawei.com> # message auto-generated for no-merge-commit merge: !297 merge bug_fix/sync_bit_rename into master include 目录下 aclshmemi_ 内容整改 Created-by: zhu-wangyi Commit-by: zhu-wangyi Merged-by: cann-robot Description: ## 描述 include 目录下 aclshmemi_ 内容整改 ## 关联的Issue https://gitcode.com/cann/shmem/issues/206 ## 测试 <!--描述进行了哪些测试来验证你的改动。包括但不限于二级冒烟、算子泛化等。--> ## 文档更新 <!--如果这个PR包含文档的更新,请在这里指出。例如:更新了README.md文件。--> ## 类型标签 <!-- [x] 表示选中 --> - [x] Bug修复 - [ ] 新特性 - [ ] 性能优化 - [ ] 文档更新 - [ ] 其他,请描述: See merge request: cann/shmem!29730 天前
aclshmem p2p sync interface Co-authored-by: Liwansi<liwansi@huawei.com> # message auto-generated for no-merge-commit merge: !79 merge p2p_sync into master aclshmem p2p sync interface Created-by: liwansi Commit-by: Liwansi Merged-by: cann-robot Description: ## 描述 对标nvshmem,支持P2P同步接口:wait类接口和test类接口 ## 关联的Issue Issue https://gitcode.com/cann/shmem/issues/57 ## 测试 新增UT: ![image.png](https://raw.gitcode.com/user-images/assets/8546182/6c2c5630-de27-48b7-842f-bf4665e7d644/image.png 'image.png') 所有UT: ![image.png](https://raw.gitcode.com/user-images/assets/8546182/6b8219c2-82c2-41fc-80f1-d32657fe8efd/image.png 'image.png') ## 文档更新 /docs/pythonAPI.md ## 类型标签 <!-- [x] 表示选中 --> - [ ] Bug修复 - [x] 新特性 - [ ] 性能优化 - [ ] 文档更新 - [ ] 其他,请描述: See merge request: cann/shmem!794 个月前
aclshmem p2p sync interface Co-authored-by: Liwansi<liwansi@huawei.com> # message auto-generated for no-merge-commit merge: !79 merge p2p_sync into master aclshmem p2p sync interface Created-by: liwansi Commit-by: Liwansi Merged-by: cann-robot Description: ## 描述 对标nvshmem,支持P2P同步接口:wait类接口和test类接口 ## 关联的Issue Issue https://gitcode.com/cann/shmem/issues/57 ## 测试 新增UT: ![image.png](https://raw.gitcode.com/user-images/assets/8546182/6c2c5630-de27-48b7-842f-bf4665e7d644/image.png 'image.png') 所有UT: ![image.png](https://raw.gitcode.com/user-images/assets/8546182/6b8219c2-82c2-41fc-80f1-d32657fe8efd/image.png 'image.png') ## 文档更新 /docs/pythonAPI.md ## 类型标签 <!-- [x] 表示选中 --> - [ ] Bug修复 - [x] 新特性 - [ ] 性能优化 - [ ] 文档更新 - [ ] 其他,请描述: See merge request: cann/shmem!794 个月前
feat(issue-280): [Task|任务]: namespace 整改 Co-authored-by: nino888<yinqiran1@huawei.com> # message auto-generated for no-merge-commit merge: !387 merge autodev/issue-280 into master feat(issue-280): [Task|任务]: namespace 整改 Created-by: nino888 Commit-by: nino888 Merged-by: cann-robot Description: ## Summary - Implement issue #280: [Task|任务]: namespace 整改 - Source issue: https://gitcode.com/cann/shmem/issues/280 - Branch: autodev/issue-280 (nino888/shmem -> cann/shmem) ## Changes - examples/dispatch_gmm_combine/include/dispatch_gmm_combine.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_init_routing_quant_v2.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_common.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_expert_token_out.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_fullload_dynamic_quant.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_fullload_quant.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_fullload_quant_base.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_gather_dynamic_quant.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_gather_quant.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_mrgsort.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_mrgsort_out.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_sort_base.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_sort_multi_core.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_sort_one_core.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_src_to_dst_and_gather.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_src_to_dst_op.h - examples/dispatch_gmm_combine/include/moe_init_routing_quant_v2/moe_v2_src_to_dst_with_capacity.h - examples/dispatch_gmm_combine/include/moe_token_unpermute.h - examples/dispatch_gmm_combine/include/select_helper.h - examples/dispatch_gmm_combine/include/sync_util.h - examples/dynamic_tiling/impl/kernel/allgather_matmul.h - examples/dynamic_tiling/impl/kernel/allgather_matmul_padding.h - examples/dynamic_tiling/impl/kernel/allgather_matmul_with_gather_result.h - examples/dynamic_tiling/impl/kernel/matmul_allreduce.h - examples/dynamic_tiling/impl/kernel/matmul_reduce_scatter.h - examples/dynamic_tiling/impl/kernel/matmul_reduce_scatter_padding_a.h - examples/dynamic_tiling/impl/kernel/matmul_reduce_scatter_padding_ab.h - examples/dynamic_tiling/impl/kernel/matmul_reduce_scatter_padding_b.h - examples/matmul_allreduce/epilogue/block/epilogue_allreduce.hpp - src/device/gm2gm/shmemi_device_rma.cpp - src/host/bootstrap/shmemi_bootstrap_config_store.cpp - src/host/data_plane/shmem_host_rma.cpp - src/host/entity/mem_entity_default.cpp - src/host/entity/mem_entity_entry.cpp - src/host/init/shmem_init.cpp - src/host/mem/heap/hybm_vmm_based_segment.cpp - src/host/mem/shmem_rma.cpp - src/host/team/shmem_team.cpp - src/host/transport/transport_manager.cpp ## Local Validation - echo 'TODO: replace with real tests, e.g. pytest -q': passed See merge request: cann/shmem!38716 小时前
include 目录下 aclshmemi_ 内容整改 Co-authored-by: zhu-wangyi<zhuwangyi@huawei.com> # message auto-generated for no-merge-commit merge: !297 merge bug_fix/sync_bit_rename into master include 目录下 aclshmemi_ 内容整改 Created-by: zhu-wangyi Commit-by: zhu-wangyi Merged-by: cann-robot Description: ## 描述 include 目录下 aclshmemi_ 内容整改 ## 关联的Issue https://gitcode.com/cann/shmem/issues/206 ## 测试 <!--描述进行了哪些测试来验证你的改动。包括但不限于二级冒烟、算子泛化等。--> ## 文档更新 <!--如果这个PR包含文档的更新,请在这里指出。例如:更新了README.md文件。--> ## 类型标签 <!-- [x] 表示选中 --> - [x] Bug修复 - [ ] 新特性 - [ ] 性能优化 - [ ] 文档更新 - [ ] 其他,请描述: See merge request: cann/shmem!29730 天前