文件最后提交记录最后更新时间
OP hash supports thread-level Co-authored-by: wang-guangbin<wgb_strive@163.com> # message auto-generated for no-merge-commit merge: !3867 merge 7.3.0 into 7.3.0 OP hash supports thread-level Created-by: wang-guangbin Commit-by: wang-guangbin Merged-by: ascend-robot Description: <!-- Thanks for sending a pull request! --> **What type of PR is this?** > Uncomment only one /kind <> line, hit enter to put that in a new line, and remove leading whitespaces from that line: > > /kind bug > /kind task > /kind feature **What does this PR do / why do we need it**: op仓的atb op复用采用线程级单例。 当前op仓的atb缓存设计为进程级别的单列,多线程场景存在op复用冲突,atb底层的setup和exector不支持相同的op在不同线程。 **Special notes for your reviewers**: See merge request: Ascend/op-plugin!38675 个月前
!2596 [DFX] More detail info in op_log and support common OP_LOG. Merge pull request !2596 from 刘嘉巍/master-dev 11 个月前
!2284 Clean code Merge pull request !2284 from wgb/clean_code 1 年前
!1242 Update copyright * Update copyright 2 年前
!2595 [Feature] Support npu_add_rms_norm_quant. Merge pull request !2595 from 刘嘉巍/log 11 个月前
!2595 [Feature] Support npu_add_rms_norm_quant. Merge pull request !2595 from 刘嘉巍/log 11 个月前
!2637 add process of high type cast for clamp, clamp_max and clamp_min Merge pull request !2637 from zhangqiongwen/master_clamp_dtype_cast 11 个月前
!2637 add process of high type cast for clamp, clamp_max and clamp_min Merge pull request !2637 from zhangqiongwen/master_clamp_dtype_cast 11 个月前
Fix avg_pool2d infershape bug Co-authored-by: wang-guangbin<wgb_strive@163.com> # message auto-generated for no-merge-commit merge: !3606 merge 7.3.0 into 7.3.0 Fix avg_pool2d infershape bug Created-by: wang-guangbin Commit-by: wang-guangbin Merged-by: ascend-robot Description: <!-- Thanks for sending a pull request! --> **What type of PR is this?** > Uncomment only one /kind <> line, hit enter to put that in a new line, and remove leading whitespaces from that line: > > /kind bug > /kind task > /kind feature **What does this PR do / why do we need it**: **Special notes for your reviewers**: See merge request: Ascend/op-plugin!36066 个月前
[fix]split_with_sizes_copy performance reduction Co-authored-by: zqwenn<zhangqiongwen@huawei.com> # message auto-generated for no-merge-commit merge: !3494 merge 7.3.0_split_fix into 7.3.0 [fix]split_with_sizes_copy performance reduction Created-by: zqwenn Commit-by: zqwenn Merged-by: ascend-robot Description: <!-- Thanks for sending a pull request! --> **What type of PR is this?** > Uncomment only one /kind <> line, hit enter to put that in a new line, and remove leading whitespaces from that line: > > /kind bug > /kind task > /kind feature **What does this PR do / why do we need it**: 融合算子只在部分场景性能有优化,非优化场景下aclnn的性能劣于小算子拼接性能。这个pr修复性能出现劣化的场景,只保留特定场景接入aclnn,其余场景走小算子。 **Special notes for your reviewers**: See merge request: Ascend/op-plugin!34946 个月前
!1713 fix ops * fix ops 1 年前
!1242 Update copyright * Update copyright 2 年前
[fix] update unfold_backward check cann version Co-authored-by: zqwenn<zhangqiongwen@huawei.com> # message auto-generated for no-merge-commit merge: merge master_unfold_backward_cann_check into master [fix] update unfold_backward check cann version Created-by: zqwenn Commit-by: zqwenn Merged-by: ascend-robot Description: <!-- Thanks for sending a pull request! --> **What type of PR is this?** > Uncomment only one /kind <> line, hit enter to put that in a new line, and remove leading whitespaces from that line: > > /kind bug > /kind task > /kind feature **What does this PR do / why do we need it**: **Special notes for your reviewers**: See merge request: Ascend/op-plugin!32158 个月前
[fix] update unfold_backward check cann version Co-authored-by: zqwenn<zhangqiongwen@huawei.com> # message auto-generated for no-merge-commit merge: merge master_unfold_backward_cann_check into master [fix] update unfold_backward check cann version Created-by: zqwenn Commit-by: zqwenn Merged-by: ascend-robot Description: <!-- Thanks for sending a pull request! --> **What type of PR is this?** > Uncomment only one /kind <> line, hit enter to put that in a new line, and remove leading whitespaces from that line: > > /kind bug > /kind task > /kind feature **What does this PR do / why do we need it**: **Special notes for your reviewers**: See merge request: Ascend/op-plugin!32158 个月前
Checkout v2r10 Co-authored-by: dilililiwhy<why.wuhuanyu@huawei.com> # message auto-generated for no-merge-commit merge: merge pr_3154_1757822748315 into master Checkout v2r10 Created-by: dilililiwhy Commit-by: dilililiwhy Merged-by: ascend-robot Description: 1. add v2r10 --- <!-- Thanks for sending a pull request! --> **What type of PR is this?** > Uncomment only one /kind <> line, hit enter to put that in a new line, and remove leading whitespaces from that line: > > /kind bug > /kind task > /kind feature **What does this PR do / why do we need it**: **Special notes for your reviewers**: See merge request: Ascend/op-plugin!31548 个月前
add converType for OptionalIntArrayRef Co-authored-by: DaiFu<daifu2@huawei.com> # message auto-generated for no-merge-commit merge: !3347 merge 2510ConvertType into master add converType for OptionalIntArrayRef Created-by: daifu1234567 Commit-by: DaiFu Merged-by: ascend-robot Description: **What type of PR is this?** > Uncomment only one /kind <> line, hit enter to put that in a new line, and remove leading whitespaces from that line: > > /kind feature **What does this PR do / why do we need it**: actual_seq_qlen is an optional input of npu_nsa_compress_attention_infer, when it is none, kernel wants a nullptr, but now it is IntArrayRef{} **Special notes for your reviewers**: 用例本地已跑通,skip是因为需要更新cann: ![image.png](https://raw.gitcode.com/user-images/assets/7403085/ded2fda2-7a06-43e7-a203-147e601f6a69/image.png 'image.png') 问题场景用例跑通: ![image.png](https://raw.gitcode.com/user-images/assets/7403085/85d959f9-02b7-4708-a6a0-43611828104a/image.png 'image.png') ![image.png](https://raw.gitcode.com/user-images/assets/7403085/868d3fdc-aa19-4be2-ab04-f08f51f9837c/image.png 'image.png') See merge request: Ascend/op-plugin!33477 个月前
add libopapi_oam.so for stress test Co-authored-by: huanyongben<huanyongben1@hisilicon.com> # message auto-generated for no-merge-commit merge: !3765 merge 7.3.0 into 7.3.0 add libopapi_oam.so for stress test Created-by: huanyongben Commit-by: huanyongben Merged-by: ascend-robot Description: <!-- Thanks for sending a pull request! --> **What type of PR is this?** > Uncomment only one /kind <> line, hit enter to put that in a new line, and remove leading whitespaces from that line: > > /kind bug > /kind task > /kind feature **What does this PR do / why do we need it**: **Special notes for your reviewers**: See merge request: Ascend/op-plugin!37655 个月前
!2789 [DFX] Delete skip in op_log Merge pull request !2789 from 刘嘉巍/log 10 个月前
!3022 [DFX] OP_EXEC_LOG for npu aten and get data_ptr for tensor Merge pull request !3022 from 刘嘉巍/log 9 个月前
support faketensor in log Co-authored-by: DaiFu<daifu2@huawei.com> # message auto-generated for no-merge-commit merge: !3442 merge 2511faketensor_log into master support faketensor in log Created-by: daifu1234567 Commit-by: DaiFu Merged-by: ascend-robot Description: support fake tensor in log 修复日志中faketensor没有data_ptr的问题 **What type of PR is this?** > Uncomment only one /kind <> line, hit enter to put that in a new line, and remove leading whitespaces from that line: > > /kind bug **What does this PR do / why do we need it**: **Special notes for your reviewers**: See merge request: Ascend/op-plugin!34426 个月前
!2284 Clean code Merge pull request !2284 from wgb/clean_code 1 年前