Skip to content

[pull] develop from PaddlePaddle:develop#268

Merged
HydrogenSulfate merged 16 commits intoHydrogenSulfate:developfrom
PaddlePaddle:develop
Jul 4, 2024
Merged

[pull] develop from PaddlePaddle:develop#268
HydrogenSulfate merged 16 commits intoHydrogenSulfate:developfrom
PaddlePaddle:develop

Conversation

@pull
Copy link

@pull pull bot commented Jul 4, 2024

See Commits and Changes for more details.


Created by pull[bot]

Can you help keep this open source service alive? 💖 Please sponsor : )

tc20042008 and others added 16 commits July 4, 2024 10:07
* dump FeedOp tensor meta

* dump pir program only once

---------

Co-authored-by: jiahy0825 <jiahongyu@baidu.com>
Co-authored-by: lawrence910426 <lawu@nvidia.com>
…65094)

* Either stream safe or async allocator

* Ignore if not enabled

* fix: ignore cuda managed

* fix: disable async allocator

* fix: either async or stream safe

* fix useless if

---------

Co-authored-by: lawrence910426 <lawu@nvidia.com>
* inference use FLAGS_enable_pir_api control pir mode

* fix ut

* fix
* add decimals for round

* set defalut value

* fix

* fix round inplace

* add round inplace func

* empty

* fix round  on onednn

* fix

* remove redundant comments

* re-run

* change calculation process

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix bug

* fix coverage

* add attr in yaml file
…it in INT8 GEMM. (#65597)

* [Inference] Refine global search optimization for cuBLASLt and apply it in INT8 GEMM
* Fix

* Fix

* Fix

* Fix

* ci
…niform.py` (#65660)


---------

Co-authored-by: Nyakku Shigure <sigure.qaq@gmail.com>
…/worker.py` (#65645)


---------

Co-authored-by: SigureMo <sigure.qaq@gmail.com>
* [DCU][XPU] add develop dockerfile for dcu and xpu

* update comments
@HydrogenSulfate HydrogenSulfate merged commit 6d3d314 into HydrogenSulfate:develop Jul 4, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.