Skip to content

[PHI] Fix cuda701 error for Pool3DKernel with big tensor#74063

Merged
huangjiyi merged 1 commit intoPaddlePaddle:developfrom
huangjiyi:fix_pool_cuda701
Jul 16, 2025
Merged

[PHI] Fix cuda701 error for Pool3DKernel with big tensor#74063
huangjiyi merged 1 commit intoPaddlePaddle:developfrom
huangjiyi:fix_pool_cuda701

Conversation

@huangjiyi
Copy link
Member

@huangjiyi huangjiyi commented Jul 16, 2025

PR Category

Operator Mechanism

PR Types

Bug fixes

Description

  • 修复 Pool3dKernel cuda701 报错(之前修完 int32 溢出后出现的报错),测试调小每个 block 的 thread_num 能过,因此当在 int32 溢出时,将 thread_num 由原本的 1024 改为 512
image

Pcard-91573

@paddle-bot
Copy link

paddle-bot bot commented Jul 16, 2025

你的PR提交成功,感谢你对开源项目的贡献!
请关注后续CI自动化测试结果,详情请参考Paddle-CI手册
Your PR has been submitted. Thanks for your contribution!
Please wait for the result of CI firstly. See Paddle CI Manual for details.

@huangjiyi huangjiyi merged commit f2c1ff9 into PaddlePaddle:develop Jul 16, 2025
72 of 73 checks passed
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.

2 participants