fix bugs when compiling with VS2022#75611
fix bugs when compiling with VS2022#75611youge325 wants to merge 27 commits intoPaddlePaddle:developfrom
Conversation
|
你的PR提交成功，感谢你对开源项目的贡献! |
|
使用cuda13在Linux上编译GPU版本的的wheel包无法链接cublas的问题已在 #75509 中解决，在Windows上的解决方案已在 #75654 中解决，cmake命令cmake .. -GNinja -DWITH_GPU=ON -DWITH_UNITY_BUILD=ON -DCUDA_ARCH_NAME=Auto -DWITH_TENSORRT=ON -DTENSORRT_ROOT=D:\TensorRT-10.13.3.9，在RTX3070上编译，驱动程序版本581.29，cuda版本13.0.1，cudnn版本9.13.1，paddle.utils.run_check()可以正常跑通 |
youge325
changed the title
|
/re-run all-failed |
|
/re-run all-failed |
|
#75684 由于代码规范原因尚未合入，这里先提交修改保证通过CI |
Codecov Report❌ Patch coverage is
❌ Your patch status has failed because the patch coverage (0.00%) is below the target coverage (90.00%). You can increase the patch coverage or adjust the target coverage. Additional details and impacted files@@ Coverage Diff @@
## develop #75611 +/- ##
==========================================
Coverage ? 0.00%
==========================================
Files ? 1
Lines ? 16
Branches ? 0
==========================================
Hits ? 0
Misses ? 16
Partials ? 0 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
|
/re-run all-failed |
|
/re-run all-failed |
|
/re-run all-failed |
5 participants
PR Category
Environment Adaptation
PR Types
Bug fixes
Description
修复了使用VS2022编译时遇到的bug，目前cmake .. -GNinja -DWITH_GPU=OFF -DWITH_UNITY_BUILD=ON可以正确编译CPU版本，paddle.utils.run_check()也能跑通，但是GPU版本对CUDA13.0的适配还有点问题，后面会再提一个PR