a question when i try to use the project

[dazuihoudaidai]

Dear author,
It is a perfect work. However, When i use FaceVerse with the code python tracking_online_cuda.py，there are some errors. The window collected to camera pop up，but it is black and no reaction. Only my click turns to cross. The errors trouble me few days. How can i solve it?
the same error all in online,offline and StyleAvatar

`[e 1101 16:30:33.663000 20 log.cc:565] cl : Command line warning D9025 : overriding '/EHa' with '/EHs'
cub_where__Ti_bool__To_int32__NDIM_2__JIT_1__JIT_cuda_1__index_t_int32_hash_21a0093dfe861907_op.cc
c:\users\miniconda3\envs\styleavatar\include\cuda\std\detail\libcxx\include__type_traits/is_constant_evaluated.h(27): error: identifier "__builtin_is_constant_evaluated" is undefined

c:\users\miniconda3\envs\styleavatar\include\cuda\std\detail\libcxx\include__type_traits/is_constant_evaluated.h(32): error: identifier "__builtin_is_constant_evaluated" is undefined

2 errors detected in the compilation of "C:/Users/.cache/jittor/jt1.3.8/cl/py3.8.16/Windows-10-10.x78/13thGenIntelRCx8b/default/cu11.2.67/jit/cub_where__Ti_bool__To_int32__NDIM_2__JIT_1__JIT_cuda_1__index_t_int32_hash_21a0093dfe861907_op.cc".
cub_where__Ti_bool__To_int32__NDIM_2__JIT_1__JIT_cuda_1__index_t_int32_hash_21a0093dfe861907_op.cc`

Thanks you!

[LizhenWangT]

Sorry, I have not seen this error before. My Jittor version is 1.3.4 or 1.3.5. Maybe you can check if your jittor works well or try an earlier version.

python -m pip install jittor
python -m jittor.test.test_core
python -m jittor.test.test_example
python -m jittor.test.test_cudnn_op

[LizhenWangT]

I reproduced this bug just now on Jittor==1.3.8.5. Please use Jittor==1.3.5.19. It seems that there is a problem with jittor's backward compatibility

[dazuihoudaidai]

Thank you so so so much!!! I will try later and I am looking forward to using your perfect project ealier.

[dazuihoudaidai]

Dear author,
It is a really odd thing.I try to use jittor 1.3.5.19 but the result is an error **”compute_89 is not defined for your gpu architecture".**The same error is all in many previous jittor versions (no matter what cuda version I try ,11.3,11.7,12.0,ect.). Only jittor 1.3.8.5 can successfully run core , test and Faceverse.But I will meet the error i mentioned yesterday.

Is it a promblem that previous jittor version fail to suit my gpu arch? Is there a way to reset my gpu arch？Is there another method to solve it ，or just wait for jittor to deal with backward compatibility problem. It seems i am really in dellima. orz

Thank you very much