Wsl2 problem to execute patch motion correction

I have an error when executing patch motion correction of an .err movies in my laptop with wsl2, I have 32gb RAM and a GPU 4080 with 12 gb. I have 280 movies.

This is the last 40 lines error of the job:

(base) sanjusare@sanjusare:~$ cryosparcm job log P5 J7 | tail -n 40
Warning, Unknown field with tag 65100 (0xfe4c) encountered.
Warning, Unknown field with tag 65100 (0xfe4c) encountered.
TIFFReadDirectory: Warning, Unknown field with tag 65100 (0xfe4c) encountered.
DIE: cuMemsetD32Async(dest=0xb0b400000, value=0x0, num_u32_elements=4194304, stream=(nil)): CUDA ERROR: ()
Total: 7.899s
iosys:1:

iosys:2:
iocb processing [hits 1]: 0.000028s 0.00% (0.038380s 0.49% with children)
iosys_sort_jobs [hits 1]: 0.000001s 0.00%
iosys_open_cacheable_fds [hits 1]: 0.000000s 0.00%
iosys_process_cached_fd_reads [hits 1]: 0.000265s 0.00% (0.000266s 0.00% with children)
io_uring fill sq [hits 1]: 0.000000s 0.00%
io_uring block [hits 1]: 0.000000s 0.00%
io_uring drain cq [hits 1]: 0.000000s 0.00%
slow_path [hits 1]: 0.038086s 0.48%

iosys:0:
iocb processing [hits 1]: 0.000028s 0.00% (6.028147s 76.31% with children)
iosys_sort_jobs [hits 1]: 0.000001s 0.00%
iosys_open_cacheable_fds [hits 1]: 0.000000s 0.00%
iosys_process_cached_fd_reads [hits 1]: 0.001043s 0.01% (0.001044s 0.01% with children)
io_uring fill sq [hits 1]: 0.000000s 0.00%
io_uring block [hits 1]: 0.000000s 0.00%
io_uring drain cq [hits 1]: 0.000000s 0.00%
slow_path [hits 1]: 6.027075s 76.30%

2026-07-21 21:28:42,128 core heartbeat INFO | ========= Updating heartbeat
2026-07-21 21:28:52,133 core heartbeat INFO | ========= Updating heartbeat
/home/sanjusare/cryosparc/cryosparc_worker/.pixi/envs/worker/lib/python3.12/site-packages/numpy/core/fromnumeric.py:3504: RuntimeWarning: Mean of empty slice.
return _methods._mean(a, axis=axis, dtype=dtype,
/home/sanjusare/cryosparc/cryosparc_worker/.pixi/envs/worker/lib/python3.12/site-packages/numpy/core/_methods.py:129: RuntimeWarning: invalid value encountered in scalar divide
ret = ret.dtype.type(ret / rcount)
2026-07-21 21:28:58,587 core.data_management _dump_job_gridfs INFO | Dumping all of job’s images in the database to /home/sanjusare/mimivirus/CS-mimivirus-sofia-ifsc/J7/gridfs_data…
2026-07-21 21:28:58,587 core.data_management _dump_job_gridfs INFO | Done. Exported 0 images in 0.00s
2026-07-21 21:28:58,589 core.data_management _dump_job_events INFO | Done. Dumped 40 files in 0.00s
2026-07-21 21:28:58,595 core.data_management dump_job INFO | Dumped Job P5-J7 (Patch Motion Correction) in 0.01s
========= main process now complete at 2026-07-21 21:28:58.612759
2026-07-21 21:29:02,199 core monitor INFO | ========= monitor detected main process exit.
2026-07-21 21:29:02,202 core monitor INFO | ========= monitor process now complete.
(base) sanjusare@sanjusare:~$

Any idea of what to do?

Thank you,

I have not tested running GPU-accelerated CryoSPARC jobs on WSL2.
Is your CUDA device recognized outside CryoSPARC? For an example test sequence of commands, see CryoSPARC5 can only recognize 6 of 8 GPUs - #2 by wtempel.