Ye Ting

@DamianSzwichtenberg thanks for reporting, we will try to reproduce.

We are able to reproduce the issue. Will investigate the root cause.

@MarekStrachacki Sorry for the late response. We have root caused it and will provide a fix soon. Once ready, will update here.

@jingxu10 Could you please help to reproduce?

Close per above comments from @SAint7579 , thanks.

@jzhoulon root caused this host memory leak issue existed in block format tensor only. We will provide a patch release based on IPEX v2.1.30+xpu soon.

We have fixed the host memory leak issue in IPEX v2.1.30+xpu patch release, the source code is available in https://github.com/intel/intel-extension-for-pytorch/commits/release/xpu/2.1.30/ (https://github.com/intel/intel-extension-for-pytorch/commit/57f2d8fcf265a1c6c29230cd9b864048fef25a54), the binary wheel is available in https://pytorch-extension.intel.com/release-whl/stable/xpu/us/intel-extension-for-pytorch/. You may...

@sobomax Great! Let me close this now. Feel free to reopen or create a new ticket if you still see the issue with the updated release. @Disty0

@BA8F0D39 @fredlarochelle we don't plan to support this. You can still allocate > 4GB with 2.0.110+xpu because we disabled the allocation in master not the previous released drop. Could you...

The issue is from compiler in 2024.2.0 basekit. It will be resolved in the upcoming 2024.2.1 basekit release. For now, the recommended basekit version to work with Intel® Extension for...