NVIDIA / NVIDIA/CUDALibrarySamples
[BUG](Cascade) The Lib does not check the data type of the input data
@ndickson-nvidia is already working on this.
Since Oct 24, 2022.
- Dominant language
- Cuda
- Stars
- 2.5k
- Forks
- 478
- PR merge metrics
- No merged PRs in 30d
Description
Describe the bug
When decompressing the Cascaded Manager gets the data type from arguments which passed to its constructor. Seems It would be far more logical to compute the data type of the input stream because the Cascade compressor stores the data type in compressed stream.
https://github.com/NVIDIA/nvcomp/tree/a6e4e64a177e07cd2e5c8c5e07bb66ffefceae84/src/CascadedKernels.cuh#L1014
Steps/Code to reproduce bug
For example I made a few modifications to the test_cascaded test:
https://github.com/ser-mk/nvcomp/commit/3b90fb0321c6c13ba3678f75dcb610d6d388f916
This test fails because the data type was changed on NVCOMP_TYPE_LONGLONG. If you comment the line https://github.com/ser-mk/nvcomp/commit/3b90fb0321c6c13ba3678f75dcb610d6d388f916#diff-5beab312a7dc19a57e8e9f0d85eb7894b48699f09ff45128401113db280bd5caR112
That test will pass.
Environment details (please complete the following information):
Tesla T4
Cuda compilation tools, release 11.1, V11.1.105
Contributor guide
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Assessment
This issue has not been assessed yet.