Does cudaHostRegister still needs the address to be aligned to 4K
Open
Nobody has claimed this yet.
question
- Dominant language
- C++
- Stars
- 9.6k
- Forks
- 2.4k
- Avg merge
- 53m
- Merged PRs (30d)
- 1
Description
The code indicates that the host memory to be registered must be aligned to 4k at both ends. AFAIK this requirement was only there around the era of CUDA 4.0.
Does this restriction still apply? Otherwise IMO it's a good time to update the comments.
Mengda
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.
Research direction
Start with Samples/0_Introduction/simpleZeroCopy/simpleZeroCopy.cu around line 165, then check the current CUDA documentation for cudaHostRegister's alignment requirements. Done means resolving whether the comment is still accurate and updating it if the documented restriction has changed.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- cpp
- Domain
- hpc
- Issue type
- Documentation
- Difficulty
- 3/5
- Estimated time
- 1-2 days
- Activity status
- Stale
- Clarity
- Mostly clear
- Newbie friendliness
- 35/100