Add syclSolverInverter to use oneMKL on Intel GPUs #7
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
ye-luo
reviewed
May 17, 2022
| const size_t m_max = ((m + tile_size - 1) / tile_size) * tile_size; | ||
| const size_t n_max = ((n + tile_size - 1) / tile_size) * tile_size; | ||
|
|
||
| return q.submit([&](sycl::handler& cgh) { |
There was a problem hiding this comment.
Could you remove all handler and use the compact form of q.parallel_for with events.
| int n, | ||
| int lda, | ||
| const TMAT* a, | ||
| const INDEX* pivot, |
There was a problem hiding this comment.
Likely we should just hard-code int64_t
There was a problem hiding this comment.
I don't feel it is better than what CUDA/HIP does due to its blocking nature. CUDA code path just transfer the pivot and diagonal terms to host asynchronously.
| { | ||
| const int N = 911; | ||
|
|
||
| #ifdef MIXED_PRECISION |
There was a problem hiding this comment.
Can we remove ifdef and always test both?
A python script (using autograd) is used to generate reference values for the values of the wavefunction and derivatives at one point in space. The test system is a helium atom with two orbitals and optionally a Jastrow factor.
Add tests for Rotated SPOs using LCAO
Update bora test scripts and remove unused test scripts
The gen_rotated_lcao_wf.py file is extended to generate QMC averages for the parameter gradients. A test for the legacy driver using one thread is introduced.
Do not use cudaDeviceProp maxTexture1D in HIP
Add orbital rotation test with legacy driver
For CI running on nitrogen
Add MPI support to ROCm legacy CI
Rewrite loop to workaround NVHPC bugs
Remove omp parallel over walkers
Add mock-up mw_accept_rejectMove in MSD.
…lone_warning Drop standalone-debug for offload builds
Fix test_structure divide by 0
ye-luo
pushed a commit
that referenced
this pull request
Jun 2, 2024
only use this overload if semantically correct
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Please review the developer documentation
on the wiki of this project that contains help and requirements.
Proposed changes
Add syclColverInveter and support functions in Platforms/SYCL.
What type(s) of changes does this code introduce?
Does this introduce a breaking change?
What systems has this change been tested on?
Checklist