VkFFT v1.3.3 release #147
Merged
VkFFT v1.3.3 release #147
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
VkFFT v1.3.2 release
-This update removes the limit of ~2^12 for R2C and R2R systems - they can all now be done in up to three uploads with coverage ~2^32 for all dimensions, same as C2C. -Added versions of all R2C and R2R algorithms, implementad as load/store callbacks. This functionality will be enchanced in the future to support arbitrary user callbacks (I just need to find out how this can be done for a multiple-API user-interaction). -Restructured internal kernel typing enumeration.
… are hanging thread blocks
… of how the type of these transforms is stroed in v1.3.2 (vincefn/pyvkfft#32)
- Fixed incorrect LUT referencing for inverse DCT/DST-II/III multi-upload algorithm -Fixed incorrect R2C layout management for non-strided axes - Fixed incorrect twiddle factors calculation for inverse DST-I
- Fixed incorrect twiddle factors calculation for inverse DST-I Bluestein transform
…d when callback algorithm was used. (vincefn/pyvkfft#32)
…istent usage of FP container types.
-Fixed inconsistent usage of shared memory in dct/dst-I Bluestein if original sequence is power of 2 and Bluestein sequence is not. -Added boundary guard for DCT/DST-IV callback version. -Added a weird boundary guard for DST-II/III that should do nothing but fixes out of nounds accesses in some strided cases.
-Fixed inconsistent usage of swapTo3Stage4Step for Bluestein sequences.
-Fixed inconsistent read check for second value in callback version -Small Bluestein sequence handler improvement
-Disabled incorrect reuse of non-strided LUT for strided axes
-Also changed how dispatch of push constants works on CUDA and HIP backends (setting constants as argument should be better than address copy).
-Also fixed an issue when switch to logical workgroup count increase sometimes being not enabled during kernel compilation.
…generations for H100/A100 (vincefn/pyvkfft#32)
Improve LevelZero detection on Ubuntu
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.
Multi-upload R2C and R2R algorithms
-This update removes the limit of ~2^12 for R2C and R2R systems - they can all now be done in up to three uploads with coverage ~2^32 for all dimensions, same as C2C.
-Added versions of all R2C and R2R algorithms, implemented as load/store callbacks. This functionality will be enhanced in the future to support arbitrary user callbacks (I just need to find out how this can be done for a multiple-API user interaction).
-Bugfixes