Arm SVE CGEMM / ZGEMM Natural Kernels #542
Merged
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
|
In fact I haven't considered writing ZGEMM kernels until I realized there's no But anyway, the kernel's out and it's better than |
|
Data from Fugaku is out. |
|
Excellent! |
|
@xrq-phys have the cgemm/zgemm kernels gotten the beta == 0 treatment? |
|
@devinamatthews Not yet. Working on |
|
Converted to draft while working on beta == 0. @xrq-phys please convert back to a regular PR when it's ready for merging. |
Pic. size seems a bit different from upstream. Generaged w/ MATLAB. Open to any change.
FMOV [hsd]M, #imm does not allow zero immediate. Use wzr, xzr instead.
|
Fixed Pushing rebased for convenience. |
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Outdated suggestions cannot be applied.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.
These new kernels (2vx10) yield a peak efficiency of around 95% on GW4 Isambard.
As GW4 Isambard is observed to be working at a frequency a bit lower than 1.8GHz, these numbers could be a little higher on SC Fugaku (which is very crowded at the moment due to end-of-term job accounting).
Here's also a comparison between
natand1m:The new kernels should compile also with Clang.