Adding flexibility for embedded devices. Optimizing A2 by ~1.5% - #329
Open
edwillys wants to merge 9 commits into
Open
Adding flexibility for embedded devices. Optimizing A2 by ~1.5%#329edwillys wants to merge 9 commits into
edwillys wants to merge 9 commits into
Conversation
Fixing bug in test
Fixing Windows build
Conclude the merge with upstream NeuralAmpModelerCore while preserving the embedded portability changes, A2 fast path, binary loading support, and host-only .nam validation boundary.
…mples that wrap the ring. This saves ~1.4% average and 0.5% peak on Cortex-M7
This file contains hidden or 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
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
PR built on top of the Windows compatibility changes from this PR.
Adding some flexibility for embedded targets, where things such as mutex, JSON parsing, exceptions and file system... are not available. Hence introducing
NAM_NO_EXCEPTIONS,NAM_NO_MUTEX,NAM_NO_JSON,NAM_NO_FILESYSTEMOptimizing
_ring_write()with lazy tail mirroring. Previously ,it copied the entire ring prefix into the mirror on every layer and every block. Most blocks do not wrap the ring, so that copy is unnecessary. This saves ~1.5% CPU on cortex M7 and is a generic implementation for all platforms.