crypto/cipher: add optimized assembly xorBytes for ARM (NEON + non-NEON) #53154
+180
−9
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Outdated suggestions cannot be applied.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.
xorBytes is used extensively in crypto and can sometimes be the
bottleneck (e.g. CTR). 32-bit ARM is still the most common arch
for IoT, it is a slower platform so crypto optimization is important,
but unfortunately crypto/cipher does not have optimized xorBytes
for ARM.
This PR adds optimized assembly language xorBytes
implementations for both NEON and non-NEON ARM, which can
result in very large performance gains.
Example comparing current tip crypto/cipher xorBytes on
ARMv7+NEON, with results after applying this PR.
Additionally, this PR improves xor_generic.go to use word copy
for xorBytes on all arches if the dst and src buffers are all aligned,
which in real world usage is often the case. Previously word copy
was used only on platforms that supported unaligned access,
which is unnecessarily strict and left performance on the table.
Fixes #53023