libressl: add assembly for aarch64, ppc*, extra assembly for arm #23655
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
q66
force-pushed
the
lssl-asm
branch
3 times, most recently
from
41e6e86
to
4758ad6
Compare
|
I have no opinion on the switch to OpenSSL, but reducing the cost to deferring that decision is an attractive option. |
This yields significant speedups (20x) on platforms with hardware crypto support (e.g. POWER8 and newer, and most aarch64) as well as enhances performance of assorted alogrithms by using optimized assembly implementations instead of portable fallbacks. As it is now, libressl only supports asm for x86(_64) and to a limited degree, 32-bit ARM. Everything was tested, test suite passes on all platforms, and benchmarks were run on relevant hardware, confirming the speedups.
q66
changed the title
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
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Outdated suggestions cannot be applied.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.
This yields significant speedups (20x) in some things (AES/ghash) on platforms with hardware crypto support (e.g. POWER8 and newer, and most aarch64) as well as enhances performance of assorted alogrithms by using optimized assembly implementations instead of portable fallbacks. E.g. on aarch64 https://gist.githubusercontent.com/q66/c75a46fc7d0c27eff21cc9d70639bf95/raw/dde1f60e304ef8f4a4049d41388d9c25e63b4a1e/gistfile1.txt
As it is now, libressl only supports asm for x86(_64) and to a limited degree, 32-bit ARM. The extra support was added by my project: https://github.com/q66/libressl-portable-asm
Everything was tested, test suite passes on all platforms, and benchmarks were run on relevant hardware, confirming the speedups.
I still think we should move back to OpenSSL, since there are many reasons to do so (even disregarding performance) and the performance is still better on it (as there are various algos that don't have optimized implementations under libressl at all) but this could be a decent stopgap solution, enabling at least basic things (like hardware AES on non-x86 systems).
Tested:
Benchmarked:
@void-linux/pkg-committers