Kyber Aarch64: assembly implementations of functions #7998
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
SparkiDev
force-pushed
the
kyber_aarch64_asm
branch
2 times, most recently
from
436df89
to
b829134
Compare
dgarske
requested changes
Sep 25, 2024
configure.ac
Outdated
| @@ -2977,7 +2977,7 @@ then | |||
| AM_CPPFLAGS="$AM_CPPFLAGS+sm4" | |||
| fi | |||
| else | |||
| AM_CPPFLAGS="$AM_CPPFLAGS -mcpu=generic+crypto" | |||
| AM_CPPFLAGS="$AM_CPPFLAGS -march=armv8.1-a+crypto" | |||
There was a problem hiding this comment.
Customers target is not v8.1. Needs to be v8.0 -march=armv8-a+crypto, however that does not allow sqrdmlsh. Is there a way to not use sqrdmlsh?
CC wolfcrypt/src/src_libwolfssl_la-dilithium.lo
wolfcrypt/src/port/arm/armv8-kyber-asm.S: Assembler messages:
wolfcrypt/src/port/arm/armv8-kyber-asm.S:266: Error: selected processor does not support `sqrdmlsh v21.8h,v29.8h,v4.h[0]'
wolfcrypt/src/port/arm/armv8-kyber-asm.S:267: Error: selected processor does not support `sqrdmlsh v22.8h,v30.8h,v4.h[0]'
wolfcrypt/src/port/arm/armv8-kyber-asm.S:274: Error: selected processor does not support `sqrdmlsh v23.8h,v29.8h,v4.h[0]'
wolfcrypt/src/port/arm/armv8-kyber-asm.S:275: Error: selected processor does not support `sqrdmlsh v24.8h,v30.8h,v4.h[0]'
wolfcrypt/src/port/arm/armv8-kyber-asm.S:282: Error: selected processor does not support `sqrdmlsh v25.8h,v29.8h,v4.h[0]'
wolfcrypt/src/port/arm/armv8-kyber-asm.S:283: Error: selected processor does not support `sqrdmlsh v26.8h,v30.8h,v4.h[0]'
wolfcrypt/src/port/arm/armv8-kyber-asm.S:290: Error: selected processor does not support `sqrdmlsh v27.8h,v29.8h,v4.h[0]'
wolfcrypt/src/port/arm/armv8-kyber-asm.S:291: Error: selected processor does not support `sqrdmlsh v28.8h,v30.8h,v4.h[0]'
FYI: When I run on ZCU102 A53 I get:
./configure --host=aarch64 CC="aarch64-linux-gnu-gcc" AR="aarch64-linux-gnu-ar" RANLIB="aarch64-linux-gnu-ranlib" --enable-sp=yes,asm --enable-keygen --enable-armasm --enable-experimental --enable-kyber --enable-dilithium --enable-keygen --enable-lms --enable-xmss --enable-curve25519 --enable-ed25519 --enable-curve448 --enable-ed448 --disable-shared --enable-static --disable-dh --disable-filesystem && make
./benchmark -kyber
------------------------------------------------------------------------------
wolfSSL version 5.7.2
------------------------------------------------------------------------------
Math: Multi-Precision: Wolf(SP) word-size=64 bits=4096 sp_int.c
Single Precision: ecc 256 384 521 rsa 2048 3072 4096 asm sp_arm64.c
wolfCrypt Benchmark (block bytes 1048576, min 1.0 sec each)
Illegal instruction
CURVE448 test passed!
ED448 test passed!
Illegal instruction
There was a problem hiding this comment.
Changed the code to not use instruction when WOLFSSL_AARCH64_NO_SQRMLSH is defined.
Using the instruction is quicker so I want it used by default.
Aarch64 assembly implementation of Kyber functions. SHA-3 assembly implementations when not hardware crypto.
SparkiDev
force-pushed
the
kyber_aarch64_asm
branch
from
b829134
to
de65778
Compare
dgarske
approved these changes
Sep 26, 2024
|
Tested on Xilinx UltraScale+ ZCU102: Master: PR 7998: Cross compiled using: |
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.
Description
Aarch64 assembly implementation of Kyber functions.
SHA-3 assembly implementations when not hardware crypto.
Testing
Tested on M1 Mac.
Inline assembly tested as well.
Checklist