Updated Support for AArch64 (markdown)

JayDDee
2023-10-22 14:08:41 -04:00
parent 1d4b6449f2
commit eed4687a0f

@@ -25,8 +25,8 @@ Uncommited changes:
The miner compiles and runs on Raspberry Pi 4B, and compiles for all version of armv8 with our without AES or SHA2 or both.
What works:
* Working wth NEON: Argon2d, Yespower, Yescrypt, Power2b, M7M, Lyra2Z330.
* Unoptimized: Sha256dt, Blake2s.
* All algorithms except Hodl are working, some optimized, see below.
* Unoptimized: Sha256dt, Blake2s, Scryptn
* Partially optimized: X17, MinotaurX.
* stratum+ssl and stratum+tcp are working, GBT is untested.
* CPU and SW feature detection and reporting is working, algo features in progress, CPU brand not yet implemented.
@@ -40,11 +40,14 @@ Known problems:
* Algos may show support for NEON even if it's disabled or not yet implemented.
* AES & SHA2 are enasbled but untested.
* Sha256 & Sha512 Parallel N-way are disabled. They work on X86_64.
* Blake256, Blake512, Blake2s, Blake2b N-way parallel hash not working.
* Simd: Multiple issues with NEON, using unoptimized version.
* Luffa: NEON not working, using unoptimized version.
* SWIFFTX: Multiple issues with NEON.
* Fugue: AES disabled.
* Scryptn2 optimzations disabled due to Sha256 issues.
* X17, MinotaurX are partially optimized.
* Blake256, Blake512, Blake2s, Blake2b N-way parallel hash not working, using liner when possible, unoptimzed otherwise.
* Simd: Multiple issues with NEON, using unoptimized.
* Luffa: NEON not working, using unoptimized
* SWIFFTX: Multiple issues with NEON,using unoptimized.
* Fugue: AES disabled, multiple issues with AES version.
* Remainder are not yet optimized for NEON but should work unoptimized.
Goal for next release: