mirror of
https://github.com/JayDDee/cpuminer-opt.git
synced 2025-09-17 23:44:27 +00:00
Updated Support for AArch64 (markdown)
@@ -4,7 +4,6 @@ This is provided as source code only and may be built on native Linux or followi
|
||||
|
||||
Bitcoin talk discussion thread: https://bitcointalk.org/index.php?topic=5226770.0
|
||||
|
||||
|
||||
Requirements:
|
||||
|
||||
* An ARM CPU supporting AArch64
|
||||
@@ -37,17 +36,17 @@ What works:
|
||||
|
||||
Known problems:
|
||||
* CPU model only shows "ARM"
|
||||
* NEON may not be represented in algo features for some algos that support it.
|
||||
* No detection of ARM minor version number.
|
||||
* NEON may not be represented in algo features for some algos that may support it.
|
||||
* Algos may show support for NEON even if it's disabled or not yet implemented.
|
||||
* 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, affects X* algos performance.
|
||||
* SWIFFTX: Multiple issues with NEON, affects X22i & X25X.
|
||||
* Lyra2 algos including allium, phi2, X22i & X25X are not working.
|
||||
* 2-way 64 bit parallel hash development is in progress.
|
||||
|
||||
Goal for next release:
|
||||
|
||||
* get as many algos as possible working even if optimizations need to be disabled,
|
||||
* get as many algos as practical working even if optimizations need to be disabled,
|
||||
* address some of the disabled optimizations,
|
||||
* enable SHA2, and AES for Groestl & Echo, untested.
|
Reference in New Issue
Block a user