mirrors/xmrig
1
0
Fork 1
mirror of https://github.com/xmrig/xmrig.git synced 2026-06-18 10:22:39 -04:00
Code Issues Projects Releases Wiki Activity
3,646 Commits 9 Branches 171 Tags
978720462d4ea84a1d07795514bbc9996bf07092
Commit Graph

36 Commits

Author SHA1 Message Date
aa022
978720462d ARM64 RandomX JIT: dataset prefetch + non-temporal loads 
Two Apple-silicon-targeted tweaks to the aarch64 RandomX JIT:

- Early dataset prefetch: when readReg2/readReg3 are finalized well before
  the end of the program body, emit the next iteration's dataset-line prefetch
  early to hide more DRAM latency on the serial scalar chain.
- Non-temporal dataset loads: each 64-byte dataset line is read once and never
  reused, so ldp -> ldnp avoids evicting the hot scratchpad, and the prefetch
  hint moves pldl2strm -> pldl1strm to match the longer lead time.

Measured ~8% hashrate gain on Apple M4 base over dev (7eadfdc9).
 2026-05-25 13:46:41 +02:00
SChernykh
720325c40f RandomX optimizations: 
- ARM64: optimized emitMovImmediate/emitMemLoad
- ARM64: disabled 32-bit literal preloading (it was slower)
- Android and Linux: added MADV_COLLAPSE support to memory allocation
 2026-05-17 21:04:02 +02:00
SChernykh
f8dd210531 ARM64 RandomX JIT: 
- optimized F/E register loading
- aligned asm code
- optimized emitAddImmediate for small negative values
- v2: optimized CFROUND
- v2: optimized AES in the main loop
 2026-05-15 17:30:05 +02:00
SChernykh
5347458fc7 ARM64 JIT: Optimize Group E register conversion 
Based on https://github.com/tevador/RandomX/pull/324
 2026-04-25 11:37:47 +02:00
SChernykh
26ee1cd291 ARM64 fixes 
- Added a check for free memory before enabling NUMA
- Removed duplicate AES tables from the ARM64 JIT compiler

Fixes #3729 and #3777
 2026-02-04 00:07:16 +01:00
SChernykh
cb6001945e RandomX v2 (ARM64) 2026-01-31 17:30:35 +01:00
Christopher Wright
93f5067999 minor Aarch64 JIT changes (better instruction selection, don't emit instructions that add 0, etc) 2025-09-20 08:32:32 -04:00
SChernykh
ef6fb728b5 Fixed ARMv8 compilation 2024-08-02 17:51:08 +02:00
SChernykh
caae7c64f0 RandomX: correct memcpy size for JIT initialization 
No buffer overflow, better fix for `_FORTIFY_SOURCE`
 2024-04-14 09:13:00 +02:00
SChernykh
f6c50b5393 Fix RandomX crash when compiled with fortify_source 2024-03-20 21:24:02 +01:00
SChernykh
5e66efabcf ARM64 JIT: don't use x18 register 
From https://developer.apple.com/documentation/xcode/writing-arm64-code-for-apple-platforms
> The platforms reserve register x18. Don’t use this register.

This PR fixes invalid hashes when running on Apple silicon with the latest macOS SDK.
 2023-10-19 17:45:15 +02:00
SChernykh
5dcbab7e3a RandomX: optimized program generation 2023-05-21 17:44:20 +02:00
SChernykh
346892e170 Update jit_compiler_a64.cpp 2021-02-01 22:52:02 +01:00
SChernykh
db03573804 ARM JIT: added missing cache flush 2021-02-01 22:42:35 +01:00
SChernykh
e74573f81f Fixed code allocation for ARM 2021-02-01 22:36:11 +01:00
SChernykh
4108428872 Fixed crashes on ARM 2021-02-01 17:07:45 +01:00
SChernykh
633aaccd9c Added config option for AVX2 dataset init 
-1 = Auto detect
0 = Always disabled
1 = Enabled if AVX2 is supported
 2020-12-19 16:18:49 +01:00
XMRig
6b331b6945 Reduce JIT memory for ARM. 2020-12-15 02:52:38 +07:00
XMRig
179f09081f Alternative secure JIT for macOS. 2020-12-12 21:32:36 +07:00
XMRig
775867fc3e Fixed secure JIT on Linux and code cleanup. 2020-12-12 19:18:47 +07:00
XMRig
497863441a Remove duplicated code. 2020-12-12 12:39:11 +07:00
XMRig
ec62ded279 Added generic secure JIT support for RandomX. 2020-12-11 23:17:54 +07:00
SChernykh
50bdaba526 Fixed Debug build in Visual Studio 2020-10-27 14:08:36 +01:00
SChernykh
166c011d37 Fixed ARM compilation 2020-10-07 18:09:42 +02:00
SChernykh
44dcded866 RandomX: added huge-pages-jit config parameter 
Set to false by default, gives 0.2% boost on Ryzen 7 3700X with 16 threads, but hashrate might be unstable on Ryzen between launches. Use with caution.
 2020-10-07 17:42:55 +02:00
SChernykh
c7476e076b RandomX refactoring, moved more stuff to compile time 
Small x86 JIT compiler speedup.
 2020-09-18 20:51:25 +02:00
SChernykh
d0df824599 Optimized dataset read for Ryzen CPUs 
Removed register dependency in dataset read, +0.8% speedup on average.
 2019-12-08 16:14:02 +01:00
SChernykh
578bebb04d Prefer sys_icache_invalidate on iOS 
Also break compilation with error if clear cache is not available
 2019-10-18 18:17:57 +02:00
SChernykh
5611249af7 Fixed __builtin___clear_cache detection 2019-10-18 18:04:13 +02:00
SChernykh
0ad992985c Update jit_compiler_a64.cpp 2019-10-18 16:36:50 +02:00
SChernykh
1a66c3f1a1 Update jit_compiler_a64.cpp 2019-10-18 16:32:01 +02:00
SChernykh
a2ef2fd9d9 Update jit_compiler_a64.cpp 2019-10-18 16:28:49 +02:00
SChernykh
998c55030a Fixed code cache cleanup on iOS/Darwin 2019-10-18 16:26:15 +02:00
XMRig
5c02cb50da Fix copy/paste typo. 2019-10-18 21:26:15 +07:00
XMRig
10d292092a #1246 Fixed build on iOS. 2019-10-18 12:02:10 +07:00
SChernykh
38f4f4f695 Added JIT compiler for RandomX on ARMv8 2019-09-21 10:10:52 +02:00