Ecosystem/xmrig
Ecosystem/xmrig
1
0
Fork 0
mirror of https://github.com/xmrig/xmrig.git synced 2025-01-25 12:06:04 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions

soft_aes: fix previous optimization

Browse source 
the best order of hash/fill/prefetch depends on hw/soft AES
only hw AES is faster after previous optimization
This commit is contained in:
cohcho 2020-09-25 14:36:06 +00:00
parent e4929d7c06
commit 9be3b69109
1 changed files with 18 additions and 6 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

24
src/crypto/randomx/aes_hash.cpp
View file

@ -260,14 +260,26 @@ void hashAndFillAes1Rx4(void *scratchpad, size_t scratchpadSize, void *hash, voi
rx_store_vec_i128((rx_vec_i128*)scratchpadPtr + k * 4 + 2, fill_state2); \ rx_store_vec_i128((rx_vec_i128*)scratchpadPtr + k * 4 + 2, fill_state2); \
rx_store_vec_i128((rx_vec_i128*)scratchpadPtr + k * 4 + 3, fill_state3); rx_store_vec_i128((rx_vec_i128*)scratchpadPtr + k * 4 + 3, fill_state3);
HASH_STATE(0); switch(softAes) {
HASH_STATE(1); case 0:
HASH_STATE(0);
HASH_STATE(1);
FILL_STATE(0); FILL_STATE(0);
FILL_STATE(1); FILL_STATE(1);
rx_prefetch_t0(prefetchPtr); rx_prefetch_t0(prefetchPtr);
rx_prefetch_t0(prefetchPtr + 64); rx_prefetch_t0(prefetchPtr + 64);
break;
default:
HASH_STATE(0);
FILL_STATE(0);
rx_prefetch_t0(prefetchPtr);
HASH_STATE(1);
FILL_STATE(1);
rx_prefetch_t0(prefetchPtr + 64);
}
scratchpadPtr += 128; scratchpadPtr += 128;
prefetchPtr += 128; prefetchPtr += 128;