fzorb/REDACTED-rig
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

RandomX: optimized loading from scratchpad

Browse source 
Prefetches scratchpad data as soon as possible to calculate data address for the next load.

Up to ~1.4% speedup on Ryzen 7 3700X @ 4.1 GHz, RAM 3200 MHz 14-14-14-28 with optimized sub-timings:
Variant|Before H/S|After H/S
-------|----------|---------
rx/0|8663|8777
rx/wow|9867|10009
rx/loki|8652|8731
This commit is contained in:
SChernykh 2019-09-11 19:10:01 +02:00
parent 01b2c952ea
commit 2322e3bcf7
9 changed files with 75 additions and 21 deletions
Download patch file Download diff file Expand all files Collapse all files

3
src/crypto/randomx/jit_compiler_x86_static.hpp
View file

@ -29,7 +29,10 @@ OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.
#pragma once
extern "C" {
void randomx_prefetch_scratchpad();
void randomx_prefetch_scratchpad_end();
void randomx_program_prologue();
void randomx_program_prologue_first_load();
void randomx_program_loop_begin();
void randomx_program_loop_load();
void randomx_program_start();