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

Pooled allocation of RandomX VMs

Browse source 
+0.5% speedup on Zen2 when the whole L3 cache is used (16 threads on 3700X/3800X, 32 threads on 3950X).
This commit is contained in:
SChernykh 2020-04-07 18:31:35 +02:00
parent 7f01c5c6f3
commit 6ae37a9519
11 changed files with 68 additions and 61 deletions
Download patch file Download diff file Expand all files Collapse all files

2
src/net/JobResults.cpp
View file

@ -116,7 +116,7 @@ static void getResults(JobBundle &bundle, std::vector<JobResult> &results, uint3
return;
}
auto vm = new RxVm(dataset, memory->scratchpad(), !hwAES, Assembly::NONE);
auto vm = new RxVm(dataset, memory->scratchpad(), !hwAES, Assembly::NONE, -1);
for (uint32_t nonce : bundle.nonces) {
*bundle.job.nonce() = nonce;