search

UTSASRG / Scaler

GNU General Public License v2.0
4 stars 0 forks source link

Implement versions that don't need thread_local #36

Closed GammaPi closed 2 years ago

GammaPi commented 2 years ago

Per-thread data-structure takes much memory. Since we don't want precise results.

We could have 3 versions:

  1. Per-thread data
  2. No per-thread data, atomic instruction
  3. No per-thread data, normal instruction
GammaPi commented 2 years ago

Currently give up