diff options
| author | Tim Rogers <[email protected]> | 2018-11-20 16:13:16 -0500 |
|---|---|---|
| committer | GitHub <[email protected]> | 2018-11-20 16:13:16 -0500 |
| commit | 6443f21d433f1b642003867e56fe1f54efae55e3 (patch) | |
| tree | 6e017904f8dbeab9925810e775a3eaf874d23fdc /CHANGES | |
| parent | 8ec70c69eb89c1fa836c233be3e4c478602d9bb7 (diff) | |
| parent | 695592593ac59be49bdc013814710e216d18a438 (diff) | |
Merge pull request #83 from purdue-aalp/dev
INT unit, sub-core-model, Increase L1 throughput and add tensor core config parameters
Diffstat (limited to 'CHANGES')
| -rw-r--r-- | CHANGES | 3 |
1 files changed, 2 insertions, 1 deletions
@@ -14,11 +14,12 @@ Version 4.0.0 (development branch) versus 3.2.3 4- debug support for wmma instruction using debug_tensorcore flag -GPU Core Performance Simulation: 1- Fermi/Pascal coalescer: coalescing on 32-bytes sectors. -2- Adding separate dp unit pipeline. +2- Adding separate int and dp units pipeline. 3- diff dual issue: allow scheduler to issue diff insts at a time 4- Fair memory issue from multiple schedulers. 5- Added tensorcore unit pipeline. 6- Corrected the bug in vector load instruction +7- Volta sub-core model (scheduler isolation) -Cache System: 1- Sector L1/L2 cache 2- Fetch-on-write and lazy-fetch-on-read write allocation policy. |