gpu-compute: Added method to compute the actual workgroup size
Review Request #3634 - Created Sept. 27, 2016 and submitted
| Information | |
|---|---|
| Alexandru Dutu | |
| gem5 | |
| default | |
| Reviewers | |
| Default | |
Changeset 11653:beaa9af9fada --------------------------- gpu-compute: Added method to compute the actual workgroup size This patch adds a method to the Wavefront class to compute the actual workgroup size. This can be different from the maximum workgroup size specified when launching the kernel through the NDRange object. Current solution is still not optimal, as we are computing these for each wavefront and the dispatcher also needs to have this information and can't actually call Wavefront::computeActuallWgSz before the wavefronts are being created. A long term solution would be to have a Workgroup class that deals with all these details.
Regressions pass.
Looks good to me. Make sure you update the summary line, i.e., gpu: -> gpu-compute:
Review request changed
Updated (Oct. 2, 2016, 10:40 p.m.)
Summary: |
|
|||||||||||||||||||||||||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
Description: |
|
|||||||||||||||||||||||||||||||||||||||
Testing Done: |
|
|||||||||||||||||||||||||||||||||||||||
Diff: |
Revision 2 (+38 -33) |
Ship It!