PGI User Forum
 SearchSearch   MemberlistMemberlist     RegisterRegister   ProfileProfile    Log inLog in 

CUDA-x86.

Possible bug in triangular loop
Goto page Previous  1, 2
 
Post new topic   Reply to topic    PGI User Forum Forum Index -> Accelerator Programming
View previous topic :: View next topic  
Author Message
mkcolg



Joined: 30 Jun 2004
Posts: 5952
Location: The Portland Group Inc.

PostPosted: Wed Oct 31, 2012 7:55 am    Post subject: Reply with quote

Hi Will,
Quote:
OpenACC vendor implementation. The original code did not yield incorrect results with CCE 8.1.1.
Most likely the difference is how they scheduled the loop where "vector" was used on the outer loop.

In general, it is not recommended to privatize scalars. Besides the issue you encountered where the change in schedule changed the behavior, privatized scalars need to be put in global memory instead of a local register (except when spilled when not enough registers are available).

- Mat
Back to top
View user's profile
Display posts from previous:   
Post new topic   Reply to topic    PGI User Forum Forum Index -> Accelerator Programming All times are GMT - 7 Hours
Goto page Previous  1, 2
Page 2 of 2

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © phpBB Group