[gomp4] teach the tracer pass to ignore more blocks for OpenACC

2015-08-26 Thread Cesar Philippidis
I hit a problem in on one of my reduction test cases where the GOACC_JOIN was getting cloned. Nvptx requires FORK and JOIN to be single-entry, single-exit regions, or some form of thread divergence may occur. When that happens, we cannot use the shfl instruction for reductions or broadcasting (if

Re: [gomp4] teach the tracer pass to ignore more blocks for OpenACC

2015-08-26 Thread Nathan Sidwell
On 08/26/15 09:57, Cesar Philippidis wrote: I hit a problem in on one of my reduction test cases where the GOACC_JOIN was getting cloned. Nvptx requires FORK and JOIN to be single-entry, single-exit regions, or some form of thread divergence may occur. When that happens, we cannot use the shfl