If the target construct can be executed in SPMD mode + it is a loop
based directive with static scheduling, we can use lightweight runtime
support.
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@340953 91177308-0d34-0410-b5e6-96231b3b80d8
Summary: In the SPMD case, we need to initialize the data sharing and globalization infrastructure. This covers the case when an SPMD region calls a function in a different compilation unit.
Reviewers: ABataev, carlo.bertolli, caomhin
Reviewed By: ABataev
Subscribers: Hahnfeld, jholewinski, guansong, cfe-commits
Differential Revision: https://reviews.llvm.org/D49188
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@337015 91177308-0d34-0410-b5e6-96231b3b80d8
In generic data-sharing mode we do not need to globalize
variables/parameters of reference/pointer types. They already are placed
in the global memory.
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@332380 91177308-0d34-0410-b5e6-96231b3b80d8
This patch adds codegen for the 'target parallel' directive on the NVPTX
device. We term offload OpenMP directives such as 'target parallel' and
'target teams distribute parallel for' as SPMD constructs. SPMD constructs,
in contrast to Generic ones like the plain 'target', can never contain
a serial region.
SPMD constructs can be handled more efficiently on the GPU and do not
require the Warp Loop of the Generic codegen scheme. This patch adds
SPMD codegen support for 'target parallel' on the NVPTX device and can
be reused for other SPMD constructs.
Reviewers: ABataev
Differential Revision: https://reviews.llvm.org/D28755
git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@292428 91177308-0d34-0410-b5e6-96231b3b80d8
Alexey Bataev