Arturo Vargas
Arturo Vargas
# Summary Adds a global function to turn on/off caliper profiling when caliper is enabled. This PR will depend on: https://github.com/LLNL/RAJA/pull/1859
# Summary This WIP PR explores high-dimensional thread teams for launch.
Since there is a discrepancy between RAJA/HIP and SYCL ordering it may be convenient to define raja device policy alias to simplify transitions between backends. It could be implemented as...
There a number of warnings that appear with GCC 12.1.1. They look something like this: ``` inlined from ‘void RAJA::internal::expt::tensorTileExec_expanded(const TILE_TYPE&, BODY&&, camp::idx_seq&, camp::idx_seq&) [with STORAGE = RAJA::expt::TensorRegister; TILE_TYPE =...
Format suggestion: Consider a format in which 1 or 2 PRs we review as a group and the rest of the time should be flexible.
# Summary This PR introduces the concept of a memory arena of launch. The underlying idea is to allow users to allocate static shared memory (or could be an external...
Drawing from the conversation in: https://github.com/LLNL/RAJA/pull/1759#discussion_r1866293836 From @trws: `` On the *loc objects, it makes sense to me to have a helper type for those that has an loc or...
Concept 1: Consider support high-dimensional thread teams in RAJA by having a RAJA launch context store a ThreadBlockLayout. ThreadBlockLayout which will store number of threads in a high dimensional block....
Semantically if we have a collection of dense matrices of the same size it may be more convenient to store them in a DenseTensor. This PR proposes that. This simplifies...