William R Tobin
William R Tobin
> > The mem-aware packing branch was functionally complete and ready-to-go, the only reason it hasn't been merged has been related to failures of integrated tests (in develop) on our...
None of this is immediately indicative of problems **with the code**, but instead with the scheduling and the test suite restartcheck baselines including the linear solver parameters which often necessarily...
@CusiniM BLT and its ability to set the C++ standard for HIP compilation have been updated to include C++20 support. The HIP/ROCM compilers we use are all clang-based and the...
@untereiner @rrsettgast and I had a discussion about dropping 9.X support. The only reason we haven't committed to making the move is the Pangea-3 CI using 9.4 as well. If...
Its a bug I fixed yesterday in a branch of lvarray. It only happens when we create an arrayofarrays or arrayofsets with number of arrays and default array sizes known...
The code that caused the bug to be found only exists in a new unit test in #3314, so it wouldn't and couldn't have caused actual failures on any platforms...
The bug has been mitigated, so it isn't critical. It is however very unlikely this will ever be addressed in the gcc-9.X release line, and sunsetting older compiler versions is...
rocmcc supports the cuda flag: `-fcuda-flush-denormals-to-zero` https://rocm.docs.amd.com/en/docs-6.0.2/reference/rocmcc.html
misclick
Nothing jumps out at me in the implementation that needs changes, really just the naming.