Julian Brown

@spophale These ones fail on an APU with HSA_XNACK enabled: ``` julbrown@pp-128-b1-2:~/work$ diff -u testresults-nopatch/sollve.sum testresults/sollve.sum | grep "^\+FAIL" +FAIL: tests/5.0/application_kernels/lsms_triangular_packing.cpp run +FAIL: tests/5.0/target/test_target_defaultmap_none.c run +FAIL: tests/5.0/target/test_target_defaultmap_to_from_tofrom.c run +FAIL: tests/5.0/target_teams_distribute_parallel_for/test_target_teams_distribute_parallel_for_collapse.c...

Fortran reduction test fix: https://github.com/ROCm/flang/pull/88

> Just to confirm, we expect `openmp5.0-tests/test_imperfect_loop_warning.cxx` to pass once the llvm patch lands? Yes (although the "test_imperfect_loop" of course xfails, in this hackish new way).

Thank you for the approval! I can't merge this though -- it _could_ wait until the upstream imperfect loop analysis option lands on amd-staging, but it probably doesn't need to...

Well, it still seems to work after the drop-dead date! At least I haven't noticed any problems after (admittedly light) testing, other than subreddits that have been restricted, etc. because...

As it is, the aomp-shellcheck script automatically modifies the files given as arguments in some circumstances (i.e. if they have failing checks in the "patchable" check list), which is probably...

I think so, something like that.

I've added a new --check-only option in the version up on https://github.com/ROCm/aomp/pull/1463 now.

This probably hasn't been tested well enough so far to merge just yet (and apparently needs rebasing already), but I'm interested in feedback on the approach. The build_offload.sh script should...

I just noticed the `show_build_dir` output is completely bogus. Oops! (edit: fixed now)